JP6683098B2 - Character specifying method, character specifying device, and character specifying program - Google Patents

Description

  The present invention relates to a character specifying technique.

  With the establishment of JIS (Japanese Industrial Standards) X0213 and the character information infrastructure development project promoted by the Information Processing Promotion Organization and the like, the number of characters that can be used in computers is increasing. Characters that can be used on a computer include obfuscated characters that cannot be read by a general user.

  Obfuscated characters may hinder the determination of whether or not they are the same data. For example, assume that there are two systems as shown in FIG. In system A, a name is registered in kanji that can be used by a computer. In system B, the user who registered the character cannot read the kanji of the name and cannot specify the kanji of the name by the kana-kanji conversion, so user-defined characters (also called external characters) are registered. In such a case, it is not possible to determine whether the two records relate to the same person, and thus information integration cannot be performed.

  When performing input work using a smart device such as a smartphone or a tablet, the size of the dictionary that can be installed is limited due to the small capacity of the storage device, and thus it becomes more difficult to specify the obfuscated character. In addition, since input methods using smart devices are limited, detailed and accurate specification is difficult. Further, since the screen is smaller than that of a personal computer or the like, it is difficult to determine a fine difference between kanji having many variants.

  As a technique for obtaining a target kanji, a certain document discloses the following kanji search device. Specifically, the kanji search device generates a search formula including the constituent parts of the target kanji and information on the relative positions of the constituent parts, and searches the dictionary with the generated search formulas.

JP, 9-237272, A

  The above technique is premised on that the search is performed in accordance with the intention of the dictionary creator, and a search omission occurs when the number of combinations prepared by the dictionary creator is small. The size of the dictionary will increase if all combinations are prepared to avoid omission of searches. Also, in the case of obfuscated characters, it is difficult to enter a search expression in the first place.

  An object of the present invention is, in one aspect, to provide a technique for improving the accuracy of identifying obfuscated characters while suppressing the size of a dictionary.

  A character specifying method according to the present invention uses a first dictionary stored in a storage area and including a character and configuration information of the character, that is, a first divided display that is a display in which an input character is divided into a plurality of parts. , When the display of one or a plurality of candidate characters, which are characters in which the first part selected according to the first divided display is replaced, is accepted, the input that the first part is not valid is accepted, 1 Alternatively, a first confirmed part that is a part other than the first part among the plurality of candidate characters is specified, and a second split display, which is a display in which the first part is further divided into a plurality of parts, is performed using the first dictionary. When performed, the selection of the second part included in the first part is accepted, the second confirmed part that is a part other than the second part in the first part is specified, and the configuration information of one or more candidate characters is used. Therefore, the configuration information of the part including the second confirmed part in the configuration information is Of the characters registered in the first dictionary, a second dictionary that includes the configuration information replaced with the configuration information of the character that includes the second confirmed part and one or more candidate characters is generated, and the second dictionary is generated. A process of extracting a candidate character including the first confirmed part and the second confirmed part and displaying the extracted candidate character is included.

  In one aspect, it is possible to improve the accuracy of identifying obfuscated characters while suppressing the size of the dictionary.

FIG. 1 is a diagram for explaining that obfuscated characters hinder the utilization of information. FIG. 2 is a diagram for explaining the input work realized by the present embodiment. FIG. 3 is a diagram for explaining the input work realized by the present embodiment. FIG. 4 is a diagram for explaining the input work realized by the present embodiment. FIG. 5 is a diagram for explaining the input work realized by the present embodiment. FIG. 6 is a functional block diagram of the information processing device. FIG. 7 is a diagram for explaining the kanji constituent descriptive characters. FIG. 8 is a diagram showing an example of a dictionary stored in the dictionary storage unit. FIG. 9 is a diagram showing a processing flow of processing executed by the processing unit. FIG. 10 is a diagram showing a processing flow of processing executed by the input support unit. FIG. 11 is a diagram showing a transition of processing at the time of an event. FIG. 12 is a diagram showing a processing flow of processing at the time of an input start event. FIG. 13 is a diagram showing an example of data displayed by the processing at the time of the input start event. FIG. 14 is a diagram showing a processing flow of processing at the time of an instruction event. FIG. 15 is a diagram showing an example of a kanji structure description character. FIG. 16 is a diagram illustrating an example of a record stored in the data storage unit as the property “candidates”. FIG. 17 is a diagram showing an example of data displayed by the processing at the time of the instruction event. FIG. 18 is a diagram showing a processing flow of processing at the time of a drilldown start event. FIG. 19 is a diagram showing an example of data displayed by the processing at the drill-down start event. FIG. 20 is a diagram showing a processing flow of processing at the time of a drilldown instruction event. FIG. 21 is a diagram showing an example of a constituent character string after drilling down. FIG. 22 is a diagram showing an example of the constituent character string generated from the constituent character string after the drill down and the constituent character string stored as the property “candidates”. FIG. 23 is a diagram illustrating an example of data stored in the data storage unit as the property “candidates”. FIG. 24 is a diagram showing an example of data displayed by the processing at the time of the drill-down instruction event. FIG. 25 is a diagram showing a processing flow of processing at the time of a confirmed event. FIG. 26 is a diagram showing an example of data displayed by the processing at the time of a confirmed event. FIG. 27 is a diagram showing a processing flow of processing at a cancel event. FIG. 28 is a diagram showing an example of data displayed by processing at the time of a cancel event. FIG. 29 is a diagram for explaining setting of each property. FIG. 30 is a diagram for explaining a set of characters to be searched. FIG. 31 is a diagram for explaining a set of characters to be searched. FIG. 32 is a diagram for explaining a set of characters to be searched. FIG. 33 is a diagram for explaining a set of characters to be searched. FIG. 34 is a diagram for explaining suppression of the data amount of the dictionary. FIG. 35 is a functional block diagram of a computer.

