JP2006185380A - Character processor with prediction function, method, recording medium and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a character processor with a prediction function capable of giving a priority to a prediction candidate in accordance with an expression usage frequency and a conversion period to expression. <P>SOLUTION: An expression converted from reading is stored together with the reading in a file 304 of input history. In addition, a pointer indicating a storage position of the expression stored in the input history is added to reading to be registered in a file 303 of a tree for retrieval in a network form. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a character processing method for inputting a reading and converting it into a notation corresponding to the reading, for example, a kana-mixed word. More specifically, when a part of the reading is input, the notation corresponding to the reading is partially converted. The present invention relates to a character processing apparatus with a prediction function, a method, a recording medium, and a program capable of presenting a candidate for notation to a user.

  Conventionally, when a reading is input and converted into a notation corresponding to the reading, for example, a kana mixed word, if a part of the reading is input, a candidate for a notation partially including the notation corresponding to the reading is presented to the user. A character processing method is known (Patent Documents 1 and 2). Obtaining convertible notation candidates from partial reading is called word prediction or notation prediction.

  Information processing for realizing the prediction function will be briefly described. An input history file as shown in FIG. 1A is prepared in a computer such as a personal computer. The input history file is a table in which a plurality of records are described with the readings input so far and the conversion result obtained by character processing as a pair of records.

  When the user reads “INTA” in the form of hiragana by reading from the keyboard and gives a prediction instruction to the computer, the computer follows the prediction program, that is, the prediction candidate that can be converted, that is, the conversion notation candidate below. Seed like this.

  When a portion of a partial reading of input is first input to the computer, the computer obtains a reading having the notation “inta” corresponding to the reading by searching the input history file. This search results in the readings “Intnet” and “Intarubaru”. Next, the notation corresponding to the acquired reading, “Internet” and “interval” are acquired and displayed on the display screen as prediction candidates. When the user selects one of the displayed prediction candidates with, for example, a mouse, the selected prediction candidate is determined as a prediction result for “Inta” and delivered to word processing software or the like.

  Since the input history file as shown in FIG. 1A describes the entire reading of the notation, when the number of readings to be registered becomes large, the data amount of the input history file also becomes enormous. In view of this, as shown in FIG. 1B, a proposal has been made to construct the reading portion of the input history file in the form of a network using a common reading.

Japanese Patent Laid-Open No. 11-238059 JP 2000-276466 A

  However, if the reading is expressed in the form of a network as shown in FIG. 1B of the conventional example, information such as the usage frequency of the notation and the conversion time to the notation cannot be reflected on the network. Therefore, there was a technical problem to be solved that the priority in the usage frequency of the notation and the conversion timing to the notation could not be given.

  Therefore, an object of the present invention is to provide a character processing device with a prediction function that can give priority to prediction candidates in accordance with the frequency of use of the notation and the conversion timing to the notation.

In order to achieve such an object, the present invention inputs a reading and outputs a notation corresponding to the reading. When a part of the reading is input, the reading is performed according to a prediction instruction. In a character processing apparatus with a prediction function that acquires a plurality of notation candidates that can be acquired from a section (FIG. 1, conventional technology)
Input history storage means for storing the notation output so far and the corresponding readings that have been input in the order of output; (FIG. 4 (B), 304, HD 203)
Pointer information that stores a plurality of readings in the form of a network and indicates a storage position of at least one of the notations stored in the input history storage means, corresponding to the stored readings And storage means for storing (FIG. 4A, 303, HD 203)
When a part of the reading is input and a prediction instruction is given, at least one or more readings including the part of the input reading are stored in the plurality of network-type readings stored in the search storage means. And first information processing means for obtaining pointer information corresponding to the obtained at least one or more readings from the search storage means (S110, CPU 201)
Display means for displaying information; (display 204)
Second information processing means for reading out the notation indicated by the acquired at least one pointer information from the input history storage means and displaying it as the candidate on the display screen of the display means (S120 to S140, CPU 201)
It is provided with.

