JP2003157257A - Dictionary device and natural language processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a dictionary device and a natural language processing system by which a high quality natural language processing result is obtained by sharing dictionary data concerning user registration keeping high-quality contents by a plurality of users. SOLUTION: The dictionary device is provided in the natural language processing system used by the plurality of users. The dictionary device is characterized in having a dictionary main body which hierarchizes a plurality of special word dictionaries by a tree structure with a general term dictionary as a route node, a dictionary changing means for changing a dictionary and/or dictionary data such as moving dictionary data among a plurality of dictionaries including a user dictionary and generating a dictionary, and a dictionary history recording means for recording the history of changing of the dictionary and/or dictionary data by the dictionary changing means. The natural language processing system is characterized in having the dictionary device and a natural language processing main body for performing natural language processing by utilizing the stored contents of this dictionary device.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dictionary device and a natural language processing system, for example, a machine translation system used by a plurality of users for translating a source language into a target language, and a dictionary device as an element of the machine translation system. Can be applied to.

[0002]

2. Description of the Related Art A machine translation system generally has a basic system dictionary and a user dictionary registered by the user. When a plurality of users use the same machine translation system, if a user dictionary is registered for each user, the same term must be individually registered by the plurality of users. For example, almost all users register a newly appeared new word that is used by almost all users.

As a conventional technique for avoiding such an inconvenience, there is one disclosed in Japanese Laid-Open Patent Publication No. 11-134339. According to this publication, when a machine translation system is shared by a plurality of users, it is possible for a plurality of users to search by searching from a user dictionary with a higher priority or by extracting a parallel translation with the latest registration update time from the user dictionary. Disclosed is a system capable of efficiently sharing and using a created user dictionary.

[0004]

In the above-mentioned prior art, it is necessary to specify the priority relationship in order to determine the priority order of the user dictionary. The priority relationship is a priority between users who own the user dictionary, and indicates that a user dictionary owned by a user having a higher priority has higher reliability.

However, it is necessary to compare the contents of the user dictionaries owned by each user and specify the priority relationship among the users.
It is necessary to have the labor and the judgment ability to confirm the parallel translation in each user dictionary and determine the superiority or inferiority. Also, if the number of registrations is large,
Ranking among users is difficult because there are advantages and disadvantages depending on the words. Furthermore, there is no guarantee that the priority relationship will be the same in the future, and the priority relationship must be reviewed frequently.

Further, in the above-mentioned conventional technique, when there are source language words in a plurality of user dictionaries, the bilingual translation with the latest registration update time is extracted. , The latest translation is not always reliable. Therefore, in order to obtain a high quality translation result, it is necessary to constantly check the translation result and a plurality of user dictionaries.

The present invention has been made in view of the above problems, and a plurality of users maintain high-quality contents.
By sharing the dictionary data related to user registration,
An object of the present invention is to provide a dictionary device and a natural language processing system that can obtain high quality natural language processing results.

[0008]

In order to solve such a problem, the first aspect of the present invention provides a dictionary device provided in a natural language processing system used by a plurality of users, wherein a plurality of technical term dictionaries for each category are provided. A dictionary body that is hierarchically structured in a tree structure with a general term dictionary as a root node, a user dictionary registration unit that sets a user dictionary in association with any of the technical term dictionary or the general term dictionary, and a plurality of user dictionaries. Set common dictionary data in the associated technical term dictionary and / or the general term dictionary, set common dictionary data in multiple lower-level dictionaries in the tree structure in the upper-level dictionary, Based on the dictionary data common to the user dictionary, a lower-level technical term dictionary is set in the technical term dictionary, and the dictionary and / or the dictionary data is set. A dictionary changing means for changing the data, characterized by having a dictionary history recording means for recording the history of changes in the dictionary and or dictionary data by the dictionary changing means.

The natural language processing system according to the second aspect of the present invention includes the dictionary device according to the first aspect of the present invention and a natural language processing main body for performing natural language processing using the stored contents of the dictionary device.

[0010]

BEST MODE FOR CARRYING OUT THE INVENTION (A) First Embodiment A first embodiment in which a dictionary device and a natural language processing system according to the present invention are applied to a machine translation system will be described in detail below with reference to the drawings.

(A-1) Configuration of the First Embodiment FIG. 1 is a block diagram showing the functional configuration of the machine translation system of the first embodiment.

In FIG. 1, the machine translation system 100 of the first embodiment has a browsing input means 101 and a control means 10.
2, machine translation means 103, dictionary management means 104, dictionary database 105 and dictionary history recording means 106.

The machine translation system 100 according to the first embodiment may be composed of a single information processing device (for example, a personal computer) having a predetermined processing program and data, or a plurality of information processing devices. It may be configured with. As the latter, the browsing input means 101 is installed in an information processing device (client terminal), and the other control means 102, machine translation means 103, dictionary management means 104, dictionary database 105 and dictionary history recording means 106 are used for other information. An example in which it is installed in a processing device (server), and an example in which the machine translation means 103 is also installed in another information processing device can be given. Also, the browsing input means 1
There may be a plurality of 01 and a plurality of machine translation means 103.

The browsing input means 101 is used by the user to input a translation request document (source language document), browse the translation result, and register a word (translation of source language and target language) by the user. This is for performing, browsing user registration contents, and browsing operation guidance messages. The input unit of the browsing input unit 101 is not limited to a general input configuration such as a keyboard and a mouse, but is also a drive configuration when a translation request document is read from a flexible disk and input.
The browsing unit of the browsing input unit 101 mainly corresponds to a display, but a printer also corresponds to a unit that prints out a translation result or the like instead of displaying it.

The control means 102 comprises a browsing input means 101,
It is connected to the machine translation unit 103 and the dictionary management unit 104, and controls which document is translated and which dictionary is used for translation.