[Outline of input work in the present embodiment]
FIG. 2 shows an example of the input work realized by this embodiment. The user wants to input the character 1k in FIG. The character 1k is not displayed as a candidate in a general IME (Input Method Editor) even if kana-kanji conversion is performed with "Ryo". Therefore, first, the user inputs “companion” whose appearance is close to the character 1k, selects “companion”, and long-tap, for example. It should be noted that no input is required when correcting a character that has already been input.

  As shown in FIG. 2A, a broken line representing the composition of the selected character is displayed according to the data registered in the dictionary. It is assumed that “companion” has two vertical columns and is registered in the dictionary as being left-sided and left-sided.

  The user selects a part different from the part of the kanji desired to be input, for example, by tapping. In the example of FIG. 2B, since the left side part which is the radical in the configuration of two vertical columns is different from the part of the character 1k, the user selects the left side part.

  Then, as shown in FIG. 2C, a list of characters whose left side part is different from "companion" is displayed. The user selects the character 1k, which is the character to be input, from the displayed characters by, for example, tapping.

  3 to 5 show another example of the input work realized by the present embodiment. The user wants to input the character 2k in FIG. Since the user does not know how to read the character 2k, the candidate cannot be displayed by kana-kanji conversion.

  Therefore, as shown in FIG. 3A, the user inputs “se” whose appearance is close to the character 2k, selects “se” and long-tap, for example. It should be noted that no input is required when correcting a character that has already been input.

  As shown in FIG. 3B, a broken line representing the composition of the selected character is displayed according to the data registered in the dictionary. It is assumed that "se" has a structure of two vertical columns and is registered in the dictionary as a left side is a sardine and a right side is a "yori".

  The user selects a part different from the part of the kanji desired to be input, for example, by tapping. In the example of FIG. 3C, the right part of the two vertical columns is different from the part of the character 2k, so the user selects the right part.

  Then, as shown in FIG. 4A, a list of characters whose right side part is different from "se" is displayed. The user searches the displayed characters for the character 2k that is the character that he wants to input, but cannot find the character 2k because there are too many characters including miso.

  Therefore, as shown in FIG. 4B, the user again selects the right part by, for example, a long tap. Hereinafter, the process started in response to the reselection operation by the user is referred to as a drill down.

  As a result of the drilldown, as shown in FIG. 4C, a broken line representing the configuration of the reselected portion is displayed according to the data registered in the dictionary. It is assumed that "yori" has a structure of two vertical columns and is registered in the dictionary as "bundle" on the left side and "page" on the right side.

  The user selects a part different from the part of the kanji desired to be input, for example, by tapping. In the example of FIG. 5A, the right part (here, the rightmost part of the three parts) in the two vertical columns is different from the part of the character 2k, so the user selects the right part. .

  Then, as shown in FIG. 5B, a list of characters whose right side part is different from "se" is displayed. The user selects the character 2k, which is the character to be input, from the displayed characters by, for example, tapping.

[Specific contents of the present embodiment]
Hereinafter, specific contents for realizing the above-described input work will be described. FIG. 6 shows a functional block diagram of the information processing device 1 according to the present embodiment. The information processing device 1 is, for example, a smart device. An application 10 (for example, a browser or a text editor) is realized by a CPU (Central Processing Unit) of the information processing device 1 executing an application program. The application 10 includes a processing unit 101 that executes original processing as an application, and an input support unit 102. The input support unit 102 includes a first event processing unit 1021, a second event processing unit 1022, a third event processing unit 1023, a fourth event processing unit 1024, a fifth event processing unit 1025, and a sixth event processing. Section 1026. The storage device (for example, a flash memory) of the information processing device 1 is provided with a dictionary storage unit 11 and a data storage unit 12.

  The first event processing unit 1021 executes processing at the time of an input start event. The second event processing unit 1022 executes processing at the time of the instruction event. The third event processing unit 1023 executes processing at the time of a drilldown start event. The fourth event processing unit 1024 executes processing at the time of a drill down instruction event. The fifth event processing unit 1025 executes processing at the time of a confirmed event. The sixth event processing unit 1026 executes processing for a cancel event.

  The kanji constituent descriptive character will be described with reference to FIG. The kanji constituent descriptive characters are used to express "characters that are not in modern times" and "characters that have not been standardized" in old documents. The broken-line block is the Ideographic Description Characters of Unicode, and is represented by a hexadecimal 4-digit character code. Regarding the kanji constituent descriptive characters, for example, 18.2 of http://unicode.org/charts/PDF/U2FF0.pdf and http://www.unicode.org/versions/Unicode8.0.0/ch18.pdf. See etc.

  FIG. 8 shows an example of a dictionary stored in the dictionary storage unit 11. In the dictionary according to the present embodiment, keys are characters and values are constituent character strings. The constituent character string includes a kanji constituent descriptive character and an element (for example, radical) included in the kanji, and is normalized by the maximum constituent.

  For example, in the case of the character “se”, instead of being divided into “sanzui”, “bunch” and “page” as a three-divided structure, it is divided into “sanzui” and “yori” as a vertically divided two-part structure. The "yori" is divided into a "bundle" and a "page" as a vertically divided structure.

  Since the third record cannot be represented by the structure of two vertical columns, the kanji structure description character representing the structure of three vertical columns is stored. The fourth record is nested because there is no character that directly represents the right part. As for the fifth record, there is no character indicating the right side portion and it cannot be expressed by nesting, so the right side portion is represented by “x”.

  Next, the operation of the information processing device 1 will be described with reference to FIGS. 9 to 34.

