JP2008217398A - Technical term classification device, technical term classification method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technical term classification device for automatically classifying technical terms. <P>SOLUTION: This technical term classification device is provided with a technical term storage part 11 for storing two or more technical terms; an adjacent word/phrase storage part 12 for storing adjacent words/phrases being words/phrases which may be adjacent to the technical terms; a retrieval part 14 for configuring retrieval terms by making the adjacent words/phrases stored in the adjacent word/phrases storage part 12 adjacent to the technical terms stored in the technical term storage part 11, and for retrieving the retrieval terms in a plurality of document information being information showing a document in a specific field; a classification part 15 for classifying the technical terms included in the retrieval terms according to the retrieval result of the retrieval part 14; a classification result information storage part 16 for storing classification result information being information showing the classification result of the classification part 15; and an output part 17 for outputting the classification result information stored in the classification result information storage part 16. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a technical term classification device for classifying technical terms.

In recent years, various methods for automating language processing that has been performed manually have been developed. For example, a method of automatically creating information indicating a technology trend indicating “what kind of technology has been used since” has been developed (see, for example, Non-Patent Document 1).
Eiwa Namba, Yuko Taniguchi, “Extracting and Visualizing Research Trend Information from Academic Paper Databases”, 12th Annual Conference of the Association for Language Processing, “Contact between Language Processing and Information Visualization”, p. 35-38, 2006

  By automating language processing, it is possible to obtain a large amount of output that has not been easily available by hand until now. However, if the amount of output is too large, the output will be very difficult for the user to understand. Therefore, it is necessary to devise, for example, to classify the output.

  For example, when the output from automated language processing is a technical term, it is possible to automatically classify the technical term of the output by using information obtained as a result of classifying the technical term in advance. . In addition, there is a problem that creating information as a result of classifying the technical terms manually requires a great deal of time and labor. In addition, such technical terms can only be classified by those who are familiar with the specialized field, and it is very difficult to secure such people.

  The present invention has been made in consideration of the above-described circumstances, and an object thereof is to provide a technical term classification device that can automatically classify technical terms.

  In order to achieve the above object, the technical term classification device according to the present invention includes a technical term storage unit that stores two or more technical terms, and an adjacent term storage unit that stores adjacent phrases that can be adjacent to the technical terms. And a plurality of pieces of information indicating a document in a specific field by configuring a search term in which the adjacent term stored in the adjacent term storage unit is adjacent to the technical term stored in the technical term storage unit. In the document information, a search unit that searches for the search term, a classification unit that classifies the technical terms included in the search term according to a search result by the search unit, and information that indicates a result classified by the classification unit A classification result information storage unit that stores certain classification result information; and an output unit that outputs the classification result information stored in the classification result information storage unit.

  With such a configuration, it is possible to classify technical terms and obtain classification result information that is a result of the classification. Therefore, by using the classification result information, for example, it is possible to classify various language processing results. Moreover, by automatically classifying the technical terms, there is a merit that a lot of time and labor are not required, and it is not necessary to secure a person who is familiar with the specialized field.

The technical term classification device according to the present invention further includes a document information storage unit in which a plurality of document information is stored, and the search unit includes a search term in the plurality of document information stored in the document information storage unit. You may search for.
With such a configuration, it is possible to classify technical terms using the literature information stored in the literature information storage unit included in the technical term classification device.

Further, in the technical term classification device according to the present invention, the adjacent phrase is a prefix that can be adjacent to the beginning of the technical term or a suffix that can be adjacent to the end of the technical term, and the classification unit is the search unit In the case where the retrieved search terms are included in the literature information, the technical terms constituting the search terms may be classified according to the adjacent terms constituting the search terms.
With such a configuration, it is possible to classify the technical terms by configuring the search terms using the adjacent words and phrases that are prefixes and suffixes, and searching the search terms.

In the technical term classification device according to the present invention, the classification unit classifies the technical terms constituting the search term when the search term searched by the search unit is included in the document information a predetermined number of times or more. You may go.
With such a configuration, it is possible to prevent classification of technical terms included in search terms with a small number of hits.

Further, in the technical term classification device according to the present invention, the adjacent words / phrases are divided into a plurality of groups, and the classification unit performs the search when the search term searched by the search unit is included in the literature information. The technical terms constituting the term may be classified according to the group of adjacent words constituting the search term.
With such a configuration, for example, technical terms included in a search term that is searched using similar neighboring words can be made to belong to the same category.

