JP5471457B2 - Information processing device - Google Patents

Description

  The present invention relates to an information processing apparatus having a character input function, and more particularly to an information processing apparatus that suitably corrects character input.

  An information processing apparatus such as a portable terminal is provided with operation keys for inputting characters and an input device such as a touch panel. These input devices cannot be provided with many input keys like a keyboard of a personal computer from the viewpoint of miniaturization of the device. For this reason, an important issue is how to perform an input operation with good operability from an input device provided within a limited range.

  Today, in order to support the user's input operation, a so-called predictive conversion function that predicts a character string intended for conversion from a character string input by the user and displays the predicted character string as an input candidate is widely used. Yes.

  Conventionally, there has been disclosed a character string prediction technique that uses this predictive conversion function to obtain a predicted character string candidate and efficiently input a character string without explicitly instructing the start of input prediction ( Patent Document 1).

  The character string prediction technique disclosed in Patent Document 1 is a candidate for a predicted character string based on an input character string after correction when the input character string input by the user is deleted and the input character string is corrected. By displaying the character string, the character string can be input efficiently.

Japanese Patent Laid-Open No. 11-212705

  Here, the user may perform a character input operation while visually checking the operation key and confirming the position of the key. In such a case, after the character input is completed, the user performs an operation of moving the line of sight to the main display in order to confirm the character input before the conversion operation. If an erroneous input is included in the input character string, the user will notice the erroneous input after the character input is completed. For this reason, the user performs an operation of deleting a character string that has already been input and an operation of inputting a correct character string.

  Here, a procedure of character input operation when the user recognizes that an erroneous input is included and corrects the character after completion of character input will be considered.

  As an example, if the user intends to convert the character string “Kondoha” to the character string “Now”, the user accidentally inputs “Kon” and “ An explanation will be given by applying the case where it is recognized after the character input is completed that "and" is erroneously input).

  The first step is to delete the character string “to” from the end of the input character string “konto” to the erroneously entered character “to”, input the correct character “do”, and the deleted character “ha” ”, A conversion operation from the character string“ Kondoha ”to the character string“ Now ”, and a conversion confirmation operation are required.

  The second procedure is to move the cursor to delete the wrong character string “to” from the end of the character string “konto”, delete the erroneously entered character “to”, correct character “ ”, A conversion operation from the character string“ Kondoha ”to the character string“ this time ”, and a conversion confirmation operation are required. The cursor indicates the editing position for character input, and the character focused by the cursor indicates the character to be deleted.

  When a character input operation is performed in the first procedure, a list of predicted character string candidates for the deleted character string “Kon” is displayed by applying the character string prediction technique of Patent Document 1 described above. can do. If the candidate includes “Now” or “Now” intended to be converted, an erroneous input can be corrected by selecting the candidate. However, even if “next” is included in the conversion candidates and the selection is made, the character string “ha” needs to be input again. Even if the conversion candidate includes “Now”, it is a conversion candidate for the character string “Kon”, so the priority of the display order is low, and conversion and confirmation operations can be performed quickly. May not be possible.

  In addition, when the character input operation is performed in the second procedure, the intended character string “Kondoha” is entered when the incorrect character “To” is deleted and the correct character “Do” is entered. Is in a completed state. Here, the position of the cursor may be used to recognize a character string that is a conversion candidate prediction target. This is because, as described above, the cursor indicates the editing position for character input, and the position of the cursor immediately after the character string input can be regarded as a delimiter for the character string to be converted. In the second procedure, when the erroneously entered character “TO” is corrected to the correct character “DO”, the cursor is positioned immediately after the character “DO”. Therefore, the character string “Kondo” is regarded as a delimiter for the character string to be converted. On the other hand, after the user moves the cursor to the end of the input character string, the character string “Kondoha” is set so that the character string “Kondoha” is smoothly converted as a conversion candidate prediction target. Needs to be predicted.

  Moreover, since the character string prediction technique of Patent Document 1 is assumed to be applied when an erroneously input character string is the last input character and this character is deleted, the end of the character string is assumed. It was not a character string prediction technique suitable for cases where erroneous input was included in addition to the characters.

  The present invention has been made in view of such circumstances, and provides an information processing apparatus that can easily perform a correction operation even when erroneous input is included between character strings. Objective.

In order to solve the above-described problem, an information processing apparatus according to the present invention is provided with a character input unit that receives input of a character string, a word dictionary unit that stores a plurality of words in association with readings, and a predetermined user operation. A determination unit that determines an erroneous input related character that is a character related to an erroneous input from the character string based on the character string generated using the character string that has received the input and the erroneous input related character; A search unit that searches the phrase dictionary unit, extracts a phrase that corresponds to a correction candidate that corrects an erroneous input of the character string that has received the input, and a display unit that selectively displays the searched phrase . The search unit is configured to search, as the correction candidate, a phrase that reads a character string in which the erroneous input related character is deleted from the character string .

In addition, the information processing apparatus according to the present invention includes a character input unit that receives input of a character string, a phrase dictionary unit that stores a plurality of words that have similar meanings, and stores the characters based on a predetermined user operation. A determination unit for determining an erroneous input related character that is a character related to an erroneous input from a sequence; and the word dictionary unit using a character string generated by using the input character string and the erroneous input related character. Locate and provided with the erroneous input related character search unit means and said erroneous input related character to search for similar terms as correction candidates to correct, and a display unit that selectably displays the phrase searched, the The search unit is configured to search, as the correction candidate, a phrase that reads a character string in which the erroneous input related character is deleted from the character string .

  The information processing apparatus according to the present invention can easily perform a correction operation even when an erroneous input is included between character strings.

The figure which shows the structure of the external appearance seen from the front at the time of the open state which opened the folding type mobile telephone which is an example of the information processing apparatus which concerns on this invention at about 180 degree | times. 1 is a schematic functional block diagram showing a first embodiment of an information processing apparatus according to the present invention. The flowchart explaining the correction candidate display process performed with the mobile telephone in 1st Embodiment. The flowchart explaining the deletion correction candidate production | generation process performed with the mobile telephone in 1st Embodiment. The flowchart explaining the insertion correction candidate production | generation process performed with the mobile telephone in 1st Embodiment. The flowchart explaining the change correction candidate production | generation process performed with the mobile telephone in 1st Embodiment. FIG. 7 is a screen transition diagram illustrating an example of character input operation to which each process of FIGS. 3 to 6 is applied. The conceptual diagram explaining the correction candidate production | generation process of FIGS. The flowchart explaining the other change correction candidate production | generation process performed with the mobile telephone in 1st Embodiment. The conceptual diagram explaining the change correction candidate production | generation process of FIG. 9 which considered the toggle input method. The flowchart explaining the correction candidate re-display process performed with the mobile telephone in 1st Embodiment. The screen transition diagram explaining the example of character input operation to which the correction candidate redisplay process of FIG. 11 is applied. The schematic functional block diagram which shows 2nd Embodiment of the information processing apparatus which concerns on this invention. The flowchart explaining the character input process before the conversion confirmation operation performed with the mobile telephone in 2nd Embodiment. The flowchart explaining the correction candidate display process before the conversion confirmation operation performed with the mobile telephone in 2nd Embodiment. The flowchart explaining the deletion correction candidate production | generation process performed with the mobile telephone in 2nd Embodiment. The flowchart explaining the insertion correction candidate production | generation process performed with the mobile telephone in 2nd Embodiment. The flowchart explaining the change correction candidate production | generation process performed with the mobile telephone in 2nd Embodiment. The screen transition diagram explaining the example of character input operation to which each process of FIGS. 14-18 is applied. The flowchart explaining the character input process after the conversion confirmation operation performed with the mobile telephone in 2nd Embodiment. The flowchart explaining the correction candidate display process after the conversion confirmation operation performed with the mobile telephone in 2nd Embodiment. The flowchart explaining the insertion correction candidate production | generation process after the confirmation operation of the conversion performed with the mobile telephone in 2nd Embodiment. The flowchart explaining the change correction candidate production | generation process after the conversion confirmation operation performed with the mobile telephone in 2nd Embodiment. The flowchart explaining the change correction candidate production | generation process performed after the conversion confirmation operation performed with the mobile telephone in 2nd Embodiment. The screen transition diagram explaining the example of character input operation to which each process of FIGS. 20-24 is applied.

  Embodiments of an information processing apparatus according to the present invention will be described with reference to the accompanying drawings. In each embodiment, a mobile phone will be described as an example of an information processing apparatus.

[First Embodiment]
A first embodiment of an information processing apparatus according to the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a diagram showing a configuration of an external appearance of a foldable mobile phone 1 which is an example of an information processing apparatus according to the present invention when viewed from the front when opened at about 180 degrees.