  First, the outline of the processing executed by the processing unit 101 of the application 10 will be described with reference to FIG. 9. The processing unit 101 determines whether a character correction instruction has been received from the user (FIG. 9: step S1). The correction instruction is, for example, a long tap of a character. When the correction instruction has not been received (step S1: No route), the process returns to step S1. On the other hand, when the correction instruction is received (step S1: Yes route), the data of the target character is transmitted to the input support unit 102 (step S3). In the following, the target character is called the base character.

  The processing unit 101 determines whether the character has been decided (step S5). When the character is confirmed, the character after confirmation is notified from the input support unit 102, and thus the determination in step S5 is performed depending on whether or not the notification is received.

  When the character is not determined (step S5: No route), the process returns to step S1. On the other hand, when the character is determined (step S5: Yes route), the base character is corrected to the determined character (step S7).

  The processing unit 101 determines whether or not there is an end instruction to the application 10 (step S9). If there is no end instruction to the application 10 (step S9: No route), the process returns to step S1. On the other hand, if there is an end instruction to the application 10 (step S9: Yes route), the process ends.

  According to the above, the processing unit 101 in the application 10 and the input support unit 102 introduced can cooperate with each other.

  Processing executed by the input support unit 102 of the application 10 will be described with reference to FIG. 10. The input support unit 102, when called by the processing unit 101, executes an initialization process (for example, a process of expanding a dictionary stored in the dictionary storage unit 11 into a memory (for example, a RAM (Random Access Memory))) ( FIG. 10: Step S11). The call from the processing unit 101 is performed, for example, when the operation of the processing unit 101 is started.

  The input support unit 102 executes an input support process (step S13). In input support processing, processing at the time of input start event, processing at the time of instruction event, processing at the time of drilldown start event, processing at the time of drilldown instruction event, processing at the time of confirmation event, and processing at the time of cancel event Processing and are executed. The transition of processing at the time of an event is as shown in FIG. The processing at the input start event is started by the notification from the processing unit 101. The process transitions from the process at the input start event to the process at the instruction event. Next, the process transitions from the process at the instruction event to the process at the drill-down start event or the process at the confirmed event. The process transitions from the process at the drill-down start event to the process at the drill-down instruction event. The process transitions from the process at the drill-down instruction event to the process at the drill-down start event, the process at the instruction event, or the process at the confirmed event. When the processing at the time of the confirmation event ends, the input support unit 102 waits until receiving the notification from the processing unit 101. In addition, when the input support unit 102 receives an instruction to delete the base character window, the process at the cancel event is executed.

  The contents of each process will be described below with reference to FIGS. 12 to 28.

  First, the processing at the input start event will be described with reference to FIGS. 12 and 13. The first event processing unit 1021 receives the base character data from the processing unit 101 (FIG. 12: step S21). The base character data is received when the user long-tap the selected character string as an input start event, for example.

  The first event processing unit 1021 initializes each property stored in the data storage unit 12 (step S23). In the present embodiment, the state of the base character is stored as the property “base”, the drill-down depth is stored as the property “depth”, and the state of the configuration character string is stored as the property “elements [depth]”. A variable (that is, a wildcard) position is stored as "position [depth]", a candidate state is stored as a property "candidates", a state of the base character window object is stored as a property "wBase", and a property "wBase" is stored. The state of the candidate window object is stored as “wCandidates”. The initial value of the property "depth" is 0. The property “position [depth]” has the origin (0) at the upper left, and is sequentially numbered clockwise. For example, in the case of a configuration with three columns, the numbers are 0, 1, 2 in the order of left, middle, and right.

  The first event processing unit 1021 stores the data of the base character in the data storage unit 12 as the property “base” (step S25).

  The first event processing unit 1021 displays the base character window on the display device of the information processing device 1 so that the base character fits (step S27). The state of the displayed base character window is stored as the property “wBase”.

  The first event processing unit 1021 acquires the constituent character string corresponding to the base character from the dictionary and stores it in the data storage unit 12 as the property “element [depth]” (step S29).

  The first event processing unit 1021 displays a broken line representing the configuration of the base character on the display device of the information processing device 1 based on the configuration character string acquired in step S29 (step S31). Then, the processing at the input start event ends.

  FIG. 13 shows an example of data displayed by the processing at the input start event. In the example of FIG. 13, a base character window 1301 and a broken line 1302 representing the configuration are displayed for the base character “SE” that has been long-tapped, for example.

  The processing at the time of the instruction event will be described with reference to FIGS. 14 to 17. By the processing at the time of the input start event, the base character is divided into a plurality of parts by the broken line, so the user taps to select a part different from the part of the character to be input. This selection corresponds to an instruction event.

  The second event processing unit 1022 receives, from the processing unit 101, position information of a part selected as different from the part of the character to be input among the plurality of parts generated by the division by the broken line (FIG. 14: step S41). For example, when the right part is selected from the two parts for “se”, the position information is “1”.

  The second event processing unit 1022 identifies a confirmed part that is not selected (that is, confirmed) and stores the position information of the confirmed part in the memory (step S42). For example, when the right side portion of “se” is selected, the confirmed portion is the left side portion (that is, Sanzui).

  The second event processing unit 1022 stores the position information of the selected body part in the data storage unit 12 as the property “position [depth]” and highlights the selected body part (step S43).

  The second event processing unit 1022 generates a search condition that makes the selected part variable (step S45). The search condition reflects the information of the confirmed part specified in step S42. For example, when the position information of the selected part of the two parts in “se” is “1”, the following search condition is generated.

  The Kanji structure descriptive character of “se” is shown in FIG. 15A, but the Kanji structure descriptive character of FIG. 15B having the same structure is also added to the search condition. Further, when the Kanji structure description character of the base character is as shown in FIG. 15C, the Kanji structure description character of FIG. 15D is also added to the search condition.

  The second event processing unit 1022 searches the dictionary with the search condition generated in step S45, and acquires a record containing a constituent character string that satisfies the search condition (step S47).