In the technical term classification device according to the present invention, the classification unit may perform classification of the technical term in association with the number of search results of the search term searched by the search unit.
With such a configuration, for example, the number of search results included in the classification result information can be used in subsequent analysis on the classification result information.

  In the technical term classification device according to the present invention, the adjacent word / phrase is a suffix that can constitute a term indicating an action by being adjacent to the end of the technical term, and the classification unit includes the search term. An acquisition unit that acquires one or more sentences from literature information, a specification unit that specifies an input word and / or an output word for the search term in one or more sentences acquired by the acquisition unit, and a specification of the specification unit Storage means for storing input words and / or output words in association with technical terms constituting the search terms, and depending on the input words and / or output words stored in the storage means, the input words and / or output words Classification means for classifying technical terms corresponding to the output words may be provided.

With such a configuration, classification is performed using input words and / or output words. For example, technical terms can be classified for each piece of information to be processed and information to be processed.
In the technical term classification device according to the present invention, the classification unit may classify technical terms common to input words and / or output words into the same classification.
With such a configuration, for example, the range of classification can be narrowed.

In the technical term classification device according to the present invention, the classification means may classify technical terms having the same concept of input words and / or output words into the same classification.
With such a configuration, for example, the range of classification can be widened.

In the technical term classification device according to the present invention, a dictionary information storage unit that stores dictionary information, which is information indicating a dictionary for determining whether or not a plurality of technical terms belong to the same classification, and the technical term storage unit A judgment unit that judges whether the stored technical term that is the classification target belongs to the same classification as the technical term that has already been classified, with reference to the dictionary information; An additional classification unit that classifies technical terms that are subject to classification into the same classification as technical terms that are already classified when it is determined that the technical terms that belong to the same technical terms that are already classified And may be further provided.
With such a configuration, technical terms can be further classified and added to the classification result information, and the classification result information can be expanded.

  In the technical term classification device according to the present invention, the dictionary information is information indicating a dictionary of synonyms, and the determination unit belongs to the same classification when two technical terms have a synonym relationship. You may judge.

  With such a configuration, technical terms that are related to synonyms are considered to belong to the same classification, but the classification result information can be expanded by using information indicating a dictionary of the synonyms. .

  In the technical term classification device according to the present invention, the dictionary information is information indicating correspondence between a broader word and a narrower word, and the determination unit includes technical terms of a plurality of narrower words corresponding to the broader terminology. However, when belonging to the same classification, it may be determined that the technical term of the broader term belongs to the classification to which the technical term of the lower term belongs.

  With such a configuration, it is considered that the broader word will also have a property common to many narrower words, for example, by using information indicating the correspondence between the broader word and the narrower word, such as a broader thesaurus. Classification result information can be expanded.

Further, the technical term classification device according to the present invention includes a technical term storage unit that stores technical terms, a classification result information storage unit that stores classification result information that is information indicating a result of classification of technical terms, A dictionary information storage unit storing dictionary information that is a dictionary for determining whether the technical terms belong to the same classification, and a technical term that is a classification target stored in the technical term storage unit A judgment unit that judges whether or not the technical term that belongs to the same category as the technical term that has already been classified by referring to the dictionary information, and the technical term that is the classification target by the judgment unit When it is determined that the term belongs to the same category as the term, an additional classifying unit that classifies the technical term to be classified into the same category as the technical term that has already been classified, and the classification result information storage unit An output unit for outputting a classification result information that is, those having a.
With such a configuration, technical terms can be further classified and added to the classification result information, and the classification result information can be expanded.

  According to the technical term classification device and the like according to the present invention, it is possible to classify technical terms and obtain classification result information which is a result of the classification. Therefore, by using the classification result information, for example, it is possible to classify various language processing results. Further, by automatically classifying the technical terms, a great deal of time and labor is not required, and it is not necessary to secure a person who is familiar with the specialized field.

  Hereinafter, a technical term classification device according to the present invention will be described using embodiments. In the following embodiments, components and steps denoted by the same reference numerals are the same or equivalent, and repetitive description may be omitted.

(Embodiment 1)
A technical term classification device according to Embodiment 1 of the present invention will be described with reference to the drawings. The technical term classification device according to the present embodiment classifies technical terms by using adjacent words.

  FIG. 1 is a block diagram showing a configuration of a technical term classification device 1 according to the present embodiment. In FIG. 1, the technical term classification device 1 according to the present embodiment includes a technical term storage unit 11, an adjacent phrase storage unit 12, a document information storage unit 13, a search unit 14, a classification unit 15, and classification result information. A storage unit 16 and an output unit 17 are provided.