The machine translation means 103 translates a source language document given via the control means 102 into a target language document and sends it back to the control means 102.
In the case of the above embodiment, any existing machine translation method may be used. The machine translation means 103 is also provided with dictionary data necessary for machine translation via the control means 102.

The dictionary management means 104 accommodates the dictionary database 105, and under the control of the control means 102, creates a user dictionary, edits dictionary data, creates a technical term dictionary, integrates, divides, creates, etc. It is a means.

The dictionary history recording means 106 is provided in association with the dictionary management means 104, and is a means for recording a history such as a registration update time of a dictionary or dictionary data for each dictionary data.

The dictionary database 105 is a technical term dictionary database that stores user dictionaries classified by field. The dictionary database 105 may also be constructed by being distributed to a plurality of information processing devices.

The dictionary database 105 has a hierarchical structure as shown in FIG. That is, the plurality of dictionaries stored in the dictionary database 105 have a tree structure, with the general term dictionary as a root node, and in the next layer, there are specialized term dictionaries in the case of major classification of specialized fields and genres. It is provided, and in the lower hierarchy, a technical term dictionary for further classifying the specialized fields and genres of the upper hierarchy is provided, and the number of branches (depth of the hierarchy) reaching the end node is not uniform, and As shown in FIG. 2, the user dictionaries Dm1 to DmN are associated with each technical term dictionary Dm.
Is provided.

Each user can create a user dictionary in the field of any technical term dictionary. The contents of this user dictionary are used not only when the created user requests translation, but also when other users request translation when the contents are transferred to the associated technical term dictionary and registered. To be done.
By the function of the dictionary management means 104 described above, the dictionary of the field of the document and the technical term dictionary belonging to the upper hierarchy of the field are used at the time of translation. In the case of the example of the hierarchical structure shown in FIG. 1, a computer dictionary, an engineering dictionary, and a general term dictionary are used when translating a document in the field of computers.

(A-2) Operation of the First Embodiment (A-2-1) User Dictionary Registration Next, referring to the flowchart of FIG. 3, the operation when the user registers a word in the user dictionary will be described. explain.

For example, the user uses the browsing / inputting means 101 to instruct a user registration mode for words. At this time,
The control means 102 and the dictionary management means 103 start the user dictionary registration processing shown in FIG. 3, and first determine whether or not the user is a person who can execute user registration of a word (step S1). Although not shown, user information that can be registered by the user is also stored in the dictionary database 105.

If the user is not a person who can execute the user registration of the word, the control means 102 returns a message to that effect and ends the user dictionary registration processing (step S2).

On the other hand, when the user is a person who can execute user registration of words, the control means 102 causes the dictionary management means 10 to operate.
4, the tree structures of the general term dictionary and the technical term dictionary are taken out, and the user dictionary registration input screen including the information of the tree structure is transmitted to the browsing input means 101 (step S
3) Wait for the registered dictionary data from the browsing / inputting means 101 (step S4).

When the user dictionary registration input screen is displayed, the user inputs the dictionary data to be registered, and
Based on the displayed tree structure of the general term dictionary and the technical term dictionary, information of the dictionary to be associated is added to the registered dictionary data and transmitted to the control unit 102. Note that a plurality of registered dictionary data may be collectively transmitted. Below, for simplification of description, it is assumed that one registration dictionary data is transmitted.

When the registered dictionary data is given, the control means 102 gives it to the dictionary management means 104, and the dictionary management means 104 relates to this user's registered dictionary data based on the stored contents of the dictionary database 105. It is determined whether or not the user dictionary (the user dictionary associated with the technical term dictionary or the general term dictionary) is already set (step S5).

If the user dictionary is not set,
After setting such a user dictionary (step S6),
If it has already been set, the browsing input means 10 is immediately entered.
The registered dictionary data given from No. 1 is registered, and the history of registering the dictionary data is recorded in the dictionary history recording means 106 (step S7), and a series of user dictionary registration processing is ended.

The dictionary database 105 has a hierarchical structure functionally as shown in FIG. 1, but in practice, for example, the registered dictionary data can be stored in the dictionary regardless of the type of dictionary. It is to be sequentially stored together with the data representing the type, and the dictionary history recording means 10 is stored in such a record.
The history of dictionary data according to 6 is recorded in association with each other.

(A-2-2) Enrichment Operation of Technical Term Dictionary Next, the operation of enriching the technical term dictionary will be described with reference to the flowchart of FIG.

Incidentally, FIG. 4 shows that when the system is activated periodically,
Alternatively, it corresponds to the case where it is activated by the system operator. Further, FIG. 4 shows processing when a certain one technical term dictionary is targeted, and the processing shown in FIG. 4 is executed for all the technical term dictionaries.

When the processing shown in FIG. 4 is started, the dictionary management means 104 fetches the information of all the user dictionaries from the dictionary database 105 while aggregating the same dictionary data for each related technical term dictionary and buffering them. Yes (step S11).

FIG. 5 shows an example of taking out data while tabulating. From the left, information (dictionary data ID) numbered for different dictionary data, headwords, translated words, and the number of registered words are shown. Note that the dictionary data IDs are sequentially numbered from 0.

Upon completion of the extraction while totaling, the dictionary management means 104 sets the dictionary data ID to 0 (initial value) (step S12).

Then, the processing loop of steps S13 to S17 is repeated until the enrichment processing is completed for all the dictionary data.

In the dictionary management means 104, the registered number is
It is determined whether or not there is dictionary data of a majority of the number of user dictionaries associated with the technical term dictionary (steps S13 and S14).

If even one such dictionary data exists, all the dictionary data are additionally registered in the technical term dictionary (step S15). At this time, it is confirmed that the dictionary data does not exist in the technical term dictionary and registered.

