JP2008305385A - Character input device, server device, dictionary download system, method for presenting conversion candidate phrase, information processing method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enhance and streamline a predictive conversion function in a character input system. <P>SOLUTION: Additional dictionary (a time priority dictionary table, a position priority dictionary table, and a buzz term priority dictionary table) is provided to a basic dictionary. Then, in response to input of a Kana (Japanese syllabary)character, the basic dictionary is used first to create a list (conversion candidate phrase table) and then, a phrase applicable to undetermined character strings at present is retrieved from the additional dictionary, and the highest priority order is set to the retrieved phrase among phrases to be registered in the conversion candidate phrase table. A conversion candidate phrase list is displayed to a GUI of character input so that the content of the conversion candidate phrase table created in this way is reflected. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a character input device for inputting characters for text creation and the like, and a method for presenting conversion candidate phrases applied in such a character input device. The present invention also relates to a server device that provides dictionary data to a character input device. The present invention also relates to a dictionary download system comprising these character input device and server device. The present invention also relates to an information processing method executed by the dictionary download system. Further, the present invention relates to a program that the character input device and the server device execute in the dictionary download system.

It is known that a Japanese character input system mounted on a mobile phone or the like employs the following predictive conversion function.
As an initial operation for determining a certain unit of character string, the user first inputs one or more characters using hiragana. For example, if the "Today (Hello)" as an attempt to enter, the user first, will enter the "child" in hiragana. That is, since it is possible to read in the pseudonym of "Today" is "Hello", is intended to enter the "child" of the first of the first character. In response to the input of “ko”, the character input system displays a plurality of words starting from the reading of “ko” as conversion candidate words in descending order of preset priority. If “today is” is displayed in the conversion candidate words displayed in this way, the user selects the conversion candidate word for “today” and performs a confirmation operation. The As a result, “Today is” is confirmed as the input character string, and the character string “Today is” is reflected in the display of the confirmed character string (sentence).
In this way, if the predictive conversion function is adopted in the character input system, it becomes possible to determine the character string to be input with a smaller number of keystrokes than to input all hiragana readings, and efficient character input Can be done. In particular, a cellular phone is a useful function due to large restrictions on the number and size of operation keys.

  Regarding the predictive conversion function, the higher the priority as a conversion candidate word / phrase for a character string that the user intends to input is, the better the usability is. However, in a predictive conversion system from the past, the priority order for conversion candidate words is basically determined based on the priority order set for each word registered in the dictionary in advance. Since such priorities are set in consideration of general terms, it may naturally occur that they do not match depending on the environment and situation when the character input operation is performed.

Therefore, techniques such as Patent Documents 1 and 2 have been proposed as a configuration for increasing the prediction accuracy of conversion candidate words in the prediction conversion system as much as possible.
In Patent Document 1, each time a character string is selected, the time zone at that time is recognized, the number of times the character string is selected is counted by time zone, and based on the count value, A configuration for setting priority is disclosed.
Patent Document 2 discloses a configuration in which a plurality of character input conversion dictionaries are stored for each region, and a character input conversion dictionary to be used is selected based on position information obtained by GPS (Global Positioning System). Has been.
That is, Patent Documents 1 and 2 try to improve the prediction accuracy by adapting to the time zone and the region, respectively.

JP 10-254872 A JP 2006-350904 A

  The invention of the present application is also intended to obtain a prediction accuracy as high as possible as a predictive conversion system, and in addition, the conversion candidate words and phrases can be predicted more efficiently than those known so far. It aims to be able to do.

In view of the above problems, the present invention is configured as a character input device as follows.
That is, basic dictionary information storage means for storing basic dictionary information formed by registering basic phrases that are words prepared as basic, and auxiliary dictionary information formed by registering words corresponding to predetermined conditions And / or auxiliary dictionary information storage means for storing auxiliary dictionary information formed by registering words that match a specific condition, and conversion indicating a conversion source character string made up of one or more predetermined types of input characters A conversion source character string information acquisition unit that acquires original character string information, and a basic phrase that is predicted from the basic dictionary information to be a confirmed character string corresponding to the conversion source character string indicated by the conversion source character string information. Conversion candidate character / phrase information generating means for generating conversion candidate word / phrase information based on searching and registering the searched basic words / phrases as conversion candidate words / phrases, and conversion source characters indicated by the conversion source character string information from the auxiliary dictionary information The auxiliary dictionary search means for searching for a phrase that is predicted to be a definite character string corresponding to the above and a result obtained by setting and registering a predetermined priority for the words searched by the auxiliary dictionary search means The conversion candidate word / phrase information changing means for changing the content of the conversion candidate word / phrase information, and the conversion candidate words / phrases registered in the conversion candidate word / phrase information are presented in a predetermined manner so as to reflect the priority order. The conversion candidate word / phrase presenting means is provided.

The server device is configured as follows.
That is, holding means for holding network data designation information that designates network data that is permitted to be used for creating high-frequency phrase dictionary information among network data that is data transmitted on the network, and the acquired network Based on the character string information collecting means for collecting character string information from the network data designated by the data designation information and the character string information collected by the character string information collecting means, words that are frequently used at that time are It is provided with registered dictionary information creation means for creating high-frequency phrase dictionary information, and transmission means for transmitting the high-frequency phrase dictionary information to the character input device for downloading via the communication network. .

The dictionary download system is configured as follows.
This dictionary download system includes a character input device and a server device.
In addition, the character input device registers basic dictionary information storage means for storing basic dictionary information formed by registering basic phrases that are words prepared as basics, and registers phrases corresponding to predetermined conditions. Auxiliary dictionary information storage means for storing auxiliary dictionary information formed and / or auxiliary dictionary information formed by registering words that match a specific condition, and conversion composed of one or more predetermined types of input characters Based on the conversion source character string information acquisition means for acquiring the conversion source character string information indicating the original character string and the basic dictionary information, it is predicted that the character string will become a confirmed character string corresponding to the conversion source character string indicated by the conversion source character string information. Conversion candidate word / phrase information generating means for generating conversion candidate word / phrase information based on a search for a basic word / phrase and registering the searched basic word / phrase as a conversion candidate word / phrase, and conversion source character string information from auxiliary dictionary information An auxiliary dictionary search means for searching for a phrase that is predicted to be a confirmed character string corresponding to the conversion source character string shown, and a predetermined priority is set and registered for the words searched by the auxiliary dictionary search means As a result, the conversion candidate word / phrase information changing means for changing the contents of the conversion candidate word / phrase information, and the conversion candidate words / phrases registered in the conversion candidate word / phrase information are set in a predetermined manner so that the priority order is reflected. The conversion candidate word / phrase presenting means to be presented according to the mode and the server device are registered as words / phrases that are frequently used at that time as auxiliary dictionary information formed by registering words / phrases that match specific conditions. Downloading means for downloading the frequency word dictionary information via the communication network and storing it in the auxiliary dictionary information storage means.
In addition, the server device holds network data designation information that designates network data that is permitted to be used for creating high-frequency phrase dictionary information among network data that is data transmitted on the network. And character string information collecting means for collecting character string information from the network data designated by the acquired network data designation information, and the high-frequency phrase dictionary information based on the character string information collected by the character string information collecting means Dictionary information creating means for creating the high-frequency phrase dictionary information and transmitting means for transmitting the high-frequency phrase dictionary information to the character input device for downloading via the communication network.

In the above configuration, first, in the character input device, basic dictionary information exists, and auxiliary dictionary information is provided as an option added thereto. Then, according to the conversion source character string, first, the conversion candidate word / phrase is searched from the basic dictionary information to generate conversion candidate word / phrase information, and then the auxiliary dictionary information is referred to by referring to the auxiliary dictionary information. The content of the conversion candidate word / phrase information is changed so as to result in setting an appropriate priority for the words / phrases retrieved from.
With such a configuration, in order to improve the prediction accuracy, an auxiliary dictionary corresponding to each specific condition setting may be prepared, and it is not necessary to change the basic dictionary.
In the above configuration, the high-frequency phrase dictionary information that is one of the auxiliary dictionaries is created by the server device, and the character input device can be downloaded and acquired from the server device. As a result, for example, the sufficient processing capability of the server device provides high-reliability high-frequency word / phrase dictionary information with high reliability.

Thereby, the present invention does not enlarge the contents of the basic dictionary more than necessary. Also, for example, by preparing auxiliary dictionary information corresponding to each different condition setting, it is possible to obtain high prediction accuracy while flexibly responding to various environments and situations.
In particular, as for the high-frequency word dictionary information that is one of the auxiliary dictionaries, it is possible to obtain content with high reliability as described above.
In this way, depending on the present invention, an efficient character input system having high prediction accuracy can be obtained.

FIG. 1 shows an internal configuration example of a cellular phone device to which a character input device as the best mode for carrying out the present invention (hereinafter referred to as an embodiment) is applied.
The antenna 11 receives radio waves from the mobile phone network, and the received signal is input to the receiving unit 12. The receiving unit 12 performs received signal processing on the input signal including, for example, multiple access reception processing, reception demodulation (conversion to baseband) processing, demodulation processing for transmission coding, error correction processing, and the like. Thus, data in the baseband signal format is obtained.
The contents of signals carried by radio waves in the cellular phone communication network are roughly divided into call voice and communication data on the Internet (network) via the gateway of the telephone communication network. When the baseband signal format data obtained by executing the received signal processing is the voice of the other party of the call, the receiving unit 12 outputs this data to the voice signal processing unit 14. When the data in the baseband signal format is communication data, the data is output to the system control unit 17.

The audio data input from the receiving unit 12 to the audio signal processing unit 14 has a format in which, for example, audio compression encoding is performed. Therefore, the audio signal processing unit 14 performs demodulation (decompression) processing corresponding to compression encoding on the input audio data, conversion to an analog audio signal, amplification, and the like, and outputs the audio from the speaker 15.
In addition, when communication data is input from the receiving unit 12, the system control unit 17 executes an appropriate process by an application corresponding to each type of communication data. As an example, when the input communication data is data of an e-mail transmitted to an e-mail address registered by the user of the mobile phone, an application called mailware responds to the reception of the e-mail. Execute the process. As a result, for example, it is possible to output a ringing tone for notifying reception of an e-mail, save the received e-mail, and view an e-mail according to a user operation.

  For example, when it is an electronic sound accompanying a user operation or when it is necessary to output a ringtone, the system control unit 17 passes the required electronic sound and ringtone signal to the audio signal processing unit 14 and the speaker 15 The control is executed so that the sound is output as sound.

Further, at the time of communication by telephone call, an audio signal obtained by collecting audio by the microphone 16 is input to the audio signal processing unit 14. The audio signal processing unit 14 performs predetermined audio signal processing such as audio compression coding on the input audio signal and outputs the processed audio signal to the transmission unit 13.
In addition, for example, when browsing web content (web content) published on the Internet or transmitting data in a network-compatible manner by sending and receiving e-mail, the system control unit 17 transmits data to be created. The output is made to the unit 13.
In the transmission unit 13, the reception unit 12 receives the audio signal data as the transmission sound input from the audio signal processing unit 14 and the network-compatible transmission data input from the system control unit 17 as described above. Transmission signal processing that is paired with signal processing is executed and transmitted as a radio wave from the antenna 11.

The system control unit 17 executes various control processes in the mobile phone device 1 and includes, for example, a CPU (Central Processing Unit).
The storage unit 18 is a storage area composed of hardware such as a ROM and a flash memory, for example. First, data of a program to be executed by the CPU in the system control unit 17 is stored. This program is, for example, an OS (Operating System) for controlling basic operations as the mobile phone device 1, and various applications (application software) started on the OS. In addition, the storage unit 18 includes various setting information used by these OSs and applications, and various types of information such as a telephone book, moving images, still images, voices, and documents corresponding to each application. User data and data files are stored in a properly managed state.
The RAM 19 is an area as a main storage device used by the CPU as the system control unit 17 and includes hardware such as DRAM and SDRAM.

  The display unit 22 is a part that displays, as an image, required contents according to the operation state of the mobile phone device 1 at that time under the control of the system control unit 17. The display device employed in the display unit 22 is not particularly limited, but currently, liquid crystal display devices are widely employed.

  The operation unit 23 includes various operators provided at predetermined sites in the main body of the mobile phone device 1, and a site for outputting an operation information signal corresponding to an operation performed on these operators to the system control unit 17. Are collectively shown. The system control unit 17 executes a required control process so as to obtain an operation in response to the operation information signal.

For example, the clock unit 20 measures the current time (time) by year, month, date, hour, minute, and second.
The GPS unit 21 is a part adapted to acquire current position information corresponding to GPS (Global Positioning System).

The cellular phone device 1 configured as described above includes a character input system. That is, in the cellular phone device 1, character input is performed by performing an operation on a predetermined operator in the operation unit 23 under a predetermined operation mode state including, for example, creation of an e-mail title, body text, and the like. It can be done.
As this character input system, a predictive conversion method under a method called a kana-kanji conversion method is adopted. The Kana-Kanji conversion method corresponds to the case where the input characters are Japanese, and refers to a method that can convert a character string entered in Kana (usually Hiragana) into Kanji-mixed phrases and sentences. . The predictive conversion method presents words (conversion candidate words) that are predicted as conversion candidates based on the input kana character string when performing kana-kanji conversion as described above. This means a method in which when a phrase that the user actually wanted to convert is included in the phrase, the conversion candidate word / phrase can be specified and confirmed as an input character string.
An object of the present embodiment is to provide a more efficient character input system on the assumption that character input is performed by such a predictive conversion method. Therefore, hereinafter, the character input system as the present embodiment will be described.

  In order to configure the character input system having the predictive conversion function as described above by the mobile phone device 1 of the present embodiment, first, a specific operator included in the operation unit 23 of the mobile phone device 1 is specified. Is assigned to the kana character input function, and the application software for realizing the operation as the character input system and information such as a dictionary used by the application software are stored (installed) in the storage unit 18 of the cellular phone device 1. It should be kept. Hereinafter, the application software that implements the character input system according to the present embodiment will be referred to as a “character input support application”.

FIG. 2 schematically shows a configuration example of a character input system in the mobile phone device 1 of the present embodiment.
First, in this figure, a character input support application 30 is shown. The character input support application 30 is a program executed by the system control unit 17 (CPU), and is stored (installed) in the storage unit 18, for example. In this figure, the character input support application 30 is shown as comprising a block (processing function block) corresponding to each processing function. As the processing function blocks, a character conversion processing unit 31, a dictionary search processing unit 32, a reference sentence processing unit 33, a morpheme analysis processing unit 34, and a confirmation result output unit 35 are shown.

The character conversion processing unit 31 is a central functional part in the character input support application 30 and executes processing so that an appropriate response operation is obtained mainly in response to a character input operation.
In response to an instruction from the character input conversion processing unit 31, the dictionary search processing unit 32 searches for words / phrases registered in the basic dictionary 51. The basic dictionary 51 is an information unit formed by registering standard and reference words as conversion candidate words, and is stored in the storage unit 18.

The reference sentence analysis processing unit 33 is a functional block for managing a sentence reference process when referring to the contents of a sentence in an e-mail or a document file for predictive conversion as described later. For example, the reference sentence analysis processing unit 33 cooperates with the character conversion processing unit 31 to describe the characters described in the data of the e-mail or the document file that is taken from the application software as the mailware 41 or the document editor 42. A column (sentence or the like) is taken as a reference sentence and passed to the morphological analysis processing unit 34.
The morpheme analysis processing unit 34 executes an analysis process for decomposing the text content of the data as the designated e-mail or document file into morphemes under the control of the reference sentence analysis processing unit 33. Information about the morpheme obtained as a result of this analysis is stored in the RAM 19 as morpheme information 64, for example.

  For example, the confirmation result output unit 35 receives notification of the confirmed word / phrase from the character conversion processing unit 31 according to the word confirmation operation and generates and outputs information on the confirmed character string. As an output mode of the confirmed character string, here, it is shown that the display unit 22 can output the confirmed character string so as to be displayed.

Hereinafter, an operation example of character input realized by the character input support application 30 having the above functional block configuration will be described. First, the character input support application 30 can execute the operation. A basic character input operation example will be described with reference to FIG. 2 and FIGS.
First, the user sets an operation mode in which characters can be input using kana and kanji by operating the cellular phone device 1 according to the present embodiment. Examples of such an operation mode include a mode for creating a character string such as a title and a body when writing an e-mail, a mode for creating a sentence, a mode for inputting a name, an address, and the like when creating a phone book.
Under the operation mode in which characters can be input using kana and kanji as described above, the character input operation is to input kana characters (Hiragana) one by one. A character string composed of one or more kana characters input in this way corresponds to the conversion source character string in the present invention. As an example here, it is assumed that an input operation of “ko” is performed as the first kana character. In other words, for example, the user performs an operation for inputting “ko” on the operator assigned to the input of the kana character “ko” among the operators provided in the operation unit 23.

In response to the input operation of the kana character “ko”, an operation information signal indicating that the kana character “ko” has been input is output from the operation unit 23 to the system control unit 17 (CPU). However, the operation information signal indicating the input of the kana character is received by the character conversion processing unit 31 in the character input support application 30 as shown in FIG.
When the character conversion processing unit 31 accepts a new input operation for a certain kana character in this way, the conversion candidate after specifying the kana character “ko” as a search condition to the dictionary search processing unit 32. Command the search for a phrase. As a response to this search command, the dictionary search processing unit 32 executes a search for the basic dictionary 51.

The basic dictionary 51 is information stored in the storage unit 18 to be used by the character input support application 30, and is dictionary information formed by collecting a large number of words selected as a standard by a predetermined structure. Here, in order to simplify the explanation, the word / phrase registered in the basic dictionary 51 is pre-selected by the vendor side. Is also possible.
In this case, the dictionary search processing unit 32 searches the basic dictionary 51 using the designated kana character “ko” as a search condition. As can be understood from the following description, the search for the basic dictionary 51 is performed by searching for a phrase that the user intends to input, that is, a conversion candidate phrase based on the input kana character being “ko”. It can be seen as a process of narrowing down. Further, such narrowing of conversion candidate words can be considered to be prediction / estimation of the most basic conversion candidate words. In this way, the dictionary search processing unit 32 searches the basic dictionary 51 to obtain a basic conversion candidate word prediction function as the character input support application 30.

The character conversion processing unit 31 creates a conversion candidate word / phrase table 62 based on the current search result returned from the dictionary search processing unit 32.
An example of the contents of the conversion candidate word / phrase table 62 created at this time is shown in FIG.
As shown in this figure, the conversion candidate word / phrase table is formed by a table structure having items of ID (phrase ID), reading, part of speech, notation, and basic priority for each conversion candidate word / phrase. . And, in this case, looking at the conversion candidate phrase in the combination of the "read-notation", in order from the top to the bottom, "Hello, hello,""tonight if - Good evening,""This,this.""In this way, this way.""Hi · today,""Good evening, tonight,""children and children" ... (continued later) has been registered. These phrases are a group of phrases having “ko” as a common leading kana character corresponding to the fact that the kana character input this time is “ko”.
In addition, the order of arrangement from the top to the bottom of the words corresponds to the priority order of the conversion candidate words registered in the conversion candidate word table, but here, as the basic priority associated with each word It follows the order with the most numbers. The basic priority is a numerical value set in advance corresponding to the probability and likelihood that the word is determined as a conversion word when “ko” is input. Therefore, the priorities of the conversion candidate words in the conversion candidate word table here are in descending order of the basic priority. In this way, in the conversion candidate word / phrase table 62, first, the priority order for the conversion candidate words / phrases is set based on the basic priority, so that the word / phrase beginning with the kana character “ko” is simply listed. The prediction accuracy will be improved.