  In the present invention, the same notation and duplication of readings are allowed with respect to the readings and notations stored in the input history storage means. Further, a pointer indicating the storage position of the notation stored in the input history storage means is added to the reading stored in the search storage means. Thus, after the notation indicated by the pointer information is read from the input history storage means, the use frequency of the same notation can be counted to determine the priority order of the prediction candidates.

  In addition, since the value of the pointer indicates the slow or fast storage in the input history storage means, it is possible to prioritize prediction candidates according to the storage order in the input history storage means based on the pointer. It becomes.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 2 shows a system configuration of a computer for realizing a character processing apparatus with a prediction function according to the present invention. As a computer to be used, various computers having an information processing function such as a personal computer, a server, and a portable terminal device such as a mobile phone can be used depending on the application.

  In FIG. 2, reference numeral 201 denotes a CPU which performs a prediction process according to the present invention by executing a character processing program. A system memory 202 includes a ROM and a RAM. The system memory 202 stores data for controlling the components of the computer in the ROM, and stores input / output data used for character processing of the CPU 201, image data to be displayed on the display 204 described later, and the like.

  A hard disk (abbreviated as HD) 203 stores a character processing program and a file (described later) used in the prediction process.

  Reference numeral 204 denotes a display that displays readings input from the keyboard 205, notations corresponding to readings, prediction notation candidates (abbreviated as prediction candidates), and the like.

  Reference numeral 205 denotes a keyboard for inputting operation instructions to the CPU 201 and readings to be subjected to character processing. A mouse 206 is used to specify a candidate to be confirmed among a plurality of prediction candidates displayed on the display screen of the display 201. The above-described components are connected to the bus 207 via an interface (not shown). The character processing program mounted on the HD 204 is recorded on a CDROM or floppy (registered trademark) disk, read by the CPU 201 via a reading drive (reading device) (not shown), and written on the HD 204.

  A program stored in the HD 204 and a file used in the prediction process will be described with reference to FIG.

  In FIG. 3, reference numeral 301 denotes a prediction processing program. In this embodiment, the prediction processing program 301 is incorporated in the character processing program. However, it goes without saying that the prediction processing program 301 may be separated from the character processing program.

  A kanji dictionary 302 is a dictionary used to convert readings into notation when character processing, in this case, prediction processing is not performed. As is well known, the kanji dictionary 302 describes a plurality of words having readings and notations, and grammatical information of the words indicated by the readings. The Kanji dictionary 302 is not directly related to the present invention.

  Reference numeral 303 denotes a search tree (file) which describes the readings used by the user in the above character processing so far in the form of a network. If the reading input by the user is new, the reading is registered in the search tree 303.

  Reference numeral 304 denotes an input history (file) which describes readings input by the user and notations converted by character processing in the order of confirmation. Here, it should be noted that the notation described in the input history allows the same notation to be duplicated (stored).

  The semantic structure of the search tree is shown in FIG. 4A, and the data format on the HD 203 is shown in FIG.

  In FIG. 4A, nodes, that is, a plurality of characters constituting a reading are connected by arcs (arrows).

  For example, if you want to read “Kyogi” on the network. Each character of “Kyogi” is connected with →, such as “ki” → “yo” → “u” → “gi”. In addition, when expressing the word “beautiful” on the network, branch from the above “ki” and connect each character as “ki” → “re” → “i” → “da” ( (See FIG. 4A).

  Each character constituting the reading is a pointer indicating the storage position of the corresponding input history (referred to as input history pointer), and a pointer indicating the storage position on the search tree where the subsequent characters are stored (referred to as subsequent character pointer). ) As attached information.

  When “today”, “no”, “competition”, “meeting”, “no”, and “venue” are stored at the positions “1” to “6” in the input history of FIG. On the search tree in FIG. 4A, a null value indicating that the corresponding notation is not in the input history is given to the value of the first pointer of the input history of “ki”. The storage position a1 on the search tree for “ょ” is given to the value of the first pointer of the character following “ki”.

  The storage position a2 on the search tree for “re” is given to the value of the second pointer after the character “ki”.

  Based on such a rule, the value of the first pointer in the input history of “yo” represents the position on the input history where the notation corresponding to “cho” is stored. In this case, Since the corresponding notation is not stored in the input history, a null value indicating that is given to the first pointer of the input history, and the value of the first pointer of the succeeding character is on the search tree for “u”. Storage location b1 is given.