  The technical term storage unit 11 stores two or more technical terms. Here, the technical terms are technical terms used in a specific field, such as technical terms, economic terms, political terms, medical terms, legal terms, and the like. The specific field is, for example, a specific research field, a specific technical field, or another specialized field. This technical term is actually “information indicating a technical term”, but is simply referred to as a technical term. The same applies to other terms and phrases.

  This technical term may be collected manually, for example, from technical terminology dictionaries, economic terminology dictionaries, etc., or mechanically obtained from multiple literature information. It may be what was done. When acquiring technical terms mechanically from multiple pieces of literature information, for example, by extracting morphological analysis of technical literature, single nouns etc. are extracted, and scoring that indicates the possibility of being technical terms for each single noun etc. And a single noun with a high score may be acquired as a technical term. Here, as a method for assigning a score, a plurality of types such as scoring based on word-making ability and scoring based on appearance frequency are known. Further, other methods may be used. About the method of acquiring a technical term, since it is disclosed by the following some literature etc. and is conventionally known, the detailed description is abbreviate | omitted. Also, TermExtract has been released as a tool for creating a technical term list (http://gensen.dl.itc.u-tokyo.ac.jp/termextract.html).

Literature: Hiroshi Nakagawa, “Automatic Term Recognition based on Statistics of Compound Nouns”, Terminology, Vol. 6, no. 2, p. 195-210, 2000
Literature: Hirokazu Ohata, Hiroshi Nakagawa, “Extraction of technical terms by the number of words connected differently”, Information Processing Society of Japan, 2000-NL-136, p. 119-126

  References: Hiroshi Nakagawa, Masanori Mori, Yasuaki Yumoto, “Extraction of technical terms based on appearance frequency and connection frequency”, natural language processing, Vol. 10 No. 1, p. 27-45, January 2003

  The technical term storage unit 11 can be realized by a predetermined recording medium (for example, a semiconductor memory, a magnetic disk, an optical disk, etc.). The process in which technical terms are stored in the technical term storage unit 11 does not matter. For example, technical terms may be stored in the technical term storage unit 11 via a recording medium, and technical terms transmitted via a communication line or the like are stored in the technical term storage unit 11. Alternatively, the technical term input via the input device may be stored in the technical term storage unit 11.

  In the adjacent phrase storage unit 12, adjacent phrases are stored. Here, the adjacent phrase is a phrase that can be adjacent to a technical term. The adjacent phrase is, for example, a phrase in which a technical term adjacent to the adjacent phrase can further become a technical term, a phrase in which a technical term adjacent to the adjacent phrase can become a term indicating an operation, or the like. The latter case will be described in the second embodiment. It should be noted that technical terms adjacent to adjacent words may not be technical terms, for example, may be inappropriate terms that are not normally used. Also, the adjacent phrase itself is not usually a technical term. An adjacent phrase that can be adjacent to the beginning of a technical term is called a prefix. An adjacent phrase that can be adjacent to the end of a technical term is called a suffix. For example, the adjacent words “system” and “device” are suffixes that can be adjacent to the end of a technical term. For example, the adjacent words “Japanese” and “English” are prefixes that can be adjacent to the beginning of technical terms. When the adjacent phrase includes both a prefix and a suffix, information indicating whether each adjacent phrase is a prefix or a suffix may be associated. Note that the adjacent words may be divided into a plurality of groups. For example, “system” and “apparatus” as suffixes are considered to be similar suffixes that constitute hardware for a certain purpose by adjoining the terminology, so they are identical. May belong to any group.

  The adjacent phrase may be collected manually, for example, or may be one manually selected from a plurality of document information. In the latter case, for example, by collecting compound nouns from document information and collecting nouns that frequently appear at the end of the compound nouns, candidates for adjacent phrases as suffixes can be obtained. On the other hand, for example, by collecting compound nouns from document information and collecting nouns that frequently appear at the top of the compound nouns, candidates for adjacent phrases as prefixes can be obtained. A list of adjacent words can be obtained by manually selecting useful ones from the candidates. It goes without saying that adjacent phrases may be collected by other methods.

  The adjacent phrase storage unit 12 can be realized by a predetermined recording medium (for example, a semiconductor memory, a magnetic disk, an optical disk, etc.). The process of storing the adjacent phrase in the adjacent phrase storage unit 12 does not matter. For example, an adjacent phrase may be stored in the adjacent phrase storage unit 12 via a recording medium, and an adjacent phrase transmitted via a communication line or the like is stored in the adjacent phrase storage unit 12. Alternatively, the adjacent phrase input via the input device may be stored in the adjacent phrase storage unit 12.

