JP6311367B2 - User dictionary management device, user dictionary management method, and user dictionary management program - Google Patents

Description

  The present invention relates to a user dictionary management device that manages user dictionary information that is referred to when a language conversion processing device performs language conversion processing.

  In recent years, various language conversion processing apparatuses that perform language conversion processing such as Japanese-to-English translation processing or kana-character to kanji conversion processing have been used. When performing these language conversion processes, these language conversion processing devices refer to a conversion dictionary having dictionary data in which a conversion target word before conversion and a conversion candidate word after conversion are associated with each other. Such conversion dictionaries include user dictionaries used individually by each user, in addition to common dictionaries commonly used by users who use the language conversion processing device. The common dictionary is a conversion dictionary related to conversion target words that are commonly used by users. On the other hand, the user dictionary is a conversion dictionary related to conversion target words that are not registered in the common dictionary and are used individually for each user due to the characteristics of the document to be converted by each user.

In a general language conversion processing apparatus, each time a user performs language conversion processing, new dictionary data is registered in the user dictionary. As the dictionary data registered in the user dictionary increases, the accuracy of language conversion processing performed by each user improves. Therefore, there is an increasing expectation for a technique for registering dictionary data more efficiently with respect to such a user dictionary.
As a technique related to such a technique, Patent Document 1 calculates similarity between the first user dictionary and the second user dictionary, and when the similarity is equal to or greater than a threshold, the first user dictionary A system for registering in the second user dictionary the dictionary data included in the second user dictionary but not in the second user dictionary is disclosed.

JP 2007-080019 A

  Usually, in the language conversion process, the number of conversion candidate words related to one conversion target word is not limited to one. Regarding the language conversion processing performed by one user, one conversion target word is converted into a different word depending on time and circumstances. Therefore, a general conversion dictionary has, as such dictionary data, a conversion instruction record in which one or more conversion candidate words are associated with one conversion target word.

  When a plurality of conversion candidate words are associated with one conversion target word, the probability of conversion into each conversion candidate word is different. Therefore, the conversion instruction record associates and stores each conversion candidate word and the value indicated by the conversion priority based on the conversion probability to the conversion candidate word, thereby improving the efficiency of the language conversion process. For example, in the Kana-Kanji conversion system, a plurality of conversion candidates, Kanji, are displayed on the screen in descending order of conversion priority. Thereby, the user can efficiently perform the kana-kanji conversion process.

  In the technique disclosed in Patent Literature 1, when dictionary data registered in the first user dictionary is newly registered in the second user dictionary, the degree of similarity regarding the first and second user dictionaries is equal to or greater than a threshold value. In this case, the dictionary data is registered uniformly. Then, the technique does not register the dictionary data uniformly when the similarity is less than the threshold value. However, even if this similarity is less than the threshold value, the language conversion processing performed by the user who uses the second user dictionary does not always use the dictionary data registered in the first user dictionary. Absent. In this case, the language conversion processing performed by the user who uses the second user dictionary may use dictionary data registered in the first user dictionary although the conversion priority is low. Therefore, the technique disclosed in Patent Literature 1 cannot be said to have sufficient flexibility when registering unregistered dictionary data in a specific user dictionary from another user dictionary.

  A main object of the present invention is to provide a user dictionary management device, a user dictionary management method, and a user dictionary management program that solve this problem.

  The user dictionary management device according to the present invention refers to a language conversion processing device when language conversion processing is performed on user document information, and includes a conversion target word that is a word before conversion and one or more conversion candidates that are converted words. The degree of similarity related to the first and second user dictionary information among a plurality of user dictionary information including a word and a conversion instruction record that stores a value indicated by the conversion priority related to the conversion candidate word. A value indicated by the conversion priority based on a value indicated by the similarity based on a value indicated by the similarity and a calculation means for calculating a value based on a predetermined criterion and information indicated by the conversion instruction record included in the first user dictionary information And registration means for registering in the second user dictionary information.