The phrase registration structure in the basic dictionary 51 conforms to the structure of the conversion candidate phrase table 62 shown in FIG. That is, the basic dictionary 51 has a structure in which items of ID, reading, part of speech, notation, and basic priority are provided for each word registered as a basic word.
For example, when the dictionary search processing unit 32 performs a search using the kana character “ko” as a search condition, words whose pronunciation begins with “ko” among the words or phrases registered in the basic dictionary (the first character in the character string of the word) 1 character is “ko”), and the information of the ID, reading, part of speech, notation, and basic priority items is returned to the character conversion processing unit 31 as information for each searched phrase. Is done. Then, the character conversion processing unit 31 uses the item information (ID, reading, part of speech, notation, basic priority) for each search phrase returned in this way, By arranging them in the order of basic priority, the conversion candidate word / phrase table 62 is created.

  The conversion candidate word / phrase table 62 created as described above has an initial content created based only on the basic dictionary 51 corresponding to the first input kana character “ko”. The character conversion processing unit 31 that created the initial conversion candidate word / phrase table 62 then refers to the priority word / phrase table 63. The priority word / phrase table 63 is formed by registering a word / phrase determined by a character input operation, and is stored in the storage unit 18 as information that can be rewritten (updated).

An example of the contents of the priority phrase table 63 is shown in FIG.
As shown in the figure, the priority word / phrase table 63 is information in which words / phrases that have been confirmed so far are registered, and corresponding to each registered word / phrase, like the conversion candidate word / phrase table 62, ID, reading , Part of speech, notation, basic priority item information.

In this case, the character conversion processing unit 31 refers to the priority word / phrase table 63 so as to search for a word / phrase in which “ko”, which is the current input kana character, is the first character (first character) in reading. Is done. Here, if the priority word / phrase table 63 actually searched and searched is the one shown in FIG. 4, the word / phrase whose reading is “children” is searched.
In accordance with this search result, the character conversion processing unit 31 changes or updates the contents of the conversion candidate word / phrase table 62 having the initial contents to the contents shown in FIG.
In the conversion candidate word / phrase table 62 of FIG. 5, the word / phrase “children” read from the priority word / phrase table 63 is moved in the order of the first arrangement. As a result, the phrase “children” has a basic priority of 50, which is lower than the basic priority of other words shown in the figure, for example, in the conversion candidate word / phrase table 62. Is given the highest priority.
In response to this, the arrangement order of the remaining words in the arrangement order before the word “children” in FIG. 3 is lowered one by one.
That is, as a result of searching the priority word / phrase table 63, if the word / phrase corresponding to the search condition exists in the priority word / phrase table 63, the character conversion processing unit 31 selects the word / phrase most in the conversion candidate word / phrase table 62. High priority is set.

  At this time, as in the case of the phrase “children” read as described above with reference to FIGS. 3 to 5, the conversion candidate phrase table 62 of the contents created earlier corresponds to the search condition. If there is a word or phrase, the highest priority order may be set in the conversion candidate word / phrase table 62 simply by rearranging the order of arrangement in the conversion candidate word / phrase table 62.

However, in some cases, it may be considered that the words searched from the priority word / phrase table 63 as corresponding to the search condition do not exist in the conversion candidate word / phrase table 62.
In such a case, first, a word corresponding to the search condition is newly registered in the conversion candidate word / phrase table 62. At this time, the highest priority in the conversion candidate word / phrase table 63 is set for the newly registered word / phrase.

  Here, the conversion candidate phrases searched from the priority phrase table 63 can be regarded as words that are frequently used for the user. From this, by setting the highest priority for the conversion candidate words searched from the priority word / phrase table 63, the prediction accuracy of the conversion candidate words / phrases can be increased corresponding to each user.

  Then, in this basic operation example, based on the contents of the conversion word table 62 of FIG. 5 updated as described above, a list of conversion candidate words in response to the input of the current kana character “ko”, It is displayed on the operation screen (character input operation screen) for character input displayed on the display unit 22. In the process for this display, for example, first, the character conversion processing unit 31 instructs the confirmation result output unit 35 to display the conversion candidate word / phrase table 62. In response to this instruction, the confirmation result output unit 35 reads the conversion candidate word / phrase table 62, executes control for generating the contents as an image of the conversion candidate word / phrase list and displaying it on the display unit 22.

FIG. 6A shows an example of the content of the operation screen on which the list of conversion candidate words is displayed. In this figure, the display screen 22A shows the screen portion of the display portion 22 that is provided on the main body of the mobile phone device 1 so as to be visible to the user. The character input operation screen displayed on the display unit 22A has a main input operation area AR1 displayed at the top and a conversion candidate word / phrase list area AR2 displayed below it.
The main input operation area AR1 is the main for character input operations, and is a character string that has been confirmed by previous operations, and an indeterminate character string that has been input in response to an operation on an operator corresponding to the current character input. Is the area where is displayed. In FIG. 6A, “I am in the present era” is a confirmed character string, and “ko” of hiragana shown with an underline is an undefined character string. This indefinite character string “ko” is displayed in response to the input of the current kana character “ko”. In the conversion candidate word / phrase list area AR2, a list of conversion candidate words / phrases having contents corresponding to the conversion candidate word / phrase table 62 shown in FIG. 5 is displayed. In this conversion candidate word list area AR2 is, "children" → "Hello" → "Good evening" → "this" (new line) "thus" → "Today" → order in the conversion candidates phrase "tonight" list display This is consistent with the notation of the words registered in the conversion candidate word / phrase table 62 shown in FIG. 5, and the list order of each word is the conversion shown in FIG. This matches the priority order of words set in the candidate word table 62. In this way, in the conversion candidate word / phrase list area AR2, the words / phrases registered in the conversion candidate word / phrase list 62 are arranged and displayed according to the priority set in the conversion candidate word / phrase table 62.
The cursor CR in the conversion candidate word / phrase list area AR2, for example, moves between conversion candidate words / phrases in the order of arrangement in accordance with an operation with respect to a predetermined feed / return operation element. Has been. This indicates that the conversion candidate word highlighted by the cursor CR is the conversion candidate word selected as the decision candidate. Further, when the conversion candidate word / phrase list area AR2 is initially displayed, the position of the cursor CR is arranged on the conversion candidate word / phrase having the highest priority. The conversion candidate word / phrase list area AR2 in FIG. 6A also corresponds to the contents of the initial display, and the cursor CR is arranged at “child” having the highest priority among the conversion candidate words / phrases.

  Then, for example, the user looks at the list displayed in the conversion candidate word / phrase list area AR2 in response to the input of the kana character “ko” this time, and if there is a word / phrase that matches the one he / she wanted to convert, After moving the cursor CR to the word / phrase, an operation for selection decision is performed. For example, here, if the user intends to input “thus”, the user moves the cursor CR to the word “thus” in the conversion candidate word / phrase displayed in the conversion candidate word / phrase list area AR2. And the selection decision operation is performed subsequently. When the character conversion processing unit 31 that receives these operations controls the display unit 22 via the confirmation result output unit 35, the character input operation screen changes as shown in FIG. 6B. . That is, in the main input operation area AR1, a fixed character string of “in this way” is displayed instead of an indeterminate character string of “ko”, and as a whole, the input of the character string up to “in this way is now” To be confirmed. In this case, the list of conversion candidate words displayed in the conversion candidate word list area AR2 is erased in response to the word confirmation operation. It should be noted that, depending on the conversion candidate phrase confirming operation, a display mode example may be considered in which the conversion candidate phrase list area AR2 itself is deleted and the entire display screen is used as the main input operation area AR1.

  Further, in response to the determination of the phrase “in this way” as described above, the character conversion processing unit 31 newly registers the determined phrase in the priority phrase table 63. . FIG. 7 shows an example of the contents of the priority phrase table 63 updated by newly registering the phrase “in this way” with respect to the contents shown in FIG.

  Also, without performing the selection decision operation (confirmation operation) of the phrase “in this way” under the state of FIG. 6A, instead of inputting “k”, which is an indefinite character string, a kana character is input. Suppose that an operation is performed. Here, it is assumed that an operation for inputting the kana character “n” is performed. Upon receiving this operation, the character conversion processing unit 31 converts a kana character string “kon”, which is obtained by connecting the previous input kana character “ko” to the current input kana character “n”, to a new indeterminate character string (conversion source). Character string). That is, a conversion candidate word / phrase table 62 corresponding to the kana character string “kon” is created using the basic dictionary 51 and the current priority word / phrase table 63 and then the conversion reflecting the contents of the priority word / phrase table 63 is performed. Processing is performed until the candidate word list area AR2 is displayed. An example of the character input operation screen according to the execution result of this process is shown in FIG.

In FIG. 8, in the main input operation area AR1, among the character strings displayed as “current era is kon”, the confirmed character string is “current era”, and the continuation “kon” is unacceptable. It is shown that it is a fixed character string. In the conversion candidate word / phrase list area AR2, a list of conversion candidate words / phrases corresponding to the indeterminate character string “kon” is arranged according to the priority order based on the basic priority order and the search result for the priority word / phrase table 63. Here, in the conversion candidate word list area AR2, conversion candidate word corresponding to the uncertain character string "crowded" is, "Hello" → "Good evening" → "Today" → "tonight" → "Tonight" → "Kondo "→" This year "→" Konkusho "is shown in the order of arrangement (priority). The selection determination operation for the conversion candidate word / phrase in the conversion candidate word / phrase list area AR2 displayed in this manner is the same as the operation procedure described above with reference to FIG.
Such a basic operation example of character input enabled by the character input support application 30 of the present embodiment is as follows.

  For example, as a general thing about character input operation in a mobile telephone device, a plurality of characters are almost always assigned to one operator (key). Therefore, providing an interface for character input operation in which all the kana characters that form the necessary readings are input by an operation on the operator and a conversion candidate is presented forces the user to perform many key operations. . Therefore, for example, as described above, in response to the input of one or more kana characters, for example, the words intended by the user are predicted and estimated, and these are presented as conversion candidate words. If the configuration is such that a confirmation operation can be performed for a word selected from among them, the number of operations until the user inputs and confirms a character string as a certain word can be reduced on average. That is, the burden on the user regarding the character input operation is reduced, and thereby a character input system with improved usability can be obtained.

And it can be said that the ease of use of the character input system having such a predictive conversion function is greatly contributed by the high prediction accuracy of the conversion candidate words. Therefore, the present embodiment is configured such that the prediction accuracy of conversion candidate words higher than the above basic operation is obtained. Furthermore, after improving such prediction accuracy, it aims at making a structure, a process, etc. as a character input system more efficient than before.
Hereinafter, an operation example as the character input system of the present embodiment and a configuration for realizing such an operation will be described.

As an operation of the character input system including the character input support application 30 of the present embodiment, in response to the input of kana characters, first, as described with reference to FIGS. A conversion candidate word / phrase table 62 having a priority determined based on the basic dictionary and the contents of the priority word / phrase table is generated. In the case of the previous basic operation, the conversion candidate word / phrase list area AR2 is displayed based on the conversion candidate word / phrase table 62 generated in this way. As described below, the contents of the conversion candidate word / phrase table 62 are changed according to at least the time, position, trending word, etc., and the contents of the conversion candidate word / phrase table 62 are changed, and then the conversion candidate word / phrase list is changed. The display in the area AR2 is performed.
In the following, regarding the processing such as changing the contents of the conversion candidate word / phrase table 62, the one according to time is referred to as “time priority”, the one according to position as “position priority”, and the one according to buzzwords. Is called “Buzz word priority”.

First, when the conversion candidate word / phrase table 62 is to be processed according to time priority, the character conversion processing unit 31 performs the current time (year / month / day / hour / minute / second) measured by the clock unit 20. And the time priority dictionary table 52A prepared in the additional dictionary table group 52 is referred to.
Here, the additional dictionary table group 52 is a generic term for dictionaries and table information that are treated as options with respect to the basic dictionary 51. Therefore, the time priority dictionary table 52A included therein is also a kind of additional dictionary / table. It will be. Actual data as additional dictionaries / tables including the time priority dictionary table 52 </ b> A is stored in the storage unit 18.

FIG. 9A shows an example of basic contents of the time priority dictionary table 52A. FIG. 9B shows a partial content example of the basic dictionary 51 in order to show the relationship and correspondence with the time priority dictionary table 52A. Here, the basic priority set for each word on the basic dictionary 51 side is not shown.
The time-priority dictionary table 52A shown in FIG. 9A is a part registered in the basic dictionary 51, as can be understood from the relationship with the basic dictionary 51 shown in FIG. 9B. Alternatively, it is formed with a structure in which information of a specific time is associated with IDs of all words. Note that the term “time” here includes not only the time and period based on the hour / minute / second unit, but also the period based on the monthly unit, year unit, and the like.
That is, the time priority dictionary table 52A is formed by registering words so as to correspond to a predetermined time set as a condition.
For example, as for some examples of word / phrase correspondence contents between the time priority dictionary table 52A and the basic dictionary 51 shown in FIG. 9, first, “reading / reading” registered in the basic dictionary 51 will be described. The phrase “notation” is “good morning / good morning” is associated with the time (04:00 to 12:00) registered in the time priority dictionary table 52A by its ID = 01011. This is because, in the case of a word whose reading / notation is “good morning / good morning” registered in the conversion candidate word / phrase table 62 as a conversion candidate word / phrase, it is stored in advance in the basic dictionary 51 in the time zone from 04:00 to 12:00. This indicates that the highest priority should be given in the conversion candidate word / phrase table 62 regardless of the prescribed basic priority. In other words, the phrase “good morning” is based on what is commonly used in the morning and morning as a general rule, 04: 00-12: It is defined by the time priority dictionary table 52A so that a high priority is given at 00.
Also, registered in the basic dictionary 51, the phrase "reading, notation" a "Hello-Hello", with its ID = 00001, time registered in the time priority dictionary table 52A (12: 01~18: 00). This, for the word reading, notation of "Hello, Hello", if it is registered in the conversion candidate word table 62 as a conversion candidate word, 12: 01-18: In the 00 time zones, in advance by the basic dictionary 51 This indicates that the highest priority should be given in the conversion candidate word / phrase table 62 regardless of the prescribed basic priority. In other words, the phrase of the reading of the "Hello", on the basis that the conventional wisdom on, is intended to be used during the day in the afternoon, a time zone that corresponds to this 12:01 to 18:00 a higher priority in As given, it is defined by the time priority dictionary table 52A.
In addition, the phrase “reading / notation” registered in the basic dictionary 51 is “Kurisuru / Christmas” with the ID = 06779, the time registered in the time priority dictionary table 52A (November to 12th). Month). This is defined in the basic dictionary 51 in advance for the period from November to December when the words whose reading / notation is “Kurisuma / Christmas” are registered in the conversion candidate word / phrase table 62 as conversion candidate words / phrases. This indicates that the highest priority should be given in the conversion candidate word / phrase table 62 regardless of the basic priority. This is based on the fact that Christmas, which is the event, is December 25th, and Christmas is often the subject of discussion about a month before, with a high priority in the corresponding period from November to December. It is defined by the time priority dictionary table 52A so that the order is given.

Then, the character conversion processing unit 31 selects a word associated with the time corresponding to the current time acquired as described above from the words registered in the conversion candidate word / phrase table 62 at the current stage. The time priority dictionary table 52A is searched. If there is a searched phrase, the contents are changed so that this phrase has the highest priority in the conversion candidate phrase table 62.
As a specific example, for example, when the input of the kana character “ko” described in the example of the basic operation is performed at a certain time included in the time zone of 12:01 to 18:00, the conversion candidate word / phrase table 62, from what is shown in Figure 5, "read-representation" is upon you as the priority of the two words the highest of the "Hello, Hello" and "Hello, is today", the rest of the For words, the order of arrangement is changed by lowering the order of arrangement. The arrangement order changed in this way is, for example, the notation by “reading / notation”,
"Hello Hello"
"Hello, today."
"Children and children"
"This, this"
"This way, this way"
"Good evening, tonight"
It becomes.
It should be noted that, in fact and to in such a case, and the phrase "reading, notation" and "Hello, Hello", to more of any of the phrase "Hello, is today", about how to set a higher priority, For example, a rule having a higher basic priority may be followed, for example, according to a predetermined rule.
When the display of the conversion candidate word / phrase list area AR2 is executed based on the contents of the conversion candidate word / phrase table 62 processed in this way, for example, the arrangement shown in FIG. from the order, and that word of the title of "Hello", "today" is displayed at the top rank.
In this way, when the time-priority conversion candidate word / phrase table processing is executed, the priority order of words / phrases that are estimated to have a high use probability is increased according to the time zone and time at that time. The That is, it is possible to improve the prediction accuracy according to the time condition.

According to FIG. 9, the time priority dictionary table 52 </ b> A has IDs of some or all of the words registered in the basic dictionary 51. That is, in this case, the phrase registered in the time priority dictionary table 52A is within the range of the phrase registered in the basic dictionary 51.
However, the time-priority dictionary table 52A may be registered contents including words and phrases that are not registered in the basic dictionary 51.
In response to the contents of the time priority dictionary table 52A, the character conversion processing unit 31 first reads the current indeterminate character string (conversion source character string) and the beginning of reading from the time priority dictionary table 52A. Are matched and the word / phrase associated with the time corresponding to the current time acquired at that time is searched. In other words, when searching for a phrase, narrowing down by the phrase existing in the conversion candidate phrase table 62 is not performed.
If there is a searched phrase, first, it is determined whether or not there is a phrase that matches the searched phrase in the already created conversion candidate phrase table 62.
Here, if it is determined that there is a phrase that matches the searched phrase, the phrase that matches the searched phrase has the highest priority in the conversion candidate phrase table 62. Change the array order of.
On the other hand, if it is determined that there is no phrase that matches the searched phrase, the searched phrase is newly registered in the conversion candidate phrase table 62 and the newly registered phrase The order of arrangement is changed so that becomes the highest priority.

  When the conversion candidate word / phrase table 62 corresponding to position priority is to be processed, the character conversion processing unit 31 acquires the current position information detected and measured by the GPS unit 21 at the current time, and then adds the additional dictionary table. The position priority dictionary table 52B prepared in the group 52 is referred to.

FIG. 10A shows an example of the contents of the position priority dictionary table 52B, and FIG. 10B shows an example of a partial content of the basic dictionary 51 corresponding to the position priority dictionary table 52B. Here, the basic priority set for each word on the basic dictionary 51 side is not shown.
As shown in this figure, the position priority dictionary table 52B is formed to have a structure in which certain specific position information is associated with the IDs of some or all of the words registered in the basic dictionary 51. Yes. The position information here is indicated by latitude and longitude.
Here again, looking at an example of the contents of word association between the position priority dictionary table 52B and the basic dictionary 51 shown in FIG. 10, first, “reading / notation” registered in the basic dictionary 51. Is associated with the position information (latitude = 36.65 °, longitude = 139.74 °) registered in the position priority dictionary table 52B by its ID = 05001. Yes. This position information indicates a position determined as a reference in an area where the Tokyo Tower, which is actually a building, exists.
In addition, the phrase “reading / notation” “Nagoya / Nagoya” registered in the basic dictionary 51 is registered with the position information (latitude = 35.11 °, (Longitude = 136.54 °). This position information indicates a position (for example, the position of the prefectural office, etc.) determined as a reference in the area where the city of Nagoya in Aichi Prefecture actually exists.
In other words, the position priority dictionary table 52B is formed by registering words in association with predetermined positions set as conditions.