  In the mobile phone 1, a first casing 12 and a second casing 13 are hingedly connected with a central hinge portion 11 as a boundary. The first housing 12 and the second housing 13 are formed to be foldable in the direction of the arrow X via the hinge portion 11. A transmitting / receiving antenna is provided at a predetermined position inside the mobile phone 1. The cellular phone 1 transmits and receives radio waves to and from a base station (not shown) via this built-in antenna.

  An operation key 14 is provided on the surface of the first housing 12. The operation key 14 includes a plurality of keys such as a numeric key 15, a cross key 16, and a confirmation key 17.

  The numeric key 15 is a key that can accept input of numbers “0” to “9”, “A” to “Wa”, and “A” to “Z” alphabets. The numeric key 15 accepts input of various characters such as kana characters, numbers, symbols, and pictographs when performing a character input operation. Among these, for the input of kana characters, the input of each line of Japanese syllabary is assigned to each key. For example, the number “1” of the number key 15 is assigned a kana character input of “a line” of the Japanese syllabary, and the kana character input of “ka line” is assigned to the number “2”. The other numeric keys 15 are also assigned a character input of the Japanese syllabary below “Sa line”. When these numeric keys 15 are pressed once, the pseudonym of “Adan”, the pseudonym of “Idan” when pressed twice, the pseudonym of “Udan” when pressed three times, and so on A so-called toggle input method is used, in which “Edan” and “Odan” are input, and when the character reaches the last character assigned to each key, the character assigned to the first character is returned to. In addition to the toggle input method, a so-called two-touch method can be used in which predetermined numbers are assigned to each line and each row of the Japanese syllabary and input is performed by a combination of two key presses.

  The cross key 16 is operated in the up / down / left / right direction to accept an operation for moving the cursor or the like displayed on the main display 21 in the up / down / left / right direction. The confirmation key 17 is pressed to accept various processing confirmation operations. In the character input operation, the confirmation key 17 accepts the confirmation operation of the inputted character string, and the cross key 16 accepts the selection operation of the conversion candidates displayed so as to be selectable.

  Side keys 22 are provided on the side surface of the first housing 12. The side keys 22 each receive an instruction for a predetermined process when pressed in the internal direction of the first housing 12 when the mobile phone 1 is in the open state and the closed state.

  The first casing 12 is provided with a microphone 24 below the operation keys 14. The mobile phone 1 collects the user's voice during a call using the microphone 24.

  On the other hand, the second housing 13 is provided with a main display 21 that occupies most of the area. The main display 21 can display a pictogram icon indicating a radio wave reception state and a remaining battery level, a current date and time, an e-mail content, a simple homepage, and the like. The main display 21 is, for example, a display constituted by an LCD (Liquid Crystal Display), an organic EL (Electro Luminescence) display, and an inorganic EL display.

  A receiver (receiver) 25 is provided at a predetermined position above the main display 21 of the second housing 13. The mobile phone 1 outputs the voice of the other party at the time of the call by the receiver 25. By using the receiver 25 and the microphone 24, the user can make a voice call.

  FIG. 2 is a schematic functional block diagram showing the first embodiment of the information processing apparatus according to the present invention.

  The mobile phone 1 is not only a W-CDMA system, but also a GSM system, a cdma2000 1x RTT system, an EVDO system, and an E-UTRA system that is a wireless access of a 3.9th generation LTE system or a plurality of wireless communication systems Can be used for voice communication and data communication. Further, the antenna 31, the wireless transmission / reception unit 32, and the signal processing unit 33 correspond to these methods.

  The antenna 31 receives a radio signal transmitted from a base station accommodated in a mobile communication network by using a predetermined communication processing system. The antenna 31 radiates a radio signal of a predetermined access method to the space so that wireless communication can be performed with a predetermined communication processing system.

  The radio transmission / reception unit 32 performs radio communication with a base station accommodated in the mobile communication network via the antenna 31 by a predetermined communication processing method. The radio transmission / reception unit 32 generates a radio signal having a carrier frequency instructed from the control unit 39 based on the modulation signal generated by the signal processing unit 33. The radio transmission / reception unit 32 receives a radio signal having a carrier frequency instructed from the control unit 39, mixes it with the local oscillation signal output from the frequency synthesizer, and converts the frequency into an intermediate frequency signal (down-conversion). Then, the radio transmission / reception unit 32 orthogonally demodulates the down-converted intermediate frequency signal and outputs a reception baseband signal. The reception result is output to the signal processing unit 33 and the control unit 39.

  The signal processing unit 33 is configured by a DSP (Digital Signal Processor) or the like, and performs predetermined signal processing on the received baseband signal to obtain received packet data in a predetermined transmission format. The signal processing unit 33 demodulates the audio signal included in the received packet data and decodes the demodulation result to obtain audio data and the like. The audio data obtained by the decoding process is supplied to the PCM codec 34. The PCM codec 34 performs PCM decoding on the audio data and outputs an analog audio data signal after PCM decoding to the reception amplifier 35. The analog audio signal is amplified by the reception amplifier 35 and then output from the receiver 25.

  On the other hand, the voice signal (analog audio signal) of the speaker (user) input to the microphone 24 is converted into a digital signal, amplified to an appropriate level by the transmission amplifier 37, and then PCM encoded by the PCM codec 34. The The audio signal after PCM encoding is input to the signal processing unit 33. The signal processing unit 33 encodes the audio signal, generates a modulation signal based on audio data and other data obtained by the encoding, and outputs the generated modulation signal to the radio transmission / reception unit 32.

  The control unit 39 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The CPU executes various processes according to programs stored in the ROM or various application programs and control programs including the operating system (OS) loaded from the storage unit 40 to the RAM. In addition, the CPU generates various control signals and supplies them to each unit to control the mobile phone 1 in an integrated manner. In the first embodiment, the predictive conversion candidate search unit 51, the correction candidate search unit 52, and the candidate sort unit 53 are realized by the control unit 39 executing a predetermined program. Details of the predictive conversion candidate search unit 51, the correction candidate search unit 52, and the candidate sort unit 53 will be described later. The RAM of the control unit 39 appropriately stores data necessary for the CPU to execute various processes.

  The storage unit 40 includes, for example, a flash memory element that is an electrically rewritable and erasable nonvolatile memory, an HDD (Hard Disc Drive), and the like. The storage unit 40 stores various application programs and various data groups executed by the CPU of the control unit 39, control programs and control data of the mobile phone 1, and identification information uniquely assigned to the mobile phone 1 or the user. . In addition, the storage unit 40 appropriately stores, for example, telephone book data in which names, telephone numbers, addresses, and the like are registered as contact information, data acquired through data communication, downloaded data, and the like. In the first embodiment, the storage unit 40 functions as a word dictionary unit 61, an input character string / history storage unit 62, and a priority evaluation data storage unit 63. The input character string / history storage unit 62 is not limited to the storage unit 40 and may be temporarily stored on the RAM side of the control unit 39.

  The power supply circuit 41 generates a predetermined operating power supply voltage Vcc based on the output of the battery 42 and supplies it to each circuit unit. The cellular phone 1 is provided with a clock circuit (timer) 43 that measures the current time.

  Next, the function of each part that supports the character input operation and the character input operation of the mobile phone 1 in the first embodiment will be described.

  Character input operations involving conversion such as kana-kanji conversion include operations such as a character string input operation, an input character string conversion operation, and a conversion confirmation operation. The character string input operation is an operation of inputting a character string composed of kana characters or the like via the operation keys 14. The character string conversion operation is an operation for converting an input character string into kanji, symbols, pictograms, and the like. The conversion confirmation operation is an operation for confirming a character string (character) whose conversion target is selected by the conversion operation.

  The mobile phone 1 according to the first embodiment determines that an erroneous input is included in the input character string when a deletion operation is performed from the end of the character string after the input operation is completed and before the conversion confirmation operation. An appropriate correction candidate for correcting this erroneous input is displayed. This is effective in that the user can correct erroneous input without performing an operation of inputting a new character. It should be noted that the “wrong input” mode in each embodiment includes a character that is different from the character string intended for input when an unintended extra character string is input, or a character that is intended to be input is not input. The case where is entered is included.

  The prediction conversion candidate search unit 51 of the control unit 39 searches the word dictionary unit 61 of the storage unit 40 for a phrase (prediction conversion candidate) whose conversion is predicted for the input character string. The predictive conversion candidate means a phrase composed of kanji or kana reading the input character string or a phrase where the input character string matches the first reading.