  In another aspect of achieving the above object, the user dictionary management method of the present invention refers to a conversion target that is a word before conversion, which is referred to when the language conversion processing device performs language conversion processing on the user document information by the information processing device. Among a plurality of user dictionary information including a word, one or more conversion candidate words that are converted words, and a conversion instruction record that associates and stores a value indicated by a conversion priority for the conversion candidate word, A value indicated by the similarity between the first and second user dictionary information is calculated based on a predetermined criterion, and the information indicated by the conversion instruction record included in the first user dictionary information is a value indicated by the similarity. And registering it in the second user dictionary information in association with the value indicated by the conversion priority.

  Further, in a further aspect of achieving the above object, the user dictionary management program according to the present invention refers to a conversion target word that is a word before conversion, which is referred to when the language conversion processing device performs language conversion processing of user document information. The first of the plurality of user dictionary information including one or more conversion candidate words that are converted words and a conversion instruction record that stores the conversion priority order related to the conversion candidate words in association with each other. The similarity indicates the calculation process for calculating the value indicated by the similarity regarding the second user dictionary information based on a predetermined criterion, and the information indicated by the conversion instruction record included in the first user dictionary information. And causing the computer to execute a registration process for registering in the second user dictionary information in association with the value indicated by the conversion priority based on the value. To.

  Furthermore, the present invention can also be realized by a computer-readable non-volatile storage medium storing such a user dictionary management program (computer program).

  The present invention makes it possible to efficiently and flexibly register dictionary information for a user dictionary used by the language conversion processing device.

It is a block diagram which shows the structure of the user dictionary management system which concerns on 1st Embodiment of this invention. It is a flowchart which shows operation | movement of the user dictionary management system which concerns on 1st Embodiment of this invention. It is a figure which illustrates the similarity between the input document data which concerns on 1st Embodiment of this invention. It is a figure which shows the structural example of the similarity management information which concerns on 1st Embodiment of this invention. It is a figure which shows the structural example of the priority management information which concerns on 1st Embodiment of this invention. It is a figure which shows the example of registration of the dictionary data to the user dictionary data based on 1st Embodiment of this invention. It is a block diagram which shows the structure of the user dictionary management apparatus which concerns on 2nd Embodiment of this invention. It is a block diagram which shows the structure of the information processing apparatus which can execute the user dictionary management apparatus concerning each embodiment of this invention.

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

<First Embodiment>
FIG. 1 is a block diagram conceptually showing the structure of the user dictionary management system 1 according to the first embodiment. The user dictionary management system 1 according to the present embodiment includes a user dictionary management device 10, a language conversion device 20, a user input document data storage unit 30, and a user output document data storage unit 40.

In the present embodiment described below, a case where processing is performed for five users (users A to E) will be described as an example. That is, the user input document data storage unit 30 includes input document data (input document information) 300 to 304 that are input data when five users A to E perform language conversion processing using the user dictionary management system 1. Is stored. That is,
User A: input document data 300,
User B: input document data 301,
User C: input document data 302,
User D: input document data 303,
User E: input document data 304

  Note that the number of users using the user dictionary management system 1 is not limited to five, but the five users are merely examples. The user input document data storage unit 30 is a storage device such as an electronic memory or a magnetic disk.

The user output document data storage unit 40 stores output document data 400 to 404, which is data output by the language conversion processing device 20 by performing language conversion processing on the input document data 300 to 304. That is,
User A: output document data 400,
User B: input document data 401,
User C: input document data 402,
User D: input document data 403,
User E: input document data 404

  The user output document data storage unit 40 is a storage device such as an electronic memory or a magnetic disk.

  The language conversion processing device 20 performs language conversion processing on the input document data 300 to 304, and outputs them as output document data 400 to 404, respectively. For example, the language conversion processing device 20 may perform translation processing such as English-Japanese translation, or may perform kana-kanji conversion processing.