The character conversion processing unit 31 is associated with position information corresponding to the current position information acquired as described above from the words registered in the conversion candidate word / phrase table 62 at the current stage. The phrase is searched from the position priority dictionary table 52B.
In this case, since the position information associated with the word / phrase in the position priority dictionary table 52B is based on latitude / longitude, for example, the current position information measured by the GPS unit 21 even if the user is in the vicinity of Tokyo Tower. It is rare that the “reading / notation” indicated by the position priority dictionary table 52B exactly matches the position information regarding the phrase “Tokyo Kyoto Tower”. Therefore, when searching from the position priority dictionary table 52B, the position range is set around the current position measured by the GPS unit 21, and the position range set in this way is used as current position information. The current position information and the position information in the position priority dictionary table 52B are compared and referenced. Even in this case, if there is a searched word / phrase, the contents are changed so that this word / phrase has the highest priority in the conversion candidate word / phrase table 62. Based on the contents of the conversion candidate word / phrase table 62 processed as described above, the conversion candidate word / phrase list area AR2 is displayed.
In this way, when the position priority conversion candidate word / phrase table processing is executed, the use probability increases according to the point and area where the user is present (the mobile phone device 1 is present). The priority of the estimated words is made higher. That is, the prediction accuracy can be improved according to the positional conditions such as the point, the area, and the place.

It should be noted that the position priority dictionary table 52B may have contents that can be registered including words and phrases that are not registered in the basic dictionary 51.
In accordance with the contents of the position priority dictionary table 52B, the character conversion processing unit 31 first reads the current indeterminate character from the position priority dictionary table 52B as in the case of the time priority dictionary table 52A. A word / phrase associated with the time corresponding to the current time acquired at the time when the column (conversion source character string) matches the beginning of the reading is searched.
Then, if there is a searched phrase, as in the case of the time priority dictionary table 52A, whether there is a phrase that matches the searched phrase in the already created conversion candidate phrase table 62. The content of the conversion candidate word / phrase table 62 is changed according to the determination result.

Next, an operation example when the conversion candidate word / phrase table processing as buzzword / phrase priority is to be performed will be described.
Also in the character conversion processing unit 31 in this case, first, in correspondence with the kana character input in the kana / kanji character input mode, the conversion candidate word / phrase table according to the basic operation example described above with reference to FIGS. 62 is created. Next, the character conversion processing unit 31 accesses the buzzword priority dictionary table 52C.

An example of the contents of the buzzword priority dictionary table 52C is shown in FIG.
In the popular word / phrase priority dictionary table 52 </ b> C shown in this figure, it is assumed that various typical words / phrases that are popular at the time are registered. Each registered word has a structure in which items of ID, reading, part of speech, and notation are associated.
That is, the buzzword / phrase priority dictionary table 52C is viewed as being formed by registering words / phrases that meet the condition that the frequency of use (frequency of use) by the user of the mobile phone device 1 is high at that time. Can do.
Note that the buzzword priority dictionary table 52C may include new words, etc., in addition to words and phrases used so far. Therefore, the buzzword priority dictionary table 52 </ b> C has contents that can include words and phrases registered in the basic dictionary 51 and can also include words and phrases that are not registered in the basic dictionary 51.
Also, in the buzzword priority dictionary table 52C, when the same word / phrase is registered in the basic dictionary 51, only the ID may be registered.

The processing of the character conversion processing unit 31 corresponding to the buzzword priority dictionary table 52C conforms to the case where words and phrases not registered in the basic dictionary 51 can be registered in the time priority dictionary table 52A and the position priority dictionary table 52B. .
That is, the character conversion processing unit 31 that has accessed the buzzword priority dictionary table 52C reads the current undefined character string (conversion source character string) and the reading from the words registered in the buzzword priority dictionary table 52C. Search for a phrase that matches the beginning and is associated with a time corresponding to the current time acquired at that time.
If there is a searched phrase, first, it is determined whether or not there is a phrase that matches the searched phrase in the already created conversion candidate phrase table 62.
When it is determined that there is a phrase that matches the searched phrase, the phrase that matches the searched phrase has the highest priority in the conversion candidate phrase table 62 so that the phrase is searched for. Change the array order.
Further, when it is determined that there is no phrase that matches the searched phrase, the searched phrase is newly registered in the conversion candidate phrase table 62, and the newly registered phrase is the highest. The arrangement order is changed so as to be the priority order.

Then, the character conversion processing unit 31 newly registers the searched word / phrase in the conversion candidate word / phrase table 62 already created so that the priority is the highest, and the conversion obtained in this way. Display control of the conversion candidate word / phrase list area AR2 in the display unit 22 is executed via the confirmation result output unit 35 so that the contents of the word / phrase table 62 are reflected.
By performing such conversion candidate word / phrase table processing, the word / phrase that is popular at that time is given priority and presented as a conversion candidate. Therefore, in this case, the current trend (trend / phrase) The prediction accuracy of the phrase can be improved by adapting to the conditions.

  As described above, in the present embodiment, the conversion candidate word / phrase table processing as the basic operation described above with reference to FIGS. 3 to 8 is executed, and then the current time, current position (point) , Area, place), the contents of the reference sentence (e-mail, document file), and the conversion candidate word / phrase table process for determining the priority order based on at least one of the popular words / phrases. As a result, the priority of conversion candidate phrases that are more appropriate in conformity with the situation at that time is made higher, so the conversion is performed in comparison with the case of only conversion candidate phrase table processing as a basic operation. The prediction accuracy for candidate words will be greatly improved.

  In the present embodiment, as a configuration for improving the prediction accuracy in this manner, the basic dictionary 51 is essential, and an additional dictionary ( A time priority dictionary table 52A, a position priority dictionary table 52B, and a buzzword priority dictionary table 52C). In other words, as described above, first, the basic dictionary 51 (and the priority word / phrase table 63) is searched to create the conversion candidate word / phrase table 62 in which basic priorities are set, and then necessary. A search from the additional dictionary is also performed, and the contents searched for from the additional dictionary are changed so that the highest priority is given to the conversion candidate word / phrase table 62.

In such a mechanism, first, the basic dictionary 51 which is indispensable in the creation of the conversion candidate word / phrase table may be formed so as to have truly basic registration contents, and this has the following advantages. Will be obtained.
For example, if the prediction accuracy equivalent to that of the present embodiment is to be obtained only by the basic dictionary 51 (and the priority phrase table 63), for example, time information, position information, It is necessary to add information indicating that it is a buzzword at that time, and the contents of the basic dictionary 51 itself are enlarged and the structure is complicated. Then, a large capacity is required for storing the basic information 51 in the mobile phone device 1, and processing such as searching for the basic dictionary 51 becomes heavy. Will also cause bad effects such as becoming heavy.
Further, for example, depending on the user, it is considered that not all priority functions of time priority, position priority, and buzzword priority are necessary, but only a part of them may be necessary. If all of the functions are given only to 51, it is necessary to incorporate in the basic dictionary 51 even priority functions that are unnecessary depending on the user. That is, the above-described adverse effects occur due to excessive priority functions, which is a considerable disadvantage for the user.
Therefore, if the priority function is provided as an additional dictionary as in the present embodiment, as described above, only basic words and phrases need to be registered in the basic dictionary 51. The capacity of the dictionary 51 is reduced, and the structure is simplified. Therefore, the retrieval process to the basic dictionary 51 for creating the initial conversion candidate word / phrase table 62 is light. Further, when searching for additional dictionaries thereafter, it is only necessary to search by accessing only the additional dictionaries corresponding to the priority function required for the user. In this way, the processing load required for dictionary search is reduced by combining the basic dictionary 51 and the additional dictionary.
In this way, in the present embodiment, by providing an additional dictionary for the basic dictionary 51, the configuration for improving the prediction accuracy in the character input system is more efficient than before, Something useful for the user can be obtained.

  In addition, what is shown as an additional dictionary in the present embodiment is a time priority dictionary table 52A, a position priority dictionary table 52B, and a buzzword priority dictionary table 52C. Among these additional dictionaries, the time-priority dictionary table 52A and the position-priority dictionary table 52B have contents that are universal to some extent. It is also reasonable to keep it. However, in the popular word / phrase priority dictionary table 52C, since the target word / phrase is based on the current trend, it is considered that the period in which the word / phrase priority dictionary table can be used effectively is very short. For this reason, although it is possible to store it as a preset in the mobile phone device 1, for example, after the mobile phone device 1 reaches the user's hand, the content updated in accordance with the current time at that time It can be said that it is preferable that the buzzword priority dictionary table 52C can be acquired sequentially.

As a configuration for acquiring the buzzword priority dictionary table 52C as described above, for example, an electronic device that is assumed to be acquired and stored in the mobile phone device 1 using the morphological analysis function described below. Morphological analysis of character string information corresponding to the text content of e-mail and document files, or text information corresponding to the text content described on the Web page (Web content) browsed using the network connection function It is conceivable that the mobile phone device 1 creates a buzzword / phrase priority dictionary table based on the analysis result by executing the process. More specifically, for example, each time a morphological analysis process is performed, a history of words obtained as a result is created. For example, words obtained with a certain frequency of appearance from the history are buzzwords. Are registered in the buzzword priority dictionary table 52C. In addition, for words / phrases that have already been registered in the popular word / phrase priority dictionary table 52C but whose appearance frequency has decreased to a certain level or less by referring to the history, if the registration is deleted, the popular word / phrase priority is given. The dictionary table 52C can always maintain the contents in which words and phrases as buzzwords adapted to the current trend are registered.
Also, a method of downloading and acquiring from a server providing data as the buzzword priority dictionary table 52C on the Internet for a fee or free of charge is also conceivable.

In the present embodiment, it is possible to change the priority order of the conversion candidate words / phrases by referring to the document contents of an e-mail or a document file. In this way, the priority function based on referring to the document content is referred to as “reference sentence priority”.
When the process of the conversion candidate word / phrase table 62 as the reference sentence priority is to be performed, the following processes and operations are executed in the character input system (character input support application 30) of the present embodiment.
In executing the reference sentence priority conversion candidate phrase table processing in the present embodiment, the target reference sentence is an individual e-mail stored in the storage unit 18 as the e-mail data 53 shown in FIG. The contents and each of the document files included in the document file group 54 similarly stored in the storage unit 18.

First, a case where an e-mail is used as a reference sentence will be described as the conversion candidate word / phrase table processing prioritizing the reference sentence.
In this case, the character conversion processing unit 31 first instructs the reference sentence analysis processing unit 33 to specify an electronic mail to be referred to and acquire it as a reference sentence. In response to this command, the reference sentence analysis processing unit 33 requests the mailware 41 for the contents of the designated electronic mail. The mailware 41 is application software for realizing e-mail transmission / reception control, e-mail storage management after reception, e-mail creation, and the like. The mailware 41 executes direct management of the e-mail data 53.
Further, in this case, when the e-mail is used as a reference sentence, for example, a case where an e-mail is created as a reply to the already received e-mail can be cited. In this case, the electronic mail that is the source of the reply is used as a reference sentence.
In response to the above request, the mailware 41 reads the content of the specified email from the email data 53 and passes it to the reference sentence analysis processing unit 33.

Here, it is assumed that the content of the electronic mail acquired by the reference sentence analysis processing unit 33 as described above is as shown in FIG. In this figure, the content of this e-mail is shown by the display mode when it is displayed on the display unit 22. On the upper side, the sender of the e-mail (From: recreation secretary general, destination (To: recreation committee)) , The title (Subject: About the bowling tournament) is shown, and the text is shown below it.

"Hello, It's cold.
Now, let's decide the prize for the bowling competition.
What is the best time?
Please contact me.
Secretary-General "

It has become.

In this way, the reference sentence analysis processing unit 33 to which the content data of the e-mail corresponding to a specific message is passed, for example, among the e-mail contents, in this case, the content of the text is converted into the morphological analysis processing unit 34. To instruct the morphological analysis of the contents of this text. In response to this instruction, the morpheme analysis processing unit 34 executes a morphological analysis process on the text content of the passed e-mail body as an analysis target. Note that there is no particular limitation on the method and algorithm to be used for the morphological analysis processing, and a method based on a technique that has been known so far or will be known in the future. Any algorithm may be employed. Then, the morpheme analysis processing unit 34 generates the result of the morpheme analysis process as the morpheme information 64, and stores this in the RAM 19, for example.
The morphological analysis in this embodiment is intended to subdivide and decompose words that can be used in a dictionary for kana conversion. For example, if there is a method other than morphological analysis, it is adopted. It doesn't matter.

  FIG. 12B shows a part of the morpheme information 64 that is generated and held in accordance with the content of the body of the electronic mail shown in FIG. The morpheme information 64 shown in this figure is information obtained by associating four items, ID, notation, part of speech, and reading, for each word decomposed as a morpheme. The ID shown here describes the ID assigned to the searched word / phrase by searching the basic dictionary 51 for a word / word / part of speech that matches the word / phrase obtained by disassembling as a morpheme. Is. Thereby, the phrase registered in the basic dictionary 51 and the phrase registered in the morpheme information 64 are associated with each other by the ID. From this, the structure of the morpheme information 64 is such that, if the same words are registered in the basic dictionary 51 among the phrases obtained as morphemes, only the ID is included as an item. Also good. In this way, for example, the data size required for the morpheme information 64 can be reduced.

As a character input operation using kana / kanji here, it is assumed that the user first inputs the kana character “shi”. FIG. 13A shows an example of contents when the conversion candidate word / phrase table 62 generated in response to this input is in accordance with the basic operation example described above with reference to FIGS. The conversion candidate word / phrase table 62 in FIG. 13 (a) has a content in which words / phrases whose first character of reading is “shi” retrieved from the basic dictionary 51 are arranged in a priority order that directly corresponds to the basic priority. (Accordingly, here, it is assumed that a word whose reading starts with “shi” is not registered in the priority word table 63).
If the reference sentence priority conversion candidate phrase table processing is enabled, the character conversion processing unit 31 creates the conversion candidate phrase table 62 by the basic operation example as described above, and then The generated morpheme information 62 for the designated reference sentence is generated and held, and the contents of the morpheme information 62 are referred to.
Here, it is assumed that the designated reference sentence is the e-mail body shown in FIG. 12A, and the morpheme information 64 generated and held corresponding to this is shown in FIG. 12B. The character conversion processing unit 31 refers to the contents of the morpheme information 62, and uses the current input kana character (indeterminate character string (conversion source character string)) among the words (morphemes) registered here. Search for a word that begins with a certain “shi”. Then, in this case, the phrase “syohin / prize” is obtained as a search result. Therefore, the character conversion processing unit 31 uses the search candidate word / phrase table 62 of FIG. 13 (a) as shown in FIG. Is changed to have the highest priority.

  Depending on the content of the e-mail body, the morpheme information 64 may include words that are not in the basic dictionary 51. When such a phrase that does not exist in the basic dictionary 51 is searched based on an indeterminate character string, the character conversion processing unit 31 newly registers the searched phrase in the conversion candidate phrase table 62. Then, the order of arrangement is changed so that the newly registered word / phrase has the highest priority.

For example, when creating a reply sentence for a received electronic mail, there is a high possibility that the character string in the body of the received electronic mail will be used again. Therefore, if the priority order of registered words in the conversion candidate word / phrase table 62 is set according to the content of the e-mail body as described above, and this is reflected in the display of the conversion candidate word / phrase list area AR2, for example, The prediction accuracy corresponding to the text creation in e-mail or the like will be improved.
It should be noted that here, in the electronic mail, only the text is used as the reference sentence, that is, only the information on the body part, but it is conceivable that the title, destination, sender name, etc. in the header part are also included in the reference sentence. In particular, for the address and sender name, if the user's profile information registered in the mobile phone device 1 or the person name, address, etc. acquired by linking with the phone book information are included in the reference sentence Will be useful.

Also, a case where a document file is used as a reference sentence will be described as reference candidate priority conversion candidate word / phrase table processing.
In this case, the character conversion processing unit 31 instructs the reference sentence analysis processing unit 33 to specify a document file to be referred to and acquire it as a reference sentence. In response to this command, the document editor 42 is requested for the designated document file. The document editor 42 is application software for performing creation / editing and storage management of document files. Here, one or more document files stored and managed by the document editor 42 in the storage unit 18 are shown as a document file group 54.
When a document file is used as a reference sentence during character input operation, for example, a document file having a certain text content and stored so as to be included in the document file group 54 is read by the document editor 42. You can mention when you are opening and editing.
In response to the above request, the document editor 42 reads the designated document file from the document file group 54 and passes it to the reference sentence analysis processing unit 33.
Then, the reference sentence analysis processing unit 33 passes the transferred document file to the morpheme analysis processing unit 34, and the case where the reference sentence is the e-mail body for the character string (sentence) that is the content of the document file. Similarly, morpheme analysis processing is executed, and morpheme information 64 is generated and held.

  14A and 14B respectively show a content example of a document file that is a target reference sentence in this case and a part of a content example of morpheme information 64 corresponding thereto. As can be seen by comparing these figures, the morpheme information 64 shown in FIG. 14B includes morphemes (phrases) obtained as a result of performing morphological analysis on the text contents in the document file shown in FIG. ) Is registered.

In the character input mode using kana / kanji here, it is assumed that the user first inputs the kana character “shi”. The conversion candidate word / phrase table 62 generated in response to this input is the same as the case where the previous mail text is used as a reference sentence as an example of contents when the basic operation example described above with reference to FIGS. Referring to FIG.
As described above, the conversion candidate word / phrase table 62 in FIG. 13A is a priority in which words / phrases whose first character of reading is “shi” retrieved from the basic dictionary 51 are directly associated with the basic priority. It has contents arranged in order (hence, in this case, it is assumed that no words starting with “shi” are registered in the priority word table 63).
If the reference sentence priority conversion candidate phrase table processing is enabled, the character conversion processing unit 31 creates the conversion candidate phrase table 62 by the basic operation example as described above, and then The generated morpheme information 62 for the designated reference sentence is generated and held, and the contents of the morpheme information 62 are referred to.
The designated reference text here is the document file shown in FIG. 14A, and the morpheme information 64 generated and held corresponding to this is shown in FIG. 14B. The character conversion processing unit 31 refers to the content of the morpheme information 62 and searches for the word / phrase (morpheme) registered here for the word / phrase beginning with “shi” which is the current input kana character. Become. Then, in this case, the phrase “reading / notation” “shohin / product” is obtained as a search result. Therefore, the character conversion processing unit 31 uses the search candidate word / phrase table 62 shown in FIG. 13 (a) as shown in FIG. Is changed to have the highest priority. Also in this case, there is a possibility that words that are not in the basic dictionary 51 are included in the morpheme information 64 according to the contents of the document file, and such words are searched based on an indefinite character string. . Also in this case, as in the case of the previous mail body, the character conversion processing unit 31 newly registers the searched phrase in the conversion candidate phrase table 62 and the newly registered phrase has the highest priority. The order of arrangement is changed so that the order becomes the same.

  In this way, the priority order of registered words / phrases in the conversion candidate word / phrase table 62 is set in accordance with the text content in the document file and reflected in the display of the conversion candidate word / phrase list area AR2, so that the document file can be edited or modified. The prediction accuracy will be improved. In other words, depending on conversion candidate word / phrase table processing as a reference sentence priority, prediction accuracy is improved by presenting at least words / phrases that are likely to be input with respect to the e-mail / document file created at that time. It will be.

  According to the description so far, the character input system of this embodiment has a priority function (priority item) for four conversion candidate words: time priority, position priority, buzzword priority, and reference sentence priority. It will be. These priority items have been described a little earlier, but the user can arbitrarily set the validity / invalidity. In other words, the user can set the validity / invalidity for each priority item by calling the setting screen for these four priority items by a predetermined operation on the mobile phone device 1 and performing the operation on this setting screen. It is said that. The contents about valid / invalid set in this way are reflected in the priority item table 61. That is, the priority item table 61 is information indicating validity / invalidity for each priority item determined in accordance with a user operation, which includes time priority, position priority, buzzword priority, and reference sentence priority, and is stored in the storage unit 18. It is what is done.