  As the value of the first pointer of the input history of “U”, a “1” value indicating the storage location of the input history of the notation “today” corresponding to the reading “Kyo” is given (see FIG. 4B), and the subsequent The storage position c1 on the search tree for “u” is given to the value of the character first pointer.

  As the input history pointer of “gi”, a value “3” indicating the storage position of the input history of the notation “competition” corresponding to the reading “Kyogi” is given (see FIG. 4B), and the first character of the succeeding character is given. The pointer is given a value D1 indicating that there is no subsequent character.

  The description so far has been made on one route “ki” → “cho” → “u” → “gi” on the network, but the other routes have the same configuration.

  By representing a plurality of readings such as “Kyo” and “Kyogi” in such a network structure, the data capacity for storing the plurality of readings can be saved.

  The novel feature of the present invention is that duplication of the same notation is allowed for the notation stored in the input history file 304, and therefore, the input history pointer of the above-mentioned reading character is also set to 1 or more. .

  FIG. 5A shows a data format for storing the search tree file 303 described above in the HD 203.

  The search tree file 304 is stored in the HD 203 in the form of a table file, and one record has data areas 501 to 506. Reference numeral 501 denotes a data area for storing reading characters, and reference numeral 502 denotes a data area for storing n input history (first to n) pointers 503. A data area 504 stores m subsequent character (first to m) pointers 505.

  Next, the input history will be described.

  In FIG. 4B, the input history file 304 has storage areas for reading and notation. Each time a user inputs a reading in character processing, converts the reading into a notation, and confirms it, the reading and the notation are stored in the input history file 304 in the order of confirmation (output). Here, “determined” means that the notation finally determined by the user is transferred (output) to an application program (for example, word processing software, spreadsheet software, etc.) using the notation.

  The data format of the input history file 304 is shown in FIG.

  In FIG. 5, one record has a reading data area 510 and a notation data area 511.

  The prediction process according to the present invention will be described with reference to FIGS. 6 to 10 are implemented in the HD 203 in the form of a program (prediction processing program 301) that can be executed by the CPU 201. The CPU 201, the system memory 202, the display 204, and the HD 203 when executing the character processing program including the prediction processing program constitute the character processing device of the present invention. The HD 204 for storing the character processing program, the CDROM for installing the character processing program in the HD 204, the system memory 202 for loading the character processing program at the time of execution, and the like correspond to the storage medium of the present invention.

(Character processing)
FIG. 6 shows the processing procedure of the character processing program. S40 and S50 described later indicate the processing procedure of the prediction processing program 301.

  In FIG. 6, in S <b> 10, the CPU 201 accepts a reading input from the keyboard 205 and temporarily stores it in the system memory 202. For example, when the user inputs a reading “Kyo” and then instructs execution of prediction using the TAB key, the CPU 201 determines whether or not there is an instruction for prediction in S20, and the determination result is YES. The procedure proceeds to S40, where a prediction process described later is performed.

  When the notation corresponding to the reading is determined to be “competition” by the prediction process, the reading “Kyogikai” corresponding to the notation and the notation is cumulatively stored in the input history 304 by the CPU 201 (S50).

  On the other hand, if the determination result in S20 is NO, that is, if it is determined that there is no prediction instruction, the procedure proceeds to S30, and character processing similar to the conventional one, specifically, the keyboard 205 in S10. Is read and temporarily stored in the system memory 202 is converted into a notation, and the conversion result is transferred to the application program (S30). Further, the conversion result notation and the reading corresponding to the notation are cumulatively stored in the input history 304 (S50).

(Prediction process)
FIG. 7 shows details of the prediction processing in S40 of FIG.

  When a reading “Kyo” is input from the keyboard 205, the CPU 201 searches the network on the search tree of the HD 303 (see FIG. 4A) from the start point, and a path including the character string “Kyo”. Specifically, the same character as the first character “ki” in the input reading is found on the network, and a subsequent character pointer (at least one) attached to the character is obtained. Is extracted from the search tree and combined with the character from the start point, in this example, "ki" and "yo" are combined to create the string "cho". Then, the subsequent pointer indicating the same character as the second character of the input reading is selected from the characters indicated by the subsequent pointer, and the path having the same character string as the input reading is thus selected. It is searched on over (S100).