After such additional registration of the dictionary data to the technical term dictionary is completed, or when there is no dictionary data having the number of registrations equal to or more than the majority of the user dictionaries, the dictionary data ID is incremented by 1. (Step S1
6) It is determined whether or not unprocessed dictionary data remains (step S17).

If unprocessed dictionary data remains, the process returns to step S13 described above, and if the enrichment process is completed for all dictionary data, the series of processes shown in FIG. 4 is completed.

In the example of FIG. 5, if the number of user dictionaries associated with the technical term dictionary is 5, the dictionary data “pencil; pencil” has three user dictionary registrations, which is more than half. Through the process of FIG.
It is registered in the target technical term dictionary.

In the above description, the registration condition for the technical term dictionary is set to be a majority of the number of user dictionaries or more, but it may be a registration requirement in which another ratio is used as a threshold value. Further, the threshold value defining the registration requirement may be different depending on the number of user dictionaries.

In the above description, the dictionary data is left in the user dictionary even after it is additionally registered in the technical term dictionary. However, the dictionary data additionally registered in the technical term dictionary is deleted from the user dictionary. May be. In the former case, in the additional registration process of step S15, the dictionary history recording means 106 records the history of additionally registering the dictionary data in the corresponding technical term dictionary, and in the latter case, the dictionary data belongs. A history indicating that the dictionary type has been changed or deleted from the user dictionary is recorded.

Further, in the above description, the registration in the technical term dictionary is judged from the registration status in all the user dictionaries associated with the technical term dictionary, but in some user dictionaries. Whether or not to register in the technical term dictionary may be determined based on the registration status. Furthermore,
In the above, the registration to the technical term dictionary was completely automated, but after displaying the dictionary data that the server operator wants to additionally register to the technical term dictionary and confirming it, it is additionally registered to the technical term dictionary. It may be done.

If a user dictionary exists in association with the general term dictionary, the general term dictionary may be expanded in the same manner as the above processing.

(A-2-3) Operation of Integrating Dictionary Information Next, the operation of integrating the information of a plurality of technical term dictionaries will be described with reference to the flowchart of FIG. The information integration of a plurality of technical term dictionaries refers to checking the information of the data of the technical term dictionary and registering the data of the lower technical term dictionary as the data of the upper technical term dictionary.

Incidentally, FIG. 6 shows that, when it is activated periodically,
Alternatively, it corresponds to the case where it is activated by the system operator. Further, FIG. 6 shows the processing in the case where one certain technical term dictionary is the target, and the processing shown in FIG. 6 is executed for all the technical term dictionaries.

When the processing shown in FIG. 6 is started, the dictionary management means 104 first extracts the dictionary data registered in all the technical term dictionaries subordinate to the target technical term dictionary (step S21), and the dictionary. The data ID is set to 0 (initial value) (step S22).

Then, it is determined whether or not there is dictionary data registered in a majority of the number of lower technical term dictionaries (steps S23 and S24).

If there is even one such dictionary data, all of these dictionary data are additionally registered in the upper technical term dictionary, and those dictionary data are deleted from the lower technical term dictionary (step S25). ). At this time, it may be possible to confirm that the dictionary data does not exist in the upper technical term dictionary and then additionally register the dictionary data.

Here, in the additional registration process of step S25, the dictionary history recording means 106 changes the dictionary type to which the dictionary data belongs to the higher technical term dictionary or deletes it from the lower technical term dictionary. A history indicating the fact is recorded.

After the above processing is completed, the dictionary data ID
Is incremented by 1 (step S26), and it is determined whether or not unprocessed dictionary data remains (step S2).
7). If unprocessed dictionary data remains, the process returns to step S22 described above, and if the integration process is completed for all dictionary data, the series of processes shown in FIG. 6 is completed.

In the example of the dictionary database 105 shown in FIG.
When the same dictionary data is registered in both the physical dictionary and the mathematical dictionary, the dictionary data is registered in the upper natural science dictionary, and the dictionary data is deleted from the physical dictionary and the mathematical dictionary.

Here, by performing the siphoning process (integration process) in order from the technical term dictionary with the lower hierarchy to the technical term dictionary with the higher hierarchy, the dictionary data (word information) that has just been siphoned is obtained. It may also be downloaded to a higher-level technical term dictionary.

In the above description, the registration condition for the upper technical term dictionary is assumed to be registered with a majority of the number of lower technical term dictionaries or more, but it may be a registration requirement with another ratio as a threshold value. . Also, depending on the number of lower technical term dictionaries,
The thresholds defining the registration requirements for higher ranks may be different. Furthermore, each technical term dictionary manages the usage frequency and usage rate of the dictionary data, and the fact that it is dictionary data (word information) that is used to some extent in multiple lower technical term dictionaries also means higher technical terms. It may be included in the dictionary registration requirement.

In the above description, when additional registration is performed in the higher technical term dictionary, the dictionary data is deleted from the lower technical term dictionary. However, the dictionary data additionally registered in the upper technical term dictionary is inferior. You may leave it in the technical term dictionary of. Further, in the above, although the one in which the registration in the higher-level technical term dictionary is completely automated is shown, the system administrator or the like displays and confirms the dictionary data to be additionally registered in the upper-level technical term dictionary, and then, It may be additionally registered in a higher technical term dictionary.

The dictionary data registered in a plurality of large-class technical term dictionaries may be registered in the general term dictionary.

(A-2-4) Dictionary Division / Generation Operation Next, the operation of dividing / generating the technical term dictionary will be described with reference to the flowchart of FIG. It should be noted that the process of FIG. 7 shows a process for a certain technical term dictionary,
The process of FIG. 7 is executed for each of the technical term dictionaries.