As an example of the structure of the priority item table 61, as shown in FIG. 15, the priority item table 61 includes information items of “time priority”, “position priority”, “position range setting”, “popular phrase priority”, and “reference sentence priority”.
The “time priority” information item stores a value indicating whether the time priority conversion candidate word / phrase table processing is valid or invalid.
The “position priority” information item stores a value indicating whether the position priority conversion candidate word / phrase table processing is valid or invalid.
The “position range setting” information item is an information item that is valid when the “position priority” information item is set to valid. In the position priority conversion candidate word / phrase table processing, the position priority dictionary table 52B is searched for position information included in a certain position range based on the current position information acquired from the GPS unit 21. The “position range setting” indicates a position range based on the current position information to be set at this time. That is, in this embodiment, when the position-priority conversion candidate word / phrase table processing is set to be valid, the position range can also be set by the user. Here, the information as the position range setting is expressed by ± x ° (x indicates an arbitrary number in a numerical range that can be set in advance).
The “popular phrase priority” information item stores a value indicating either valid / invalid with respect to the setting of the conversion candidate phrase table process with trendy phrase priority.
The “reference sentence priority” information item stores a value indicating whether the reference sentence priority conversion candidate phrase table process is valid or invalid. Note that with reference sentence priority in this embodiment, the document to be referred to is either an email or a document file. Correspondingly, conversion candidate word / phrase table processing with reference sentence priority is enabled. When set to, it may be possible to select and set a document to be referred to from at least one of an electronic mail and a document file.

  For example, the character conversion processing unit 31 is set to valid with reference to the priority item table 61 each time a kana character is input in the kana / kanji character input mode and the conversion candidate word / phrase table is newly created. A priority function is recognized, and conversion candidate word / phrase table processing corresponding to the recognized priority function is executed. For example, when the conversion candidate word / phrase table processing of “time priority”, “position priority”, and “buzz word priority” is enabled and the reference sentence priority is disabled, “time priority”, The conversion candidate word / phrase table processing of “position priority” and “buzzword priority” is sequentially executed. If “position priority” is enabled as in this case, the position range specified in the position range setting item is set starting from the current position measured by the GPS unit 21. The position information in the time priority dictionary table 52B is searched.

The flow of operation of the character input system according to the present embodiment described so far is shown in the flowchart of FIG. The processing shown in this figure is an operation corresponding to one-time (single-character sentence) kana character input. That is, if it corresponds to the case in FIG. 3 to FIG. 8, a series of steps after step S102 shown in FIG. 16 is performed in response to the operation of inputting the kana character “ko” first. If the procedure is executed and, further, an operation for inputting “n” as shown in FIG. 8 after the kana character “ko” is performed, the sequence of steps after step S102 is executed again. The
Note that the processing shown in this figure can be considered to be mainly a procedure executed by the character conversion processing unit 31 in the character input support application 30. In addition, this means that the procedure of FIG. 16 is a process obtained by the CPU in the system control unit 17 executing the program as the character input support application 30 supported by the character input system of the present embodiment. It also becomes.

In this figure, first, in step S101, it waits for an operation to input a kana character of one character sentence to be detected. As long as a negative determination result is obtained here, the process proceeds to step S128, and a determination operation is waited as will be described later. For example, if the character conversion processing unit 31 detects an input operation of a kana character of a single character sentence, the process proceeds to the steps after step S102. Further, along with this detection, it is recognized what kind of kana character is input.
Next, in step S102, under the control of the character conversion processing unit 31, the uncertain character set as the search condition using the current uncertain character string including the newly input kana character for the first character as the search condition. The basic dictionary 51 is searched for a phrase having the column as a reading character string at the beginning (beginning) portion. Based on the search result, a new conversion candidate word / phrase table 62 is created in step S103 as described above with reference to FIG.

In subsequent step S104, the character conversion processing unit 31 refers to the priority word / phrase table 63 to search for a phrase having the uncertain character string as the search condition as a reading character string in the first portion, and in step S105, It is determined whether or not a word or phrase corresponding to the current search condition has been searched. If a negative determination result is obtained here, step S106 is skipped and the process proceeds to step S107. If an affirmative determination result is obtained, the procedure of step S106 is executed and then step S107 is performed. Proceed to
In step S106, as described with reference to FIGS. 4 and 5, the contents of the conversion candidate word / phrase table 62 are set so that the priority of the word / phrase retrieved from the priority word / phrase table 63 in step S104 is highest (first). change.

In step S107, the character conversion processing unit 31 reads and references the priority item table 61, and determines whether or not “time priority” is set to be valid.
If a negative determination result is obtained assuming that “time priority” is set to be invalid, the procedure of steps S108 to S111 is skipped and the process proceeds to step S112. On the other hand, if an affirmative determination result is obtained assuming that “time priority” is set to be valid, the procedure of steps S108 to S111 is executed.

In step S <b> 108, the character conversion processing unit 31 acquires the current time from the clock unit 20. Next, referring to the time priority dictionary table 52A in step S109 and step S110, the words / phrases registered in the current conversion candidate word / phrase table 62 (here, the words / phrases registered in the time priority dictionary table 52A are stored in the basic dictionary 51). Search for a phrase (corresponding phrase) associated with a time including the current time acquired in step S108, and the corresponding phrase exists. Whether or not is determined. Alternatively, the time-priority dictionary table 52A is a word / phrase in which the uncertain character string obtained in step S101 matches the initial character string and the time associated with the current time obtained in step S108 is associated as the corresponding phrase. To determine whether or not the corresponding phrase exists.
If a negative determination result is obtained in step S110 that there is no corresponding phrase, the procedure of step S111 is skipped. On the other hand, if a positive determination result is obtained because the corresponding phrase exists, the procedure of step S111 is executed, and the process proceeds to step S112.
In step S111, the contents of the conversion candidate phrase table 62 are changed so that the corresponding phrase searched in step S110 has the highest priority.

  In step S112, the priority item table 61 is referenced to determine whether or not “position priority” is set to be valid. If a negative determination result is obtained, the procedure from steps S113 to S116 is performed. Skip to step S117. On the other hand, if a positive determination result is obtained in step S112, the process proceeds to step S113. If it is assumed that “position priority” is set to “valid”, and the process proceeds from step S112 to step S113, the character conversion processing unit 31 also reads and acquires the “position range setting” information shown in FIG. To be done.

  In step S113, current position information is acquired. For this purpose, first, the current location information, such as latitude and longitude, measured by the GPS unit 21 is acquired. In addition, the position range specified by the “position range setting” information in the priority item table is obtained using the position information acquired from the GPS unit 21 as a starting point. The position range obtained in this way becomes the current position information to be acquired in step S113.

In subsequent steps S114 and S115, with reference to the position priority dictionary table 52B, the indeterminate character string obtained in step S101 matches the initial character string, and is included in the current position information obtained in step S113. It is determined whether or not there is a registered phrase (corresponding phrase) associated with the position information. If the determination result is negative, step S116 is skipped and the process proceeds to step S117. On the other hand, if a positive determination result is obtained, the process proceeds to step S116.
In step S116, the content of the conversion candidate word / phrase table 62 is changed so that the highest priority is given to the corresponding word / phrase obtained in correspondence with step S115.

  In the next step S117, with reference to the priority item table 61, it is determined whether or not “buzzword phrase priority” is set to be valid. If a negative determination result is obtained, steps S118 to S120 are performed. The procedure is skipped and the process proceeds to step S121. On the other hand, if a positive determination result is obtained in step S117, the process proceeds to step S118.

In step S118, referring to the buzzword / phrase priority dictionary table 52C, the buzzword / phrase priority dictionary table 52C includes an indeterminate character string obtained according to the current character input operation as the first character string of the reading. (Corresponding word / phrase) is searched, and whether or not the corresponding word / phrase exists as a search result is determined in step S119. Here, if a negative determination result is obtained because there is no corresponding phrase, the procedure of step S120 is skipped and the process proceeds to step S121. If a positive determination result is obtained, step S120 is performed. To be executed.
In step S120, the content of the conversion candidate word / phrase table 62 is changed so that the highest priority is given to the searched word / phrase. At this time, the corresponding phrase searched from the buzzword priority dictionary table 52C is only in the buzzword priority dictionary table 52C, and the conversion candidate word / phrase table 62 at that time is not registered (in the basic dictionary 51). If not registered), as described above, after newly registering the corresponding phrase in the conversion candidate phrase table 62, the priority of the corresponding phrase is set to be the highest. .

  In step S121, with reference to the priority item table 61, it is determined whether or not “reference sentence priority” is set to be valid. If a negative determination result is obtained, steps S122 to S126 are performed. The procedure is skipped and the process proceeds to step S127. On the other hand, if a positive determination result is obtained in step S21, the process proceeds to step S122.

  In step S122, it is determined whether there is a reference sentence to be used. For example, in an e-mail creation mode as a reply to an e-mail or an edit character input mode of a document file, there is a reference sentence to be used, and a positive determination result is obtained. On the other hand, a negative determination result is obtained particularly when there is no related electronic mail or document file and there is no reference sentence to be used. If a negative determination result is obtained, the procedure from step S123 to step S126 is skipped and the process proceeds to step S127. If a positive determination result is obtained, the process proceeds to step S123.

  In step S123, a morpheme information creation process is executed. That is, as described above with reference to FIGS. 12 and 14, an e-mail or document file as a reference sentence is acquired and the morpheme analysis processing unit 34 is caused to execute the morpheme analysis process. When the completion of the morpheme analysis processing is notified to the character conversion processing unit 31 from the morpheme analysis processing unit 34 via the reference sentence analysis processing unit 33, for example, the character conversion processing unit 31 proceeds to step S124. move on. In step S124, the morpheme information 64 created and held in accordance with step S123 is referred to, and the indeterminate character string obtained in accordance with the current character input operation is read from this morpheme information 64. The phrase (corresponding phrase) included as is searched, and whether or not the corresponding phrase exists is determined in step S125. If a negative determination result is obtained in step S125, step S126 is skipped and the process proceeds to step S127. If a positive determination result is obtained, the process proceeds to step S127 via the procedure of step S126.

  In step S126, the content of the conversion candidate word / phrase table 62 is changed so that the highest priority is given to the searched corresponding word / phrase. Even in this case, the corresponding phrase retrieved from the morpheme information 64 is registered only on the morpheme information 64 side, and is not registered in the conversion candidate phrase table 62 at that time (registered in the basic dictionary 51). If not, first, the corresponding word / phrase is newly registered in the conversion candidate word / phrase table 62, and the priority of the corresponding word / phrase is set to be the highest.

At the stage of reaching step S127, starting with the creation of the basic conversion candidate word / phrase table 62 by the basic dictionary 51 (and the priority word / phrase table 63), priority is given to the items specified to be executed by the priority item table 61. This means that the content change of the conversion candidate word / phrase table 62 has been completed.
It should be noted that, in the actual conversion candidate word / phrase table 62, there is only one highest priority in reality, so the process of changing the contents of the conversion candidate word / phrase table 62 corresponding to a plurality of priority functions is performed. In the case of execution, there is a case where the conversion candidate word / phrase corresponding to which priority function should be given the highest priority and given the highest priority.
For example, if the procedure up to step S127 described so far is followed as it is, the highest priority is set for the corresponding phrase at each step S106, S111, S116, S120, S126. In this case, the highest priority is given to the corresponding phrase obtained by the reference sentence priority. That is, a higher priority is given to the processing corresponding to the priority function as the processing order is later. Therefore, when the priority order is given according to the difference in the priority function, the processing order for the process corresponding to the priority function may be determined in correspondence with the required priority order.
Alternatively, when priority is set for the corresponding phrase in S111, S116, S120, and S126, a certain rule may be given so that the actual priority is set according to this rule. . For example, it is possible to consider a configuration in which different weightings are given to the respective priority functions, and the higher priority is set for the corresponding phrase as the weighting value is larger.

  In step S127, the contents of the conversion candidate word / phrase table 62 in which the priority order of the conversion candidate words / phrases is changed and set by the process so far are reflected on the conversion candidate word / phrase list area AR2. Control for displaying a list of conversion candidate words is executed.

  After the list display of conversion candidate words / phrases is performed in the conversion candidate word / phrase list area AR2 in step S127, the user is displayed in the conversion candidate word / phrase list area AR2 as described above with reference to FIG. It is possible to perform operations for selecting and confirming conversion candidate words. Therefore, in step S128, the process waits for the confirmation operation for the conversion candidate word to be performed. If the confirmation operation is performed, the process proceeds to step S129. Although not shown here, depending on the determination operation being performed, as described in the transition example from FIG. 6A to FIG. 6B, the determination operation is performed. The conversion candidate words are displayed in the main input operation area AR1 as a confirmed character string.

  In step S129, whether the confirmation operation performed in response to step S128 was a confirmation operation for conversion candidate words / phrases other than the first candidate, that is, conversion candidate words / phrases whose priority is not the first, If the determination result is negative and a positive determination result is obtained, the conversion candidate word / phrase that is the object of the final confirmation operation is newly registered in the priority word / phrase table 63 in step S130. If a negative determination result is obtained in step S129, the procedure of step S130 is skipped and the process is temporarily exited. For example, the procedure returns to step S101 again.

By the way, as described above, the buzzword priority dictionary table, which is one of the additional dictionaries (auxiliary dictionaries), has a characteristic that the words and phrases registered therein are so-called buzzwords. It is preferable that updated contents are obtained sequentially after reaching the hand.
As one solution for this, a server that provides data of the buzzword priority dictionary table 52C on the Internet is provided, and the mobile phone device 1 downloads data of the buzzword priority dictionary table 52C from this server. It is also possible to adopt as described above.

By adopting a configuration for downloading the buzzword priority dictionary table from the server in this way, for example, there are the following advantages.
For example, as described above, the mobile phone device 1 itself decomposes text contents such as e-mails, document files, and web contents into morphemes, and creates a buzzword priority dictionary table using the morpheme information. It is possible to update. However, if this buzzword priority dictionary table is configured to be downloaded from the server, there is no need to provide the mobile phone device 1 with a configuration for creating and updating the buzzword priority dictionary table 52C. The operation of the mobile phone device 1 can be lightened.
If a buzzword / phrase priority dictionary table is prepared at the server, a buzzword / phrase priority dictionary table can be created by processing a large amount of information that cannot be handled by the mobile phone device 1. Therefore, for example, the trendy word / phrase priority dictionary table is more sensitive to fashion and can easily have abundant vocabulary.
Further, if the system allows the mobile phone device 1 to download the buzzword priority dictionary table from the server, a business model can be given for the download of the buzzword priority dictionary table.

Therefore, hereinafter, a configuration example of a network system (dictionary download network system) for downloading a buzzword / phrase priority dictionary table constructed including the mobile phone device 1 described so far will be described. .
As will be understood from the following description, the network system here includes not only the “buzz phrase preferential dictionary table” but also the “word corresponding priority phrase table” reflecting the buzz phrase from the server. Downloadable.

FIG. 17 schematically shows an overall construction example of the dictionary download network system of this embodiment.
As shown in this figure, the dictionary download network system of this embodiment can be regarded as comprising a mobile phone device 1, a dictionary server 2 (server device), and a gateway server 3.
The cellular phone device 1 can input characters by the character input support application 30 as described above. In this case, the buzzword priority dictionary table 52C, which is one of the additional dictionary tables in the additional dictionary table group 52 of FIG.
Here, only one mobile phone device 1 is shown, but in reality, there are many in this dictionary download network system.

The dictionary server 2 holds data of the buzzword priority dictionary table, and can execute data transmission of the buzzword priority dictionary table corresponding to download in response to a request from the mobile phone device 1.
In this case, the buzzword priority dictionary table is automatically created and automatically updated.
Furthermore, a process for charging the user can be executed in accordance with the download of the buzzword priority dictionary table.
In this case, the dictionary server 2 circulates (crawls) Web contents such as homepages and blogs existing on the Web (Internet) to automatically create a buzzword priority dictionary table. Content data can also be collected. In this figure, Web contents to be circulated by the dictionary server 2 are shown as a Web content group 4.

The gateway server 3 has a function of converting a communication protocol of the so-called mobile phone device 1 and a communication protocol corresponding to the network such as the Internet so that the mobile phone device 1 can communicate via the network. Have.
Each mobile phone device 1 is capable of transmitting / receiving e-mails via a network, or accessing Web contents represented by a home page, a blog, and the like through the gateway server 3.
In addition, the gateway server 3 in this case also has a function as a mail server corresponding to electronic mail transmitted and received by the mobile phone device 1.
The gateway server 3 in this case is linked to the dictionary server 2 and communicates with the dictionary server 2 as necessary, as will be described later. The communication with the dictionary server 2 may be performed via a dedicated line or a network, for example.

If the dictionary server 2 is provided on a network such as the Internet, in actual operation, communication between the dictionary server 2 and the mobile phone device 1 also passes through a gateway such as the gateway server 3. Here, for convenience of explanation, the dictionary server 2 and the mobile phone device 1 are illustrated as being capable of directly communicating without the intervention of a gateway.
In any case, in the present embodiment, each communication path between the mobile phone device 1, the dictionary server 2, and the gateway server 3 shown in FIG. The communication path to the network and the communication path between the dictionary server 2 and the network can be viewed as a communication path in a single communication network.

  FIG. 18 shows an internal configuration example of the dictionary server 2. As shown in this figure, the dictionary server 2 is assumed to include an arithmetic processing unit 71, a ROM 72, a RAM 73, a mobile phone compatible communication unit 74, a gateway compatible communication unit 75, and a storage unit 76. Can do.

The dictionary server 2 is actually configured with, for example, hardware and software as a computer system, but the arithmetic processing unit 71 is formed having a central processing unit (CPU). Various controls and processes for realizing functions necessary as the dictionary server of the present embodiment are executed by arithmetic processing by executing a program.
The RAM 72 serves as a main storage device when the CPU in the arithmetic processing unit 71 executes arithmetic processing.
The ROM 73 stores various data necessary for the CPU in the arithmetic processing unit 71 to execute arithmetic processing.

The mobile phone compatible communication unit 74 executes communication with the mobile phone device 1 according to the control of the arithmetic processing unit 71, and can be regarded as being formed by hardware and software for that purpose.
In addition, the mobile phone compatible communication unit 74 executes communication with the mobile phone device 1 in accordance with the control of the arithmetic processing unit 71, and can be regarded as being formed by hardware and software for that purpose.
Moreover, the gateway corresponding | compatible communication part 75 performs communication with the gateway server 3 according to control of the arithmetic processing part 71, and can be seen as what is formed with the hardware and software for it.
In addition, as described above, in practice, it is possible to adopt a configuration in which communication between the dictionary server 2 and the mobile phone device 1 can be performed via a gateway. In this case, communication with the mobile phone device 1 can also be realized by the gateway-compatible communication unit 75.

The storage unit 76 is configured by, for example, an HDD (hard disk drive) or the like, and is used as an auxiliary storage device for the arithmetic processing unit 71.
The storage unit 76 stores a program to be executed by the CPU as the arithmetic processing unit 71. In addition, various information and data necessary for the operation execution of the dictionary server 2 are stored.

FIG. 19 shows an example of functions and operations of the dictionary server 2 based on software in a block configuration.
In this figure, functions and operations realized when the dictionary server 2 mainly executes a program include an integrated management unit 81, a crawling unit 82, a morphological analysis processing unit 83, a statistical information processing unit 84, and a dictionary creation processing unit 85. The functional blocks of the communication control unit 86 and the charging processing unit 87 are shown.