  The correction candidate search unit 52 functions as a determination unit that determines an erroneous input related character that is a character related to an erroneous input from a character string based on a predetermined user operation. Further, the correction candidate search unit 52 searches the phrase dictionary unit using a character string generated using the input character string and the erroneous input related character, and corrects the erroneous input of the input character string. It functions as a search unit that extracts words corresponding to correction candidates. The predetermined user operation in the first embodiment is an operation of deleting a character from the end of the character string for which the input has been accepted. The deletion operation is accepted via a predetermined key (such as a clear key) of the operation keys 14 as a deletion operation unit. The correction candidate search unit 52 determines the first character of the deleted character string as an erroneous input related character. The correction candidate search unit 52 searches for correction candidates when an operation for deleting characters from the end is detected. On the other hand, the prediction conversion candidate search unit 51 searches for a prediction conversion candidate each time a character string is input or when a cursor movement operation is performed according to the position of the cursor.

  The candidate sorting unit 53 sorts the display order of the prediction conversion candidates and the correction candidates based on the data stored in the priority evaluation data storage unit 63 of the storage unit 40. The candidate sorting unit 53 reduces the number of candidate selection operations of the user by sorting the prediction conversion candidates and the correction candidates so that the candidate conversion candidates and the correction candidates are preferentially displayed from the candidates that are frequently used by the user. Further, the candidate sorting unit 53 displays the sorted words as candidates on the main display 21 as a display unit so as to be selectable.

  The word dictionary unit 61 of the storage unit 40 is a dictionary in which a plurality of words used when searching for prediction conversion candidates and correction candidates are registered, and stores these words in association with readings. The word registered in the word dictionary unit 61 includes not only a word as a minimum unit of a phrase but also a word as a break of the phrase.

  The input character string / history storage unit 62 stores the character string before the conversion operation input by the user. Hereinafter, the input character string before the conversion operation is referred to as “input character string”. The stored input character string is updated each time a character input operation and a deletion operation are performed by the user. The input character string / history storage unit 62 stores the character string before the deletion as the input history when the last character string is deleted from the character string before the conversion operation input by the user.

  The priority evaluation data storage unit 63 stores data related to the frequency of use of the words / phrases stored in the word dictionary unit 61 by the user.

  Next, details of the correction candidate display process will be described using a flowchart.

  FIG. 3 is a flowchart for explaining a correction candidate display process executed by the mobile phone 1 according to the first embodiment.

  In step S1, the correction candidate search unit 52 determines whether or not the last character in the input character string has been deleted. If it is determined that the last character string has been deleted (YES in step S1), the correction candidate search unit 52 determines whether or not the correction candidate display mode is in effect in step S2. The “correction candidate display mode” is a mode set when correction candidate display processing is executed and correction candidates are being displayed. When it is determined that the correction candidate search unit 52 is in the correction candidate display mode (YES in step S2), the correction candidate search unit 52 proceeds to input character string update step S5.

  On the other hand, if it is determined that the correction candidate display mode is not set (NO in step S2), the correction candidate search unit 52 sets the correction candidate display mode in step S3. The correction candidate display mode is set in step S3 when the last character string of the input character string is deleted first.

  In step S4, the correction candidate search unit 52 stores the input character string immediately before deletion in the input character string / history storage unit 62 as an input history. In step S5, the correction candidate search unit 52 updates the input character string stored in the input character string / history storage unit 62 to a new input character string immediately after deletion.

  In step S6, the correction candidate search unit 52 generates a correction candidate based on the input history and the input character string stored in the input character string / history storage unit 62. Details of the correction candidate generation process will be described later with reference to FIGS. The correction candidate search unit 52 registers the generated correction candidates in a predetermined storage area.

  In step S7, the candidate sorting unit 53 sorts the display order of the correction candidates based on the data stored in the priority evaluation data storage unit 63, and causes the main display 21 to display the correction candidates in the sorted order. . The candidate sorting unit 53 performs sorting in order from correction candidates that are most likely to be selected by the user, based on data regarding general usage frequency and accumulated user usage frequency. In step S8, the control unit 39 displays the character string after the deletion operation is accepted in step S1 on the main display 21.

  On the other hand, if the correction candidate search unit 52 determines in step S1 that the last character has not been deleted (NO in step S1), that is, if a new character input operation has been performed, the correction candidate display is performed in step S9. Cancel the mode setting. If the correction candidate display mode is not set in step S9, the process proceeds to step S10 without any particular cancellation.

  In step S10, the correction candidate search unit 52 updates the input character string stored in the input character string / history storage unit 62 to a new input character string based on the new character input operation. Thereafter, the process proceeds to step S8, where the control unit 39 displays a new input character string on the main display 21.

  Next, the correction candidate generation process in step S6 of the correction candidate display process will be described. The correction candidate generation process has three generation processes. The three correction candidate generation processes may be executed in parallel, or any one or two processes may be executed.

  The first generation process is a process for generating an effective correction candidate when an erroneous input mode is input as an extra character that is not intended to be input as a character determined as an erroneous input related character. Hereinafter, the first generation process is referred to as “deletion correction candidate generation process”, and the candidate generated by the first generation process is referred to as “deletion correction candidate”.

  FIG. 4 is a flowchart for explaining deletion correction candidate generation processing executed by the mobile phone 1 in the first embodiment.

  In step S11, the correction candidate search unit 52 generates a character string to be searched. The search target character string includes the current input character string (character string after deletion) stored in the input character string / history storage unit 62 and the characters after the first character of the deleted character string (character). It is a character string consisting of columns. That is, the search target character string is a character string in which the first character of the deleted character string determined as the erroneous input related character is deleted from the character string as the input history.

  The deleted character string is obtained by comparing the input history stored in the input character string / history storage unit 62 with the input character string. In other words, a deleted character string can be obtained by removing the input character string from the input history.

  In step S <b> 12, the correction candidate search unit 52 searches the word dictionary unit 61 for a word whose reading is a character string obtained as a search target. The correction candidate search unit 52 registers the searched word in a predetermined storage area.

  Next, the second generation process will be described.

  The second generation process is a process of generating an effective correction candidate when the mode of erroneous input is not input immediately before the character that is intended to be input is determined as the erroneous input related character. Hereinafter, the second generation process is referred to as “insertion correction candidate generation process”, and the candidate generated by the second generation process is referred to as “insertion correction candidate”.

  FIG. 5 is a flowchart for explaining insertion correction candidate generation processing executed by the mobile phone 1 in the first embodiment.

  In step S21, the correction candidate search unit 52 generates a character string to be searched. The character string to be searched includes the current input character string (character string after deletion) stored in the input character string / history storage unit 62, an arbitrary character string (character), and the deleted character string (character ). In other words, the character string to be searched is arbitrarily set between the character string in front of the first character of the deleted character string determined as the erroneous input related character in the input history and the erroneous input related character. This is a character string in which the character string (character) is inserted.

  This arbitrary character string (character) is not particularly limited as long as the number of characters is one or more. However, in consideration of search processing efficiency, it is preferable to limit the number of characters to two characters or less.

  In step S <b> 22, the correction candidate search unit 52 searches the word dictionary unit 61 for a word whose reading is a character string obtained as a search target. The correction candidate search unit 52 registers the searched word in a predetermined storage area.

  Next, the third generation process will be described.

  The third generation process is a process of generating an effective correction candidate when an erroneous input mode is input as a character that is different from a character string intended for input, as a character determined as an erroneous input related character. Hereinafter, the third generation process is referred to as “change correction candidate generation process”, and the candidate generated by the third generation process is referred to as “change correction candidate”.

  FIG. 6 is a flowchart for explaining change / modification candidate generation processing executed by the mobile phone 1 according to the first embodiment.

  In step S31, the correction candidate search unit 52 generates a character string to be searched. The character string to be searched includes the current input character string (character string after deletion) stored in the input character string / history storage unit 62 and the first character of the deleted character string (character). This is a character string composed of an arbitrary different character string (character) and characters after the first character of the deleted character string (character). In other words, the search target character string is a character string that is changed to an arbitrary character string (character) in which the first character of the deleted character string determined as an erroneous input related character in the input history is different. is there.

  An arbitrary character string (character) among the character strings to be searched is not particularly limited as long as the number of characters is one or more. However, in consideration of search efficiency, it is preferable to limit the number of characters to 2 characters or less.

  In step S <b> 32, the correction candidate search unit 52 searches the word dictionary unit 61 for a word that reads the character string obtained as a search target. The correction candidate search unit 52 registers the searched word in a predetermined storage area.

  Next, a specific character input operation example is applied to each process of FIGS.

  As an example of a character input operation, a correction candidate display process that is performed after the character string “Tokusoku” is input will be described.