  The document information storage unit 13 stores a plurality of document information. Here, the document information is information indicating documents in a specific field. The specific field is, for example, a specific research field, a specific technical field, or another specialized field. The document may be any document as long as it has a certain unit such as a paper, a patent document (for example, an application publication gazette, a patent gazette, etc.), a web page, an official gazette, and a book. The fields related to the plurality of document information stored in the document information storage unit 13 may all be the same or different. In the latter case, for example, document information in a plurality of fields such as technology, economy, politics, and law is stored in the document information storage unit 13. One document information includes information (data) of a corresponding one document. However, one document information may only include a predetermined range of information. For example, in the case of literature information of an academic paper, the literature information may be information having only a summary of the academic paper. The document information may be a text file, or may be a file of a unique format such as word processing software.

  In the present embodiment, a configuration in which the technical term classification device 1 includes the literature information storage unit 13 will be described, but the technical term classification device 1 may not include the literature information storage unit 13. Even if the technical term classification device 1 does not include the literature information storage unit 13, the technical term classification device 1 can access literature information held in a server (not shown) or other devices. And

  The document information storage unit 13 can be realized by a predetermined recording medium (for example, a semiconductor memory, a magnetic disk, an optical disk, etc.). The process of storing a plurality of document information in the document information storage unit 13 does not matter. For example, a plurality of document information may be stored in the document information storage unit 13 via a recording medium, and a plurality of document information transmitted via a communication line or the like is stored in the document information storage unit 13. Alternatively, a plurality of document information input via the input device may be stored in the document information storage unit 13.

  The search unit 14 configures a search term in which the adjacent term stored in the adjacent term storage unit 12 is adjacent to the technical term stored in the technical term storage unit 11. When the adjacent word / phrase is a prefix, the search unit 14 configures the search term by adjoining both in the order of the adjacent word / phrase and the technical term. In addition, when the adjacent word / phrase is a suffix, the search unit 14 configures the search term by causing both words to be adjacent in the order of the technical term and the adjacent word / phrase. And the search part 14 searches the comprised search term in the some literature information memorize | stored in the literature information storage part 13. FIG.

  In the present embodiment, a case will be described in which the search unit 14 directly searches for a configured search term in a plurality of document information stored in the document information storage unit 13. It is also possible to give a search instruction to the device or other components. In particular, when the technical term classification device 1 does not include the document information storage unit 13, the search processing performed by the search unit 14 transmits information that instructs the device including the document information storage unit 13 to perform a search using a search term. Then, a process of receiving a search result returned in response thereto may be used.

  The classification unit 15 classifies the technical terms included in the search terms according to the search result by the search unit 14. When the search term searched by the search unit 14 is included in the literature information, the classification unit 15 may classify the technical terms constituting the search term according to the adjacent terms constituting the search term. . The classification according to the adjacent phrase may be, for example, classification for each adjacent phrase, or classification for each group of adjacent phrases. Note that classifying technical terms means adding technical terms to classification result information to be described later so as to be distinguished for each classification.

  Moreover, the classification | category part 15 may perform the classification | category of the technical term which comprises the search term, when the search term searched by the search part 14 is contained in literature information more than predetermined times. For example, by setting the predetermined number of times to an appropriate value of 2 or more, it is possible not to classify technical terms included in a search term with a small number of hits. For example, in a small number of document information stored in the document information storage unit 13, there is a case where an inappropriate description exists, and therefore, a search term that is not originally hit by a search may be hit. However, since it is considered unlikely that such inappropriate description exists in large quantities, the number of hits in that case is considered to be small. Therefore, the classification according to an inappropriate search result can be avoided by not classifying the technical terms included in the search terms having a small number of hits. Note that the predetermined number of times serving as the threshold depends on, for example, the scale of the document information stored in the document information storage unit 13, but in general, once, twice, three times, five times, The number is preferably as small as 10 times. In addition, a predetermined ratio may be used instead of the predetermined number of times. For example, the classification unit 15 may classify the technical terms constituting the search term when the search term searched by the search unit 14 is included in the document information by a predetermined ratio or more.

  In addition, when the adjacent terms are divided into a plurality of groups, the classification unit 15 selects the technical terms constituting the search terms when the search terms searched by the search unit 14 are included in the document information. You may classify | categorize according to the group of the adjacent phrase which comprises a search term. For example, when the adjacent terms “system” and “device” belong to the same group and the search term “morpheme analysis system” is included in the literature information, the classification unit 15 specializes in configuring the search term. The term “morpheme analysis” may be classified in association with a group of adjacent words “system” and “device”.