  The second event processing unit 1022 stores the record acquired in step S47 in the data storage unit 12 as the property "candidates" (step S49).

  FIG. 16 shows an example of a record stored in the data storage unit 12 as the property “candidates”. In the example of FIG. 16, a character and a character string of the character are stored.

  The second event processing unit 1022 displays the character (hereinafter referred to as a candidate) stored in the data storage unit 12 as the property “candidates” and the candidate window on the display device of the information processing device 1 (step S51). Then, the process ends.

  FIG. 17 shows an example of data displayed by the processing at the time of the instruction event. In the example of FIG. 17, the selected part of the two parts in the base character window 1301 is highlighted. Further, a candidate window 1701 and candidates contained in the candidate window 1701 are displayed. If the number of candidates is equal to or greater than a predetermined number, “...” is used to omit the display.

  The processing at the drill-down start event will be described with reference to FIGS. 18 and 19. One or more candidates are displayed by the process at the time of the instruction event, but when the candidate that matches the character the user wants to input is not displayed, the user selects the part of the base character different from the part of the character to be input. For example, long tap to reselect. This reselection corresponds to the drilldown start event.

  The third event processing unit 1023 receives the position information of the part selected as the target of the drilldown from the processing unit 101 (FIG. 18: step S61).

  The third event processing unit 1023 determines whether the part to be drilled down matches any of the characters registered in the dictionary (step S63). For example, if the part to be drilled down is "yori", which is the part on the right side of "se", then "yorai" is registered in the dictionary, so the process proceeds to the Yes route of step S65.

  If the part to be drilled down does not match any of the characters registered in the dictionary (step S65: No route), the drill down cannot be executed, and the process ends. On the other hand, when the part to be drilled down matches any of the characters registered in the dictionary (step S65: Yes route), the third event processing unit 1023 executes the following process. Specifically, the third event processing unit 1023 erases the display of the candidate window and the candidate (step S67).

  The third event processing unit 1023 deletes the highlighted display of the part to be drilled down (step S69).

  The third event processing unit 1023 increments the property “depth” stored in the data storage unit 12 by 1 (step S71).

  The third event processing unit 1023 acquires the constituent character string corresponding to the part to be drilled down from the dictionary, and in the acquired constituent character string, the property “elements [depth-1]” (that is, corresponding to the base character). The portion corresponding to the part to be drilled down is updated (step S73). For example, in the case of "se", the "yori" portion of the constituent character string is replaced with the constituent character string of the second record in FIG. The updated constituent character string corresponds to the property “elements [depth]”.

  The third event processing unit 1023 displays a broken line representing the configuration of the base character based on the property “elements [depth]” (step S75). Then, the process ends.

  FIG. 19 shows an example of data displayed by the processing at the drill-down start event. In the example of FIG. 19, broken lines 1302 and 1901 are displayed for the base character “se” reselected by long tap, for example.

  Processing at the time of the drill-down instruction event will be described with reference to FIGS. 20 to 24. By the processing at the time of the drill-down start event, the part of the base character selected by the user is further divided into a plurality of parts by the broken line, so the user taps a part different from the part of the character to be input among the plurality of parts. Select more. This selection corresponds to the drill down instruction event.

  The fourth event processing unit 1024 receives, from the processing unit 101, the positional information of the part selected as different from the part of the character to be input among the plurality of parts generated by the division by the broken line (FIG. 20: step S81). For example, when the right part is selected from the three parts for “se”, the position information is “2”.

  The fourth event processing unit 1024 identifies a confirmed part that is not selected (that is, confirmed) and stores the position information of the confirmed part in the memory (step S82). For example, when the right side portion of the “se” is selected, the confirmed portions are the left side portion and the center portion (that is, the sushi and the “bundle”).

  The fourth event processing unit 1024 stores the position information of the selected site in the data storage unit 12 as the property “position [depth]” and highlights the selected site (step S83).

  The fourth event processing unit 1024 generates a search condition that makes the selected part variable (step S85). The search condition reflects the information of the confirmed part specified in step S82. For example, when the position information of the selected part of the three parts in “se” is “2”, the following search conditions are generated.

  The Kanji structure descriptive character of “se” is shown in FIG. 15A, but the Kanji structure descriptive character of FIG. 15B having the same structure is also added to the search condition. Further, when the Kanji structure description character of the base character is as shown in FIG. 15C, the Kanji structure description character of FIG. 15D is also added to the search condition.

  The fourth event processing unit 1024 searches the dictionary under a search condition that makes the selected portion of the drill-down target variable, and generates a constituent character string after the drill-down (step S87). For example, when the target of the drill down is "request" and the selected site is "page", the following search conditions are used in step S87.

  By the process of step S87, a constituent character string as shown in FIG. 21, for example, is generated.

  The fourth event processing unit 1024 updates the part corresponding to the part to be drilled down in the constituent character string of the property “candidates” with the constituent character string after drilling down generated in step S87 (step S89).

  For example, it is assumed that the constituent character string after the drill down is the constituent character string shown in FIG. 21, and the constituent character string of the property “candidates” is the constituent character string shown in FIG. In such a case, the record r01 in FIG. 16 is updated by the record r04 in FIG. 21, the record r02 in FIG. 16 is updated by the record r05 in FIG. 21, and the record r03 in FIG. 16 is updated by the record r06 in FIG. Will be updated. As a result, a constituent character string as shown in FIG. 22 is generated. Record r07 in FIG. 22 is a record generated from record r01 and record r04, record r08 is a record generated from record r02 and record r05, and record r09 is record r03 and record r06. It is a generated record.

  The fourth event processing unit 1024 identifies records satisfying the search condition from the records stored in the data storage unit 12 as the property “candidates” (step S91).