  The correction candidate search unit 52 sets the correction candidate display mode when detecting that the user has deleted the last character “RU” in the character string (YES in step S1 in FIG. 3) (step S3). ). Immediately after the input of the character string “Tokusoku”, since the last character has not yet been deleted, the correction candidate display mode is not set (NO in step S2).

  The correction candidate search unit 52 stores the input character string “deleting” before deletion in the input character string / history storage unit 62 as an input history (step S4). Further, the correction candidate search unit 52 updates the input character string stored in the input character string / history storage unit 62 to the input character string “Tokukusu” after deletion (step S5).

  When the correction candidate generation process of FIGS. 4 to 6 is performed by the correction candidate search unit 52 and a predetermined correction candidate is generated (step S6), the candidate sorting unit 53 displays the correction candidates on the main display 21 (step S7). ). In addition, the control unit 39 displays the input character string “Tokukusu” on the main display 21 (step S8).

  Further, when the character “s” and the character “s” are further deleted from the current input character string “Tokusou” (YES in Step S1), each process of Steps S2 to S8 is performed.

  FIG. 7 is a screen transition diagram for explaining an example of character input operation to which the processes of FIGS. 3 to 6 are applied.

  As shown in FIG. 7A, the character string “Tokusoku” is input from the user via the operation keys 14 and displayed in the character editing area 70. The underline of the character string in the figure indicates that the character string conversion has not been finalized. After that, the three characters from the character string “Tokusoku” to the character string “To” are sequentially deleted. At this time, the input history stored in the input character string / history storage unit 62 is the character string “Toku”, and the input character string is the character string “Toku”.

  Since “Yes” is deleted from the character string “Toku-Soku”, the correction candidate search unit 52 determines the first character “O” as the erroneous input related character in the deleted character string. The correction candidate search unit 52 regards that the deletion of the character string “do” is an operation performed to correct an erroneous input related to the character “o”, and each of the correction candidate generation processes shown in FIGS. I do.

  FIG. 8 is a conceptual diagram illustrating the correction candidate generation process of FIGS.

  The correction candidate search unit 52 acquires a character string deleted from the input history in order to generate a character string to be searched. As shown in FIG. 8A, the character string “Toku-to” as the input history is compared with the character string “Toku-to” as the input character string (step S41), and the deleted character string “to” is selected. "Is acquired (step S42). Further, the correction candidate search unit 52 acquires the first character string “o” of the deleted character string as the erroneous input related character (step S43), and the first and subsequent characters of the deleted character string. Is acquired (step S44).

  The correction candidate search unit 52 performs the deletion correction candidate generation process (FIG. 4) assuming that the erroneous input mode is an extra character that is not intended to be input. As shown in FIG. 8B, the correction candidate search unit 52 performs a character string “Toku” consisting of the current input character string “Toku” and the character string “Yes” after the first second character of the deleted character string. "As a search target character string (step S11 in FIG. 4), and the word dictionary unit 61 is searched for a phrase that reads the character string" Tokuru "(step S12 in FIG. 4). As a result, the phrase “obtain” or the like is extracted as a deletion correction candidate.

  Further, the correction candidate search unit 52 assumes that the character string (character) intended to be input is not input before the erroneous input related character “O” as the mode of erroneous input, and insert correction candidate generation processing (FIG. 5) is performed. As shown in FIG. 8C, the correction candidate search unit 52 performs the current input character string “Toku”, an arbitrary character string (character string “Bai” in the figure), and the deleted character string “ Is generated as a character string to be searched (step S21 in FIG. 5), and the character string “Toku + arbitrary character string + do” is read as a word. Is searched from the word dictionary unit 61 (step S22 in FIG. 5). As a result, the word “sale” is extracted as an insertion correction candidate.

  Further, the correction candidate search unit 52 performs the change correction candidate generation process (FIG. 6) assuming that the erroneous input mode is a character different from the character string intended for input by the erroneous input related character “O”. . As shown in FIG. 8D, the correction candidate search unit 52 deletes the current input character string “TOKU” and an arbitrary character string (character “MO” in the figure) different from the erroneous input related character. 6 is generated as a character string to be retrieved (step S31 in FIG. 6). ), The word dictionary unit 61 is searched for a phrase that reads the character string “Toku + an arbitrary character string different from the character“ to ”+ do” (step S32 in FIG. 6). As a result, the phrase “to obtain” is extracted as a change correction candidate.

  When a character string is input and deleted, the predictive conversion candidate search unit 51 appropriately searches for a predictive conversion candidate corresponding to the input character string and the position of the cursor.

  FIG. 7B is an example of a screen on which the searched correction candidates are displayed. For example, a correction candidate display area 71 for displaying correction candidates is provided at the lower left of the character input screen, and a prediction conversion candidate display area 72 for displaying prediction conversion candidates is provided at the lower right of the screen. The correction candidates and the prediction conversion candidates are displayed in the respective areas so that they can be identified by the user. In addition to providing the correction candidate display area 71 and the prediction conversion candidate display area 72 separately, the correction candidate display area 71 and the prediction conversion candidate display area 72 are switched with tabs, or the correction candidate display area 71 and the prediction conversion candidate display are displayed. A prediction conversion candidate and a correction candidate having different character colors and character background colors can be mixed and displayed without being distinguished from the region 72.

  When the user selects a desired candidate, the input character string “Toku” is converted into the selected word and the input is confirmed. For example, as shown in FIG. 7C, when the correction candidate “Get” is selected, the character string “Toku” in the character editing area 70 is converted to the word “Get” and displayed.

  According to this mobile phone 1, when a user deletes a character string from the end, it is determined that an erroneous input is included in the character string, and the input includes a character string that has already been deleted. The correction candidates are searched and extracted in consideration of the history. For this reason, the correction candidate which a user intends can be shown more, and the operativity of the user at the time of incorrect input correction can be improved.

  In particular, when an incorrect input of one character is included in the middle of the input character string, when the correction is performed by sequentially deleting from the last character, this deleted character string is taken into account when searching for correction candidates. Therefore, it is possible to present a correction candidate that more reflects the user's intention.

  Here, when the above-described toggle input method is used as the input method using the operation keys 14, “A” and “I” are determined according to the number of times the operation key 14 to which each line of the Japanese syllabary is assigned is pressed. ”,“ Udan ”,“ Edan ”, and“ Odan ”. For this reason, when a stage before the stage originally intended for input is input, it may be noticed that an erroneous input is included in the already input character string. For example, the character intended to be input is a “stage” character that is input by pressing the operation key 14 twice. However, when the user presses the operation key 14 once, the input is incorrect. Is assumed to have been included.

  In the mobile phone 1 according to the first embodiment, in particular, assuming that the last character of the input character string is another character that can be input with the same operation key 14 as the character intended by the user, It is also possible to appropriately display the correction candidate desired by the user. Specifically, when searching for correction candidates, a word or phrase that reads a character string whose last character in the input history matches the character in the other row of the same line is included in the search target.

  Hereinafter, as an example, the change correction candidate generation process for generating a change correction candidate will be described. However, a deletion correction candidate generation process for generating a deletion correction candidate or an insertion correction candidate generation process for generating an insertion correction candidate may be applied.

  FIG. 9 is a flowchart for explaining another change correction candidate generation process executed by the mobile phone 1 according to the first embodiment.

  In step S51, the correction candidate search unit 52 generates a character string to be searched. The character string to be searched includes the current input character string (character string after deletion) stored in the input character string / history storage unit 62 and the first character of the deleted character string (character). It is a character string consisting of a different arbitrary character string (character), a character string from the first character to the second character from the end of the deleted character string (character), and a character on the same line as the last character. . That is, the character string to be searched is changed to an arbitrary character string (character) in which the first character of the deleted character string determined as an erroneous input related character with respect to the input history is different, and the input history It is a character string whose last character is the same line.

  In step S <b> 52, the correction candidate search unit 52 searches the word dictionary unit 61 for a word whose reading is a character string obtained as a search target. The correction candidate search unit 52 registers the searched word in a predetermined storage area.

  FIG. 10 is a conceptual diagram illustrating the change correction candidate generation process of FIG. 9 in consideration of the toggle input method.

  In FIG. 10, the user intends to input the phrase “fun” with the character string “enjoyment” as a reading, but the character “ta” is mistaken for the character “ka” during the input of the last character “mi”. This is an example in which correction is started when the character “Okashima” is input as a result of recognizing that it has been input. Note that the last character “ma” is a character that is input when the operation key 14 assigned “ma line” is pressed once in the toggle input method, and the character “mi” is “ma line”. This is a character that is input when the assigned operation key 14 is pressed twice.