The language conversion processing device 20 includes a user dictionary data storage unit 21. The user dictionary data storage unit 21 stores dictionary data (dictionary information) 210 to 214 which are user dictionaries used when the users A to E perform language conversion processing using the user dictionary management system 1. . That is,
User A: dictionary data 210,
User B: dictionary data 211,
User C: dictionary data 212,
User D: dictionary data 213
User E: dictionary data 214

  The language conversion processing device 20 refers to the dictionary data 210 to 214 when performing the language conversion processing on the input document data 300 to 304, respectively. The language conversion processing device 20 is a general-purpose server device or the like, or an information processing device that exclusively performs language conversion processing. The user dictionary data storage unit 21 is a storage device such as an electronic memory or a magnetic disk.

  The user dictionary management device 10 is a device that updates and manages the dictionary data 210 to 214. The user dictionary management device 10 includes a calculation unit 11 and a registration unit 12. The calculation unit 11 and the registration unit 12 may be electronic circuits or may be realized by a computer program and a processor that operates according to the computer program.

  The calculation unit 11 calculates the value indicated by the similarity between the input document data for the input document data 300 to 304 based on a predetermined criterion.

  FIG. 3 illustrates the similarity between input document data according to the present embodiment. Here, it is assumed that the input document data 300 to 302 of the users A to C are English sentences having the contents shown in FIG. At this time, the language conversion processing device 20 is a device that performs English-Japanese translation. The calculation unit 11 detects a word included in at least any two of the input document data 300 to 302 with respect to the words included in the English sentences of the input document data 300 to 302.

  In the case of the example shown in FIG. 3, two words “Philippines” and “typoon” are commonly included in the input document data 300 of user A and the input document data 301 of user B. There are no words included in at least any two of the input document data 300 to 302 other than the two words. In this case, a value indicated by the similarity between the input document data 300 of the user A and the input document data 301 of the user B that includes the two words in common is large. On the other hand, the value indicated by the similarity between the input document data 300 of the user A and the input document data 302 of the user C and between the input document data 301 of the user B and the input document data 302 of the user C Will be small. For example, the calculation unit 11 calculates a value indicated by the similarity based on a ratio of words shared by the two input document data in the two input document data.

  Note that the calculation unit 11 may calculate a value indicated by the similarity by creating a document vector using an algorithm of tf (term frequency) -idf (inverse document frequency). Alternatively, the calculation unit 11 may calculate a value indicated by the similarity by using latent semantic analysis (Lent Semantic Analysis).

  In the input document data 300 to 304, the calculation unit 11 calculates a value indicated by the similarity regarding the combination of all the two input document data, and generates similarity management information 110 that is the calculated result. A configuration example of the similarity management information 110 is shown in FIG. In FIG. 4, the unit of the value indicated by the similarity management information 110 is a percentage. In the example illustrated in FIG. 4, for example, the value indicated by the similarity between the input document data 300 of the user A and the input document data 301 of the user B is 80%.

  Each time the calculation unit 11 detects that at least one of the input document data 300 to 304 or at least one of the dictionary data 210 to 214 has been updated, and performs the detection, the similarity management information 110 may be generated. Alternatively, the calculation unit 11 may generate the similarity management information 110 periodically or triggered by an instruction from a system administrator or the like.

  The registration unit 12 generates priority management information 120 based on the similarity management information 110 generated by the calculation unit 11. When the registration unit 12 registers the dictionary data that is not registered in the user dictionary data related to a specific user from the user dictionary data related to another user, the priority management information 120 is used to store dictionary data from which user dictionary data. This is information indicating whether or not to register with a higher conversion priority.

  A configuration example of the priority management information 120 is shown in FIG. In the similarity management information 110 shown in FIG. 4, the values indicated by the similarity between the input document data 301 to 304 regarding the input document data 300 of the user A are 80%, 10%, 41%, and 5%. Therefore, when the input document data 301 to 304 are arranged in the order of similarity with the input document data 300 of the user A, the input document data 301 of the user B, the input document data 303 of the user D, and the input document data of the user C 302, the input document data 304 of the user E. As a result, the registration unit 12 generates a record in the priority management information 120 indicating that the conversion priority regarding the user A is the order of the user B, the user D, the user C, and the user E. The registration unit 12 similarly generates records in the priority management information 120 for the users B to E.

  The registration unit 12 registers unregistered dictionary data in the dictionary data 400 to 404 based on the generated priority management information 120. An example of dictionary data registration for user dictionary data is shown in FIG.