  Further, the classification unit 15 may classify the technical terms in association with the number of search results (number of hits) of the search terms searched by the search unit 14. In this case, the number of search results is included in the classification result information described later. The number of search results included in the classification result information can be used in subsequent analysis related to the classification result information. For example, it is also possible to acquire more likely classification result information by sorting the classification result information in descending order of the number of search results and selecting only those having the highest number of search results.

  In the classification result information storage unit 16, classification result information is stored. Here, the classification result information is information indicating a result classified by the classification unit 15. In this classification result information, technical terms are distinguished and included for each classification. Each category may or may not be given a category name indicating the category. The classification name may be, for example, an adjacent word / phrase used in a search for a technical term included in the classification, a group name of the adjacent word / phrase, or another name. Further, as described above, the number of search results may be associated with the technical terms included in the classification result information. The classification result information storage unit 16 can be realized by a predetermined recording medium (for example, a semiconductor memory, a magnetic disk, an optical disk, etc.).

  The output unit 17 outputs the classification result information stored in the classification result information storage unit 16. As a result of this output, for example, the user can know the classification result, or the classification result can be used by another device or the like. Here, the output may be, for example, display on a display device (for example, a CRT or a liquid crystal display), transmission via a communication line to a predetermined device, printing by a printer, or audio output by a speaker. However, it may be stored in a recording medium. Further, this output may be, for example, a process of giving classification result information to a function call. In message passing, the classification result information is passed to an object of the next process (for example, SendMessage or the like is used). The classification result information may be written in a shared memory used by another component.

  The output unit 17 may or may not include an output device (for example, a display device or a printer). The output unit 17 may be realized by hardware, or may be realized by software such as a driver that drives these devices.

Note that the storage in the technical term storage unit 11, the adjacent phrase storage unit 12, the document information storage unit 13, and the classification result information storage unit 16 may be temporary storage in a RAM or the like, or may be long-term storage. Further, the technical term storage unit 11, the adjacent phrase storage unit 12, the document information storage unit 13, and the classification result information storage unit 16 may be realized by the same recording medium or by different recording media. May be. In the former case, for example, an area storing technical terms is the technical term storage unit 11, and an area storing adjacent words is the adjacent word storage unit 12.
Next, the operation of the technical term classification device 1 according to the present embodiment will be described using the flowchart of FIG.

  (Step S101) The classification unit 15 determines whether to start the classification process. And when starting, it progresses to step S102, and when that is not right, it repeats the process of step S101 until it judges that it starts. For example, when the technical term classification device 1 receives an instruction to start classification, the classification unit 15 may determine to start classification, or may determine to start classification at other timing. This determination process may be performed by components other than the classification unit 15.

(Step S102) The retrieval unit 14 sets the counter i to 1.
(Step S103) The retrieval unit 14 sets the counter j to 1.

(Step S <b> 104) The search unit 14 reads the i-th technical term from the technical term storage unit 11.
(Step S <b> 105) The search unit 14 reads the j-th adjacent phrase from the adjacent phrase storage unit 12.

(Step S106) The search unit 14 configures the search term by causing the adjacent term read in step S105 to be adjacent to the technical term read in step S104. When the adjacent word / phrase is a prefix, the search unit 14 causes the adjacent word / phrase to be adjacent to the beginning of the technical term. When the adjacent word / phrase is a suffix, the search unit 14 selects the adjacent word / phrase. Adjacent to the end of the terminology.
(Step S <b> 107) The search unit 14 searches the search term in the plurality of document information stored in the document information storage unit 13.

  (Step S108) The search unit 14 determines whether or not there has been a predetermined number of hits as a result of the search for the search term. If there has been a predetermined number of hits, the process proceeds to step S109. Otherwise, the process proceeds to step S110. Note that information indicating the predetermined number of times is stored in a recording medium (not shown), and the search unit 14 may perform the determination process by reading the information. In addition, by setting the predetermined number to “1”, at least the technical terms included in the search terms having hits can be classified. Therefore, the predetermined number is 1 or an integer of 2 or more.

(Step S109) The classification unit 15 classifies the technical terms included in the search terms searched by the search unit 14. Specifically, the classification unit 15 adds the technical terms included in the search terms searched by the search unit 14 to the classification result information stored in the classification result information storage unit 16. For example, the classification unit 15 may add the technical term included in the search term to the classification result information stored in the classification result information storage unit 16 in association with the adjacent phrase included in the search term. Further, the classification unit 15 may add the technical terms included in the search terms to the classification result information in association with the number of search results.
(Step S110) The search unit 14 increments the counter j by 1.