  The user performed an operation of deleting “Kanoshima” from the character string “Okashima”. The correction candidate search unit 52 determines that the first character “ka” of the deleted character string is the erroneous input related character because “kashima” is deleted from the character string “okashima”. The correction candidate search unit 52 regards the deletion of the character string “Kanoshima” as an operation performed to correct an erroneous input related to the character “ka”, and performs a change correction candidate generation process.

  The correction candidate search unit 52 acquires a character string deleted from the input history in order to generate a character string to be searched. As shown in FIG. 10A, the character string “Okashima” as the input history is compared with the character “O” that is the input character string (step S61), and the deleted character string “Kanoshima” is compared. "Is acquired (step S62). Further, the correction candidate search unit 52 acquires the first character string “ka” of the deleted character string as the erroneous input related character (step S63), and the second and subsequent characters of the deleted character string. Is acquired (step S64). Further, the correction candidate search unit 52 acquires the last character “MA” and the other character string “NOSH” from the character string “NOSHIMA” after the first second character of the deleted character string ( Steps S65 and S66).

  The correction candidate search unit 52 selects the current input character string “o”, an arbitrary character string that is different from the erroneous input related character (character “ta” in the figure), and the last character from the first character of the deleted character string. The character string “Noshi” up to the second character and the character string “Toku” + “Character” consisting of the characters “M”, “M”, “M”, “M”, “M” on the same line as the last character. A phrase that reads “an arbitrary character string different from“ + ”and“ line character ”is generated as a search target, and a change correction candidate is searched from the word dictionary unit 61 (step S52 in FIG. 9). As a result, as shown in FIG. 10B, the phrase “enjoyment” or the like that reads the character string “enjoyment” is searched for as a modification candidate.

  Since this mobile phone 1 searches for correction candidates in consideration of the input process unique to the toggle input method, it is possible to more suitably search for and present correction candidates desired by the user.

  Next, a description will be given of a process for regenerating and displaying a correction candidate that is likely to be selected by the user when a character input operation is performed by the user while the correction candidate is being displayed in the correction candidate display mode.

  FIG. 11 is a flowchart for explaining a correction candidate redisplay process executed by the mobile phone 1 according to the first embodiment.

  Among the correction candidate redisplay processes, the deletion determination step S71 to the character string display step S78 are substantially the same as the deletion determination step S1 to the character string display step S8 in FIG.

  If the correction candidate search unit 52 determines in step S71 that the last character has not been deleted (NO in step S71), the correction candidate search unit 52 is in the correction candidate display mode and a character is input by the user in step S79. It is determined whether or not. If it is determined that the correction candidate search unit 52 is not in the correction candidate display mode or that character input has not been performed (NO in step S79), for example, if an input character string confirmation operation has been performed, in step S80 Cancel the correction candidate display mode. Since the mode cancellation step S80 and the update step S81 are substantially the same processing as the mode cancellation step S9 and the update step S10 of FIG. 3, detailed description thereof is omitted.

  On the other hand, when it is determined that the correction candidate search unit 52 is in the correction candidate display mode and character input has been performed (YES in step S79), the correction candidate is regenerated in consideration of the character input in step S82. To do. Specifically, the correction candidate search unit 52 searches the correction word / phrase generated by the insertion correction candidate generation process of FIG. 5 and the change correction candidate generation process of FIG. Regenerate as a correction candidate. When the character string is deleted in order from the end and then another character is input, the input error is when the character intended for input has not been input or when a character different from the character string intended for input is input. Since it can be assumed that this is the case, the correction candidate search unit 52 regenerates correction candidates from only the insertion correction candidates and change correction candidates corresponding to these erroneous input modes.

  In step S83, the candidate sorting unit 53 sorts the display order of the correction candidates based on the data stored in the priority evaluation data storage unit 63, and displays the correction candidates on the main display 21 in the sorted order. . Thereafter, the process proceeds to the character string display step S78.

  Next, a specific character input operation example is applied to a part of the correction candidate redisplay process in FIG.

  FIG. 12 is a screen transition diagram illustrating an example of character input operation to which the correction candidate redisplay process of FIG. 11 is applied.

  As shown in FIG. 12 (A), after the character string “Tokusuru” is input, the character string “Yes” is sequentially deleted, and the processing from the deletion determination step S71 to the character string display step S78 is performed. . Then, as shown in FIG. 12B, correction candidates, prediction conversion candidates, and input character strings for the character string “Toku” are displayed.

  At this time, the user who intends to input the phrase “special sale” inputs the character “BA” following the character string “TOKU”. The correction candidate search unit 52 determines that a character has been input during the correction candidate display mode (YES in step S79), and regenerates the correction candidate in consideration of the input character “B” (step S82). . Specifically, a phrase corresponding to an insertion correction candidate that reads the current input character string “Toku + newly entered character“ B ”+ arbitrary character string + do”, and “Toku + newly entered” The word corresponding to the change / candidate with the reading “let's“ letter ”+ an arbitrary character string different from the character“ ha ”+” is regenerated as a correction candidate. As for the change correction candidate, the newly input character “BA” can be regarded as an arbitrary character string different from the character “NO”.

  For example, as shown in FIG. 12C, the phrase “sale” already generated as an insertion correction candidate is generated as a corresponding correction candidate. Further, it is assumed that the regenerated correction candidate is likely to be selected, and is displayed with priority over other candidates. In the example of FIG. 12C, the correction candidates are displayed in the upper stage, and the prediction conversion candidates are displayed in the lower stage, thereby reducing the user operation when selecting candidates.

  In this correction candidate redisplay process, after a character string is deleted, if a character is further input, the correction candidate is regenerated in consideration of the input character, thereby facilitating the user's correction operation. Can be done.

[Second Embodiment]
2nd Embodiment of the information processing apparatus which concerns on this invention is described based on an accompanying drawing.

  In the correction candidate display process in the first embodiment, the deletion operation from the end of the character string is applied to the user operation for determining the erroneous input related character from the input character string. On the other hand, the correction candidate display process in the second embodiment applies a designation operation using a cursor that moves on a character string to a user operation for determining an erroneous input related character from the input character string.

  FIG. 13 is a schematic functional block diagram showing a second embodiment of the information processing apparatus according to the present invention.

  In addition, about the structure and part corresponding to 1st Embodiment, the same code | symbol is attached | subjected and the overlapping description is abbreviate | omitted.

  In the second embodiment, the predictive conversion candidate search unit 101, the correction candidate search unit 102, and the candidate sort unit 103 are realized by the control unit 120 executing a predetermined program. The storage unit 121 is provided with a word / idiom dictionary unit 111, a symbol / pictogram dictionary unit 112, and a priority evaluation data storage unit 113.

  The predictive conversion candidate search unit 101 searches the word / idiom dictionary unit 111 of the storage unit 121 for words that are predicted to be converted for the input character string. Predictive conversion candidates refer to words and phrases made up of kanji characters that read the input character string, words that match the first reading.

  The correction candidate search unit 102 functions as a determination unit that determines an erroneous input related character that is a character related to an erroneous input from a character string based on a predetermined user operation. Further, the correction candidate search unit 102 searches the phrase dictionary unit using a character string generated using the input character string and the erroneous input related character, and corrects the erroneous input of the input character string. It functions as a search unit that extracts words corresponding to correction candidates. The predetermined user operation in the second embodiment is a cursor movement operation for designating any character in the character string for which input has been accepted. The movement operation of the cursor is accepted through the cross key 16 of the operation keys 14, for example. The correction candidate search unit 102 determines the character designated by the cursor as the erroneous input related character.

  The candidate sorting unit 103 sorts the display order of the prediction conversion candidate and the correction candidate based on the data stored in the priority evaluation data storage unit 113 of the storage unit 121. The candidate sorting unit 103 sorts the prediction conversion candidate and the correction candidate so that they are displayed preferentially from the candidates that are frequently used by the user.

  The word / idiom dictionary unit 111 of the storage unit 121 is a dictionary in which words used when searching for prediction conversion candidates and correction candidates are registered, and stores various words in association with readings. The word / idiom dictionary unit 111 is a dictionary in which idioms used when searching for correction candidates are registered. The idiom registered in the word / idiom dictionary unit 111 is a phrase composed of two or more kanji characters or a phrase that is used as a single word by combining two or more words. A word is a minimum unit word composed of kanji and kana. Further, similar words / idioms stored in the word / idiom dictionary unit 111 are registered in association with each other. For example, words classified by genre as particles such as “no”, “ni”, “ha”, “ha” are registered in association with each other as similar words.

  The symbol / pictogram dictionary unit 112 is a dictionary in which symbols and pictograms are registered. The symbol is, for example, a kanji character such as a parenthesis, an exclamation mark, an arithmetic symbol, a kana character, or a character / sign other than a number. The pictogram is a visual symbol that is used in place of a character or symbol and is used supplementarily when emphasizing emotions. Symbols and pictograms are registered, for example, by genre according to their similar meaning contents.