The details of the processes and operations performed by the respective parts will be described later. Here, the operations of these parts will be described briefly.
The integrated management unit 81 executes processing as integrated management for functions and operations executed by the dictionary server 2. For example, the linkage of operations of the following parts is also controlled by the integrated management unit 81.
The crawling unit 82 executes control and processing for crawling Web content as described later.
The morpheme analysis processing unit 83 executes morpheme analysis processing on the sentence content information (character information data) as the crawling data 91 to obtain morpheme information 92 obtained by dividing the sentence content into morphemes.
The statistical information processing unit 84 performs generation and storage management of the latest statistical information 93 and storage management of comparison statistical information.
The dictionary creation processing unit 85 executes a process for creating and managing a buzzword priority dictionary table and a buzzword corresponding priority word table.
The communication control unit 86 is a part that executes communication control processing by software as the mobile phone compatible communication unit 74 and the gateway compatible communication unit 75 in FIG.
The billing processing unit 87 executes processing for billing the user of the mobile phone device 1 in accordance with the download of the buzzword priority dictionary table and the buzzword corresponding priority phrase table.

For example, the morpheme analysis processing executed by the morpheme analysis processing unit 83 may be basically the same as the morpheme analysis processing unit 34 in the mobile phone device 1 shown in FIG. However, the processing capacity of the dictionary server 2 is much higher than that of the cellular phone device 1. Therefore, even as the morpheme analysis processing executed by the morpheme analysis processing unit 83, higher processing capability and performance than the morpheme analysis processing unit 34 in the mobile phone device 1 can be provided. Thereby, for example, more reliable morphological analysis processing can be performed at higher speed.
For the same reason, for example, the processing related to the creation of the buzzword / phrase priority dictionary table and buzzword / phrase priority word / phrase table such as the statistical information processing unit 84 and the dictionary creation processing unit 85 can be given high speed and high performance. . Therefore, for example, a buzzword priority dictionary table and a buzzword-preferred priority phrase table can be created by an advanced algorithm after using a huge amount of information. As a result, depending on the dictionary server 3, a buzzword / phrase priority dictionary table and a buzzword / phrase priority word / phrase table having a very substantial content can be obtained.

19, crawling data 91, morpheme information 92, latest statistical information 93, comparative statistical information 94, basic dictionary 95, buzzword phrase information 96, buzzword phrase priority dictionary table 97, buzzword phrase priority phrase table 98, contract information 99, the crawling content list 100 is information or data acquired or generated in the dictionary server 2, and is stored in the storage unit 76 or held in the RAM 72. As will be described below, the dictionary server 2 uses these pieces of information to create functions such as the buzzword priority dictionary table 97 and the buzzword corresponding priority phrase table 98, download, etc. Is realized.
Note that the above information, data contents, and the like will be described below as appropriate.

  FIG. 20 shows an internal configuration example of the gateway server 3. In this figure, a configuration of each functional part realized by cooperation of hardware and software is shown. The hardware configuration of the gateway server 3 is a computer system including a CPU, a ROM, a RAM, an HDD and the like having performance corresponding to the server function, for example, in the same manner as the dictionary server 2.

  The gateway server 3 shown in this figure includes a protocol conversion processing unit 111, a mobile phone compatible communication unit 112, an Internet compatible communication unit 113, a mail server unit 114, and a dictionary server compatible communication unit 115.

The protocol conversion processing unit 111 first performs so-called protocol conversion for enabling the mobile phone device 1 to communicate via the Internet as its main function.
For example, when transmitting data from a mobile phone device 1 to an address (IP address) on the Internet 6 as Internet communication by a mobile phone, first, the data to be transmitted is determined according to the communication protocol and communication method between the mobile phone devices. The transmission signal is transmitted to the gateway server 3 via the mobile phone network 5. In the gateway server 3, the transmitted signal is received by the mobile phone compatible communication unit 112, for example, a predetermined process such as a demodulation process corresponding to transmission modulation is performed to extract data to be transmitted, and a protocol conversion process is performed. To the unit 111. The protocol conversion processing unit 11 converts the received data into data in a format corresponding to a communication protocol corresponding to the Internet, and transmits the data to the Internet 6 from the Internet compatible communication unit 113. As a result, the data transmitted from the mobile phone device 1 is transmitted to an appropriate address on the Internet 6.
Further, data transmitted from a certain address on the Internet 6 to the mobile phone device 1 is received by the Internet compatible communication unit 113 of the gateway server 3 and passed to the protocol conversion processing unit 111.
The protocol conversion processing unit 111 converts the received data into data in a format corresponding to a communication protocol corresponding to communication between mobile phone devices and outputs the data to the mobile phone compatible communication unit 112. The mobile phone compatible communication unit 112 performs transmission modulation corresponding to the mobile phone communication network 5 on the input data and transmits it to the mobile phone device 1 as a transmission signal. The cellular phone device 1 receives and demodulates the signal transmitted in this way. As a result, data transmitted via the Internet can be obtained.

Further, the gateway server 3 in this case is provided with a mail server function corresponding to electronic mail transmitted and received by the mobile phone device 1. The mail server unit 114 is provided for this mail server function.
For example, an e-mail transmitted from the mobile phone device 1 is received by the mobile phone compatible communication unit 112, and in this case, is passed to the protocol conversion processing unit 111. The protocol conversion processing unit 111 cooperates with the mail server unit 114, for example, to send a mail transmitted from the mobile phone device 1 according to a predetermined mail transmission / reception protocol via the Internet 6 to a destination corresponding to a destination. Send to.
In addition, an e-mail transmitted to the mobile phone device 1 (user) via the Internet 6 is received by the Internet-compatible communication unit 113. In this case, the e-mail is transmitted via the protocol conversion processing unit 111. Passed to the mail server unit 114. The mail server unit 114 stores the received mail in a mailbox corresponding to the destination mobile phone device 1 (user), and from then on, the protocol conversion processing unit 111 carries it in response to a request from the mobile phone device 1. An e-mail is transmitted via the telephone communication unit 112.
For example, in the case of mail transmission / reception between the mobile phone devices 1 of the same carrier, the mail server unit 114 does not pass through the Internet 6 but passes through the mobile phone network 5 from the mobile phone compatible communication unit 112. Thus, e-mails are sent and received.

  By providing the gateway server 3 in this way, the cellular phone device 1 is configured to be able to perform Internet communications such as Web access and electronic mail transmission / reception.

The dictionary server compatible communication unit 115 is provided to enable communication between the gateway server 3 and the dictionary server 2.
For example, when the communication between the dictionary server 2 and the gateway server 3 is an independent network such as LAN or WAN, or communication is performed using a dedicated line, a protocol corresponding to such a network or dedicated line is used. It is configured to be able to communicate with each other. Note that the communication between the dictionary server 2 and the gateway server 3 can be considered as communication via the Internet, but in this case, it can be realized by communication via the Internet compatible communication unit 113. The unit 115 can be omitted.
Regardless of which communication network is used, in the present embodiment, communication between the dictionary server 2 and the gateway server 3 is enabled, and the dictionary server 2 is described later. In this way, necessary information can be received from the gateway server 3 side.

Subsequently, the flow of processing for creating the buzzword priority dictionary table and the buzzword corresponding priority phrase table by the dictionary server 2 will be described.
Here, the buzzword / phrase priority dictionary table of the present embodiment is basically prepared individually for each user of the mobile phone device 1, and the buzzword / phrase derived and created together with the buzzword / phrase priority dictionary table. The corresponding priority phrase table is also basically prepared individually for each user of the mobile phone device 1.
In addition, as will be understood from the following description, the buzzword priority dictionary table and the buzzword-corresponding priority phrase table of the present embodiment are different from each other so as to be adapted to each user of the mobile phone device 1. Things are obtained. In this way, by providing the buzzword priority dictionary table and buzzword corresponding priority phrase table of contents suitable for each user, in this embodiment, the added value for dictionaries related to buzzwords is increased, In addition, the kana-kanji conversion more convenient for the user can be performed.
Hereinafter, the buzzword / phrase priority dictionary table and the buzzword / phrase-corresponding priority word / phrase table will be referred to as a “buzzword / phrase-related dictionary” unless otherwise distinguished.

  First, the user of the cellular phone device 1 who wants to receive a buzzword-related dictionary download service performs an appropriate contract procedure. As an actual procedure of the contract, for example, the mobile phone device 1 accesses the contract site provided by the dictionary server 2 or another server on the Internet, and the user can operate the mobile phone device 1 to operate the contract site. An input to a predetermined item is performed. Alternatively, it is also conceivable that the user submits a document or the like in which necessary items are written and performs the procedure, and the operator of the dictionary server 2 registers the received document content in the dictionary server 2.

  In any case, as a result of the user's contract for the buzzword related dictionary download service, as one of the contract information 99 of the dictionary server 2 in FIG. Will be registered. The contract information 99 is an information unit composed of a set of user individual contract information. For example, the contract information 99 is stored (held) in the storage unit 76 with a structure and format in which the user individual contract information for each contracted user is stored in a database. To be done.

FIG. 21 shows a partial content example of the individual user contract information that is supposed to form the contract information 99.
Here, the user individual contract information is composed of a header and a data part.
The header is assumed to have various required data items for the corresponding user, and here, in particular, it is indicated that the user ID, the mobile phone ID, and the mail address are stored in the header. The user ID is an identifier uniquely assigned to the user. This user ID is also stored in the cellular phone device 1 used by the user. The mobile phone ID is an identifier assigned to be unique to each mobile phone device 1, and this mobile phone ID is also stored in the mobile phone device 1. Further, the e-mail address is an e-mail address set in the mobile phone device 1 by the corresponding user.

  Here, in the data part, a plurality of contract items and examples of contract contents (contract contents) set for each contract item are shown. The setting of the contract contents for each contract item is performed by the user corresponding to the user individual contract information, and is performed, for example, at the time of the contract procedure described above.

The contract items to be stored in the data section of the user individual contract information, that is, the contract items required for the contract of the download service of the buzzword related dictionary, can be considered in various ways. Contract items related to the creation and download of buzzword-related dictionaries by the server 2 are extracted and shown.
The contract items shown here include the provided email list, email disclosure permission items (1) to (n), search keyword provision permission, web content disclosure permission, crawling designation, buzzword dictionary update frequency, priority It is the phrase dictionary update frequency.

When creating a buzzword-related dictionary that is individual for each user, in order to acquire buzzwords registered in the buzzword-related dictionary, information on the sentence content (character string data) that is the source is required However, in the present embodiment, the buzzword acquisition source should be specified by the user and provided. Among these contract items, the provided mail list, the mail provision permission items (1) to (n), the search keyword provision permission, the disclosure permission of the accessed Web content, and the crawling designation are contract information specifying the buzzword acquisition source and the contract information. Become.
The contract contents shown for each piece of contract information correspond to network designation information.

First, as the buzzword acquisition source, it is assumed that the text content in the mail transmitted and received by the user can be used. For example, the reason is that it is considered that there is a high possibility that what is in fashion at that time will become a hot topic in the contents of e-mail exchange.
In addition, one or more e-mails from which buzzwords are acquired can be designated by the mail address of the user or the other party other than the user.
The provided mail list is a list of e-mail addresses designated by the user as buzzword acquisition sources. In FIG. 21, as shown as contract contents,
(1) aaa@bbb.co.jp
(2) 123abc@bbb.co.jp
(3) eee@fff.co.jp
The three e-mail addresses are set.

In addition, items in which text is shown in electronic mail can generally be divided into a title and a body. Allowed / prohibited to provide the buzzword acquisition source corresponding to each of the three email addresses registered in the provided email list for the contract items of the email provision allowed items (1), (2), and (3). This indicates the item to be performed.
In this case, the mail provision permission item is set to a value of 1 or 0 for each of the title and the text. 1 indicates permission and 0 indicates prohibition.
Email provision permission item (1) is a setting item corresponding to the email address of (1) aaa@bbb.co.jp, but the title is set to [1] and the body is set to [1] Thus, it is shown that both the title and the text are permitted to be provided.
The mail provision permission item (2) is a setting item corresponding to the email of the mail address of (2) 123abc@bbb.co.jp. With regard to this e-mail, [1] is set in the title and [0] is set in the text, which indicates that the title is permitted to be provided but the text is prohibited from being provided.
The mail provision permission item (3) is a setting item corresponding to the email of the email address of (3) eee@fff.co.jp. With respect to this e-mail, [0] is set for the title and [1] is set for the body, which indicates that the title is prohibited from being provided but the body is set to permit the provision.

Further, in the present embodiment, as a buzzword acquisition source, it is also possible to use a character string input for a search on a search window in a site on the Internet accessed by the mobile phone device 1. As a general rule, the higher rank in the so-called search phrase ranking on portal sites on the Internet is likely to be a phrase related to a person, thing, etc. that are attracting attention at that time. From this, the search phrase that is input by the user in the search window and transmitted as character string information to the site is also likely to be a popular phrase.
The search keyword provision permission contract item indicates permission / prohibition of using a character string input to such a search window as a buzzword acquisition source. In this case, 1 indicates permission, 0 indicates prohibition. Here, setting [1] indicates that it is permitted to use the search phrase input in the search window as a buzzword acquisition source.

  The disclosure permission of the accessed Web content indicates permission or prohibition of using sentence contents in the Web content accessed by the mobile phone device 1 contracted and held by the user as a buzzword acquisition source. There is a possibility that people and things that are popular at that time are written in the text as Web contents. In particular, it can be said that there is a high possibility that many popular phrases are used in sites that introduce popular ones or blogs.

The dictionary server 2 also crawls (circulates) the Web content specified by the user, collects the text content information described in the Web content, and uses the crawling data as a buzzword. A related dictionary can be generated. In other words, Web content can be used as a buzzword acquisition source.
The contract item of the crawling designation indicates the Web content that the user designates to be crawled as a buzzword acquisition source.
Here, as the contract contents of the crawling designation,
HP http://www.aaa.co.jp/
HP http://www.bbb.com/
Blog http://sionin.solog-abc.com/nikki
RSS http://aaa.com/index.xml
Each of them is shown.
Here, “HP” represents a so-called website. Blog represents a website generally called a blog. RSS indicates RSS (Rich Site Summary, Really Simple Syndication, RDF Site Summary). In the contract content of FIG. 21, two homepages, one blog, and one RSS data are designated as crawling targets, and addresses (URL: Unified Resource Locator) for the web contents as the crawling targets are specified. It is what is shown.

Further, the buzzword phrase dictionary update frequency of the remaining contract items is determined when the user receives a download service of buzzword-related dictionaries (buzzword phrase priority dictionary table, buzzword phrase corresponding priority phrase table). This specifies the frequency of downloading and updating.
The contract contents in this case are described as “one time / one day”, which indicates that the buzzword phrase priority dictionary table should be updated once a day.
There are other options for setting the update frequency. For example, the unit of the update cycle can be selected in units of weeks or months, such as one week, one month, two months, etc. in addition to one day. It is also possible to add an annual unit to the options. Also, a value representing the number of updates may be set to a numerical value of 1 or more. That is, for example, the update frequency represented by the number of updates / update cycle can be set by combining the number of updates selected by the user and the update cycle.
In addition, it is also conceivable to provide a contract item with an update frequency individually for the buzzword priority dictionary table and the buzzword corresponding priority phrase table.

  Here, the integrated management unit 81 creates an individual crawling content list for each user based on the user unit contract information stored as the contract information 99 as described above. The information stored in the storage unit 76 by making the crawling content list for each user created in this way into a database, for example, becomes the crawling content list in FIG.

FIG. 22 shows an example of the contents of a user-specific crawling content list created based on the user unit contract information shown in FIG.
As shown in this figure, the crawling content list for each user has a structure in which each Web content (crawling content) to be crawled is associated with its type and address. In the type column, “HP” represents a homepage, “Blog” represents a blog, and “RSS” represents RSS data.
The contents of the user individual crawling list shown in this figure are the Web content address shown in the crawling designation contract contents shown in the user unit contract information in FIG. 21 and the mail address shown in the contract contents of the provided mail list. However, it is reflected as it is.

Next, a series of processing operation examples for the dictionary server 2 to create a buzzword-related dictionary will be described with reference to FIGS. Further, in this description, main data and information (crawling data 91, morpheme information 92, latest statistical information 93, comparative statistical information 94, basic dictionary 95) other than the contract information 99 and the crawling content list 100 handled by the dictionary server 2 are described. Examples of the contents of the buzzword information 96, the buzzword priority dictionary table 97, the buzzword corresponding priority word table 98, etc. will be described.
As described later, as buzzword-related dictionaries, a buzzword / phrase priority dictionary table and a buzzword / phrase-related priority phrase table can be created for each user and for each user. Basic and main ones correspond to individual users. Therefore, in the following description, a case where a buzzword-related dictionary for each user is first created will be described.

  As will be described later, in order for the dictionary server 2 to create a buzzword-related dictionary for each user, the specified crawling content, e-mail, or Web in accordance with the contents indicated in the user unit contract information Collect character string information based on search keywords entered on the site and use this character string information.

Here, as an example, the web content specified by a user individual crawling content list (which is an element of the crawling content list 100) corresponding to a certain user (here, user A) is crawled at a certain timing. It is assumed that character strings are collected by crawling by 82. And one of the character string information collected as a result is

“Kemtaku was the best at SMOP's concert today!”

Suppose that In actuality, more character string information is collected by the crawling process by the crawling unit 82, but here, for convenience of simplifying the description, of the collected character string information, Only the above sentence is shown as an example. The crawling process executed by the crawling unit 82 will be described later.

The character string information collected as described above is stored in the RAM 72, for example, as individual crawling data (user individual crawling data) by the user A by the crawling unit 82, for example. The crawling data 91 is a set unit of crawling data for each user that is held in the RAM 72 at that time.
Next, in the dictionary server 2, the morphological analysis processing unit 83 accesses the crawling data 91, reads the user individual crawling data corresponding to the user A, and reads the morpheme about the character string information as the read crawling data. Perform analysis processing. That is, a process of decomposing the text content of the character string information into morphemes is executed. The morpheme information obtained as a result of the morpheme analysis process is held in, for example, the RAM 72 as at least a part of the morpheme information 92. The morpheme information 92 can be viewed as a set of such individual user morpheme information.

FIG. 23 is a character string subjected to the current morphological analysis process.

“Kemtaku was the best at SMOP's concert today!”

3 shows an example of the contents of individual morpheme information generated for each user and held in the morpheme information 92.
The morpheme information shown in this figure is information obtained by associating three items of notation, part of speech, and reading for each word decomposed as a morpheme. Here, at this stage, no ID is assigned to each word. This is because the morpheme information 92 in the dictionary server 2 is information used to create a buzzword-related dictionary to the last, so it is not very meaningful to assign an ID used in the kana-kanji conversion process at this stage. It depends. On the other hand, for example, each word / phrase in the morpheme information corresponding to FIG. 12 (b) is directly used for changing the priority order in the conversion candidate word / phrase table 62, and in this respect, the ID for each word / phrase. Is required. Of course, if necessary, at the stage of generating user-specific morpheme information as the morpheme information 92, it is possible to add an ID to each word as in the case of FIG.

  As described above, the morpheme information 92 sequentially holds the morpheme information obtained based on the document content of the character string information obtained by the crawling process for each user. As will be described later, depending on the content of the user unit contract information, the text content in the sent / received mail, the text content in the web content accessed by the mobile phone device 1, the search character string input to the web site, etc. Words that have been decomposed by performing morpheme analysis processing are also integrated and held as morpheme information 92.

Therefore, the statistical information processing unit 84 in this case reads the individual morpheme information 92 for the user A every time the individual morpheme information 92 is newly obtained for the user A, for example. Then, using the individual morpheme information 92 for the read user A, individual statistical information is generated and updated for the user A.
The statistical information obtained in this way becomes the latest statistical information 93. The latest statistical information 93 may be stored in the RAM 72, but it is often necessary to store the latest statistical information 93 for a relatively long time. In the present embodiment, the storage unit 76 stores the same as in the case of comparison statistical information 94 described later. In this case, the latest statistical information 93 shown in FIG. 19 is also a unit of the latest statistical information for each user.