  FIG. 6 shows an example in which the registration unit 12 registers unregistered dictionary data from the dictionary data 211 to 214 in the dictionary data 210 of the user A. The dictionary data 210 to 214 include conversion instruction records in which conversion target words and conversion candidate words are associated with each other. This conversion instruction record stores one or more words as conversion candidate words in association with the conversion priority.

  The dictionary data 210 of the user A defines the translation word 1-1 as the conversion candidate word for the source word 1 that is the conversion target word, and the conversion priority order has a high value as the conversion candidate word for the source word 2 that is the conversion target word. It is assumed that the translated word 2-1 and the translated word 2-2 are defined in order. The registration unit 12 searches the dictionary data 211 to 214 for combinations of conversion target words and conversion candidate words that are not defined by the dictionary data 210 of the user A.

  In the example shown in FIG. 6, as conversion candidate words that are not defined by the dictionary data 210 of the user A relating to the source language 1, the dictionary data 212 of the user C defines the translation 1-3, and the dictionary data 213 of the user D is the translation 1-2 is defined, and dictionary data 214 of user E defines translated words 1-4. The registration unit 12 refers to the priority management information 120 and confirms that the conversion priority of these users regarding the user A is the order of the user D, the user C, and the user E.

  Then, the registration unit 12 additionally defines the translated word 1-2 defined by the dictionary data 213 of the user D as a conversion candidate word having the second conversion priority for the original word 1 in the dictionary data 210 of the user A. The registration unit 12 additionally defines the translated words 1-3 defined by the dictionary data 212 of the user C as conversion candidate words with the conversion priority of the third place for the original word 1 in the dictionary data 210 of the user A. The registration unit 12 additionally defines the translated words 1-4 defined by the dictionary data 214 of the user E as conversion candidate words having the fourth conversion priority for the original word 1 in the dictionary data 210 of the user A.

  In the example shown in FIG. 6, user B dictionary data 211 and user D dictionary data 213 define source language 3 as conversion target words not defined by user A dictionary data 210. The dictionary data 211 of the user B defines a translation word 3-1 as a conversion candidate word for the original word 3. The dictionary data 213 of the user D defines a translation word 3-2 as a conversion candidate word for the original word 3. The registration unit 12 refers to the priority management information 120 and confirms that the conversion priority of these users regarding the user A is the order of the user B and the user D.

  Then, the registration unit 12 additionally defines the original language 3 as a conversion target word in the dictionary data 210 of the user A. The registration unit 12 additionally defines the translation 3-1 defined by the dictionary data 211 of the user B as a conversion candidate word having the first conversion priority for the original word 3 in the dictionary data 210 of the user A. The registration unit 12 additionally defines the translation 3-2 defined by the dictionary data 213 of the user D as a conversion candidate word having the second conversion priority for the original word 3 in the dictionary data 210 of the user A.

  Furthermore, although not shown in FIG. 6, for example, for a specific conversion target word, at least one of the dictionary data 211 to 214 is undefined in the dictionary data 210 of the user A, and a plurality of conversion priorities are different. Consider a case where conversion candidate words are defined. In this case, the registration unit 12 additionally defines the plurality of conversion candidate words as conversion candidate words related to the specific conversion target word in the dictionary data 210 of the user A while maintaining the order indicated by the conversion priority order. .

  As described above, an example of the operation in which the registration unit 12 registers the unregistered dictionary data in the dictionary data 210 of the user A from the dictionary data 211 to 214 has been described with reference to FIG. Then, the registration unit 12 performs the same processing as when registering unregistered dictionary data in the dictionary data 211 to 214 of each of the users B to E, even when registering unregistered dictionary data.

  The registration unit 12 generates the similarity management information 110 by detecting that the calculation unit 11 has updated at least one of the input document data 300 to 304 or at least one of the dictionary data 210 to 214. You may perform the process which registers dictionary data each time. Alternatively, the registration unit 12 may perform a process of registering dictionary data periodically or triggered by an instruction from a system administrator or the like. At this time, the registration unit 12 may perform a process of registering dictionary data in synchronization with the timing at which the calculation unit 11 generates the similarity management information 110, or may perform a process related to asynchronously.