When the processing shown in FIG. 7 is started, the dictionary management means 104 fetches the information of the related user dictionary from the dictionary database 105 while collecting the same dictionary data and buffering it (step S31).

FIG. 8 shows an example in which data is collected from the user dictionary related to the sports dictionary while being collected. From the left, the dictionary data ID, the entry word (key), the translated word, and the number of registered user dictionaries are shown.

When the extraction while the tabulation is completed, the dictionary management means 104 sets the dictionary data ID to 0 (initial value) (step S32). Then, the processing loop in steps S33 to S38 is repeated for all the dictionary data.

The dictionary managing means 104 confirms whether or not there is dictionary data having the same headword (key) and different translation (value) in the registered dictionary data of the user dictionary associated with the technical term dictionary, In some cases, the number of user dictionaries registering each dictionary data is 20% of the total number of user dictionaries associated with the technical term dictionary (note that
It is determined whether or not the ratio is not less than this (steps S33 and S34).

If there is user-registered dictionary data with the same headword (key) and different translations (values), and the registration ratio of each dictionary data is 20% or more, the dictionary management means 10
4 determines whether or not there is a lower technical term dictionary with respect to the technical term dictionary (step S35).

When there is no subordinate technical term dictionary, the dictionary management means 104 is subordinate to the number of types of dictionary data having the same headword (key) but different translations (values) and registration rates of 20% or more. A technical term dictionary is created and each dictionary data is registered (step S36). Here, the user dictionary including such dictionary data is the dictionary data I
It may be directly associated with the technical term dictionary defined by D, or may be associated with a newly created technical term dictionary in which its own registered dictionary data is registered. Further, the creation of the subordinate technical term dictionary here may not be complete creation but may be provisional creation.

In the processing of step S36 described above, the history of the creation of the new technical term dictionary as described above and the history of the registration of certain dictionary data in the dictionary are recorded by the dictionary history recording means 106. It is also recorded.

On the other hand, when there is a lower-level technical term dictionary, the dictionary management means 104 stores dictionary data having the same entry word (key) but different translation words (values) and a registration rate of 20% or more. , To a lower technical term dictionary (step S37). Here, the user dictionary including such dictionary data may be directly associated with the original technical term dictionary or may be associated with the moved lower technical term dictionary.

In the above-described processing of step S37, the history of the dictionary data being moved to the lower technical term dictionary is also recorded by the dictionary history recording means 106.

The type of the lower technical term dictionary to which the dictionary data is moved is determined by recognizing the field or genre, for example, by referring to the contents of the backward dictionary database having the translated word as the index word. In addition, for example, a part of or all of the character string in the translated word of the current dictionary data is moved to the lower technical term dictionary on the side where the dictionary data including the translated word character string exists. The movement of the dictionary data to the lower technical term dictionary here may be a temporary movement instead of a complete movement.

The dictionary managing means 104 completes the creation of the lower technical term dictionary or the movement of the predetermined dictionary data to the lower technical term dictionary as described above, or the translated word (value) with the same headword (key). ) Are different, there is no user-registered dictionary data, or even if they exist, if the registration ratio of any of the dictionary data is less than 20%, the dictionary data I
After incrementing D by 1 (step S38), it is determined whether or not dictionary data remains (step S3).
9).

If the dictionary data remains, the process returns to step S33 described above, and if the dictionary division / generation processing is completed for all the dictionary data, the series of processing shown in FIG. 7 is completed.

In the dictionary division / generation processing, complete automation processing is difficult, and it is necessary to perform post-processing by an operator on the system operator side (hereinafter referred to as system operator) at the end of the processing of FIG. For example, if a newly created technical term dictionary is assigned a field or genre, if the creation in step S36 is a temporary creation, it is determined whether or not the main creation is necessary, and if the move in step S37 is a temporary move. It is necessary to determine whether or not the main movement is necessary, and to move other dictionary data from the upper technical term dictionary when a new lower technical term dictionary is created.

For example, assuming that the number of user dictionaries associated with the sports dictionary is 10 in the example of FIG. 8, two types of dictionary data, the headword is "Pitcher" and the translated words are "pitcher" and "7th iron". In both cases, since the number of user registrations is 3, the requirement for newly creating a lower technical term dictionary or the requirement for moving dictionary data of a lower technical term requirement is satisfied.

In this case, if there is no subordinate technical term dictionary in the sports dictionary, two subordinate technical term dictionaries are created, and the dictionary data "Pitc" is set in one of them.
"her; 7 iron" is registered, and dictionary data "Pitcher;pitcher" is registered on the other side. Note that the post-processing by the system operator will give the former a golf dictionary and the latter a baseball dictionary.

On the other hand, when a golf dictionary and a baseball dictionary exist under the sports dictionary, the dictionary data "Pitcher;7-iron" is registered in the golf dictionary, and the dictionary data "Pitcher;pitcher" is baseball. Registered in the dictionary. This is because, for example, if another dictionary data whose translation word is "iron" exists in the golf dictionary, a common character string with the current translation word "7 iron" exists. This is also because, for example, if the baseball dictionary has other dictionary data whose translation word is "right wing", a common character string with the current translation word "pitcher" exists.

FIG. 9 shows a change in the dictionary database 105 when a golf dictionary and a baseball dictionary are newly created under the sports dictionary through the process of FIG. 7 from the state where only such a sports dictionary is prepared. FIG. 10 shows the state, and FIG. 10 shows the movement of the associated user dictionary when the technical term dictionary is divided / generated as shown in FIG.

In the above description, the presence / absence of creation of the lower technical term dictionary is judged only from the registered information of the user dictionary, but from the registered information of the user dictionary and the registered information of the technical term dictionary, It may be possible to determine whether or not a lower technical term dictionary is created. For example, when user-registered dictionary data having the same headword as the registered information of the technical term dictionary but different translations is registered in a predetermined number or more of user dictionaries, a lower technical term dictionary is created. May be performed.