  The priority evaluation data storage unit 113 stores data related to the frequency of use of the words / phrases stored in the word / idiom dictionary unit 111 and the symbol / pictogram dictionary unit 112.

  Next, processing from when a character input operation is performed until a confirmation operation is performed will be described.

  FIG. 14 is a flowchart illustrating the character input process before the conversion confirmation operation executed by the mobile phone 100 according to the second embodiment.

  In step S <b> 101, the control unit 120 accepts input of characters from the operation keys 14. When a character input is accepted, for example, the cursor operated with the cross key 16 can be moved between the input character strings before the conversion confirmation operation is performed.

  In step S102, the correction candidate search unit 102 displays correction candidates corresponding to the character designated by the cursor. Details of the correction candidate display process in step S102 will be described later with reference to FIG.

  In step S103, the correction candidate search unit 102 determines whether one correction candidate is selected from the displayed correction candidates. If the correction candidate search unit 102 determines that the correction candidate has been selected (YES in step S103), the process proceeds to step S106.

  On the other hand, if the correction candidate search unit 102 determines that no correction candidate has been selected (NO in step S103), in step S104, the correction candidate search unit 102 determines whether a cursor movement operation has been accepted between character strings. . If it is determined that the cursor movement operation has been accepted (YES in step S104), the correction candidate search unit 102 returns to the correction candidate display step S102 and displays the correction candidate corresponding to the designated character.

  On the other hand, when the correction candidate search unit 102 determines that the cursor movement operation has not been received (NO in step S104), in step S105, the correction candidate search unit 102 determines whether or not an input confirmation operation for the input character string has been received. Do. If the correction candidate search unit 102 determines that the confirmation operation has not been accepted (NO in step S105), that is, if the confirmation operation has not been performed on the input character string and further character input has been accepted, character input Returning to step S101, the subsequent processing is repeated.

  On the other hand, if the correction candidate search unit 102 determines that the confirmation operation has been accepted (YES in step S105), in step S106, the control unit 120 converts the input character string and the character string into the selected correction candidate. The converted character string and correction candidates are output to the main display 21.

  Next, the details of the correction candidate display step S102 in the character input process of FIG. 14 will be described.

  FIG. 15 is a flowchart for explaining the correction candidate display process before the conversion confirmation operation executed by the mobile phone 100 according to the second embodiment.

  In step S111, the correction candidate search unit 102 determines whether or not the character designated by the cursor is the last character in the input character string. Note that the cursor specifies the input character immediately after the character is input, and if the cursor moves one direction toward the beginning of the character string, specifies the second character from the end. . When it is determined that the designated character is the last character (YES in step S111), the correction candidate search unit 102 proceeds to the insertion correction candidate generation step S113.

  On the other hand, when the correction candidate search unit 102 determines that the designated character is a character other than the last character (NO in step S111), in step S112, the character designated by the cursor (character related to erroneous input) is determined. Based on this, a deletion correction candidate for the input character string (character) before the conversion is determined is generated. In step S113, the correction candidate search unit 102 generates an insertion correction candidate for the input character string (character) before the conversion is confirmed based on the designated character. Furthermore, in step S114, the correction candidate search unit 102 generates change correction candidates for the input character string (character) before the conversion is confirmed based on the designated character. The correction candidate search unit 102 registers each correction candidate generated in steps S112 to S114 in a predetermined storage area.

  Details of the deletion correction candidate generation process in step S112, the insertion correction candidate generation process in step S113, and the change correction candidate generation process in step S114 will be described later with reference to FIGS.

  In step S115, the candidate sorting unit 103 sorts the display order of the correction candidates based on the data stored in the priority evaluation data storage unit 113, and displays the correction candidates on the main display 21 in the sorted order. . The candidate sorting unit 103 performs sorting in order from correction candidates that are most likely to be selected by the user, based on data relating to general usage frequency and accumulated user usage frequency.

  Next, each correction candidate generation process in steps S112 to S114 in the correction candidate display process of FIG. 15 will be described. Each correction candidate generation process includes three generation processes corresponding to the deletion correction candidate generation process (FIG. 4), the insertion correction candidate generation process (FIG. 5), and the change correction candidate generation process (FIG. 6) described in the first embodiment. Have In the following description, an example in which three correction candidate generation processes are sequentially executed will be described. However, these processes may be executed in parallel, or only one or two processes are executed. Also good.

  FIG. 16 is a flowchart for explaining deletion correction candidate generation processing executed by the mobile phone 100 according to the second embodiment.

  In step S121, the correction candidate search unit 102 generates a character string to be searched. The character string to be searched is the character string consisting of the character string (character) in front of the character specified with the cursor and the character string (character) after the character specified with the cursor. Is a column. That is, the character string to be searched is a character string in which the character specified by the cursor determined as the erroneous input related character is deleted from the input character string.

  In step S <b> 122, the correction candidate search unit 102 searches the word / idiom dictionary unit 111 for a phrase that reads a character string obtained as a search target. The correction candidate search unit 102 registers the searched word / phrase in a predetermined storage area.

  Next, the insertion correction candidate generation process in step S113 in the correction candidate display process will be described.

  FIG. 17 is a flowchart for explaining insertion correction candidate generation processing executed by the mobile phone 100 according to the second embodiment.

  In step S131, the correction candidate search unit 102 generates a character string to be searched. The character string to be searched includes the character string in front of the character specified with the cursor, the arbitrary character string (character), and the character string after the character specified with the cursor. A character string consisting of In other words, the character string to be searched is a character in which an arbitrary character string (character) is inserted between the character string in front of the character specified by the cursor determined as the erroneous input related character and the erroneous input related character. Is a column.

  In step S132, the correction candidate search unit 102 searches the word / idiom dictionary unit 111 for a phrase that reads the character string obtained as a search target. The correction candidate search unit 102 registers the searched word / phrase in a predetermined storage area.

  Next, the modification / modification candidate generation process in step S114 in the modification candidate display process will be described.

  FIG. 18 is a flowchart for explaining change / modification candidate generation processing executed by the mobile phone 100 according to the second embodiment.

  In step S141, the correction candidate search unit 102 generates a character string to be searched. The search target character string is the character string in front of the character specified with the cursor, any character string (character) that is different from the character specified with the cursor, and the character string after the character specified with the cursor. A character string consisting of That is, the character string to be searched is a character string that is changed to an arbitrary character string (character) that is different from the character specified by the cursor that is determined as an erroneous input related character with respect to the input character string.

  In step S142, the correction candidate search unit 102 searches the word / idiom dictionary unit 111 for a phrase that reads the character string obtained as a search target. The correction candidate search unit 102 registers the searched word / phrase in a predetermined storage area.

  Next, a specific character input operation example is applied to each process of FIGS.

  As an example of the character input operation, a correction candidate display process that is executed after the character string “Konchiha” is input will be described.

  After the user receives an input of the character string “Konchiha” via the operation key 14 (step S101 in FIG. 14), the correction candidate search unit 102 displays a correction candidate corresponding to the character specified by the cursor. (Step S102).

  Immediately after the input of the character string “Kon-ha”, the character designated by the cursor is the last character “ha” (YES in step S111 in FIG. 15), so that the correction candidate search unit 102 designates the character “ Is determined to be an erroneous input-related character. The correction candidate search unit 102 generates an insertion correction candidate and a change correction candidate for correcting an erroneous input related to the character “ha” (steps S113 and S114). The candidate sorting unit 103 displays the correction candidates generated by the correction candidate search unit 102 on the main display 21 (step S115).

  When the displayed correction candidate is not selected (NO in step S103 in FIG. 14), and the cursor is moved toward the head of the character string “Kon-ha” and the character “C” is designated (YES in step S104). Each process after the correction candidate display step S102 of FIG. 14 is performed.

  Since the designated character is not the last character (step S111 in FIG. 15), the correction candidate search unit 102 deletes and corrects to correct the erroneous input related to the character “Chi” determined as the erroneous input related character. Candidates, insertion correction candidates, and change correction candidates are generated (steps S112 to S114).

  The correction candidate search unit 102 performs deletion correction candidate generation processing (FIG. 16) assuming that the erroneous input mode is an extra character that is not intended to be input. The correction candidate search unit 102 includes a character string “ha” including a character string “kon” in front of “chi” designated by the cursor as an erroneous input related character and a character “ha” behind the erroneous input related character “chi”. A phrase that reads “Konha” is generated as a character string to be searched (step S121), and a phrase that reads the character string “Konha” is searched from the word / idiom dictionary unit 111 (step S122). As a result, the phrase “haha” or the like is searched for as a deletion correction candidate.