(Step S <b> 111) The search unit 14 determines whether or not the j-th adjacent phrase exists in the adjacent phrase storage unit 12. And when it exists, it returns to step S105, and when that is not right, it progresses to step S112.
(Step S112) The search unit 14 increments the counter i by 1.

(Step S113) The search unit 14 determines whether or not the i-th technical term exists in the technical term storage unit 11. And when it exists, it returns to step S103, and when that is not right, it progresses to step S114.
(Step S114) The output unit 17 outputs the classification result information stored in the classification result information storage unit 16. Then, the process returns to step S101.

  In the flowchart of FIG. 2, the process is terminated by powering off or a process termination interrupt. Further, in the flowchart of FIG. 2, the case has been described in which processing is performed for all technical terms stored in the technical term storage unit 11, but some technical terms stored in the technical term storage unit 11 Processing may be performed on technical terms.

  Further, in the flowchart of FIG. 2, the case has been described in which processing is performed on all the adjacent phrases stored in the adjacent phrase storage unit 12, but some of the adjacent phrases stored in the adjacent phrase storage unit 12 Processing may be performed on the adjacent words.

  In the flowchart of FIG. 2, the case where the classification result information is output after the processing for all the technical terms has been described has been described, but the timing for outputting the classification result information is not limited. For example, classification result information may be output in response to an instruction from the user.

Next, the operation of the technical term classification device 1 according to the present embodiment will be described using a specific example.
In this specific example, it is assumed that the technical term classification device 1 is a personal computer. In this specific example, the technical term storage unit 11 stores the technical terms shown in FIG. In this specific example, it is assumed that the adjacent phrase storage unit 12 stores the adjacent phrases shown in FIG. It is assumed that all the adjacent phrases are suffixes. Also, “10 times” is set as the predetermined number of times used in the process of step S108 in the flowchart of FIG.

  First, it is assumed that the user inputs an instruction to start classification into the technical term classification device 1 by operating a mouse, a keyboard, or the like. Then, the technical term classification device 1 accepts the instruction, and the classification unit 15 determines to start the classification process (step S101). Then, the search unit 14 reads out the first technical term “morpheme” stored in the technical term storage unit 11 and reads out the first adjacent phrase “system” stored in the adjacent term storage unit 12 ( Steps S102 to S105). The search unit 14 configures a search term “morpheme system” in which the technical term “morpheme” is adjacent to the adjacent word “system” as a suffix (step S106), and is stored in the literature information storage unit 13. The search term is searched in the existing document (step S107). In this case, if no hit is found, the search unit 14 determines that the hit is less than the predetermined number of times (= 10 times) (step S108), reads the next adjacent phrase “device”, A new search term “morpheme device” is configured (steps S110, S111, S105, S106). In this way, search term composition and search are sequentially performed.

  For the technical term “morpheme”, when the search in which all adjacent words are adjacent is completed, the search unit 14 determines that the next adjacent word does not exist (steps S110 and S111), and the second technical term “machine translation” Is performed (steps S112, S113, S104). Similarly to the above description, the search unit 14 reads the technical term “machine translation” and the adjacent phrase “system” (steps S104 and S105), and configures the search term “machine translation system” using them (steps S104 and S105). Step S106). And the search part 14 searches the search term "machine translation system" in literature information (step S107). In this case, 50,000 hits are assumed. Then, the search unit 14 determines that the hit is a predetermined number (= 10) or more (step S108), and associates the technical term “machine translation” included in the search term with the adjacent phrase “system”. Classify (step S109). Specifically, the classification unit 15 adds the technical term “machine translation” to the classification result information in association with the adjacent word “system”. The number of search results (= 50,000) is also added to the classification result information. The first record of the classification result information shown in FIG. 5 is added in this way. Such processing is performed for all technical terms.

Assume that the classification result information stored in the classification result information storage unit 16 is as shown in FIG. 5 when the processing for all technical terms is completed. Then, the output unit 17 reads out the classification result information and displays it on a display (not shown) (step S114). By the display, the user can know how each technical term is classified.
In this specific example, the case where the adjacent word / phrase is a suffix has been described. However, even when the adjacent word / phrase is a prefix, the technical terms can be classified in the same manner.