  FIG. 24A shows an example of the contents of the latest statistical information 93 individually for user A, which is assumed to be generated based on individual morpheme information for user A. The latest statistical information 93 can be updated by, for example, adding a word or adding the number of appearances after being newly created within a predetermined creation period, for example. However, it is assumed that the latest statistical information 93 individually for the user A shown in this figure is finally obtained, for example, corresponding to the time point when the creation period has passed.

As can be seen by the user A looking at the individual latest statistical information 93, the latest statistical information has a structure in which the notation, the part of speech, the reading, and the number of appearances are associated with each word / phrase decomposed into morphemes.
The number of appearances here is, for example, the total number of appearances of each word / phrase obtained as individual morpheme information 92 by the user A within a predetermined period.
This is the rule for accumulating the number of appearances. For example, if a certain “phrase a” appears twice in the text content of the character string information imported from a certain website by crawling processing, It is assumed that “word a” is obtained twice as a morpheme depending on the morphological analysis processing for the sentence content. Therefore, the value of the number of appearances is 2 only with this one character string information.
In addition, “phrase a” appears twice in the character string information imported from a certain website, and “phrase a” is 1 in the character string information imported from another website or e-mail. Assuming that it appears twice, the value of the number of appearances for “word a” is 3 (= 2 + 1).
In accordance with such a rule, the value obtained by accumulating the number of appearances for each word based on the individual morpheme information to the user A obtained sequentially within a certain fixed period is the appearance in the latest statistical information 93. Number of times.
Further, the latest statistical information 93 shown in FIG. 24A is in a form in which the word / phrase array is sorted in the descending order of appearance frequency.

  Here, it is assumed that the dictionary server 2 sets the latest statistical information 93 individually for the user A to be created once a day. For example, this corresponds to the buzzword dictionary update frequency specified in the user unit contract information. In this case, as shown in FIG. 24A, the latest statistical information 93 is created on January 1, yyyy.

Then, as the latest statistical information 93 individually for the user A, there is information created on December 31, xxxx the day before the current one is created. The statistical information processing unit 84, for example, as of this time, according to the timing when the latest statistical information 93 on January 1, yyyy is newly generated, is the previous date of December 31, xxxx. The latest statistical information 93 is changed to the comparative statistical information 94 and managed. As a result, the latest statistical information 93 corresponding to the user A can be stored and managed on January 1, yyyy without duplication with the previous one.
At this time, the statistical information that has been managed as the comparative statistical information 94 before the latest statistical information 93 on December 31, xxxx can be deleted. Alternatively, since it may be necessary for some reason later, it may be preliminarily stored in the storage unit 76 of the dictionary server 2.

FIG. 24 (b) shows the statistics of December 31, xxxx, which are the comparison statistics information 94 corresponding to the new generation and storage of the latest statistical information 93 of January 1, yyyy as described above. An example of information content is shown.
Since the comparative statistical information 94 was originally generated as the latest statistical information 93, the notation, part of speech, reading, and number of appearances for each word / phrase decomposed into morphemes, as with the latest statistical information 93. It has a corresponding structure. In addition, here, as in FIG. 24 (a), the words are arranged in the order of appearance frequency.

Next, the dictionary creation processing unit 85 creates the buzzword-related dictionary of the present embodiment using the latest statistical information 93 and the comparative statistical information 94.
First, the dictionary creation processing unit 85 reads, for example, individual latest statistical information 93 and comparative statistical information 94 into the user A, and temporarily creates buzzword phrase candidate information as shown in FIG.
This buzzword candidate information includes, for example, notation, part of speech, reading, comparison appearance count (α), latest appearance count (β), and difference appearance count (β−α) for each phrase as shown in the figure. The phrases registered in the trendy phrase candidate information are the phrases registered in the latest statistical information 93 in FIG. 24A and the phrases shown in the comparative statistical information 94 in FIG. It is obtained by logical sum. That is, in the buzzword candidate information, all the words registered in the latest statistical information 93 in FIG. 24A and the words shown in the comparative statistical information 94 in FIG. 24B are registered.
Of the comparative appearance count (α), the latest appearance count (β), and the difference appearance count (β−α) corresponding to each phrase, the comparative appearance count (α) is used to generate current trending phrase candidate information. The number of appearances (α) for the corresponding word / phrase shown in the comparison statistical information 94 is shown. Moreover, the latest appearance frequency (β) indicates the appearance frequency (β) for the corresponding word / phrase shown in the latest statistical information 93 used for generating the current popular word / phrase candidate information. The difference appearance count (β−α) indicates a difference value between the comparative appearance count (α) and the latest appearance count (β), which is expressed by β−α.
Note that in FIG. 25, the comparison appearance number (α) and the latest appearance number (β) are shown in consideration that the method of obtaining the difference appearance number (β−α) is visually easily understood. However, in the actual buzzword candidate information, among the comparative appearance count (α), the latest appearance count (β), and the difference appearance count (β−α), the comparative appearance count (α) and the latest appearance count ( Only the difference appearance frequency (β−α) obtained from β) needs to be defined as valid information.

The buzzword candidate information shown in FIG. 25 is sorted in descending order of the difference appearance count (β−α).
Here, in the trendy phrase candidate information, the phrase having a large difference appearance frequency (β−α) is the next day, January 1, yyyy, compared to the previous day, December 31, xxxx. It can be said that this is a phrase that frequently appears. In this way, words that have not been used so far in Web content, etc., but have been frequently used at a certain time, will be noticed in the furrow as a trigger It can be presumed that the phrase corresponds to the person or event that became. In the present embodiment, based on this, such a phrase is treated as one that can be a buzzword, that is, a buzzword candidate.

In addition, as the present embodiment, for example, a threshold is set for the difference appearance frequency (β−α) indicated in the buzzword candidate information. Then, from the words registered in the buzzword candidate information, the words having the difference appearance frequency (β−α) equal to or greater than the set threshold are narrowed down and selected, and these selected words are set as buzzwords. .
For example, in the case of FIG. 25, if the threshold value is set to 400, as shown in the figure, the notation indicating the number of occurrences of a difference of 400 or more (β-α) is “Mameme”, “New Year”, “New Year”, “SMOP”. "5" and "Kemtaku" are set.

Next, the dictionary creation processing unit 85 creates buzzword information 96 based on the buzzword selected as described above, and stores it in the storage unit 76, for example.
FIG. 26 shows an example of the contents of the buzzword phrase information 96 created corresponding to the buzzword phrase candidate information shown in FIG. Note that the buzzword phrase candidate information shown in FIG. 25 is individual for user A, so the buzzword information 96 shown in FIG. 26 is also individual for user A.
As shown in this figure, the buzzword information 96 has a structure in which items such as ID, notation, part of speech, and reading are associated with each word selected from the buzzword candidate information. In the present embodiment, in the process of creating the buzzword-related dictionary so far, an ID is assigned to each registered word at the stage of creating the buzzword information.

  In the above case, the buzzword is determined by comparing the difference appearance number (β−α) obtained from the comparison appearance number (α) and the latest appearance number (β) with a threshold value. It is just an example. For example, the buzzword may be determined by comparing the ratio of the comparative appearance count (α) and the latest appearance count (β) with a threshold value. Furthermore, it is also conceivable to determine a buzzword by performing a calculation using a more complicated calculation formula after using the comparison appearance count (α) and the latest appearance count (β). At this time, it is also possible to employ one or more past ones as the number of comparison appearances (α). That is, the buzzword may be determined by a predetermined algorithm based on the appearance frequency of the phrase.

Here, the basic dictionary 95 is assumed to have the same registration contents as the basic dictionary 51 stored in the mobile phone device 1, for example. For example, when the basic dictionary 51 on the mobile phone device 1 side is upgraded due to version upgrade or the like, the version upgrade is performed so that the basic dictionary 95 matches the basic dictionary 51 and has the same contents. Is done.
Of the IDs that should be assigned to each phrase in the popular phrase information 96, the same IDs that are already registered in the basic dictionary 95 may be assigned to the phrases that are already registered in the basic dictionary 95.
On the other hand, the words registered in the buzzword information 96 may include coined words, new words, etc. that have never existed so far, as can be understood from the history of the formation. Such a phrase is not included in the basic dictionary 95, for example. Therefore, there is a possibility that a phrase included in the basic dictionary 95 and a phrase (new trending phrase) not included in the phrase registered in the trending phrase information 96 are mixed.
For such new buzzwords, for example, IDs other than those attached to the words registered in the basic dictionary 95 may be set and attached according to a predetermined rule.

  Next, the dictionary creation processing unit 85 creates a buzzword-related dictionary using the buzzword information 96 created and stored as described above. That is, the buzzword priority dictionary table 97 and the buzzword corresponding priority phrase table 98 are created.

  FIGS. 27 and 28 show an individual buzzword priority dictionary table 97 and a buzzword corresponding priority phrase table 98 for user A created using the individual buzzword information 96 for user A shown in FIG. FIG. The buzzword priority dictionary table 97 and the buzzword corresponding priority word / phrase table 98 have a structure in which the ID, notation, part of speech, and reading are associated with each registered word / phrase similarly to the buzzword / phrase information 96.

First, the buzzword priority dictionary table 97 is created by registering words that are not registered in the basic dictionary 95 among the words registered in the buzzword information 96.
On the other hand, the buzzword corresponding priority phrase table 98 is created by registering words registered in the basic dictionary 95.

Here, in the trendy phrase information 96 shown in FIG. 26, the phrases registered in the basic dictionary 95 have two notations “New Year” and “New Year”, and the phrases not registered have the notation “ It is assumed that they are “no use”, “SMOP”, and “chemtaku”.
In this case, the dictionary creation processing unit 85 creates and registers the buzzword / phrase priority dictionary table 97 by registering three words / phrases “not used”, “chemtak”, and “SMOP” as shown in FIG. To do. Also, the buzzword corresponding priority word / phrase table 98 is created by registering words / phrases whose expressions are “New Year” and “New Year”.

  The buzzword / phrase priority dictionary table 97 and the buzzword / phrase priority phrase table 98 for the user A created as described above are stored in the storage unit 76. Note that if the individual popular buzzword priority dictionary table 97 and the buzzword corresponding priority word / phrase table 98 have already been stored in the storage unit 76 for the user A before this, the individual trend for the user A newly created this time is already stored. Rewriting is performed by the phrase priority dictionary table 97 and the buzzword corresponding priority phrase table 98. In other words, the update may be performed.

  And the integrated management part 81 of the dictionary server 2 is a user with respect to the mobile telephone apparatus 1 of the user A with the appropriate timing based on the buzzword phrase dictionary update frequency shown in the user unit contract information corresponding to the user A, for example. An individual buzzword-related dictionary (a buzzword / phrase priority dictionary table 97, a buzzword / phrase-corresponding priority phrase table 98) is transmitted to A. Note that the buzzword priority dictionary table 97 and the buzzword corresponding priority phrase table 98 may be transmitted at different timings.

In the mobile phone device 1 of the user A, the individual buzzword priority dictionary table 97 and the buzzword corresponding priority phrase table 98 are received and downloaded to the user A thus transmitted.
That is, when the mobile phone device 1 receives the buzzword priority dictionary table 97, it stores the received data of the buzzword priority dictionary table 97 as the buzzword priority dictionary table 52C in the additional dictionary table group 52 of FIG. . At this time, if the buzzword / phrase priority dictionary table 52C has already been stored, it is rewritten with the data of the buzzword / phrase priority dictionary table 97 received this time and stored again as the buzzword / phrase priority dictionary table 52C. Is done.
In addition, when the cellular phone device 1 receives the buzzword corresponding priority word / phrase table 98, it additionally registers the words / phrases in the received buzzword / phrase corresponding priority word / phrase table 98 in the priority word / phrase table 63 of FIG.

  In this way, by downloading the individual buzzword priority dictionary table 97 to the user A to the mobile phone device 1 of the user A, the buzzword corresponding priority phrase table 52C is converted in conformity with the user A. It is possible to obtain a registered content that is likely to be selected as a phrase. In addition, the contents of the buzzword corresponding priority word table 98 are reflected in the priority word table 63 as described above, so that a conversion candidate word table in which the priority order of the buzzword is set high is created. . That is, in the conversion candidate word / phrase list area AR2 shown in FIG. 6, a list of words / phrases can be presented in such a manner that the trending word / phrase has a higher priority. At this time, the priority order of the buzzwords obtained in the conversion candidate word list area AR2 also matches the user A and is likely to be selected as a conversion word.

  It should be noted that the buzzword-related dictionaries (buzzword priority dictionary table 97 and buzzword-preferred phrase table 98) that are individual for each other user are also created in the process of creating individual buzzword-related dictionaries for user A so far. According to this, it is created and stored in the storage unit 76. Then, each buzzword-related dictionary (the buzzword priority dictionary table 97 and the buzzword-preferred priority phrase table 98) can be transmitted to the mobile phone device 1 of the corresponding user and downloaded.

  As described above, in the present embodiment, the dictionary server 2 creates individual buzzword-related dictionaries (buzzword phrase priority dictionary table 97 and buzzword phrase-corresponding priority phrase table 98) for each contracted user. The user's mobile phone device 1 is made to download. As a result, it is possible not only to provide a dictionary storing buzzwords but also to provide a dictionary with contents that are individualized according to user preferences and the like. By using such a dictionary, the user can perform more efficient kana-kanji conversion operations than before.

In the present embodiment, a buzzword-related dictionary for all users (a buzzword priority dictionary table for all users, a buzzword corresponding priority phrase table for all users) is also created for all users. Shall.
The creation of the buzzword / phrase-related dictionary for the entire user is performed in the following manner, for example, in parallel with the series of processes described so far for the creation of the buzzword / phrase-related dictionary for each user.

  For example, in order to create (update) a buzzword-related dictionary for a certain user, the morpheme analysis processing unit 83 executes morpheme analysis processing on character string information collected at that time to obtain morpheme information 92. Next, the statistical information processing unit 84 creates the latest statistical information 93 by using the morpheme information 92, and at this time, the information that has been regarded as the latest statistical information 93 may be changed to the comparative statistical information 94. Done. This point is as described above.

When creating the buzzword / phrase-related dictionary for all users, first, the latest statistical information 93 for each user is reflected in accordance with the newest statistical information 93 for each user being newly created and held as described above. The latest statistical information 93 of all the users who have been made is created. For example, if there is a word that is registered in the latest statistical information 93 for each user but is not registered in the latest statistical information 93 for all users, this word is newly registered in the latest statistical information 93 for all users. . In addition, the word / phrase already registered is updated by adding the value indicated by the number of appearances indicated in the latest statistical information 93 for each user to the number of appearances.
In this way, the latest statistical information 93 for all users is changed every time the latest statistical information 93 for each user is created and updated, regardless of the user. In addition, the latest statistical information 93 for the entire user is changed to the comparative statistical information 94 for the entire user for each predetermined period, and the latest statistical information 93 for the entire user is newly created from the beginning at this timing. It will be a thing.

  In addition, the dictionary creation processing unit 85, for example, at a predetermined time interval (for example, a timing for changing the above-described latest statistical information 93 for the entire user to the comparative statistical information 94 for the entire user) Using the latest statistical information 93 and the comparative statistical information 94 in general, the buzzword information 96 of the user as a whole is created in accordance with the explanations corresponding to FIG. 24, FIG. 25, and FIG. Then, using the buzzword information 96 of the general user, the buzzword priority dictionary table 97 and the buzzword-corresponding priority phrase table 98 of the general user are stored in the same manner as described above with reference to FIGS. Create and remember.

  As a result, the buzzword / phrase priority dictionary table 97 and the buzzword corresponding priority phrase / table 98 of the entire user created in this way result in the user-specific buzzword / phrase priority dictionary table 97 and the buzz / phrase for almost all contracted users. The contents of the word / phrase priority word / phrase table 98 are integrated, and the word / phrase priority word / phrase table 98 corresponds to a popular word / phrase generally used at that time.

  Note that, for example, in the user unit contract information of a certain user, it is considered that there is a possibility that Web content inappropriate for use in creating a buzzword related dictionary for all users is specified. Even if an individual buzzword related dictionary is created and updated for a certain user in accordance with such Web contents, filtering is performed on the dictionary server 2 side to create and update a buzzword related dictionary for all users. May not be used. That is, the buzzword / phrase-related dictionary for the entire user may be created using at least character string information obtained based on the user unit contract information.

The buzzword / phrase-related dictionary for all users obtained in this way is not specifically shown in FIG. 21, for example, but according to the update frequency specified by the user unit contract information, the cellular phone device 1 of each contract user. It is possible to provide a service by sending it to and downloading it.
In addition, for example, on the assumption that contract users are provided with a service of downloading a buzzword-related dictionary for each user, non-contracted users can use buzzword-related dictionaries for all users for free or for a fee. It may be possible to provide a differentiated service between a contract user and a non-contract user, such as enabling download.

  In addition, as for both user-specific and general user-related buzzword-related dictionaries, in the present embodiment, buzzword / phrase information 96 illustrated in FIG. Absent. In this case, the buzzword / phrase priority dictionary table 97 includes not only words / phrases not included in the basic dictionary 95 but also words / phrases included in the basic dictionary 95, so the buzzword / phrase priority word / phrase table 98 is necessary. Disappears.

  Next, an example of a processing procedure corresponding to creation and download of a buzzword-related dictionary executed by the dictionary server 2 will be described with reference to flowcharts of FIGS. Note that the processing shown in these drawings can be considered to be realized by the CPU as the arithmetic processing unit 71 executing a program stored in the storage unit 76 or the ROM 73, for example.

First, FIG. 29 shows a procedure for creating and updating the latest statistical information 93 using electronic mail.
For example, in the dictionary download network system shown in FIG. 17, the gateway server 3 receives an email sent from the mobile phone device 1 or an email sent to the mobile phone device 1 via the network. When acquired, the received and acquired e-mail is managed by the mail server unit 114 as described above. At the same time, the received and acquired electronic mail is transmitted to the dictionary server 2 via the dictionary server compatible communication unit 115. The e-mail transmitted from the gateway server 3 to the dictionary server 2 in this manner is referred to as a notification e-mail here.

  The integrated management unit 81 of the dictionary server 2 waits for the notification e-mail transmitted from the gateway server 3 to be received in step S201 of FIG. Then, when the notification e-mail is received, the procedure proceeds to step S202 and subsequent steps.

In step S202, for example, the integrated management unit 81 of the dictionary server 2 identifies a contract user corresponding to the notification email received this time.
For this purpose, for example, if the current notification e-mail is transmitted from the mobile phone device 1, it is described in the e-mail address of the transmission source and in the header of the user unit contract information in the contract information 99 By collating with the e-mail address, the user unit contract information corresponding to this notification e-mail is specified. In this way, the user is specified here by specifying the user unit contract information corresponding to the notification e-mail.
If the current notification e-mail is sent to the mobile phone device 1, the e-mail address of the destination and the e-mail address described in the header of the contract information for each user in the contract information 99 Are identified, the user unit contract information corresponding to the current notification e-mail is specified.

In step S203, referring to the provided mail list in the user unit contract information of the user specified in step S202, it is first determined whether or not there is a designated provided mail list. If a negative determination result is obtained, the process is terminated. If a positive determination result is obtained, the process proceeds to step S203.
In step S204, it is determined whether or not the notification e-mail received and acquired this time matches the e-mail specified by the e-mail address indicated by the contract content in the contract item of the provided mail list.
If a negative determination result is obtained here, the process is exited. If a positive determination result is obtained, the process proceeds to step S205.