  Further, the correction candidate search unit 102 assumes that the character string (character) intended to be input before the erroneous input related character “Chi” has not been input as an erroneous input mode, and insert correction candidate generation processing (FIG. 17) is performed. The correction candidate search unit 102 uses a character string “kon” in front of the erroneous input related character “chi”, an arbitrary character string (character), and a character string “chiha” after the erroneous input related character “chi”. A phrase that reads “Kon + arbitrary character string + chiha” as a search target character string is generated as a search target character string (step S131), and a phrase that reads the character string “kon + arbitrary character string + chiha” is read. Is searched from the word / idiom dictionary unit 111 (step S132). As a result, such as "Today" phrase to read "Hello" is searched as the insertion suggestions.

  Further, the correction candidate search unit 102 assumes that the erroneous input mode is a character that is different from the character string (character) that the erroneous input-related character “Chi” is intended to input, and changes correction candidate generation processing (FIG. 18). )I do. The correction candidate search unit 102 includes a character string “kon” in front of the erroneous input related character “chi”, an arbitrary character string (character) different from the erroneous input related character “chi”, and an erroneous input related character “chi”. A phrase that reads the character string “Kon + any character string + ha different from the character“ Chi ”+” is generated as a search target character string (Step S141). The word / idiom dictionary unit 111 is searched for a phrase that reads “an arbitrary character string + ha different from the character“ chi ”” (step S142). As a result, the phrase “this time” with “Kondoha” as a change correction candidate is searched.

  When a character string is input, the prediction conversion candidate search unit 101 appropriately searches for a prediction candidate corresponding to the input character string and the position of the cursor.

  FIG. 19 is a screen transition diagram illustrating an example of a character input operation to which each process of FIGS. 14 to 18 is applied.

  As shown in FIG. 19A, for example, a correction candidate display area 71 for displaying correction candidates is provided at the lower left of the character input screen, and a prediction conversion candidate display area 72 for displaying a prediction conversion candidate at the lower right of the screen. Is provided. The correction candidate and the predictive conversion candidate are displayed in the respective areas so that the user can identify them. In addition to changing the display area, the display of the correction candidate and the prediction conversion candidate can be switched by using a tab, or the prediction conversion candidate and the correction candidate can be mixed and displayed in different colors.

  When the user selects a desired candidate (YES in step S103 in FIG. 14), the input character string “Konchiha” is converted into the selected conversion candidate. For example, when the correction candidate “Today is” is selected from the correction candidate display area 71, as shown in FIG. 19B, the input character string “Konha” in the character editing area 70 is the phrase “Today”. And is output to the main display 21 (step S106).

  According to the mobile phone 100, even if an input character includes an erroneous input, the erroneous input is corrected only by a cursor operation without requiring a delete operation and a new character input operation. be able to.

  The correction candidate display process of FIG. 14 described above has been described by applying the case where it is performed before the input character string conversion finalizing operation. However, as another example of the correction candidate display process, it may be performed after the conversion confirmation operation.

  First, the character input process after the character string conversion confirmation operation will be described.

  FIG. 20 is a flowchart illustrating the character input process after the conversion confirmation operation executed by the mobile phone 100 according to the second embodiment.

  In step S151, the control unit 120 accepts an input of a sentence. Note that the term “sentence” here refers to a character string in which a conversion operation and a confirmation operation are performed after the character string is input. Further, the sentence includes not only characters composed of kana and kanji but also symbols and pictograms input through a predetermined operation.

  In step S152, the correction candidate search unit 102 accepts a cursor movement operation on the sentence. The correction candidate search unit 102 detects a character designated according to the cursor movement operation as an erroneous input related character. In step S153, the correction candidate search unit 102 displays correction candidates corresponding to the designated character. Details of the correction candidate display process in step S153 will be described later with reference to FIG.

  In step S154, the correction candidate search unit 102 determines whether one correction candidate is selected from the displayed correction candidates. If the correction candidate search unit 102 determines that a correction candidate has been selected (YES in step S154), in step S155, the control unit 120 finalizes the conversion to the selected correction candidate and displays the correction candidate in the main display. To 21.

  On the other hand, if the correction candidate search unit 102 determines that no correction candidate has been selected (NO in step S154), in step S156, the correction candidate search unit 102 determines whether a cursor movement operation has been accepted between character strings. . If it is determined that the cursor movement operation has been accepted (YES in step S156), the correction candidate search unit 102 returns to the correction candidate display step S153 and displays the correction candidate corresponding to the designated character.

  On the other hand, if the correction candidate search unit 102 determines that a cursor movement operation has not been received (NO in step S156), the process ends.

  Next, details of the correction candidate display step S153 in the character input process of FIG. 20 will be described.

  FIG. 21 is a flowchart for explaining a correction candidate display process after the conversion confirmation operation executed by the mobile phone 100 according to the second embodiment.

  In step S161, the correction candidate search unit 102 extracts a word including the character specified by the cursor as the erroneous input related character. For example, the word is a minimum unit that can divide a sentence, and the word is extracted by referring to the word / idiom dictionary unit 111 and the symbol / pictogram dictionary unit 112. In addition to the kanji and kana characters, one symbol / pictogram and one set of symbols (such as parentheses) are treated as words. In addition, the correction candidate search unit 102 determines the character type of the extracted word. The character type of the word is determined by the character code of the erroneous input related character. Types include words, idioms, numbers, symbols, pictograms, and the like.

  In step S162, the correction candidate search unit 102 generates an insertion correction candidate based on the word and the character type including the designated character. In step S163, the correction candidate search unit 102 generates a change correction candidate based on the word and type including the designated character. The correction candidate search unit 102 registers each correction candidate generated in steps S162 and S163 in a predetermined storage area.

  In addition, although the example which produces | generates an insertion correction candidate and a change correction candidate and does not produce | generate a deletion correction candidate after conversion confirmation operation was demonstrated, you may produce | generate a deletion correction candidate as needed. Details of the insertion correction candidate generation process in step S162 and the change correction candidate generation process in step S163 will be described later with reference to FIGS.

  In step S164, the candidate sorting unit 103 sorts the display order of the correction candidates based on the data stored in the priority evaluation data storage unit 113, and causes the main display 21 to display the correction candidates in the sorted order. . The candidate sorting unit 103 performs sorting in order from correction candidates that are most likely to be selected by the user, based on data relating to general usage frequency and accumulated user usage frequency.

  Next, details of the correction candidate generation process (steps S162 and S163) will be described. The insertion correction candidate generation process of FIG. 22 and the change correction candidate generation process of FIG. 23 are mainly performed when the character type of the erroneous input related character is a kana or kanji.

  FIG. 22 is a flowchart for explaining the insertion correction candidate generation process after the conversion confirmation operation executed by the mobile phone 100 according to the second embodiment.

  In step S171, the correction candidate search unit 102 generates a word to be searched. The search target phrase consists of a character string in front of the character specified by the cursor in the extracted word, an arbitrary character string (character), and a character string after the character specified by the cursor in the word. It is a phrase. That is, the search target word is an arbitrary character string between the character string in front of the character specified by the cursor determined as the erroneous input related character and the erroneous input related character for the extracted word ( (Character).

  In step S <b> 172, the correction candidate search unit 102 searches the word / phrase dictionary unit 111 for a phrase corresponding to the phrase obtained as a search target. The correction candidate search unit 102 registers the searched word / phrase in a predetermined storage area.

  FIG. 23 is a flowchart for explaining a change correction candidate generation process after the conversion confirmation operation executed by the mobile phone 100 according to the second embodiment.

  In step S181, the correction candidate search unit 102 generates a character string to be searched. The character string to be searched is a character string in front of the character specified by the cursor in the word, an arbitrary character string (character) different from the character specified by the cursor, and the character specified by the cursor Is a character string consisting of That is, the character string to be searched is a phrase that is changed to an arbitrary character string (character) in which the character specified by the cursor determined as the erroneous input related character is different from the extracted word.

  In step S <b> 182, the correction candidate search unit 102 searches the word / idiom dictionary unit 111 for a phrase corresponding to the phrase obtained as a search target. The correction candidate search unit 102 registers the searched word / phrase in a predetermined storage area.

  FIG. 24 is a flowchart for explaining change correction candidate generation processing executed after the conversion confirmation operation executed by the mobile phone 100 according to the second embodiment.

  This change correction candidate generation process is different from the change correction candidate generation process of FIG. 23, and similar words are corrected as candidates according to the character type of the erroneous input related character determined in step S161 of the correction candidate generation process of FIG. Is generated.