Further, in the above description, the lower technical term dictionary is created only when the lower technical term dictionary does not exist. However, even if the lower technical term dictionary exists, it is in the same row with that. It is also possible to create a lower technical term dictionary of the same level. That is, regardless of the presence or absence of the lower technical term dictionary, another lower technical term dictionary may be created if the condition is satisfied (in other words, the process of moving to the lower technical term may be omitted). For example, even if there is a judo dictionary or a land dictionary under the sports dictionary, as described above, “Pitche” is associated with the sports dictionary.
In some cases, "r; 7th iron" or "Pitcher;pitcher" is registered as a user. In this case, a golf dictionary or a baseball dictionary (naming will be done later) is created as a lower technical term dictionary.

(A-2-5) Dictionary Browsing Operation Next, the operation of browsing the contents stored in the dictionary will be described with reference to the flowchart of FIG.

The user may want to browse the dictionary contents in advance, such as when registering dictionary data or when confirming what translated word a source language word is translated into before machine translation processing. is there.

In such a case, the user inputs the browsing request by the browsing input means 101 and controls the control means 102.
To be transmitted (step S41).

Here, the browsing request may be a request including only the headword (original language word), may be a request including only the translated word, and further includes a pair of the headword and the translated word. It may be one. Further, it may be a browsing request that limits the type of dictionary to be searched. When a technical term dictionary is limited as a search target dictionary, only the technical term dictionary may be the search target, and the technical term dictionary and all user dictionaries associated with it may be the search target. It may be done.

The control means 102 gives the received browsing request to the dictionary management means 104, and the dictionary management means 104 searches the dictionary data in accordance with the browsing request, and the history information relating to the dictionary data obtained by the search is also dictionary. It is taken out from the history recording means 106 (step S42).

After that, the dictionary management means 104 controls the dictionary data (including history information) obtained by the search to the control means 102.
A reply is sent to the browsing input means 101 via the, and the browsing input means 101 displays or prints out (step S43).

(A-2-6) Machine Translation Process Next, the machine translation operation for the input sentence or the input document will be described with reference to the flowchart of FIG.

If there is a sentence or a document desired to be machine translated, the user inputs it into the browsing input means 101 and controls it.
02, and the control means 102 gives the received input sentence or input document to the machine translation means 103 (step S5).
1). When the input document is an HTML document, the control unit 102 cuts out only the text portion and gives it to the machine translation unit 103.

When the user wants to limit the dictionary used for machine translation, he / she inputs the designated dictionary information to the browsing / inputting means 101 and causes it to be transmitted to the control means 102, and the control means 1
Dictionary management means 104 to which designated dictionary information is given from 02.
Validates only the relevant dictionary in the dictionary database 105 (step S52).

As described above, the designated dictionary and the technical term dictionary and general term dictionary belonging to the upper hierarchy of the field are validated, and are associated with the validated technical term dictionary and general term dictionary. The user dictionary of the user is valid. In the example of the machine translation process shown in FIG. 12, the user specifies the dictionary to be used. However, the existing specialized field recognition method is applied and the dictionary management unit 104 uses the input sentence or the input document. The dictionary to be used may be automatically recognized.

After that, the machine translation means 103 performs machine translation while referring to the stored contents of the activated dictionary (step S53). Here, all the stored contents of the activated dictionary may be given to the machine translation means 103 to execute the machine translation, and the machine translation means 103 may cause the control means 1 to execute the machine translation.
The machine translation may be executed by referring to the stored contents of the valid dictionary each time via 02 and the dictionary management means 104.

If the dictionary history recording means 106 or the dictionary database 105 manages the usage frequency and usage rate of dictionary data, the usage frequency and usage rate of dictionary data used in machine translation is updated. The usage frequency and usage rate of the dictionary data may be browsed at the time of the browsing request for the dictionary data.

The obtained translation result is machine translation means 103.
Is given to the control means 102, and the control means 102 replies to the browsing input means 101 (step S54). When the input document (translation request document) is a document such as an HTML document and the control means 102 cuts out only the text portion and gives it to the machine translation means 103, the control means 102.
Reconstructs the HTML document from the translation result and sends it back to the browsing input means 101.

(A-3) Effect of First Embodiment As described above, since a plurality of users share a dictionary database that maintains high-quality contents, high-quality translation results can be obtained, and , The dictionary database changes frequently to improve quality, so it is executed frequently, so it can be used in the latest state at all times, and the maintenance man-hours can be reduced.

Further, since the user can not only browse the dictionary data of such a dictionary database but also the update history thereof, not only can the user know the necessity of user registration, but also what other users can do. It is possible to know when it was updated and when it was downloaded to the technical term dictionary, and to recognize the current state of the world for each term.

(B) Second Embodiment Next, a second embodiment in which the dictionary device and the natural language processing system according to the present invention are applied to a machine translation system will be described in detail with reference to the drawings.

(B-1) Configuration of the Second Embodiment FIG. 13 is a block diagram showing the functional configuration of the machine translation system of the second embodiment, and FIG. 1 according to the first embodiment. The same or corresponding parts are designated by the same reference numerals.

In FIG. 13, the machine translation system 100A of the second embodiment has the same browsing input means 101, control means 102, machine translation means 103 as in the first embodiment,
It has a dictionary management unit 104, a dictionary database 105, and a dictionary history recording unit 106, and further has a network 201, a communication unit 202, and an update information request database (update information request DB) 203.

That is, in the machine translation system 100A of the second embodiment, the browsing / inputting means 101 is the network 2
01 to other system components 102 to 106, 2
It is connected to the 02 and 203 sides.