  In step S205, for example, the morphological analysis processing unit 83 refers to the contract details of the mail provision permission item corresponding to the notification email received this time, and recognizes what the mail provision permission item is. Next, in step S205, the character string information in the items permitted to be provided by the information of the mail provision permission item recognized in step S206 is selected from the items (title, text, etc.) of the notification email received this time. Extract. The morpheme analysis processing unit 83 creates morpheme information 92 by executing morpheme analysis processing using this character string information, and stores it in the RAM 72, for example.

  Step S207 is processing of the statistical information processing unit 84, for example. The statistical information processing unit 84 creates and updates the latest statistical information 93 using the morpheme information 92 created in step S206. Here, as the latest statistical information 93 to be newly created or updated, first, the latest statistical information 93 for the user specified in step S202 is used. Further, the latest statistical information 93 for all users is created and updated so as to reflect the contents of the latest statistical information 93 for each user.

FIG. 30 shows an example of a processing procedure for creating and updating the latest statistical information 93 according to Web access executed by the mobile phone device 1.
For example, when Web access is performed by the mobile phone device 1, data transmission / reception is performed as access to a specified URL via the gateway server 3. Therefore, the gateway server 3 can identify the mobile phone device 1 that performs communication for Web access. Therefore, in the gateway server 3, if communication for Web access is started by a certain mobile phone device 1, the mobile phone device together with information for specifying the mobile phone device 1 (for example, mobile phone ID, user ID, etc.) 1 sends a notification to the dictionary server 2 that Web access has started (Web access notification).

  In step S301, the integrated management unit 81 of the dictionary server 2 waits for reception of the Web access notification transmitted from the gateway server 3 as described above, and if Web access is received, Proceed to step S302 and subsequent steps.

In step S302, the user unit contract information corresponding to the mobile phone device 1 that has performed web access in accordance with the web access notification received in step S301 is specified.
As described above, the Web access notification is transmitted together with an identifier such as a mobile phone ID. Therefore, if the mobile phone ID received together with this Web access notification and the mobile phone ID described in the header of the user unit contract information in the contract information 99 are collated and searched for a match, the user unit contract information is obtained. Is identified.

  In step S303, the contract content set for the contract item “permission of disclosure of accessed web content” in the user unit contract information specified in step S302 is referred to, and the web access corresponding to the current web access notification is performed. For the mobile phone device 1 to be accessed, it is determined whether or not disclosure of the accessed Web content is permitted. If a negative determination result is obtained here, the processing shown in this figure is terminated. If a positive determination result is obtained, the process proceeds to step S304.

In step S304, the Web content data that the mobile phone device 1 is accessing is fetched in response to the current Web access notification. For this purpose, for example, the integrated management unit 81 of the dictionary server 2 requests the gateway server 3 for Web content data corresponding to the current Web access notification. In response to this request, the gateway server 3 also transmits to the dictionary server 2 the Web content data transmitted to the mobile phone device 1 that is currently accessing the Web.
The integrated management unit 81 receives and takes in the Web content data transmitted in this way.

  In step S305, the morpheme analysis processing unit 83 executes morpheme analysis processing using character string information in the Web content data captured in step S304 to create morpheme information 92, which is stored in the RAM 72, for example.

  In step S306, for example, the statistical information processing unit 84 creates and updates the latest statistical information 93 using the morpheme information 92 created in step S305. Here, as the latest statistical information 93 to be newly created or updated, one is the latest statistical information 93 individually created for the user corresponding to the user unit contract information specified in step S302 and this time (updated). It becomes the latest statistical information 93 of all users created and updated so that the contents of the latest statistical information 93 for each user are reflected.

FIG. 31 shows an example of a processing procedure for creating and updating the latest statistical information 93 in response to the input of a search keyword (search character string) on the website executed by the mobile phone device 1.
In the gateway server 3, it is possible to take in and analyze Web content data accessed by the mobile phone device 1. Such an analysis can be performed by interpreting a document of these formats based on, for example, that the Web content is described in the HTML format or the XML format.
Through such Web content analysis, the gateway server 3 can recognize the presence of the search character string input box in the Web content page. Further, based on this, when the gateway server 3 inputs a search character string to the search character string input box in response to an operation from the mobile phone device 1 and executes a search on the Web server side, Can obtain information of this search character string.
In this way, when the mobile phone device 1 accesses the Web content and inputs the search character string to instruct the search, the gateway server 3 performs, for example, the mobile phone device 1 that instructed the search. A search character string input notification is transmitted to the dictionary server 2 together with the telephone ID.

  In step S402, the user unit contract information associated with the mobile phone device 1 corresponding to the search character string input notification received in step S401 of this time, for example, the mobile phone ID received together with the search character string input notification, etc. Use to identify.

  In step S403, referring to the contract contents set for the “search keyword provision permission” contract item in the user unit contract information specified in step S402, either permission or prohibition for search keyword provision is set. To determine if they are. If a negative determination result is obtained here, the processing shown in this figure is exited. If a positive determination result is obtained, the process proceeds to step S404.

  In step S404, for example, the integrated management unit 81 requests the gateway server 3 to receive and acquire character string information as a search character string corresponding to the current search character string input notification, and then the morphological analysis processing unit 83 Thus, the morpheme analysis process for the acquired character string information is executed to create the morpheme information 92, which is stored in the RAM 72, for example.

  In step S405, for example, the statistical information processing unit 84 creates and updates the latest statistical information 93 using the morpheme information 92 created in step S404. Here, as the latest statistical information 93 to be newly created or updated, the latest statistical information 93 individually for the user corresponding to the user unit contract information specified in step S402 and the latest individual user information that has been created (updated) this time. It becomes the latest statistical information 93 for all users created and updated so that the contents of the statistical information 93 are reflected.

FIG. 32 shows an example of a processing procedure for creating and updating the latest statistical information 93 according to the crawling process.
Here, the dictionary server 2 sets a time (time) at which the crawling process is to be started for each contract user, and holds the information associated with the user ID as crawling process start time information. To do. The crawling process start time can be set based on, for example, a buzzword dictionary update frequency, a priority word dictionary update frequency, etc. in the user unit contract information.
And the integrated management part 81 of the dictionary server 2 discriminate | determines whether there exists the contract user who reached the crawling process start time based on the crawling process start time information in step S501, for example for every predetermined time interval. To do. Then, if a positive determination result is obtained that there is a contract user who has reached the crawling processing start time, the process proceeds to step S502 and subsequent steps.

In step S502, for example, the crawling unit 82 reads the crawling content list 100 associated with the user ID of the contract user who has reached the crawling process start time in step S501.
In step S503, the contents of the read crawling content list 100 are referred to, and it is determined whether or not Web content to be crawled is registered here. Here, for example, if no web content list is registered in the crawling content list, a negative determination result is obtained, and in this case, the processing shown in this figure is exited. On the other hand, if the Web content is registered in the read crawling content list, the process proceeds to step S504.

  In step S504, the crawling unit 82 executes crawling processing for Web content registered in the crawling content list. That is, the crawling unit 82 accesses the URL of the registered Web content indicated in the crawling content list, and acquires and acquires the Web content data. Further, the crawling unit 82 creates and holds crawling data 91 from the acquired Web content data. In step S504, such processing is sequentially repeated for each registered Web content indicated in the crawling content list.

  The crawling data obtained as described above is assumed to have at least character string information as data forming a Web content, for example. In step S506, the morpheme analysis processing unit 83 creates morpheme information 92 using the character string information in the crawling data 91 obtained in step S505, and stores it in the RAM 72, for example.

  In step S506, the statistical information processing unit 84 creates and updates the latest statistical information 93 using the morpheme information 92 created in step S505. Here, as the latest statistical information 93 to be newly created or updated, the latest statistical information 93 individually created for the contract user identified as the second highest crawling processing start time in step S501, and the individual individual created (updated) this time The latest statistical information 93 for all users created and updated so that the contents of the latest statistical information 93 are reflected.

  The processing shown in FIG. 29 to FIG. 32 is executed by the dictionary server 2. First, the latest statistical information 93 for each user includes transmission / reception of e-mail designated by user unit contract information, access to Web content, Depending on the search by inputting a search character string on the Web site on the Web browser of the mobile phone device 1, the crawling of the Web content specified as the crawling content specified by the user unit contract information, The update after the new creation will be performed. Further, the latest statistical information 93 for all users is newly created and updated in accordance with the latest creation and updating of the individual latest statistical information 93 for each user as described above.

Further, as described above, the latest statistical information 93 for each user and all users created and stored in the storage unit 76 as described above is changed to comparative statistical information 94 at a certain timing. . The flowchart of FIG. 33 shows an example of a processing procedure for this purpose.
Here, a case where the latest statistical information 93 for each user is changed to the comparative statistical information 94 is taken as an example.
In FIG. 33, first, in step S601, for example, the statistical information processing unit 84 waits for the timing to change the latest individual statistical information 93 corresponding to one contracted user to the comparative statistical information 94. Yes. For example, the timing for changing the latest statistical information 93 for each user to the comparative statistical information 94 can be set based on the buzzword dictionary update frequency of the user unit contract information corresponding to the contract user.
If it is determined that the timing for changing the latest statistical information 93 to the comparative statistical information 94 has been reached, the process proceeds to step S602.

In step S602, data stored in the storage unit 76 as individual user latest statistical information 93 corresponding to the contract user is managed in the storage unit 76 as corresponding to the same user. The data of the comparison statistical information 94 for each user is updated. As a result, the latest statistical information 93 so far is changed to the comparative statistical information 94.
Note that, for example, in this processing, the description by illustration is omitted, but the latest statistical information 93 that is the change source is deleted or the contents thereof are initialized. Thereafter, the processes shown in FIGS. 29 to 32 are appropriately executed for this user, so that new latest statistical information 93 is created, stored, and updated thereafter.

  The procedure shown in FIG. 33 is also applied to the case where the latest statistical information 93 for all users is changed to the comparative statistical information 94. At this time, the timing at which the latest statistical information 93 should be changed to the comparative statistical information 94 can be set based on the period of updating the buzzword-related dictionary for the user as a whole.

  34, in the dictionary server 2, the dictionary creation processing unit 85 creates buzzword-related dictionaries (buzzword phrase priority dictionary table 97, buzzword phrase priority phrase table 98) using the comparative statistical information 94 and the latest statistical information 93. An example of a processing procedure for doing this is shown. Here, a case where an individual buzzword-related dictionary is created for a single user is taken as an example. However, when a buzzword-related dictionary for all users is created, the processing conforms to the following description. .

For example, the integrated management unit 81 of the dictionary server 2 determines the update timing (period, time) of the buzzword related dictionary for each contract user based on the buzzword dictionary update frequency of the user unit contract information corresponding to the contract user. It is assumed that the set update timing information is held and managed.
In step S701, for example, the integrated management unit 81 determines whether or not there is a buzzword related dictionary for each user that has reached the update timing, for example, at regular intervals based on the update timing information about the buzzword related dictionary. It shall be determined. If it is determined that there is a buzzword-related dictionary for each user that has reached the update timing in step S701 and a positive determination result is obtained, the process proceeds to step S702 and subsequent steps.

The procedure after step S702 is executed by the dictionary creation processing unit 85.
First, in step S702, by using the individual statistical comparison information 94 and the latest statistical information 93 that have been updated in step S701, as described with reference to FIGS. The buzzword phrase information 96 is created and stored in the storage unit 76. At this time, if the buzzword information 96 previously stored individually for the same user is already stored in the storage unit 76, the buzzword information stored in the storage unit 76 by the buzzword information 96 created this time. 96 updates will be performed.

  Next, the dictionary creation processing unit 85 responds by using the buzzword information 96 created in step S702 as described above with reference to FIGS. 26, 27, and 28 in steps S703 and S704. A buzzword / phrase priority dictionary table 97 and a buzzword / phrase correspondence priority phrase table 98 for each user are created, updated, and stored in the storage unit 76.

Here, when the mobile phone device 1 downloads the buzzword related dictionary (buzzword priority dictionary table 97, buzzword corresponding priority phrase table 98) from the dictionary server 2, the operation method of charging the download is described as follows. Can think.
In this case, the charge for downloading the buzzword-related dictionary may be uniform among users, but it may be possible to set different charges between users as follows.
As can be understood from the above description, the buzzword-related dictionary includes the crawling content specified by the user based on the user unit contract information, the title and text of the e-mail permitted to be provided, the data of the accessed web content, the web content Created based on the search keywords entered above. This can also mean that a buzzword-related dictionary can be created by providing information of the contract user. That is, it is possible to view that a buzzword-related dictionary download service has become possible with the cooperation of providing information by contract users.
Therefore, as the present embodiment, on the basis of this, the download fee for the buzzword-related dictionary is set so that the contracted user with a substantial amount of information provision is reduced so that the fee is reduced. To do.

  FIG. 35 shows an example of a processing procedure for downloading a buzzword-related dictionary, which is to be executed in the dictionary server 2, and for a billing process in accordance therewith. In the billing process here, the fee setting based on the above-described concept is reflected.

First, in step S801, the integrated management unit 81 in the dictionary server 2 waits for an instruction to download an individual buzzword-related dictionary to a mobile phone device 1 corresponding to the user. Yes.
For example, in another processing routine in the integrated management unit 81, a time (time) for downloading and transmitting a buzzword-related dictionary, for example, corresponding to each user is set based on the buzzword-related dictionary update frequency information in the user unit contract information. Thus, it is determined whether or not there is a user who has reached the time to download and transmit the buzzword-related dictionary on a regular basis. Then, in response to the determination result that there is a user who has reached the time to download and transmit the buzzword related dictionary, in step S801, a download instruction for the buzzword related dictionary is obtained. The process proceeds from S802 onward.

The accounting processing unit 87 executes steps S802 to S806.
In step S802, the user unit contract information corresponding to the user of the mobile phone device 1 who should download and transmit the buzzword related dictionary is read.
Next, in step S803, the contract content of the contract item related to the information provision is recognized in the user unit contract information read in step S802, and the reduction value α is determined according to the recognized result. The reduction value α1 is set to be higher as there are more elements for which information is provided (or permitted) in the user unit contract information.
For example, in the case of correspondence with FIG. 21, the contract items related to information provision are provided mail list, mail provision permission items (1) to (n), search keyword provision permission, disclosure permission of accessed Web content, crawling Specified.
First, the greater the number of e-mail addresses indicated in the provided mail list, the greater the number of elements for which information is provided. Accordingly, first, points (points) as the reduction value α1 are added according to the number of e-mail addresses indicated in the provided mail list.
In addition, in the mail provision permission items (1) to (n), it is possible to set permission / prohibition for the title and body items. To be added. In addition, the amount of character string information can be larger in the text in the title and text. In view of this, it is conceivable to set a higher point when allowing the body text than when allowing the title.
Also, regarding permission to provide search keywords and permission to disclose the accessed Web content, if permission is set for these items, points corresponding to the respective items are added.
Further, in crawling designation, points corresponding to the number of designated web contents are added.
Note that the contract items related to the provision of information are considered to have different levels of importance for the creation of the preferred word / phrase-related dictionary of the present embodiment, and therefore differ for each contract item depending on the level of importance. An addition point can also be set.
In this way, the reduction value α1 is set based on the breakdown of contract items related to information provision in the user unit contract information.

  By the way, for example, even if a large number of items and elements for providing information in the user unit contract information are set, for example, e-mail transmission / reception is not performed so much or Web access does not seem to be performed much. Accordingly, the amount of character string information that can be collected for the creation of the priority phrase related dictionary is reduced. On the other hand, if the user frequently performs e-mail and Web access, a large amount of character string information can be collected even if there are few items and elements for which information is provided. That is, it is also possible to adopt a concept of increasing the reduction value as the amount of character string information substantially collected for creating the priority word / phrase-related dictionary increases.

  Therefore, the integrated management unit 81 in this case integrates the amount of character string information collected within a certain period in order to create and update the priority word / phrase-related dictionary in correspondence with each user. It is made to hold | maintain as d. As described above with reference to FIGS. 29 to 32, the priority phrase related dictionary can be updated at any time according to e-mail transmission / reception, Web access, input of a search phrase on a Web content, periodic crawling processing, and the like. . Therefore, whenever the update is performed, if the amount of character string information used at that time is accumulated, this is substantially collected for use in the creation of a priority word-related dictionary within a certain period of time. This is the amount of character string information. This is the actual information provision amount d.

  In step S804, the value of the substantial information provision amount d corresponding to the user of the mobile phone device 1 who should download and transmit the buzzword related dictionary obtained at that time is recognized. In step S805, a reduction value α2 is determined based on the recognized real information provision amount d. This determination can be made by a predetermined calculation using the substantial information provision amount d, for example. Alternatively, it can be obtained by referring to this table information as having table information associating the substantial information provision amount d with the reduction value α2.

Next, in step S806, for example, using the basic amount β when downloading the buzzword-related dictionary and the reduction values α1, α2 determined in steps S803 and S805,
For example, for download fee γ,
γ = β- (α1 + α2)
It calculates with the formula represented by these.

  In step S807, processing for downloading and transmitting individual buzzword-related dictionaries to the corresponding user is executed. That is, the integrated processing unit 81 reads out individual buzzword-related dictionaries (buzzword / phrase priority dictionary table 97, buzzword / phrase-corresponding priority phrase table 98) to the corresponding user and passes them to the communication control unit 86. The communication control unit 86 transmits the data of the buzzword-related dictionary passed to the mobile phone device 1 of the corresponding user as the transmission destination.

  For example, when the download transmission is completed, the charging processing unit 87 charges the user with the download fee γ calculated in step S806 in the next step S808.

  In the above case, when obtaining the download fee γ, only one of the reduction values α1 and α2 may be used. Alternatively, a discount rate may be set instead of each of the reduction values α1 and α2, and the download fee γ may be obtained based on the discount rate. That is, a reduction element based on the breakdown of contract items related to information provision in the user unit contract information, and a reduction element corresponding to the amount of character string information substantially collected for use in creating a preferred term related dictionary It is sufficient that at least one of the above is used, and there are several ways of setting the two reduction elements.

Here, the correspondence between main words and phrases in the present invention and this embodiment is as follows, for example.
The “auxiliary dictionary information” in the present invention includes the “time priority word / phrase table 52A”, “position priority dictionary table 52B”, “buzz word / phrase priority dictionary table 52C”, “buzz / phrase priority dictionary table 97”, and “buzz word / phrase priority word / phrase”. “Table 98”, “Buzzword information 96”, and the like correspond. Further, as can be understood from the above description, the “storage” of “auxiliary dictionary information” is not only stored in the auxiliary storage device such as the storage unit 18 or the storage unit 76 but also stored in the RAMs 19 and 72. The case of temporarily storing and holding in the corresponding main storage area is also included.
The “high frequency word / phrase dictionary information” in the present invention includes the “buzz word / phrase priority dictionary table 52C”, “buzz word / phrase priority dictionary table 97”, “buzz word / phrase priority word / phrase table 98” and “buzz word / phrase information 96” in the present embodiment. Correspond.
In addition, the “first high-frequency phrase dictionary information” and “second high-frequency phrase dictionary information” in the present invention are the “popular phrase priority dictionary table 97” and “popular phrase” created by the dictionary server 2 in the present embodiment, respectively. This corresponds to the phrase corresponding priority phrase table 98 ".