  The fourth event processing unit 1024 updates the property “candidates” with the identified record (step S93).

  FIG. 23 shows an example of data stored in the data storage unit 12 as the property “candidates”. In the example of FIG. 23, a character including the confirmed part specified in step S42 and the confirmed part specified in step S82, and the character string of the character are stored.

  The fourth event processing unit 1024 displays the characters (that is, candidates) stored in the data storage unit 12 as the property “candidates” and the candidate window on the display device of the information processing device 1 (step S95). Then, the process ends.

  FIG. 24 shows an example of data displayed by the processing at the time of the drill-down instruction event. In the example of FIG. 24, for example, the selected part of the two parts in the long-tapping part “yori” is highlighted. Further, a candidate window 2401 and candidates contained in the candidate window are displayed. If the number of candidates is equal to or greater than a predetermined number, “...” is used to omit the display.

  The process at the time of the confirmed event will be described with reference to FIGS. Although one or a plurality of candidates are displayed by the processing at the time of the instruction event, when a candidate that matches the character the user wants to input is displayed, the user taps the candidate. This selection corresponds to a confirmation event. Further, although one or a plurality of candidates are displayed by the processing of the drill-down instruction event, when a candidate that matches the character the user wants to input is displayed, the user taps the candidate. This selection also corresponds to the confirmation event.

  The fifth event processing unit 1025 receives an index (for example, a character identification number) of the selected candidate from the displayed one or a plurality of candidates from the processing unit 101 (FIG. 25: step S101).

  The fifth event processing unit 1025 transmits the data of the selected character to the processing unit 101 (step S103). In response to this, the processing unit 101 replaces the character.

  The fifth event processing unit 1025 initializes each property stored in the data storage unit 12 (step S105).

  The fifth event processing unit 1025 erases the display of the base character window (step S107).

  The fifth event processing unit 1025 erases the display of the candidate window (step S109). Then, the process ends.

  FIG. 26 shows an example of data displayed by the processing at the time of a confirmed event. In the example of FIG. 26, the character 2k input by the user is displayed, and the character 2k is underlined. When the user further inputs a confirmation instruction (for example, clicks the enter key), the input of characters is completed.

  Processing at the time of a cancel event will be described with reference to FIGS. 27 and 28. The user swipes the base character window off the screen when performing the input operation of the present embodiment, when giving up the input of the character or when noting that the character already input is correct. This operation corresponds to a cancel event.

  The sixth event processing unit 1026 receives an instruction to delete the base character window from the processing unit 101 (FIG. 27: step S111).

  The sixth event processing unit 1026 transmits to the processing unit 101 that the base character window deletion instruction has been received (step S113).

  The sixth event processing unit 1026 erases the display of the base character window (step S115).

  The sixth event processing unit 1026 erases the display of the candidate window (step S117).

  The sixth event processing unit 1026 initializes each property stored in the data storage unit 12 (step S119). Then, the process ends.

  FIG. 28 shows an example of data displayed by the processing at the time of the cancel event. In the example of FIG. 28, the characters on the screen have returned to the state before the input work of this embodiment was started.

  A description of setting each property will be added with reference to FIG.

  The property "base" is set by the processing at the input start event, and is deleted by the processing at the confirmation event and the processing at the cancel event.

  The property "depth" is set to 0 by the process at the input start event, incremented by 1 by the process at the drilldown start event, and deleted by the process at the confirmation event and the process at the cancel event.

  The property "elements []" is added by the process at the input start event and the process at the drilldown start event, and is deleted by the process at the confirmation event and the process at the cancel event.

  The property "position []" is set to empty (that is, empty) by the processing at the time of the input start event, and is added by the processing at the time of the instruction event and the processing at the time of the drilldown instruction event, and the processing at the time of the confirmation event and the cancellation event. It is erased by the process of time.

  The property "candidates" is set to empty by the processing at the time of the input start event, is set by the processing at the time of the instruction event, is narrowed down by the processing at the time of the drilldown instruction event, and is processed by the processing at the time of the confirmation event and the processing at the time of the cancel event. Erased.

  The property "wBase" is displayed by the processing at the time of the input start event, highlighted by the processing at the instruction event, the broken line is added by the processing at the drilldown start event, and highlighted by the processing at the drilldown instruction event. It is erased, and is erased by the process at the time of the confirmation event and the process at the time of the cancel event.

  The property "wCandidates" is displayed by the processing at the time of the instruction event, is deleted by the processing at the time of the drilldown start event, is displayed by the processing at the time of the drilldown instruction event, and is deleted by the processing at the time of the confirmation event and the processing at the time of the cancel event. To be done.

  A set of characters to be searched will be described with reference to FIGS. 30 to 33. In the present embodiment, the search is performed so that the set of characters to be searched becomes gradually smaller. For example, when the base character is “se”, first, the set of characters to be searched is the entire dictionary. When the part corresponding to “yori” in “se” is selected as a part different from the part of the character to be input, the L1 set as shown in FIG. 31 becomes the search target. Further, when the drill-down is performed on "yori", the constituent character strings of the characters included in the L1 set are refined as shown in FIG. Then, when the part corresponding to “page” in “request” is selected as a part different from the part of the character to be input, the L2 set as shown in FIG. 33 becomes the search target.

  By dynamically generating the sub-dictionary at the time of execution and performing the search as in the present embodiment, it is possible to suppress the consumption of the storage device capacity and reduce the CPU time for the search. . By reducing the CPU time, power consumption can be suppressed. Therefore, according to the present embodiment, it becomes possible to execute a process for inputting a character even in a smart device or the like having limited hardware resources while suppressing the influence on the process of other applications.