For example, the network 201 is the Internet, the system components 102 to 106, 202 and 203 other than the browsing / inputting means 101 are installed in a server (for example, a provider server), and the browsing / inputting means 101 is , An information processing terminal having a browser function is applicable.

The first embodiment described above may also be one in which the network as in the second embodiment is interposed, except for the update information request database 203.

The communication means 201 communicates between the browsing input means 101 and the control means 102 via the network 201.

An update information request database 202 is connected to this communication means 201, and the communication means 201 is connected.
Has a function of confirming the contents of the update information request database 202 when accessed from any of the browsing input means 101 (in other words, from any user). The access that triggers the confirmation of the content of the update information request database 202 may be limited to the access of the process related to the machine translation (including not only the machine translation process but also the dictionary browsing request process). It may be an access to a process not related to (for example, a simple document search process).

The update information request database 202 stores update information type data relating to the dictionary database 105, which each user requests automatic transmission (automatic notification).

FIG. 14 shows the update information request database 20.
2 shows an example of the stored contents of No. 2. In the case of the example in FIG. 14, when the user 1 updates the computer dictionary and the mathematical dictionary (may or may not include the user dictionary associated with them) in the dictionary database 105, , The user is requesting automatic transmission of the dictionary data related to the update, and the user 2 requests automatic transmission of the dictionary data related to the update when the natural science dictionary in the dictionary database 105 is updated. When the dictionary data of the headword “pencil” in the dictionary database 105 is updated, the user 3 requests automatic transmission of the dictionary data related to the update. The update information request database 202 may also record the previous automatic transmission time in order to clarify up to which time the dictionary data related to the update has already been transmitted.

Due to the relationship with the components newly added in the second embodiment as described above, the functions of the browsing input means 101, the dictionary management means 104, etc. are slightly different from those of the first embodiment. However, this point will be clarified in the operation description described later.

(B-2) Operation of the Second Embodiment The machine translation system 100A of the second embodiment also executes the various processes described in the first embodiment.

Further, the machine translation system 100A of the second embodiment also performs a process of automatically notifying the user of update information such as dictionary data updated in the dictionary database 105.

Hereinafter, such update information notification processing will be described.
This will be described in detail with reference to the flowchart of FIG. Figure 1
Reference numeral 5 indicates a process for one user.

Although not shown in the flow chart, the user inputs the type data of the update information requesting the automatic transmission of the update information from the browsing input means 101, and is provided in advance in association with the communication means 202. The update information request database 203 is registered in advance. The registration in the update information request database 203 may be performed when the system operator concludes a user contract or the like instead of the method based on the input operation by the user as described above.

When a user accesses the communication means 202 from the browsing input means 101 for some processing (for example, machine translation processing), the processing is finished before the processing is started or the processing is ended. After that, the process of FIG. 15 is started, and first, the communication unit 202 determines whether or not there is update information type data (update information request) in the update information request database 203 for the user (step S61). .

The processing shown in FIG. 15 is replaced with the communication means 202, the control means 102 or the dictionary management means 104 instead of the method started by the access by the user as described above.
However, a method of activating periodically may be applied, or a method of automatically activating the dictionary database 105 by recognizing the update of the dictionary database 105 may be applied. In the case of the two activation methods cited as the modified examples, the browsing input means 10
Since there is a possibility that the power supply of No. 1 is turned off, the notification operation of step S65 described later is, for example, e-mail sending.

When the type of update information (update information request) is not stored in the update information request database 203 for the accessing user, the processing shown in FIG. 15 is immediately terminated.

On the other hand, when the update information type data (update information request) is stored in the update information request database 203 for the user who has accessed, the communication means 202 cooperates with the dictionary management means 104. By confirming that the dictionary database 105 has been updated, the dictionary history recording means 106 is searched to determine whether or not there is update information for which the user is requesting automatic transmission of update information (steps S62-). S64).

More specifically, it is confirmed to the user that the dictionary database 105 has been updated after the time of the previous automatic transmission, and the update time of the user is the time after the last automatic transmission. This is performed by searching the dictionary history recording unit 106 for dictionary data such as dictionary data that is specified by the update information type data stored in the update information request database 203.

For example, if the user who made the access is shown in FIG.
In the case of the user 2 shown in FIG.
If it is confirmed that No. 5 has been updated, it is determined whether or not the dictionary data of the natural science dictionary has updated dictionary data after the time automatically transmitted to the user 2 last time. It will be determined by searching 106.

If the dictionary database 105 has not been updated (negative result in step S63) or there is no dictionary data to be notified (negative result in step S65), the process shown in FIG. 15 is terminated.

If there is the corresponding dictionary data to be notified to the user, the communication means 202 notifies the browsing input means 101 of the dictionary data and / or update history, and ends the processing shown in FIG.

Note that FIG. 15 shows the case where the update information is retrieved at the timing of automatic transmission. However, when the dictionary database 105 is updated, each user is notified based on the contents stored in the update information request database 203. Update information to be automatically transmitted is copied to the update information request database 203, and at the timing of automatic transmission, the update information is stored in the update information request database 203 without accessing the dictionary database 105 and the dictionary history recording means 106. The update information may be automatically transmitted.

(B-3) Effects of Second Embodiment The second embodiment can also achieve the same effects as those of the first embodiment described above.

Furthermore, according to the second embodiment, since the update information of the dictionary database can be automatically obtained by the user according to his / her wish, even if the dictionary database is updated by user registration by another user or the like. That can be grasped early. As a result, the user can prevent unnecessary user registration and can accurately understand the translation quality.

(C) Third Embodiment Next, a third embodiment in which the dictionary device and the natural language processing system according to the present invention are applied to a machine translation system will be described in detail with reference to the drawings.