  Next, the CPU 201 sequentially acquires the characters after the character string “Kyo” found so far along the path found by the search, and combines them in the system memory 202.

  When the character on the path is acquired, the CPU 201 checks whether or not the value of the input history pointer is a null value. A null value indicates that there is no notation corresponding to the reading in the input history. If it is not a null value, the current reading character is the end of the notation (so-called word) reading character string, and the input history Indicates that there is a notation corresponding to reading. For this reason, when the determination result of the CPU 201 is a null value, subsequent characters on the path are further acquired.

  On the other hand, if the determination result of the CPU 201 is not a null value, characters are combined in the system memory 202, and the indicated character string is extracted from the data area on the input history (file 304) indicated by the input history pointer, Temporarily stored as prediction candidates in a dedicated area in the system memory 202.

  In the example of FIG. 4A, since the value of the input history pointer of the character “U” following “KYO” is “1”, it is determined that the character “U” is not a null value. Combined with “Kyo”, “Kyo” is created. Further, the notation of the storage position indicated by the value of the input history pointer (at least 1) of “u”, specifically, the notation “today” (see FIG. 4B) on the input history file 304 is the system memory. The data is temporarily stored in the dedicated area 202 as a prediction candidate. When there are a plurality of input history pointers, all notations indicated by the input history pointers are stored in the dedicated area.

  The CPU 201 acquires the subsequent character “gi” on the path on the search tree based on the subsequent character pointer of “u”, and performs the same processing as described above. Since the character “gi” indicates that the input history pointer is not a null value, “gi” is created by combining “gi” with the character string “kyo” created so far. In addition, the CPU 201 extracts the notation “game” indicated by the storage position indicated by the input history pointer from the input history file 304 and additionally stores it as a prediction candidate in the dedicated area of the system memory 202. At this stage, the prediction candidates are “today” and “competition”. Thereafter, the CPU 201 combines characters to the end of the path, and takes out the character string indicated by the input history pointer from the input history each time an input history pointer that is not a null value is found. The extracted character string is stored in the system memory 202.

  In the example of FIG. 4A, “today” and “competition” are obtained as prediction candidates as a prediction result for the reading “Kyo” (S120).

Next, the CPU 201 determines the priority order of the acquired prediction candidates (S130). The following two methods can be considered as a method for determining the priority order.
(1) Give a high ranking to the frequently used notation in the input history.
(2) Give a high ranking to a notation that is stored late in the input history.

  The user may determine which method is to be adopted as appropriate. Alternatively, the above two methods may be prepared and the method used by the user may be selected.

  Processing procedures for concretely implementing the above two methods will be described later with reference to FIGS.

  When the priority order is determined, the CPU 201 rearranges the prediction candidates “today” and “game” in the dedicated area of the system memory 202 according to the priority order and displays them on the display screen of the display 204. The user uses the mouse 206 to select a desired candidate on the display screen and confirms it using the ENTER key on the keyboard 205. Thereby, each localized character string is delivered (output) to the application program that uses the character string as an input character string (S140 → S150).

(Registering readings in the search tree)
A procedure for newly registering a reading in the search tree is shown in FIG.

  In S10 of FIG. 6, when a reading is input, the processing procedure of FIG. 8 is started.

  The CPU 201 searches the search tree in FIG. 4A for the input reading (S200). Next, it is determined whether or not the same character string is on the path (S210). If there is an identical character string on the route, there is no need for registration, so the procedure moves to the end, and this processing procedure ends. On the other hand, if a determination result indicating that there is no identical character string on the path is obtained (the determination result in S210 is YES), the input reading is added to the search tree in FIG. Specifically, the same character on the search tree is traced along the path from the first character of the input reading character string, and the character string part not registered on the network in the input reading character is newly added. Register with. For example, if “Kyo” is registered in the network but “Kyogi” is not registered, “Gi” is registered after the last character “U” on the route of “Kyo”. . At this time, the subsequent character pointer of “u” is set to a value indicating the storage position of “gi” (S220).