  For example, when the conventional technique described in the section of background art is used, it has been confirmed that the data amount of the dictionary increases several times to several tens of times as compared with the dictionary of the present embodiment. A more specific description will be given with reference to FIG. For example, with regard to the kanji of the first record, when the dictionary data for satisfying all the search conditions assumed to be entered by the user is compared with the data amount of the constituent character strings in the dictionary of this embodiment, it is about three times as large. is there. The kanji in the second record is about 7 times as large as the data amount of the constituent character strings in the dictionary of the present embodiment when the dictionary data for dealing with all the search conditions assumed to be entered by the user is compared. . The kanji of the third record is about 21 times as large as the data amount of the constituent character strings in the dictionary of the present embodiment, when the dictionary data for satisfying all the search conditions assumed to be input by the user is compared. .

  In addition, since the search conditions are input in stages, it is possible to prevent the screen from being occupied by the input of the search conditions.

  In addition, the user can reach the input character by inputting similar characters and selecting a different portion, so that the burden on the user can be reduced.

  Further, since the search is performed based on the objective “composition” rather than the dictionary creator's subjective criteria, the possibility of omission of search is low even when a user with low kanji skill is used.

  Further, since the data can be correctly input in the computer system, the data consistency can be improved at the time of information integration.

  Further, since the handwritten search that has been conventionally used is a statistical method, it may not be possible to specify the target character depending on the writing method. As a result, there is a possibility of making an erroneous determination that the corresponding character cannot be used in the computer. On the other hand, according to the present embodiment, the character can be searched without using a statistical method, and therefore the reliability of the search result is high.

  Further, since the variant character search conventionally used searches for characters having a national relationship, it is not possible to specify similar characters. On the other hand, according to the present embodiment, it is possible to broadly search for "similar characters" even if they have no national language relationship.

  Although one embodiment of the present invention has been described above, the present invention is not limited to this. For example, the functional block configuration of the information processing device 1 described above may not match the actual program module configuration.

  Further, the data structure described above is an example, and the data structure does not have to be as described above. Further, also in the processing flow, it is possible to change the order of processing if the processing result does not change. Further, it may be executed in parallel.

  In the example described above, the functions of the present embodiment are implemented in the application 10, but may be implemented in the IME of the OS (Operating System), or may be implemented as a web application by allowing the server to have a dictionary. Good.

  The information processing device 1 may be a desktop or notebook personal computer.

  The input by the user may be a handwriting input.

  The information processing device 1 described above is a computer device, and as shown in FIG. 35, a display control unit 2507 connected to a RAM (Random Access Memory) 2501, a CPU 2503, a flash memory 2505, and a display device 2509. An input device 2515 and a communication control unit 2517 for connecting to a network are connected by a bus 2519. An operating system (OS) and application programs for executing the processing in this embodiment are stored in the flash memory 2505, and are read from the flash memory 2505 to the RAM 2501 when being executed by the CPU 2503. The CPU 2503 controls the display control unit 2507 and the communication control unit 2517 according to the processing content of the application program to cause a predetermined operation. Further, the data in the process of processing is mainly stored in the RAM 2501, but may be stored in the flash memory 2505. In the embodiment of the present invention, the application program for performing the above-described processing is installed in the flash memory 2505 via the network such as the Internet and the communication control unit 2517. Such a computer realizes the various functions described above by organically cooperating the hardware such as the CPU 2503 and the RAM 2501 described above with programs such as the OS and application programs.

  The embodiments of the present invention described above can be summarized as follows.

  The character identifying method according to the first aspect of the present embodiment uses (A) a first dictionary (for example, a dictionary according to the embodiment) stored in a storage area and including characters and configuration information of the characters. , A first divided display in which the input character is divided into a plurality of parts and a display of one or a plurality of candidate characters in which the first part selected according to the first divided display is a replaced character When performed, the input that the first part is not valid is accepted, (B) the first confirmed part which is a part other than the first part among the one or more candidate characters is identified, and (C) the first part When the second division display, which is a display in which the first part is further divided into a plurality of parts, is performed using the dictionary, the selection of the second part included in the first part is accepted, and (D) the first part Identify a second confirmed site that is a site other than the second site, and (E) 1 or more The configuration information of the candidate character, wherein the configuration information of the part including the second confirmed part in the configuration information is replaced with the configuration information of the character including the second confirmed part in the characters registered in the first dictionary. A second dictionary (for example, the property "candidates" updated in step S89) including the configuration information and one or more candidate characters is generated, and (F) the first determined part and the second confirmed part are generated from the second dictionary. A process of extracting a candidate character including a part and displaying the extracted candidate character is included.

  By extracting candidate characters using a second dictionary generated separately from the first dictionary, it is possible to identify obfuscated characters while suppressing omission of search even when computer resources are limited. become.

  Further, in the process of performing the first split display, (a1) in the process of identifying the configuration information of the input character from the first dictionary, generating the first split display according to the identified configuration information, and performing the second split display, (C1) The configuration information of the character corresponding to the first part may be specified from the first dictionary, and the second split display may be generated according to the specified configuration information. Appropriate division will be performed according to the configuration information.

  In addition, in the process of generating the second dictionary, (e1) the third dictionary including one or more candidate characters and the configuration information of the one or more candidate characters using the first dictionary (for example, in step S49 Property "candidates") is generated, and (e2) the configuration information of the part including the second confirmed part in the configuration information included in the third dictionary, and the second confirmed part among the characters registered in the first dictionary. The second dictionary may be generated by substituting the constituent information of the character and including one or more candidate characters and the constituent information after the replacement. Since the second dictionary generated by such a method is smaller in scale than the first dictionary, candidate characters can be specified while performing narrowing down step by step.

Further, in the process of displaying one or a plurality of candidate characters, (a2) the first part is highlighted, and in the process of displaying the extracted candidate characters, (f1) the second part is highlighted. May be. It becomes possible to support the user's input operation.