  Next, the operation (process) of the user dictionary management system 1 according to the present embodiment will be described in detail with reference to the flowchart of FIG.

  The calculation unit 11 generates similarity management information 110 based on the user A to E input document data 300 to 304 (step S101). The registration unit 12 generates priority management information 120 based on the similarity management information 110 (step S102).

  The process enters a loop process up to step S107 with respect to X (X is any letter of A to E) (step S103). The registration unit 12 confirms whether the combination of the conversion target word and the conversion candidate word that is not registered in the dictionary data of the user X exists in other user dictionary data (step S104).

  If the combination does not exist in other user dictionary data (No in step S105), the process proceeds to step S107. When the combination exists in other user dictionary data (Yes in step S105), the registration unit 12 determines the combination of the unregistered conversion target word and conversion candidate word based on the information indicated by the priority management information 120. The user X is registered in the dictionary data (step S106).

  When X is not E, the registration unit 12 executes the process from step S103 for the next X, and when X is E, the entire process ends (step S107).

  The user dictionary management system 1 according to the present embodiment can efficiently and flexibly register dictionary information for the user dictionary used by the language conversion processing device. The reason is that the calculation unit 11 calculates the similarity between the user dictionary data, and the registration unit 12 extracts the dictionary data not registered in the specific user dictionary data from other user dictionary data. This is because it is registered in the specific user dictionary data in association with the value indicated by the conversion priority order based on the value indicated by.

  In a language conversion processing system, by registering dictionary data that is not registered in a user dictionary used by a specific user from a user dictionary used by another user, the accuracy of the language conversion processing processed by the specific user is improved. improves. At that time, in order to avoid registering unnecessary dictionary data in the user dictionary that is the registration destination, the value indicated by the similarity between the registration destination and the user dictionary that is the registration source related to the dictionary data is greater than or equal to the threshold value. There are systems that only register in some cases. However, in such a system, there is a possibility that registration of dictionary data that may be used by a user dictionary that is a registration destination may be excluded, and flexibility regarding registration of dictionary data is lacking.

  On the other hand, in the user dictionary management system 1 according to the present embodiment, the value indicated by the similarity is not used as a criterion for determining whether or not to register dictionary data in the registration destination user dictionary. This is used as a criterion for determining the conversion priority in the user dictionary of the registration destination for dictionary data. That is, the registration unit 12 according to the present embodiment increases the value indicated by the conversion priority in the user dictionary that is the registration destination of the dictionary data from the user dictionary that is the registration source having a high value indicated by the similarity. Register as follows. The registration unit 12 does not exclude the dictionary data from the user dictionary that is the registration source having a low value indicated by the similarity, but the value indicated by the conversion priority in the user dictionary that is the registration destination is low. sign up. Thereby, the user dictionary management system 1 according to the present embodiment can efficiently and flexibly register dictionary data for the user dictionary used by the language conversion processing device.

  Note that the calculation unit 11 according to the present embodiment calculates the value indicated by the similarity between the input document data for the input document data 300 to 304, but each user is based on the dictionary data 210 to 214. A value indicated by the similarity between the dictionary data may be calculated.

  Further, the user dictionary management system 1 according to the present embodiment registers dictionary data in the user dictionary every time at least one of the input document data 300 to 304 or at least one of the dictionary data 210 to 214 is updated. Dynamic registration processing can be performed. Alternatively, the user dictionary management system 1 according to the present embodiment can also perform static registration processing for registering dictionary data periodically or triggered by an instruction from a system administrator or the like. The user dictionary management system 1 may perform dynamic registration processing when it is desired to improve the accuracy related to the language conversion processing as much as possible. On the other hand, the user dictionary management system 1 may perform static registration processing when it is desired to reduce the system load caused by the registration processing. That is, the user dictionary management system 1 can perform flexible operation regarding the process of registering dictionary data in the user dictionary in accordance with the required specifications for the system.