  The CPU 201 waits for the confirmation of the notation corresponding to the input reading by the loop processing of S230, and when the determination processing of S230 detects that the confirmation notation is stored in the input history file 304 in S50 of FIG. The storage position of the notation stored in the file 304 is set to the value of the input history pointer of “gi”, and the pointer information addition processing in S240 is ended.

(Determining the priority order of prediction candidates according to usage frequency)
FIG. 9 shows a priority determination process according to the storage order of the notation. In FIG. 9, when an input reading is given, the CPU 201 finds a route on the network having the reading and the same character string on the route by the CPU 201 (S300). Next, as described above, the character at the end of the notation is found by determining the value of the input history pointer while adding characters after the last character of the same character string on the path (S310 → S320). When the terminal character is found, the notation of the storage position indicated by the value of the input history pointer (at least 1 or more) is stored in the dedicated area of the system memory 202 (S325). By continuing this process repeatedly (the loop process of S310 to S330), the CPU 201 can acquire the notations stored in the plurality of reading and input history files 304 existing on one path. Here, it should be noted that the same notation is redundantly stored in the input history file 304. When the notation indicated by the input history pointer is taken out to the system memory 202, a plurality of identical notations are stored in the dedicated area. Therefore, the usage frequency of the notation is counted by counting the number of identical notations (S325). In other words, since the usage frequency of the notation can be counted without searching the input history file 304, this embodiment can greatly reduce the calculation time of the usage frequency.

  Finally, the CPU 201 rearranges a plurality of notations (prediction candidates) according to the usage frequency. For the processing for rearrangement, an information processing method called sorting may be used.

(Determining the priority order of prediction candidates according to the storage order of input history)
FIG. 10 shows a process for determining the priority order of the prediction candidates according to the storage order of the input history. In FIG. 10, the CPU 201 sets a path having an input reading, and finds a character at the end of one notation while adding characters along the path from the character at the start point (loop processing from S400 to S420 → S440). When the terminal character is found in the determination process of S420, the input history pointer associated with the character and the notation of the input history indicated by the pointer are registered in the rank list in the system memory. Here, the notation with a small input history pointer value (the storage order is fast) is stored at the lower position of the ranking list, and the notation with a large input history pointer value (the storage order is slow) is stored at the upper position of the ranking list (S430). ).

  When the above process is repeated until the end of the path, the prediction list is stored in the order in which the input history is stored in the order of prediction history.

In addition to the above embodiment, the following embodiment can be implemented.
1) In the above-described embodiment, a plurality of prediction candidates are displayed on the display screen, and the prediction candidate to be determined by the user is selected. As an alternative embodiment, instead of displaying a plurality of prediction candidates, the CPU 201 initially displays the highest prediction candidate on the display screen of the display, and predicts the display by the CPU 201 in response to a user key operation on the keyboard. The candidates may be switched and displayed according to the priority order.
2) In order to store the search tree (network) shown in FIG. 4A on the hard disk, a data format other than the data format described in this embodiment can be used.
3) In the embodiment described above, the character processing method from Japanese reading to notation has been described, but the present invention can also be applied to the character processing method from Chinese reading to notation.
4) In the above-described embodiment, the form in which the character processing device is constructed on a single computer has been described. However, the processing may be executed in a distributed manner by a plurality of computers, and further, the system shown in FIG. It may be constructed on a plurality of computers.
5) In the above-described embodiment, only one search tree and input history are prepared. However, the search tree and input history for each application for which character input is performed, for example, for each spreadsheet software or word processor software. May be provided. Further, a search tree and an input history can be provided for each technical field, and the technical field can be designated by the user.
6) In the above-described embodiment, when a part of reading is input and prediction is instructed, a search method called a forward matching search is used to find a prediction candidate on the search tree. The present invention is not limited to this, and a full search may be performed. The full search is a method in which when a part of a reading is input and a word included in the part of the reading is also included in the input history, the word is used as a search result. For example, if “today”, “competition” and “competition” are stored in the input history and “Kyogi” is given as part of the reading, “competition” will be the search result (predict Candidate). In all searches, “today”, “competition”, and “competition” are obtained as search results (prediction candidates).