  Further, the character is a kanji, and the configuration information included in the first dictionary may include a kanji configuration description character and an element included in the kanji. Since it is possible to use a dictionary that is based on the objective fact that the characters are composed, rather than a dictionary whose contents change depending on the dictionary creator's thinking and ability, it is not possible to specify the character because it does not match the dictionary creator's thinking. Such a situation will not occur.

  The input character may be a handwritten input character.

  Further, the computer may be a mobile terminal having a touch panel. Then, in the process of accepting an input indicating that the first part is not valid, (a3) a long tap for the first part is detected, and in the process of accepting the selection of the second part, (c1) a tap for the second part is detected. You may. The obfuscated character can be specified even in a mobile terminal or the like having a limited storage device capacity.

  A character identification device according to a second aspect of the present embodiment divides an input character into a plurality of parts by using a first dictionary stored in a (G) storage area and including a character and configuration information of the character. The first processing unit (for example, the first event processing unit 1021 in the embodiment) that performs the first split display, which is the displayed display, is replaced with (H) the first site selected according to the first split display. A second processing unit (for example, the second event processing unit 1022 in the embodiment) that displays one or a plurality of candidate characters that are different characters, and (I) receives an input that the first part is not valid. A first division part which is a part other than the first part among one or a plurality of candidate characters is specified, and the first part is further divided into a plurality of parts by the second division A third processing unit that performs display (for example, in the embodiment 3 event processing unit 1023), and (J) accepts the selection of the second site included in the first site, identifies the second confirmed site that is a site other than the second site in the first site, and determines one or more The configuration information of the candidate character, wherein the configuration information of the part including the second confirmed part in the configuration information is replaced with the configuration information of the character including the second confirmed part in the characters registered in the first dictionary. A second dictionary including the configuration information and one or more candidate characters is generated, candidate characters including the first confirmed part and the second confirmed part are extracted from the second dictionary, and the extracted candidate characters are displayed. And a fourth processing unit (for example, the fourth event processing unit 1024 in the embodiment) that performs the processing.

  A program for causing the processor to perform the processing according to the above method can be created, and the program is stored in a computer-readable storage medium or storage device such as a semiconductor memory. The intermediate processing result is temporarily stored in a storage device such as a main memory.

  With regard to the embodiments including the above-described examples, the following additional notes are disclosed.

(Appendix 1)
On the computer,
According to the first divided display, which is a display in which the input character is divided into a plurality of parts, using the first dictionary stored in the storage area and containing the characters and the configuration information of the characters, When the selected first part is displayed with one or more candidate characters that are replaced characters, the input that the first part is not valid is accepted,
Identify a first confirmed part that is a part other than the first part among the one or more candidate characters,
When the first division is used to perform a second divided display that is a display in which the first portion is further divided into a plurality of portions, the selection of the second portion included in the first portion is accepted,
Identify a second confirmed site that is a site other than the second site in the first site,
The configuration information of the one or more candidate characters, wherein the configuration information of the part including the second confirmed part in the configuration information includes the second confirmed part of the characters registered in the first dictionary. Generating a second dictionary including the configuration information replaced with the configuration information of the characters and the one or more candidate characters,
Candidate characters including the first confirmed part and the second confirmed part are extracted from the second dictionary, and the extracted candidate characters are displayed.
A character identification program that causes processing to be performed.

(Appendix 2)
In the process of performing the first split display,
Identifying the configuration information of the input character from the first dictionary, and generating the first split display according to the identified configuration information,
In the process of performing the second split display,
Character configuration information corresponding to the first part is specified from the first dictionary, and the second split display is generated according to the specified configuration information.
The character identification program described in Appendix 1.

(Appendix 3)
In the process of generating the second dictionary,
Using the first dictionary, a third dictionary including the one or more candidate characters and the configuration information of the one or more candidate characters is generated,
Of the configuration information included in the third dictionary, the configuration information of the part including the second confirmed region is replaced with the configuration information of the character including the second confirmed region among the characters registered in the first dictionary, Generating the second dictionary including the one or more candidate characters and the configuration information after replacement,
The character identification program according to appendix 1 or 2.

(Appendix 4)
In the process of displaying the one or more candidate characters,
Highlight the first part,
In the process of displaying the extracted candidate characters,
Highlighting the second part,
The character identification program according to any one of appendices 1 to 3.

(Appendix 5)
The characters are Kanji,
The configuration information included in the first dictionary includes Kanji configuration description characters and elements included in the Kanji,
The character identification program according to any one of appendices 1 to 4.

(Appendix 6)
The input character is a handwritten input character,
The character identification program according to any one of appendices 1 to 5.

(Appendix 7)
The computer is a mobile terminal having a touch panel,
In the process of accepting the input that the first part is not valid,
Detecting a long tap for the first part,
In the process of accepting the selection of the second part,
Detecting a tap on the second part,
The character identification program according to any one of appendices 1 to 6.

(Appendix 8)
Computer
According to the first divided display, which is a display in which the input character is divided into a plurality of parts, using the first dictionary stored in the storage area and containing the characters and the configuration information of the characters, When the selected first part is displayed with one or more candidate characters that are replaced characters, the input that the first part is not valid is accepted,
Identify a first confirmed part that is a part other than the first part among the one or more candidate characters,
When the first division is used to perform a second divided display that is a display in which the first portion is further divided into a plurality of portions, the selection of the second portion included in the first portion is accepted,
Identify a second confirmed site that is a site other than the second site in the first site,
The configuration information of the one or more candidate characters, wherein the configuration information of the part including the second confirmed part in the configuration information includes the second confirmed part of the characters registered in the first dictionary. Generating a second dictionary including the configuration information replaced with the configuration information of the characters and the one or more candidate characters,
Candidate characters including the first confirmed part and the second confirmed part are extracted from the second dictionary, and the extracted candidate characters are displayed.
Character identification method to perform processing.