(C-1) Configuration of the Third Embodiment FIG. 16 is a block diagram showing the functional configuration of the machine translation system of the third embodiment, and FIG. 13 according to the second embodiment. The same or corresponding parts are designated by the same reference numerals.

In FIG. 16, a machine translation system 100B of the third embodiment has a translation history recording means 301 and a translation result database 302 in addition to the configuration of the second embodiment described above.

The machine translation system 100B of the third embodiment is based on the premise that a plurality of browsing / inputting means 101 request translation of the same document.
For example, if the network 201 is the Internet,
A plurality of browsing / inputting means 101 may retrieve the same document from a document retrieval device (not shown) by search, and further, a plurality of browsing / inputting means 101 may request machine translation for the same retrieval document. The machine translation system 100B of the third embodiment is premised on such a network system.

The translation result database 302 stores the translation request document (original text), the latest translation result for the translation request document, and the registration time (may be the update time) when the latest translation result is recorded in the translation result database 302. Is what you are doing.

The translation history recording means 301 comprises the control means 10
Under the control of item 2, the translation result database 302 is updated and the like.

The function of the control means 102 and the like is also the second.
The embodiment is slightly different. The function of updating the translation result will be clarified in the operation description described later.

In addition, the control means 102 of the third embodiment.
Also has the following functions. Control means 102
When there is a machine translation request for a document from the browsing / inputting means 101, it confirms via the translation history recording means 301 whether or not the translation result of the document is registered in the translation result database 302. When the translation result of the document is not registered in the translation result database 302, the machine translation unit 103 causes the document to be machine translated and the translation result is returned to the browsing input unit 101, and the translation result is recorded by the translation history recording unit 301. When the translation result of the translation request document is registered in the translation result database 302, it is retrieved and returned to the browsing input means 101.

(C-2) Operation of the third embodiment The machine translation system 100B of the third embodiment also executes the various processes described in the second embodiment.

Furthermore, the machine translation system 100B of the third embodiment also performs a process of updating the translation result that has already been machine translated.

The translation result updating process according to the third embodiment will be described in detail below with reference to the flowchart of FIG.

For example, when the received translation result does not reflect the dictionary data registered by the user, or when it is desired to improve the quality of the received translation result, a document (translation request document or translation result document A good result or an instruction to update the translation result accompanied by the document specifying information (for example, URL) is issued to the browsing / inputting means 101 (step S71).

As a result, the specific updating process of the translation result as in step S72 and thereafter is started. It should be noted that the system administrator may perform a specific translation result update process starting from step S72, and automatically when the dictionary management unit 104 recognizes the update of the dictionary data in the dictionary database 105. It may be performed automatically or automatically at a predetermined cycle. Further, it may be automatically performed when a new machine translation request is made for the source language document for which the translation result is registered.

When an instruction to update the translation result in which the translation result is directly or indirectly designated is given from the viewing input means 101 side, the control means 102 causes the translation history recording means 301 to specify the designated translation result. Is a translation result database 3
No. 02 exists (steps S72 and S7).
3). If not, the process shown in FIG. 17 is terminated (note that
(Reply to the browsing input means 101).

If the specified translation result is in the translation result database 302, the control means 102 sets the update time immediately before the translation result (may be the initial registration time) via the translation history recording means 301. With getting (step S
74), the dictionary history recording means 106 is accessed through the dictionary management means 104 to search dictionary data whose update time is later than the update time (step S75), and the dictionary satisfying such update time is satisfied. It is determined whether or not data exists (step S76). If there is no such dictionary data, the processing shown in FIG. 17 is terminated (note that the browsing / inputting means 101 is informed that it cannot be updated).

If dictionary data satisfying the condition of the update time exists, the control means 102 accesses the translation result database 302 via the translation history recording means 301 to search for a sentence including such dictionary data in the source language document. A search is performed (step S77), and it is determined whether such a sentence exists (S78). If there is no source language sentence including the dictionary data that satisfies the update time condition, the process shown in FIG. 17 is terminated (note that there is no response to the browsing / inputting unit 101 that there is no source language sentence affected by the dictionary data update. Do).

If there is a source language sentence including dictionary data satisfying the condition of the update time, the control means 102 fetches the source language sentence (all if there is a plurality) from the translation result database 302 and gives it to the machine translation means 103. Machine translation (steps S79 and S80). After that, the control means 102
Gives the translation result for the sentence cut out by the machine translation means 103 and the update time to the translation history recording means 301 to update the translation result database 302 (step S81), and ends the processing shown in FIG. 17 ( The fact that the information has been updated is returned to the browsing input means 101.

If the machine translation means 103 translates in consideration of the context and the like, not only the sentence including the updated dictionary data but also the entire source language document may be given for machine translation. good. Further, the updated translation result may be sent back to the browsing / inputting means 101 instead of the update.

It should be noted that the update information request database 203
In addition, set the translation result update to be automatically sent,
By the notification processing of the update information described in the second embodiment, the update of the translation result is performed by the corresponding user (viewing input means 1).
01) may be notified.

(C-3) Effects of Third Embodiment The third embodiment can also achieve the same effects as those of the above-described second embodiment.

Further, according to the third embodiment, when the translation result is cached, it is possible to translate only the sentence including the phrase updated after the cached time, so that useless translation is performed. Instead, it is possible to update the translation result by the latest dictionary.

(D) Other Embodiments In the description of each of the above embodiments, various modified embodiments have been mentioned, but further modified embodiments exemplified below can be mentioned.

In the description of the first embodiment, it is described that the contents of the user dictionaries of other users can be browsed, but the browsing of the user dictionaries of other users may not be permitted. However, when it is downloaded to the technical term dictionary, it can be viewed naturally.