The forward matching search is suitable when the user's input operation is desired to be reduced, and the full search is suitable when a reading error or the like is desired to be found.
7) In registering readings in the search tree, if the same reading exists on the search tree, a pointer to the input history may be added.

FIG. 1A and FIG. 1B are explanatory diagrams showing a prediction warp of the prior art. It is a block diagram which shows the system configuration | structure of this invention embodiment. It is explanatory drawing which shows the program and file concerning this invention. FIG. 4A is an explanatory diagram showing the structure of the search tree. FIG. 4B is an explanatory diagram showing the contents stored in the input history. FIG. 5A is an explanatory diagram showing the data format of a record in the search tree. FIG. 5B is an explanatory diagram showing the data format of the record of the input history. It is a flowchart which shows the character processing procedure from reading to description. It is a flowchart which shows the detail of a prediction process. It is a flowchart which shows the registration process to the tree for reading retrieval. It is a flowchart which shows the procedure for determining the priority of the prediction candidate based on use frequency. It is a flowchart which shows the procedure for determining the priority of the prediction candidate based on the memory | storage order to an input history.

Explanation of symbols

201 CPU
202 System memory 203 Hard disk (HD)
204 Display 205 Keyboard 206 Mouse

Claims (32)

Enter a reading and output the notation corresponding to the reading.
In a character processing device with a prediction function that acquires a plurality of notation candidates that can be acquired from a part of the reading according to a prediction instruction when a part of the reading is input,
A storage means for input history that stores the notation output so far and the corresponding readings that have been input in the order of output and a plurality of readings are stored in the form of a network, and is a notation corresponding to the stored readings, A search storage means for storing pointer information indicating a storage position of at least one notation stored in the input history storage means;
When a part of the reading is input and a prediction instruction is given, at least one or more readings including the part of the input reading are stored in the plurality of network-type readings stored in the search storage means. And first information processing means for obtaining pointer information corresponding to the obtained at least one reading from the search storage means,
Display means for displaying information;
A prediction function comprising: a second information processing unit that reads out the notation indicated by the acquired at least one pointer information from the input history storage unit and displays it as the candidate on the display screen of the display unit Character processing unit.   2. The character processing device according to claim 1, wherein the notation stored in the input history storage means is allowed to store the same notation in duplicate, and the notation stored redundantly includes a plurality of the pointer information. Is stored in the search storage means.   In the character processing apparatus with a prediction function according to claim 2, for the notation to be displayed as the candidate, the notation corresponding to the pointer information is acquired from the input history storage means, the number of the same notation is counted, A character processing apparatus with a prediction function, further comprising third information processing means for determining a priority order of candidates to be displayed in descending order of counting results.   4. The character processing apparatus with a prediction function according to claim 3, wherein the second information processing means puts all of the candidate notations on the display screen of the display means in the priority order determined by the third information processing means. A character processing device with a prediction function characterized by displaying.   4. The character processing apparatus with a prediction function according to claim 3, wherein the second information processing means displays only the highest priority notation in the priority order determined by the third information processing means. A character processing device with a prediction function.   The character processing apparatus with a prediction function according to claim 1, wherein when there are a plurality of different candidates, the candidate with the later storage time indicated by the pointer information corresponding to the plurality of candidates is prioritized. A character processing device with a prediction function, further comprising a fourth information processing means for determining the priority order of   7. The character processing apparatus with a prediction function according to claim 6, wherein the second information processing means puts all of the notation of the candidates on the display screen of the display means in the priority order determined by the fourth information processing means. A character processing device with a prediction function characterized by displaying.   7. The character processing device with a prediction function according to claim 6, wherein the second information processing means displays only the highest priority notation in the priority order determined by the fourth information processing means. A character processing device with a prediction function. Prediction that inputs a reading and outputs a notation corresponding to the reading and acquires a plurality of notation candidates that can be acquired from a part of the reading according to a prediction instruction when a part of the reading is input In the prediction method of the functional character processing device,
The prediction function-equipped character processing device includes:
Input history storage means for storing the notation output so far and the corresponding readings that have been input in the order of output;
Pointer information that stores a plurality of readings in the form of a network and indicates a storage position of at least one of the notations stored in the input history storage means, corresponding to the stored readings Storage means for storing
Information processing means;
Display means, and