(Appendix 9)
A first processing unit that performs a first split display that is a display in which an input character is divided into a plurality of parts using a first dictionary that is stored in a storage area and that includes a character and configuration information of the character;
A second processing unit for displaying one or a plurality of candidate characters which are characters in which the first part selected according to the first divided display is replaced,
An input indicating that the first part is not valid is accepted, a first confirmed part that is a part other than the first part among the one or more candidate characters is identified, and the first dictionary is used to identify the first part. A third processing unit that performs a second split display, which is a display in which one site is further divided into a plurality of sites;
The selection of the second part included in the first part is accepted, the second confirmed part that is a part other than the second part in the first part is specified, and the second part is the configuration information of the one or more candidate characters. And configuration information in which the configuration information of the part including the second confirmed region in the configuration information is replaced with the configuration information of the character including the second confirmed region among the characters registered in the first dictionary, A second dictionary including the one or a plurality of candidate characters is generated, candidate characters including the first confirmed part and the second confirmed part are extracted from the second dictionary, and the extracted candidate characters are displayed. A fourth processing unit for performing
Character identifying device having.

DESCRIPTION OF SYMBOLS 1 Information processing device 10 Application 11 Dictionary storage unit 12 Data storage unit 101 Processing unit 102 Input support unit 1021 First event processing unit 1022 Second event processing unit 1023 Third event processing unit 1024 Fourth event processing unit 1025 Fifth event processing Part 1026 6th event processing part

Claims (8)

On the computer,
According to the first divided display, which is a display in which the input character is divided into a plurality of parts, using the first dictionary stored in the storage area and containing the characters and the configuration information of the characters, When the selected first part is displayed with one or more candidate characters that are replaced characters, the input that the first part is not valid is accepted,
Identify a first confirmed part that is a part other than the first part among the one or more candidate characters,
When the first division is used to perform a second divided display that is a display in which the first portion is further divided into a plurality of portions, the selection of the second portion included in the first portion is accepted,
Identify a second confirmed site that is a site other than the second site in the first site,
The configuration information of the one or more candidate characters, wherein the configuration information of the part including the second confirmed part in the configuration information includes the second confirmed part of the characters registered in the first dictionary. Generating a second dictionary including the configuration information replaced with the configuration information of the characters and the one or more candidate characters,
Candidate characters including the first confirmed part and the second confirmed part are extracted from the second dictionary, and the extracted candidate characters are displayed.
A character identification program that causes processing to be performed. In the process of performing the first split display,
Identifying the configuration information of the input character from the first dictionary, and generating the first split display according to the identified configuration information,
In the process of performing the second split display,
Character configuration information corresponding to the first part is specified from the first dictionary, and the second split display is generated according to the specified configuration information.
The character identification program according to claim 1. In the process of generating the second dictionary,
Using the first dictionary, a third dictionary including the one or more candidate characters and the configuration information of the one or more candidate characters is generated,
Of the configuration information included in the third dictionary, the configuration information of the part including the second confirmed region is replaced with the configuration information of the character including the second confirmed region among the characters registered in the first dictionary, Generating the second dictionary including the one or more candidate characters and the configuration information after replacement,
The character identification program according to claim 1. In the process of displaying the one or more candidate characters,
Highlight the first part,
In the process of displaying the extracted candidate characters,
Highlighting the second part,
The character identification program according to any one of claims 1 to 3. The characters are Kanji,
The configuration information included in the first dictionary includes Kanji configuration description characters and elements included in the Kanji,
The character identification program according to any one of claims 1 to 4. The input character is a handwritten input character,
The character identification program according to any one of claims 1 to 5. Computer
According to the first divided display, which is a display in which the input character is divided into a plurality of parts, using the first dictionary stored in the storage area and containing the characters and the configuration information of the characters, When the selected first part is displayed with one or more candidate characters that are replaced characters, the input that the first part is not valid is accepted,
Identify a first confirmed part that is a part other than the first part among the one or more candidate characters,
When the first division is used to perform a second divided display that is a display in which the first portion is further divided into a plurality of portions, the selection of the second portion included in the first portion is accepted,
Identify a second confirmed site that is a site other than the second site in the first site,
The configuration information of the one or more candidate characters, wherein the configuration information of the part including the second confirmed part in the configuration information includes the second confirmed part of the characters registered in the first dictionary. Generating a second dictionary including the configuration information replaced with the configuration information of the characters and the one or more candidate characters,
Candidate characters including the first confirmed part and the second confirmed part are extracted from the second dictionary, and the extracted candidate characters are displayed.
Character identification method to perform processing. A first processing unit that performs a first split display that is a display in which an input character is divided into a plurality of parts using a first dictionary that is stored in a storage area and that includes a character and configuration information of the character;
A second processing unit for displaying one or a plurality of candidate characters which are characters in which the first part selected according to the first divided display is replaced,
An input indicating that the first part is not valid is accepted, a first confirmed part that is a part other than the first part among the one or more candidate characters is identified, and the first dictionary is used to identify the first part. A third processing unit that performs a second split display, which is a display in which one site is further divided into a plurality of sites;
The selection of the second part included in the first part is accepted, the second confirmed part that is a part other than the second part in the first part is specified, and the second part is the configuration information of the one or more candidate characters. And configuration information in which the configuration information of the part including the second confirmed region in the configuration information is replaced with the configuration information of the character including the second confirmed region among the characters registered in the first dictionary, A second dictionary including the one or a plurality of candidate characters is generated, candidate characters including the first confirmed part and the second confirmed part are extracted from the second dictionary, and the extracted candidate characters are displayed. A fourth processing unit for performing
Character identifying device having.

Images