  Note that the user dictionary management system 1 according to the present embodiment may be constructed as a cloud service type system in which all the components shown in FIG. 1 are included in the server device. Alternatively, the user dictionary management system 1 may be constructed as a system in which user dictionary data, user input document data, and user output document data are included in a client terminal device used by each user. In this case, the user dictionary management system 1 according to the present embodiment may calculate the value indicated by the similarity between user input document data on the server device side, or a peer-to-peer type between client terminal devices The communication may be performed on the client terminal device side.

<Second Embodiment>
FIG. 7 is a block diagram conceptually showing the structure of the user dictionary management apparatus 50 of the second embodiment.

  The user dictionary management device 50 according to the present embodiment includes a calculation unit 51 and a registration unit 52.

  The calculation unit 51 calculates a value indicated by the similarity regarding the first user dictionary information 600 and the second user dictionary information 601 among the plurality of user dictionary information based on a predetermined criterion. The plurality of user dictionary information includes a conversion instruction record. The conversion instruction record is a record that is referred to when the language conversion processing device 60 performs language conversion processing on user document information. Then, the conversion instruction record stores the conversion target word that is the word before conversion, one or more conversion candidate words that are the converted word, and the value indicated by the conversion priority regarding the conversion candidate word. ing.

  The registration unit 52 associates the information indicated by the conversion instruction record included in the first user dictionary information 600 with the value indicated by the conversion priority based on the value indicated by the similarity, and sets the second user dictionary information. Register at 602.

  The user dictionary management device 50 according to the present embodiment can efficiently and flexibly register dictionary information for the user dictionary used by the language conversion processing device. The reason is that the calculation unit 51 calculates the similarity between the user dictionary information, and the registration unit 52 extracts the dictionary information that is not registered in the specific user dictionary information from other user dictionary information. This is because it is registered in the specific user dictionary information in association with the value indicated by the conversion priority order based on the value indicated by.

<Hardware configuration example>
In each embodiment described above, each unit illustrated in FIG. 1 and FIG. 7 can be realized by a dedicated HW (Holdware) (electronic circuit). Further, at least the calculation units 11 and 51 and the registration units 12 and 52 can be regarded as a function (processing) unit (software module) of the software program. However, the division of each part shown in these drawings is a configuration for convenience of explanation, and various configurations can be assumed for mounting. An example of the hardware environment in this case will be described with reference to FIG.

  FIG. 8 is a diagram illustrating an exemplary configuration of an information processing apparatus 900 (computer) that can execute the user dictionary management apparatus according to the exemplary embodiment of the present invention. That is, FIG. 8 shows a configuration of a computer (information processing apparatus) that can realize the user dictionary management apparatus shown in FIGS. 1 and 7, and a hardware environment that can realize each function in the above-described embodiment. Represents.

  The information processing apparatus 900 illustrated in FIG. 8 includes a CPU 901, a ROM (Read_Only_Memory) 902, a RAM (Random_Access_Memory) 903, a hard disk 904 (storage device), and a communication interface 905 (Interface: hereinafter referred to as “I / F”). A reader / writer 908 capable of reading and writing data stored in a storage medium 907 such as a CD-ROM (Compact_Disc_Read_Only_Memory), and an input / output interface 909, which are connected via a bus 906 (communication line). General computer.

  The present invention described using the above-described embodiment as an example is the block configuration diagram (FIG. 1 and FIG. 7) referenced in the description of the embodiment for the information processing apparatus 900 illustrated in FIG. 8. After supplying a computer program capable of realizing the functions of the calculation units 11 and 51, the registration units 12 and 52, or the flowchart (FIG. 2), the computer program is read and interpreted by the CPU 901 of the hardware. Is achieved. The computer program supplied to the apparatus may be stored in a readable / writable volatile storage memory (RAM 903) or a nonvolatile storage device such as the hard disk 904.

  In the above-described case, the computer program can be supplied to the hardware by a method of installing in the apparatus via various storage media 907 such as a CD-ROM, or an external method via a communication line such as the Internet. A general procedure can be adopted at present, such as a method of downloading more. In such a case, it can be understood that the present invention is configured by a code constituting the computer program or a storage medium 907 in which the code is stored.