When a part of the reading is input and a prediction instruction is given, at least one or more readings including the part of the input reading are stored in the plurality of network-type readings stored in the search storage means. A first information processing step of acquiring pointer information corresponding to the acquired at least one reading by the information processing means from the search storage means;
A second information processing step of reading out the notation indicated by the acquired at least one pointer information from the input history storage means and displaying it as a candidate on the display screen of the display means by the information processing means. A prediction processing method for a character processing device with a prediction function as a feature.   The predictive processing method of the character processing device according to claim 9, wherein notations stored in the input history storage unit are allowed to be stored in duplicate, and multiple notations stored in duplicate are stored. The pointer information is stored in the search storage means, and the prediction processing method of the character processing device with a prediction function is characterized in that:   11. The prediction processing method for a character processing device with a prediction function according to claim 10, wherein for the notation to be displayed as the candidate, a notation corresponding to the pointer information is acquired from the input history storage means by the information processing means, and is the same. A prediction processing method for a character processing apparatus with a prediction function, further comprising: a third information processing step for counting the number of notations in the order and determining the priority order of candidates to be displayed in descending order of the counting results.   12. The prediction processing method for a character processing device with a prediction function according to claim 11, wherein all of the notation of the candidate is displayed by the information processing means in the priority order determined by the third information processing step. A predictive processing method for a character processing device with a predictive function, characterized by:   12. The prediction processing method for a character processing device with a prediction function according to claim 11, wherein only the highest priority notation in the priority order determined by the third information processing step is displayed by the information processing means. A prediction processing method for a character processing device with a prediction function.   The prediction processing method of the character processing device with a prediction function according to claim 9, wherein when there are a plurality of different candidates, the candidate with the later storage time indicated by the pointer information corresponding to the plurality of candidates is prioritized. A prediction processing method for a character processing device with a prediction function, further comprising a fourth information processing step for determining the priority order of the plurality of candidates by the information processing means.   15. The prediction processing method for a character processing device with a prediction function according to claim 14, wherein all of the candidate notations are displayed by the information processing means in the priority order determined by the fourth information processing step. A predictive processing method for a character processing device with a predictive function, characterized by:   15. The prediction processing method for a character processing device with a prediction function according to claim 14, wherein only the highest priority notation in the priority order determined in the fourth information processing step is displayed by the information processing means. A prediction processing method for a character processing device with a prediction function. Prediction that inputs a reading and outputs a notation corresponding to the reading and acquires a plurality of notation candidates that can be acquired from a part of the reading according to a prediction instruction when a part of the reading is input In the storage medium of the functional character processing device,
The prediction function-equipped character processing device includes:
Input history storage means for storing the notation output so far and the corresponding readings that have been input in the order of output;
Pointer information that stores a plurality of readings in the form of a network and indicates a storage position of at least one of the notations stored in the input history storage means, corresponding to the stored readings Storage means for storing
Information processing means;
Display means, and
When a part of the reading is input and a prediction instruction is given, at least one or more readings including the part of the input reading are stored in the plurality of network-type readings stored in the search storage means. A first information processing step of acquiring pointer information corresponding to the acquired at least one reading by the information processing means from the search storage means;
A computer-executable comprising: a second information processing step of reading the notation indicated by the acquired at least one pointer information from the input history storage means and displaying the information on the display screen of the display means as the candidate by the information processing means A storage medium for a character processing device with a prediction function, comprising a program.   18. The storage medium of the character processing device according to claim 17, wherein the notation stored in the input history storage means is allowed to store the same notation in duplicate, and the notation stored redundantly includes a plurality of notations. A storage medium for a character processing apparatus with a prediction function, wherein the pointer information is stored in the search storage means.   The storage medium of the character processing device with a prediction function according to claim 18, wherein the program obtains a notation corresponding to the pointer information from the input history storage means by the information processing means for the notation to be displayed as the candidate. A storage medium for a character processing device with a prediction function, further comprising a third information processing step of counting the number of the same notation and determining the priority order of candidates to be displayed in descending order of the counting results .   