  As described above, according to the technical term classification device 1 according to the present embodiment, technical terms can be classified by using adjacent words. Therefore, by using the classification result information, for example, it is possible to classify various language processing results. In addition, by automatically classifying the technical terms, it does not require much time and labor as in the case of manual classification, and it is also necessary to secure people who are familiar with specialized fields. There is also a merit that there is no.

(Embodiment 2)
A technical term classification device according to Embodiment 2 of the present invention will be described with reference to the drawings. The technical term classification device according to the present embodiment classifies technical terms by using, as an adjacent phrase, a suffix that can constitute a term indicating an operation by being adjacent to the end of the technical term.

  FIG. 6 is a block diagram showing a configuration of the technical term classification device 2 according to the present embodiment. In FIG. 6, the technical term classification device 2 according to the present embodiment includes a technical term storage unit 11, an adjacent phrase storage unit 12, a document information storage unit 13, a search unit 14, a classification unit 21, and classification result information. A storage unit 16 and an output unit 17 are provided. The configuration other than the classification unit 21 is the same as that of the first embodiment except that the adjacent phrase stored in the adjacent phrase storage unit 12 is different, and the description thereof is omitted. As described above, the adjacent phrase storage unit 12 stores an adjacent phrase as a suffix that can constitute a term indicating an action by being adjacent to the end of the technical term. Specifically, this adjacent phrase is “do” or “do”. Also, the “term indicating the motion” is, for example, a term that indicates the motion when “do” is placed after the noun or a noun.

  The classification unit 21 classifies the technical terms included in the search terms according to the search result by the search unit 14, and includes an acquisition unit 31, an identification unit 32, a storage unit 33, and a classification unit 34. .

  The acquisition unit 31 acquires one or more sentences including the search term from the document information stored in the document information storage unit 13. For example, the acquisition unit 31 may acquire all sentences including the search terms, or may acquire some sentences among the sentences including the search terms. In FIG. 6, the search unit 14 and the acquisition unit 31 are separately described. However, the search by the search unit 14 and the acquisition by the acquisition unit 31 may be integrated. That is, retrieval of a search term and acquisition of a sentence including the search term may be executed as one process.

  The specifying unit 32 specifies an input word and / or an output word for a search term in one or more sentences acquired by the acquiring unit 31. That is, the specifying unit 32 may specify only an input word, may specify only an output word, or may specify an input word and an output word. Here, the input word is information to be processed corresponding to the search term. An output word is information that is a result of processing corresponding to a search term. For example, when the search term is “machine translate” and the acquired sentence is “translate Japanese into English”, the specifying unit 32 specifies “Japanese” as an input word and outputs it. Specify "English" as the word. Specifying an input word or the like may be acquiring the input word or the like and holding it on a recording medium or the like (not shown), and recording information (for example, a pointer or the like) indicating the position of the input word or the like (not shown) It may be held in a medium or the like, or may be set in correspondence with the input word etc., and if the input word or output word can be specified as a result, the specific method Does not matter.

  Here, a method for detecting input words and output words will be described. It is thought that the input words often correspond to wo case and color case. Also, the output words are likely to correspond to the second case, the made case, and the first case. Specifically, the following sentence can be considered. In each example sentence, “Machine translation” is a search term, “Japanese” is an input word, and “English” is an output word. However, Example 5 has no output word, and Example 6 has no input word.

Example 1: Machine translation from Japanese to English.
Example 2: Machine translation from Japanese to English.
Example 3: Machine translation from Japanese to English.
Example 4: Machine translation from Japanese to English.
Example 5: Machine translation of Japanese.
Example 6: Machine translation into English.

  As described above, the specifying unit 32 can detect the input word by detecting the wo case or the color case in the sentence acquired by the acquiring unit 31. Further, the specifying unit 32 can detect the output word by detecting the second case, the made case, and the first case in the sentence acquired by the acquiring unit 31. Detection of wo etc. can be performed by detecting wo case etc. related to a search term by performing syntax analysis on the sentence acquired by the acquisition means 31. These processes are already known and will not be described in detail.

  In addition, you may detect an input word etc. using methods other than the above-mentioned. For example, when the search term is configured using an adjacent phrase “do”, a single noun or compound noun before the particle “no” related to the search term may be detected as an input word. For example, when the acquisition unit 31 acquires “Machine translation in Japanese”, the noun “Japanese” before the particle “no” related to the search term “Machine translation” is detected as an input word. Also good. As described above, the detection method is not limited to the above description as long as the input word and the output word can be appropriately detected.

  The specifying means 32 holds information such as which case corresponds to the input word and which case corresponds to the output word in a recording medium (not shown), and specifies the input word or the like using the information. You may go.