  In step S191, the correction candidate search unit 102 generates a word (phrase, symbol, pictograph) to be searched. The search target word is a word that has the same type and the same meaning as the erroneous input related character that is the character specified by the cursor.

  In step S192, the correction candidate search unit 102 searches the word / idiom dictionary unit 111 or the symbol / pictogram dictionary unit 112 for a word corresponding to the word obtained as a search target. The correction candidate search unit 102 registers the searched word in a predetermined storage area. If the character type of the erroneous input related character is a kana and the word is classified by genre as a particle such as “no”, “ni”, “ha”, “wa”, the correction candidate search unit 102 The words associated as the particle particles are searched for as similar phrases. Further, if the character type of the erroneous input related character is a symbol and the word is classified into genres as parentheses such as “()” and “{}”, the correction candidate search unit 102 is associated as another parenthesis. Search for similar words.

  Next, a character input process after the conversion confirmation operation will be described by applying a specific example.

  FIG. 25 is a screen transition diagram illustrating an example of a character input operation to which each process of FIGS. 20 to 24 is applied.

  As shown in FIG. 25A, for example, after accepting an input of a sentence “Today is” (step S151 in FIG. 20), the correction candidate search unit 102 accepts a cursor movement operation on the sentence. The correction candidate search unit 102 detects that the character “now” is designated as the erroneous input related character (step S152). Then, the correction candidate search unit 102 displays a correction candidate corresponding to the designated character (step S153).

  The correction candidate search unit 102 extracts words including the erroneous input related character “now”. Further, the character type of the erroneous input related character is determined (step S161 in FIG. 21). In the example of FIG. 25A, the word “today” including the character “now” is extracted, and the character type of the character “now” is determined to be kanji. The correction candidate search unit 102 searches the word / idiom dictionary unit 111 for an insertion correction candidate and a change correction candidate for the erroneous input related character “now” of the idiom “today” (steps S162 and S163). For example, “Today”, “Tomorrow”, “Yesterday”, and the like are extracted as change correction candidates (FIG. 23). The generated correction candidates are displayed by the candidate sorting unit 103 (step S164 in FIG. 21).

  If it is determined that the correction candidate has been selected (YES in step S154 in FIG. 20), the conversion is determined and output to the selected correction candidate (step S155). FIG. 25A shows an example in which the word “today” including the erroneous input related character “now” is converted to the idiom “today” generated as a correction candidate.

  In the example of FIG. 25 (B), a pictograph indicating a star is designated as an erroneous input related character based on the cursor movement operation on the input sentence. The correction candidate search unit 102 extracts a pictograph indicating a star as a word, and determines the character type of the erroneous input related character as a pictograph (step S161 in FIG. 21). The correction candidate search unit 102 generates a word that is of the same type as the pictograph indicating the star and has a similar meaning as a search target word (step S191 in FIG. 24), and searches (step S192). For example, a pictograph indicating the moon, a pictograph indicating the sun, and the like registered in advance in the symbol / pictogram dictionary unit 112 as words similar to and similar to the pictograph indicating the star are extracted as change correction candidates. FIG. 25B shows an example in which a pictograph indicating a star as an erroneous input related character is converted into a pictograph indicating the sun, which is a pictograph generated as a correction candidate.

  Furthermore, in the example of Fig. 25C, "(" is designated as the erroneous input related character based on the cursor movement operation on the input sentence. The parenthesis “()” is extracted, and the character type of the erroneous input related character is determined as a symbol (step S161 in FIG. 21) The correction candidate search unit 102 is of the same type as the parenthesis symbol and has a similar meaning 24 is generated as a word to be searched (step S191 in FIG. 24) and is searched (step S192) For example, a symbol / pictogram dictionary unit as a word of the same type and similar to the parenthesis “()”. ““ ”(Quotes)”, “{} (curly brackets)”, etc., which are registered in 112, are searched for change correction candidates, as shown in FIG. ”But correction Examples that have been converted to the generated symbol "" "" As is illustrated.

  The mobile phone 100 can easily correct a word including a character designated as an erroneous input related character even after the conversion confirmation operation.

  In addition, since similar ones having similar meanings are registered in advance in the respective dictionary units 111 and 112 and similar words are searched from the character type when an erroneous input related character is designated, the erroneous input related You can also correct letters to the same type of words with different nuances.

  In the second embodiment, the mobile phone 100 including the operation key 14 that performs an input operation when pressed is described as an example. However, instead of the operation key 14, the cursor is directly moved to a desired position by clicking. The present invention can also be applied to an information processing apparatus that includes a touch panel that can be operated as an input device. In the case of a touch panel or the like, since there is no need to move the cursor character by character, it is possible to further easily correct an erroneous input.

  The information processing apparatus according to the present invention has been described as applied to the mobile phone 100. However, the present invention can also be applied to a PDA (Personal Digital Assistant), a personal computer, and other information processing apparatuses that perform character input operations.

DESCRIPTION OF SYMBOLS 1,100 Mobile phone 14 Operation key 15 Number key 16 Cross key 17 Confirmation key 21 Main display 39,120 Control part 40,121 Storage part 51,101 Prediction conversion candidate search part 52,102 Correction candidate search part 53,103 Candidate sort Unit 61 Word dictionary unit 62, 112 Input character string / history storage unit 63, 113 Priority evaluation data storage unit 70 Character editing region 71 Correction candidate display region 72 Predictive conversion candidate display region 111 Word / idiom dictionary unit 112 Symbol / pictogram dictionary Part

Claims (10)

A character input unit that accepts input of a character string;
A phrase dictionary part that stores a plurality of phrases in association with readings;
A determination unit that determines an erroneous input-related character that is a character related to an erroneous input from the character string based on a predetermined user operation;
It corresponds to a correction candidate that searches the phrase dictionary unit using a character string generated using the character string that has received the input and the erroneous input related character, and corrects the erroneous input of the character string that has received the input. A search unit that extracts words;
A display unit for selectively displaying the searched words / phrases ,
The information processing apparatus , wherein the search unit searches, as the correction candidate, a phrase that reads a character string in which the erroneous input related character is deleted from the character string . The predetermined user operation is a deletion operation for deleting the character string received from the end,
The information processing apparatus according to claim 1, wherein the determination unit determines the first character of the deleted character string as the erroneous input related character. The predetermined user operation is a cursor movement operation for designating any character of the character string that has received the input,
The information processing apparatus according to claim 1, wherein the determination unit determines a character designated by the cursor as the erroneous input related character. The search unit further reads a word or a character string in which an arbitrary character or a character string is inserted between the character string in front of the erroneous input related character and the erroneous input related character with respect to the character string. The information processing apparatus according to claim 1, wherein the information is searched as the correction candidate. 4. The search unit according to claim 1, further comprising: searching for a word or phrase that is read as an arbitrary character or a character string that is changed to a character string in which the erroneous input-related character is different from the character string as the correction candidate. The information processing apparatus according to any one of claims. The character input unit accepts input of the character string by a toggle input method,
The information processing apparatus according to any one of claims 1 to 3, wherein the search unit performs a search including a phrase that reads a character string whose last character is the same line as the character string that has received the input.   The information processing apparatus according to claim 2, wherein the search unit re-searches the correction candidate including the new character when receiving an input of a new character from the character input unit after the deletion operation. The predetermined user operation is a cursor movement operation for designating any character in a character string whose input is confirmed,
The determination unit determines a character designated by the cursor as the erroneous input related character,
The search unit extracts a word including the erroneous input-related character in the character string, and an arbitrary character string between the character string in front of the erroneous input-related character and the erroneous input-related character with respect to the word. The information processing apparatus according to claim 1, wherein a word or phrase including a character or a character string into which a character string is inserted is searched as the correction candidate. The predetermined user operation is a cursor movement operation for designating any character in a character string whose input is confirmed,
The determination unit determines a character designated by the cursor as the erroneous input related character,
The search unit extracts a word including the erroneous input-related character in the character string, and a word or phrase composed of an arbitrary character or a character string in which the erroneous input-related character is different from the word The information processing apparatus according to claim 1 or 2 , wherein the information is searched as the correction candidate. A character input unit that accepts input of a character string;
A phrase dictionary unit that stores a plurality of phrases that are associated with similar meanings;
A determination unit that determines an erroneous input-related character that is a character related to an erroneous input from the character string based on a predetermined user operation;
A search unit that searches the phrase dictionary unit for a phrase that has a similar meaning to the erroneous input related character as a correction candidate for correcting the erroneous input related character;
A display unit for selectively displaying the searched words / phrases ,
The information processing apparatus , wherein the search unit searches, as the correction candidate, a phrase that reads a character string in which the erroneous input related character is deleted from the character string .

Images