20. The storage medium of the character processing device with a prediction function according to claim 19, wherein all of the notation of the candidate is displayed on the display screen of the display means by the information processing means in the priority order determined by the third information processing step. A storage medium for a character processing device.   20. The storage medium of the character processing device with a prediction function according to claim 19, wherein only the highest priority notation in the priority order determined by the third information processing step is displayed by the information processing means. A storage medium for a character processing device with a prediction function.   The storage medium of the character processing device with a prediction function according to claim 17, wherein when there are a plurality of different candidates, the program selects a candidate with a later storage time indicated by pointer information corresponding to the plurality of candidates. A storage medium of a character processing device with a prediction function, further comprising a fourth information processing step for prioritizing the plurality of candidates by the information processing means.   23. The storage medium of the character processing apparatus with a prediction function according to claim 22, wherein all of the notation of the candidate is displayed on the display screen of the display means by the information processing means in the priority order determined by the fourth information processing step. A storage medium of a character processing device with a prediction function, characterized by displaying.   23. The storage medium of a character processing device with a prediction function according to claim 22, wherein only the highest priority notation in the priority order determined in the fourth information processing step is displayed by the information processing means. A storage medium for a character processing device with a prediction function. Prediction that inputs a reading and outputs a notation corresponding to the reading and acquires a plurality of notation candidates that can be acquired from a part of the reading according to a prediction instruction when a part of the reading is input In a prediction program for a functional character processing device,
The prediction function-equipped character processing device includes:
Input history storage means for storing the notation output so far and the corresponding readings that have been input in the order of output;
Pointer information that stores a plurality of readings in the form of a network and indicates a storage position of at least one of the notations stored in the input history storage means, corresponding to the stored readings Storage means for storing
Information processing means;
Display means, and
When a part of the reading is input and a prediction instruction is given, at least one or more readings including the part of the input reading are stored in the plurality of network-type readings stored in the search storage means. A first information processing step of acquiring pointer information corresponding to the acquired at least one reading by the information processing means from the search storage means;
A second information processing step of reading out the notation indicated by the acquired at least one pointer information from the input history storage means and displaying the notation on the display screen of the display means as the candidate by the information processing means. A computer-executable prediction program for a character processing device with a prediction function.   26. The prediction program for a character processing device according to claim 25, wherein notations stored in the input history storage means are allowed to be stored in duplicate, and a plurality of notations stored in duplicate are stored. A prediction program for a character processing device with a prediction function, wherein the pointer information is stored in the search storage means.   27. The prediction program for a character processing device with a prediction function according to claim 26, wherein for the notation to be displayed as the candidate, the notation corresponding to the pointer information is obtained from the input history storage means by the information processing means, and the same A prediction program for a character processing device with a prediction function, further comprising a third information processing step of counting the number of notations and determining a priority order of candidates to be displayed in descending order of counting results.   28. The prediction program for a character processing device with a prediction function according to claim 27, wherein all of the notation of the candidate is displayed on the display screen of the display means by the information processing means in the priority order determined by the third information processing step. A prediction program for a character processing device with a prediction function.   28. The prediction program for a character processing device with a prediction function according to claim 27, wherein only the highest priority notation in the priority order determined in the third information processing step is displayed by the information processing means. A prediction program for a character processing device with a prediction function.   26. In the prediction program for a character processing device with a prediction function according to claim 25, when there are a plurality of different candidates, priority is given to a candidate with later storage time indicated by pointer information corresponding to the plurality of candidates. A prediction program for a character processing apparatus with a prediction function, further comprising a fourth information processing step for determining a priority order of a plurality of candidates by the information processing means.   31. The prediction program for a character processing device with a prediction function according to claim 30, wherein all of the notation of the candidate is displayed on the display screen of the display means by the information processing means in the priority order determined by the fourth information processing step. A prediction program for a character processing device with a prediction function, characterized by displaying.   The prediction program for a character processing device with a prediction function according to claim 30, wherein only the highest priority notation in the priority order determined in the fourth information processing step is displayed by the information processing means. A prediction program for a character processing device with a prediction function.

Images