Further, the technical term dictionary includes not only those classified in so-called specialized fields but also those classified in categories related to hobbies and the like.

In the second embodiment, the update history is automatically transmitted. However, in response to the update history transmission request (excluding specified information such as fields), the dictionary data of the type desired by the user is displayed. The update history may be transmitted.

In each of the above-mentioned embodiments, the user cannot directly register the technical term dictionary, but the user may directly register and operate the technical term dictionary. Even in this case, registration in some user dictionaries may be a condition.

In the third embodiment, the update of the translation result record includes the translation result reflecting the new dictionary data,
The translation result that has not been reflected may be presented to the user or the system operator and confirmed after the confirmation.

In each of the above embodiments, the case where the dictionary device and the natural language processing system of the present invention are used in the machine translation system has been shown, but the application target of the present invention is not limited to this. For example, the dictionary device of the present invention can be applied to a kana-kanji conversion dictionary used for voice recognition or the like.

[0146]

As described above, according to the present invention, a plurality of users can obtain high quality natural language processing results by sharing dictionary data relating to user registration, which maintains high quality contents. It is possible to realize a dictionary device and a natural language processing system that can be used.

[Brief description of drawings]

FIG. 1 is a block diagram showing a functional configuration of a machine translation system according to a first embodiment.

FIG. 2 is an explanatory diagram showing a relationship between a user dictionary and a technical term dictionary according to the first embodiment.

FIG. 3 is a flowchart showing a user dictionary registration process of the first embodiment.

FIG. 4 is a flow chart showing a technical term dictionary enrichment process of the first embodiment.

5 is an explanatory diagram of user-registered dictionary data registered in a technical term dictionary by the processing of FIG. 4. FIG.

FIG. 6 is a flowchart showing an integration process of technical term dictionary information according to the first embodiment.

FIG. 7 is a flowchart showing a division / generation process of a technical term dictionary according to the first embodiment.

8 is an explanatory diagram of user-registered dictionary data in which a lower technical term dictionary is generated by the process of FIG. 7. FIG.

9 is an explanatory diagram showing a structure of a dictionary database before and after a lower technical term dictionary is generated by the processing of FIG. 7. FIG.

10 is an explanatory diagram showing movement of a user dictionary when a lower technical term dictionary is generated by the processing of FIG. 7. FIG.

FIG. 11 is a flowchart showing a dictionary browsing process according to the first embodiment.

FIG. 12 is a flowchart showing a machine translation process of the first embodiment.

FIG. 13 is a block diagram showing a functional configuration of a machine translation system according to a second embodiment.

FIG. 14 is an explanatory diagram showing an example of stored contents of an update information request database according to the second embodiment.

FIG. 15 is a flowchart showing a notification process of update information of a dictionary according to the second embodiment.

FIG. 16 is a block diagram showing a functional configuration of a machine translation system according to a third embodiment.

FIG. 17 is a flowchart showing a translation result updating process according to the third embodiment.

[Explanation of symbols]

100, 100A, 100B ... Machine translation system, 10
1 ... Browsing input means, 102 ... Control means, 103 ... Machine translation means, 104 ... Dictionary management means, 105 ... Dictionary database, 106 ... Dictionary history recording means, 201 ... Network, 202 ... Communication means, 203 ... Update information request database , 301 ... Translation history recording means, 302 ... Translation result database.

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 5B009 ME02 ME12 ME14 MF03 MF04                       VB01                 5B075 ND03 ND35 UU03                 5B091 CC14 CC15 CC17

Claims (6)

[Claims] 1. A dictionary device provided in a natural language processing system used by a plurality of users, wherein a plurality of technical term dictionaries for each category are hierarchized in a tree structure with a general term dictionary as a root node. , A user dictionary registration means for setting a user dictionary in association with any of the technical term dictionary and / or the general term dictionary, and the technical term dictionary or the general term dictionary associated with common dictionary data for a plurality of user dictionaries Set to
In the tree structure, set common dictionary data to multiple lower-level dictionaries to the upper-level dictionary, or set lower-level technical term dictionaries to the above-mentioned technical term dictionary based on dictionary data common to multiple user dictionaries associated with it. A dictionary device, including dictionary changing means for changing the dictionary and / or the dictionary data, and dictionary history recording means for recording a history of changes in the dictionary and / or dictionary data by the dictionary changing means. . 2. A browsing request fetching unit for fetching a browsing request for dictionary data from a browsing input unit operated by a user, and dictionary data is retrieved from the dictionary body according to the browsing request and the dictionary data is also retrieved. 2. The dictionary device according to claim 1, further comprising a browsing request responding unit that retrieves a change history corresponding to the above from the dictionary history recording unit and sends back to the browsing input unit. 3. A notification type setting means for setting a type of change history to be notified to each user, and a notification to each user from the dictionary history recording means based on the type information set in the notification type setting means. 3. The dictionary device according to claim 1, further comprising: a change history notifying unit that takes out a change history to be performed and notifies the browsing input unit of each user. 4. A natural language processing system, comprising: the dictionary device according to claim 1; and a natural language processing body that performs natural language processing by utilizing stored contents of the dictionary device. system. 5. A natural language processing result recording means for recording the natural language processing result executed by the natural language processing main body or the updated processing result together with recording time information, and recorded in the natural language processing result recording means. Natural language processing is performed in the natural language processing body so that the update time recorded and managed by the dictionary history recording means reflects the dictionary data in the dictionary device after the recording time of the processed result of the natural language. And a natural language processing reviewing means for updating the natural language processing result recorded in the natural language processing result recording means according to the result. 4. The natural language processing system according to item 4. 6. The natural language processing executed by the main body of natural language processing is machine translation processing.
Alternatively, the natural language processing system according to item 5.