In addition, as this Embodiment, it is not limited to the structure as embodiment described so far.
For example, in the above description, among the additional dictionaries, the time priority dictionary table 52A and the position priority dictionary table 52B are preset information. For example, these can be acquired by downloading or the like. , You can think of operations. In particular, with regard to the position priority dictionary table 52B, an operation in which a dictionary table in which a place name, a person name, etc. peculiar to each sightseeing spot are registered and provided from a server or the like is also conceivable.
Further, according to FIGS. 9 and 10, the time priority dictionary table 52A and the position priority dictionary table 52B are associated with time and position information for only those already registered in the basic dictionary 51. It is also conceivable to individually register words / phrases that are table information but are not registered in the basic dictionary 51 together with their readings / notations / parts of speech. For example, the position priority dictionary table for each sightseeing spot described above is considered to register many words / phrases that are not registered in the basic dictionary 51.
In the embodiment, three additional dictionaries are a time priority dictionary table, a position priority dictionary table, and a buzzword phrase priority dictionary table. However, the present invention is not limited to this. In other words, in addition to time, position, and trendy phrases, a priority dictionary table in which words that should be prioritized according to a specific situation or environment may be registered as additional dictionaries. Even so, it would be beneficial for the user not only to improve the operability of character input, but also to enhance the entertainment, so that a variety of such additional dictionaries are prepared and made available to the user. is there.
In addition, up to now, as an additional dictionary, words that should be set higher in priority according to a predetermined condition are registered, but on the contrary, the priority is set lower according to a predetermined condition. The contents may be such as to register the words to be registered.

  Further, in creating a buzzword-related dictionary by the dictionary server 2, in the embodiment, the difference value of the number of occurrences of the phrase in two pieces of statistical information (latest statistical information 93, comparative statistical information 94) having different creation times A buzzword is selected and extracted by comparison with a threshold value. However, this is only an example, and as long as character string information that can be acquired when using Web content (including e-mail in this application) by a mobile phone device is used, buzzwords are extracted by other methods. It does not matter if it is used. For example, a word extraction algorithm such as tf-ifd or other statistical methods can be employed.

In the embodiment, the character input system based on the present invention is applied to a mobile phone device. Of course, the effect of the character input system based on the present invention can be effectively exhibited by applying it to a mobile phone device in which the number of operation elements is limited. Any device or system can be generally applied. Such a system includes a character input support application installed in a personal computer or the like.
In the embodiment, a character input system for Japanese is given as an example. For example, a predictive conversion function is useful by using a mixture of phonograms and ideograms. If there are other languages, the present invention can cope with such languages.

  Further, a program (character input support application) for realizing the character input system according to the present invention is stored in advance in the apparatus at the time of manufacture, for example. Alternatively, it is conceivable to store the program in a removable storage medium and install the program from the storage medium to the hardware as a character input device corresponding to the present application. Alternatively, it is also conceivable that the program is stored in a server on the network and downloaded via the network by a character input device corresponding to the present application.

It is a block diagram which shows the structural example of the mobile telephone apparatus as embodiment of this invention. It is a block diagram which shows typically the functional structural example of the character input assistance application which is a program corresponding to the character input system of this Embodiment. It is a figure which shows the example of the content of the conversion candidate word / phrase table produced corresponding to the input of kana character "ko". It is a figure which shows the structure and content example of a priority word / phrase table. It is a figure which shows the content example (content change from FIG. 3) of the conversion candidate word table after a priority word table reference. FIG. 6 is a diagram showing a character input operation screen corresponding to the contents of the conversion candidate word / phrase table of FIG. 5 and a display mode example after a confirmation operation on the character input operation screen. It is a figure which shows the example of a content of the priority word / phrase table changed corresponding to the confirmation operation shown by FIG.6 (b) (content change from FIG. 4). It is a figure which shows the example of a display mode of the character input operation screen at the time of performing kana character input further from the state of the character input operation screen shown by Fig.6 (a). It is a figure which shows the structural example of a time priority dictionary table. It is a figure which shows the structural example of a position priority dictionary table. It is a figure which shows the structural example of a buzzword phrase priority dictionary table. It is a figure which shows the example of the content of an electronic mail, and the example of the content of the morpheme information obtained corresponding to this. It is a figure which shows the example of the content of the conversion candidate word / phrase table corresponding to FIG. It is a figure which shows the example of the content of a document file, and the example of the content of the morpheme information obtained corresponding to this. It is a figure which shows the example of the content of a priority item setting table. It is a flowchart which shows the example of an operation | movement procedure of the character input system as embodiment. It is a figure which shows the construction example of the dictionary download network system of this embodiment. It is a block diagram which shows the structural example of a dictionary server. It is a block diagram which shows the function and operation | movement example of a dictionary server based on software by a block configuration. It is a block diagram which shows the structural example of a gateway server. It is a figure which shows the example of the content of user unit contract information. It is a figure which shows the example of the content of a user separate crawling content list. It is a figure which shows the example of the content of the morpheme information produced in a dictionary server. It is a figure which shows the example of the content of the newest statistical information and comparison statistical information which are created in a dictionary server. It is a figure which shows the example of the content of the buzzword phrase candidate information produced in a dictionary server. It is a figure which shows the example of the content of the buzzword information produced in a dictionary server. It is a figure which shows the example of the content of the buzzword phrase priority dictionary table produced in a dictionary server. It is a figure which shows the example of the content of the buzzword corresponding priority phrase table created in a dictionary server. It is a flowchart which shows the example of a process sequence for creation and update of the newest statistical information using an email by a dictionary server. It is a flowchart which shows the example of a process sequence for creation and update of the newest statistical information according to Web access which a mobile telephone apparatus performs by a dictionary server. It is a flowchart which shows the example of a process sequence for creation and update of the newest statistical information according to the input of the search keyword (search character string) on the Web site which a mobile telephone apparatus performs by a dictionary server. It is a flowchart which shows the example of a process sequence for preparation and update of the newest statistical information according to a crawling process. It is a flowchart which shows the example of a process sequence for changing the newest statistical information into comparative statistical information. It is a flowchart which shows the example of a process sequence for producing a buzzword phrase related dictionary using the comparison statistical information and the newest statistical information by a dictionary server. It is a flowchart which shows the example of a process sequence for the download of a buzzword related dictionary assumed to be performed with a dictionary server, and the accounting process according to this.

Explanation of symbols

  1 cellular phone device, 2 dictionary server, 3 gateway server, 4 Web content group, 11 antenna, 12 receiving unit, 13 transmitting unit, 14 audio signal processing unit, 15 speaker, 16 microphone, 17 system control unit, 18 storage unit, 19 RAM, 20 Clock unit, 21 GPS unit, 22 Display unit, 23 Operation unit, 30 Character input support application, 31 Character conversion processing unit, 32 Dictionary search processing unit, 33 Reference sentence analysis processing unit 33, 34 Morphological analysis processing unit , 35 confirmation result output unit, 41 mailware, 42 document editor, 51 basic dictionary, 52 additional dictionary table group, 53 e-mail data, 54 document file, 61 priority item table, 62 conversion candidate phrase table, 63 priority phrase table, 64 morpheme information, 71 arithmetic processing unit 72 RAM, 73 ROM, 74 mobile phone compatible communication unit, 75 gateway compatible communication unit, 76 storage unit, 81 integrated management unit, 82 crawling unit, 83 morpheme analysis processing unit, 84 statistical information processing unit, 85 dictionary creation processing unit, 86 communication control unit, 87 accounting processing unit, 91 crawling data, 92 crawling content list, 93 latest statistical information, 94 comparative statistical information, 95 basic dictionary, 96 buzzword phrase information, 97 buzzword phrase priority dictionary table, 98 buzzword phrase correspondence priority Phrase table, 99 contract information, 100 crawling content list, 111 protocol conversion processing unit, 112 mobile phone compatible communication unit, 113 Internet compatible communication unit, 114 mail server unit, 115 dictionary server compatible communication unit

Claims (18)

Basic dictionary information storage means for storing basic dictionary information formed by registering basic phrases that are prepared as basic phrases;
Auxiliary dictionary information storage means for storing auxiliary dictionary information formed by registering words corresponding to a predetermined condition and / or auxiliary dictionary information formed by registering words matching a specific condition;
Conversion source character string information acquisition means for acquiring conversion source character string information indicating a conversion source character string composed of one or more predetermined types of input characters;
From the basic dictionary information, a basic phrase that is predicted to be a definitive character string corresponding to the conversion source character string indicated by the conversion source character string information is searched, and the searched basic word / phrase is registered as a conversion candidate word / phrase. Conversion candidate phrase information generating means for generating conversion candidate phrase information based on the above,
From the auxiliary dictionary information, auxiliary dictionary search means for searching for a phrase that is predicted to be a confirmed character string corresponding to the conversion source character string indicated by the conversion source character string information;
Conversion candidate word / phrase information changing means for changing the content of the conversion candidate word / phrase information so that a result obtained by setting and registering a predetermined priority order for the words / phrases searched by the auxiliary dictionary search means;
Conversion candidate phrase presenting means for presenting the conversion candidate phrases registered in the conversion candidate phrase information in a predetermined manner so that the priority order is reflected;
A character input device comprising: Auxiliary dictionary information formed by registering words that match the specific condition is high-frequency word dictionary information that is formed by registering words that meet the condition that the frequency of use is high at that time. so,
Download means for receiving and acquiring the high-frequency phrase dictionary information created by the server device and transmitted from the server device,
The character input device according to claim 1, further comprising: The auxiliary dictionary information formed by registering words corresponding to the predetermined condition is formed by registering words corresponding to the predetermined time,
The auxiliary dictionary search means includes:
Search for words and phrases registered in the auxiliary dictionary information in association with the predetermined time corresponding to the acquired current time,
The character input device according to claim 1 or 2. The auxiliary dictionary information formed by registering words corresponding to the predetermined condition is formed by registering words corresponding to the predetermined position,
The auxiliary dictionary search means includes:
Priorities are set so that the priority of words registered in the auxiliary dictionary information is higher than other words in association with a predetermined position that corresponds to the current position indicated by the acquired current position information. I do,
The character input device according to claim 1 or 3. An acquisition means for acquiring character string information as a reference sentence;
About the character string information as the reference sentence, a character string decomposing means for decomposing the unit into word units;
Auxiliary dictionary information for creating auxiliary dictionary information formed by registering words obtained by decomposing by the character string decomposing means as auxiliary dictionary information formed by registering words that match the predetermined condition And a creation means,
The character input device according to claim 1 or 4. Holding means for holding network data designation information that designates network data permitted to be used for creating high-frequency phrase dictionary information among network data that is data transmitted on the network;
Character string information collecting means for collecting character string information from network data designated by the network data designation information;
Based on the character string information collected by the character string information collecting means, dictionary information creating means for creating the high frequency word dictionary information in which words that are frequently used at that time are registered,
Transmitting means for transmitting the high-frequency phrase dictionary information to the character input device for downloading via the communication network;
A server device comprising: The dictionary information creating means
The character string information collected by the character string information collecting means is decomposed into word units, and the words / phrases to be registered in the high-frequency word / phrase dictionary information are determined based on the appearance frequencies of the words / phrases obtained by the decomposition. And using this determined phrase to create high-frequency phrase dictionary information,
The server device according to claim 6. The dictionary information creating means
As the high-frequency phrase dictionary information, individual high-frequency phrase dictionary information in which words that are frequently used at that time are registered for individual users corresponding to one character input device is created. In creating the high-frequency phrase dictionary information, using character string information collected only from the network data designated by the network data designation information corresponding to the one character input device,
The transmission means transmits the individual high-frequency phrase dictionary information to the character input device of the user corresponding to the individual high-frequency phrase dictionary information.
The server device according to claim 6 or 7. The dictionary information creating means
While creating the first high frequency phrase dictionary information consisting of words that are not registered in the basic dictionary used in the character input device,
A second high frequency that consists of words registered in the basic dictionary used in the character input device and can be used to determine the priority order of the words when generating the conversion candidate word information in the character input device Create word dictionary information,
The server device according to claim 6 or 8. The dictionary information creating means
As the high-frequency phrase dictionary information, general high-frequency phrase dictionary information in which words that are frequently used at that time are registered for all users of the character input device is created. In creating information, character string information collected from network data specified by multiple network data specification information is used.
The server device according to claim 7 or 9. In response to transmitting the high-frequency phrase dictionary information to the character input device by the transmission means, setting a fee for charging for the user corresponding to the character input device of the transmission destination,
A charge setting unit configured to set the fee based on the amount of character string information that can be substantially collected by the character string information collecting unit according to the network data designation information corresponding to the character input device of the transmission destination;
The server device according to claim 6 or 10. In response to transmitting the high-frequency phrase dictionary information to the character input device by the transmission means, setting a fee for charging for the user corresponding to the character input device of the transmission destination,
Further comprising a fee setting means for setting the fee based on a breakdown of network data designated by the network data designation information corresponding to the destination character input device;
The server device according to claim 6 or 11. It consists of a character input device and a server device,
The character input device
Basic dictionary information storage means for storing basic dictionary information formed by registering basic phrases that are prepared as basic phrases;
Auxiliary dictionary information storage means for storing auxiliary dictionary information formed by registering words corresponding to a predetermined condition and / or auxiliary dictionary information formed by registering words matching a specific condition;
Conversion source character string information acquisition means for acquiring conversion source character string information indicating a conversion source character string composed of one or more predetermined types of input characters;
From the basic dictionary information, a basic phrase that is predicted to be a definitive character string corresponding to the conversion source character string indicated by the conversion source character string information is searched, and the searched basic word / phrase is registered as a conversion candidate word / phrase. Conversion candidate phrase information generating means for generating conversion candidate phrase information based on the above,
From the auxiliary dictionary information, auxiliary dictionary search means for searching for a phrase that is predicted to be a confirmed character string corresponding to the conversion source character string indicated by the conversion source character string information;
Conversion candidate word / phrase information changing means for changing the content of the conversion candidate word / phrase information so that a result obtained by setting and registering a predetermined priority order for the words / phrases searched by the auxiliary dictionary search means;
Conversion candidate phrase presenting means for presenting the conversion candidate phrases registered in the conversion candidate phrase information in a predetermined manner so that the priority order is reflected;
From the server device, as auxiliary dictionary information formed by registering words that match specific conditions, high-frequency word dictionary information in which words that are frequently used at that time are registered is downloaded via a communication network. Download means to be stored by the auxiliary dictionary information storage means,
With
The server device is
Holding means for holding network data designation information that designates network data permitted to be used for creating high-frequency phrase dictionary information among network data that is data transmitted on the network;
Character string information collecting means for collecting character string information from network data designated by the acquired network data designation information;
Dictionary information creating means for creating the high-frequency phrase dictionary information based on the character string information collected by the character string information collecting means;
Transmitting means for transmitting the high-frequency phrase dictionary information to the character input device for downloading via the communication network;
A dictionary download system comprising: A conversion source character string information acquisition procedure for acquiring conversion source character string information indicating a conversion source character string composed of one or more input characters of a predetermined type;
A basic word / phrase that is predicted to become a fixed character string corresponding to the conversion source character string indicated by the conversion source character string information from basic dictionary information formed by registering a basic word / phrase that is prepared as a basic word / phrase Conversion candidate phrase information generation procedure for generating conversion candidate phrase information based on the search and registering the searched basic phrases as conversion candidate phrases,
Conversion indicated by the conversion source character string information from auxiliary dictionary information formed by registering words corresponding to a predetermined condition and / or auxiliary dictionary information formed by registering words matching a specific condition An auxiliary dictionary search procedure for searching for a phrase that is predicted to be a confirmed character string corresponding to the original character string;
A conversion candidate word / phrase information changing procedure for changing the contents of the conversion candidate word / phrase information so as to be a result of setting and registering a predetermined priority for the word / phrase searched by the auxiliary dictionary search procedure;
A conversion candidate phrase presenting procedure for presenting the conversion candidate phrases registered in the conversion candidate phrase information in a predetermined manner so that the priority order is reflected;
To present conversion candidate phrases. Character string information is collected from the network data specified by the network data specification information that specifies the network data that is permitted to be used for creating high-frequency phrase dictionary information among the network data that is transmitted over the network. Collecting character string information
Based on the character string information collected by the character string information collection procedure, a dictionary information creation procedure for creating the high frequency phrase dictionary information in which words that are frequently used at that time are registered,
A transmission procedure for transmitting the high-frequency phrase dictionary information to the character input device for downloading via the communication network;
Is an information processing method that is executed on a server device. As an information processing method in a dictionary download system consisting of a character input device and a server device,
In the above character input device,
Auxiliary dictionary information storage procedure for storing auxiliary dictionary information formed by registering words corresponding to predetermined conditions and / or auxiliary dictionary information formed by registering words matching a specific condition;
A conversion source character string information acquisition procedure for acquiring conversion source character string information indicating a conversion source character string composed of one or more input characters of a predetermined type;
A basic word / phrase that is predicted to become a fixed character string corresponding to the conversion source character string indicated by the conversion source character string information from basic dictionary information formed by registering a basic word / phrase that is prepared as a basic word / phrase Conversion candidate phrase information generation procedure for generating conversion candidate phrase information based on the search and registering the searched basic phrases as conversion candidate phrases,
Conversion indicated by the conversion source character string information from auxiliary dictionary information formed by registering words corresponding to a predetermined condition and / or auxiliary dictionary information formed by registering words matching a specific condition An auxiliary dictionary search procedure for searching for a phrase that is predicted to be a confirmed character string corresponding to the original character string;
A conversion candidate word / phrase information changing procedure for changing the contents of the conversion candidate word / phrase information so as to be a result of setting and registering a predetermined priority for the word / phrase searched by the auxiliary dictionary search procedure;
A conversion candidate phrase presenting procedure for presenting the conversion candidate phrases registered in the conversion candidate phrase information in a predetermined manner so that the priority order is reflected;
From the server device, as auxiliary dictionary information formed by registering words that match specific conditions, high-frequency word dictionary information in which words that are frequently used at that time are registered is downloaded via a communication network. Download procedure to be stored by the auxiliary dictionary information storage means,
Run
In the server device,
Character string information is collected from the network data specified by the network data specification information that specifies the network data that is permitted to be used for creating high-frequency phrase dictionary information among the network data that is transmitted over the network. Collecting character string information
Dictionary information creation procedure for creating the high-frequency word dictionary information based on the character string information collected by the character string information collection procedure;
A transmission procedure for transmitting the high-frequency phrase dictionary information to the character input device for downloading via the communication network;
Information processing method to execute. A conversion source character string information acquisition procedure for acquiring conversion source character string information indicating a conversion source character string composed of one or more input characters of a predetermined type;
A basic word / phrase that is predicted to become a fixed character string corresponding to the conversion source character string indicated by the conversion source character string information from basic dictionary information formed by registering a basic word / phrase that is prepared as a basic word / phrase Conversion candidate phrase information generation procedure for generating conversion candidate phrase information based on the search and registering the searched basic phrases as conversion candidate phrases,
Conversion indicated by the conversion source character string information from auxiliary dictionary information formed by registering words corresponding to a predetermined condition and / or auxiliary dictionary information formed by registering words matching a specific condition An auxiliary dictionary search procedure for searching for a phrase that is predicted to be a confirmed character string corresponding to the original character string;
A conversion candidate word / phrase information changing procedure for changing the contents of the conversion candidate word / phrase information so as to be a result of setting and registering a predetermined priority for the word / phrase searched by the auxiliary dictionary search procedure;
A conversion candidate phrase presenting procedure for presenting the conversion candidate phrases registered in the conversion candidate phrase information in a predetermined manner so that the priority order is reflected;
That causes a character input device to execute. Character string information is collected from the network data specified by the network data specification information that specifies the network data that is permitted to be used for creating high-frequency phrase dictionary information among the network data that is transmitted over the network. Collecting character string information
Based on the character string information collected by the character string information collection procedure, a dictionary information creation procedure for creating the high frequency phrase dictionary information in which words that are frequently used at that time are registered,
A transmission procedure for transmitting the high-frequency phrase dictionary information to the character input device for downloading via the communication network;
A program that causes a server device to execute.

Images