  The present invention has been described above using the above-described embodiment as an exemplary example. However, the present invention is not limited to the above-described embodiment. That is, the present invention can apply various modes that can be understood by those skilled in the art within the scope of the present invention.

DESCRIPTION OF SYMBOLS 1 User dictionary management system 10 User dictionary management apparatus 11 Calculation part 110 Similarity management information 12 Registration part 120 Priority management information 20 Language conversion processing apparatus 21 User dictionary data storage part 210 thru | or 214 Dictionary data 30 User input document data storage part 300 Through 304 input document data 40 user output document data storage unit 400 through 404 output document data 50 user dictionary management device 51 calculation unit 52 registration unit 60 language conversion processing device 600 first user dictionary information 601 second user dictionary information 900 information Processing device 901 CPU
902 ROM
903 RAM
904 Hard disk 905 Communication interface 906 Bus 907 Storage medium 908 Reader / writer 909 Input / output interface

Claims (7)

The language conversion processing device refers to the user document information when performing language conversion processing. The conversion target word that is the word before conversion, one or more conversion candidate words that are the converted word, and the conversion priority for the conversion candidate word Based on a predetermined criterion, a value indicated by the similarity regarding the first and second user dictionary information among a plurality of user dictionary information including a conversion instruction record that stores the value indicated by the rank in association with each other is calculated. A calculation means;
When the conversion candidate word in which different values are registered in the plurality of first user dictionary information with respect to the specific conversion target word is registered in the second user dictionary information, Registration means for registering each of the conversion candidate words so that the value indicated by the conversion priority is higher in the order indicating the degree of similarity of the user dictionary information is greater ;
A user dictionary management device comprising: The calculation means detects that the first user dictionary information has been updated, calculates a value indicated by the similarity,
The registration unit updates the second user dictionary information each time the calculation unit detects that the first user dictionary information is updated.
The user dictionary management apparatus according to claim 1 . The calculation means calculates a value indicated by the similarity at a first predetermined time,
The registration means updates the second user dictionary information at a second predetermined time.
The user dictionary management apparatus according to claim 1 or 2 . Said calculating means includes a first said user document information converted using the first user dictionary information, and the second of the user document information is converted using said second user dictionary information To calculate a value indicated by the similarity,
The user dictionary management apparatus according to any one of claims 1 to 3 . The calculation means calculates the similarity based on a ratio of the conversion target word that the first and second user document information includes in common to the first and second user document information. Calculate the value shown,
The user dictionary management apparatus according to claim 4 . Depending on the information processing device,
The language conversion processing device refers to the user document information when performing language conversion processing. The conversion target word that is the word before conversion, one or more conversion candidate words that are the converted word, and the conversion priority for the conversion candidate word Based on a predetermined criterion, a value indicated by the similarity with respect to the first and second user dictionary information among a plurality of user dictionary information including a conversion instruction record that is stored in association with the value indicated by the rank is calculated. ,
When the conversion candidate word in which different values are registered in the plurality of first user dictionary information with respect to the specific conversion target word is registered in the second user dictionary information, Register each of the conversion candidate words so that the value indicated by the conversion priority is higher in the order indicating that the similarity with respect to the user dictionary information is greater ;
User dictionary management method. The language conversion processing device refers to the user document information when performing language conversion processing. The conversion target word that is the word before conversion, one or more conversion candidate words that are the converted word, and the conversion priority for the conversion candidate word Based on a predetermined criterion, a value indicated by the similarity regarding the first and second user dictionary information among a plurality of user dictionary information including a conversion instruction record that stores the value indicated by the rank in association with each other is calculated. Calculation process,
When the conversion candidate word in which different values are registered in the plurality of first user dictionary information with respect to the specific conversion target word is registered in the second user dictionary information, A registration process for registering each of the conversion candidate words such that the value indicated by the conversion priority is higher in an order indicating that the similarity with respect to the user dictionary information is greater ;
User dictionary management program that causes a computer to execute.

Images