  The storage unit 33 stores the input words and / or output words specified by the specifying unit 32 in a predetermined recording medium in association with the technical terms constituting the search terms searched by the search unit 14. The recording medium is, for example, a semiconductor memory, an optical disk, a magnetic disk, or the like, and may be included in the storage unit 33 or may exist outside the storage unit 33. Moreover, this recording medium may memorize | store temporarily an input word and / or an output word, and may not be so.

  The classifying unit 34 classifies technical terms corresponding to the input words and / or output words according to the input words and / or output words stored in the storage unit 33. The classifying unit 34 may classify, for example, the technical terms common to the input words and / or output words stored in the storage unit 33 into the same classification, and the classification of the input words and / or output words stored in the storage unit 33 may be performed. Technical terms that share a common concept may be classified into the same category. The concept of the two phrases is that the meanings of the two phrases are similar. More specifically, for example, the two phrases may belong to the same concept group in a concept system dictionary or the like, and are lower terms corresponding to the same broader term in a broader thesaurus. There may be. When using a conceptual system dictionary or a higher-order lower-order thesaurus, the dictionary or thesaurus is stored in a recording medium accessible by the classification means 34. The recording medium may or may not be included in the classification unit 34. In the present embodiment, a case will be described in which technical terms are classified so that technical terms common to the input words and / or output words stored in the storage means 33 are the same. In addition, classifying a technical term means adding a technical term to the classification result information so as to be distinguished for each classification.

  Next, operation | movement of the technical term classification device 2 by this Embodiment is demonstrated using the flowchart of FIG. In the flowchart of FIG. 7, the processes other than steps S201 and S202 are the same as those of the flowchart of FIG. 2 except that the adjacent phrases stored in the adjacent phrase storage unit 12 are different, and the description thereof is omitted.

  (Step S201) The acquisition unit 31, the specification unit 32, and the storage unit 33 acquire a sentence including a search term, specify an input word and / or output word, store the specified input word and / or output word, and the like. Perform the process. Details of this processing will be described later with reference to the flowchart of FIG.

  (Step S202) The classification means 34 classifies technical terms corresponding to the input words and / or output words in accordance with the input words and / or output words stored in the storage means 33. Details of this processing will be described later with reference to the flowchart of FIG.

FIG. 8 is a flowchart showing details of the process in step S201 in the flowchart of FIG.
(Step S301) The acquisition unit 31 sets the counter k to 1.
(Step S <b> 302) The acquisition unit 31 acquires a sentence corresponding to the k-th search result from the search results by the search unit 14 from the document information storage unit 13.

  (Step S303) The specifying unit 32 specifies an input word and / or an output word corresponding to a search term that is a term indicating an action in the sentence acquired by the acquiring unit 31. As described above, this specification is performed by specifying a case corresponding to the search term.

  (Step S304) The storage unit 33 determines whether the input word and / or output word specified in step S303 is stored in association with the technical term included in the search term. If it has been accumulated, the process proceeds to step S306, and if not, the process proceeds to step S305.

(Step S305) The storage unit 33 stores the input word and / or output word specified by the specifying unit 32 in a predetermined recording medium in association with the technical term included in the search term.
(Step S306) The acquisition unit 31 increments the counter k by 1.

  (Step S307) The acquisition unit 31 determines whether or not the kth search result exists. If it exists, the process returns to step S302; otherwise, the process returns to the flowchart of FIG.

  In the flowcharts of FIGS. 7 and 8, the case where the search process by the search unit 14 and the acquisition process by the acquisition unit 31 are executed independently has been described, but as described above, the search is performed. , You may get. That is, in the search of the search term, when the search term is hit for the first time, the sentence containing the search term is acquired, and when the search term is hit for the second time, the sentence containing the search term is retrieved. You may repeat and perform the process of performing acquisition.

Further, in the flowchart of FIG. 8, a case has been described in which sentences corresponding to all search results are acquired, but this need not be the case. That is, you may acquire the sentence corresponding to a part of search results. For example, when the number of search results is enormous (for example, hundreds of thousands), if all sentences are acquired and input words are specified, the amount of processing can be enormous. In some cases, only a part of search results selected at random may be used.
In step S303, if both the input word and the output word cannot be specified, the process may jump to step S306.

  FIG. 9 is a flowchart showing details of the process in step S202 in the flowchart of FIG. Note that the classification process described with reference to the flowchart of FIG. 9 is an example, and it goes without saying that other processes may be performed. In the flowchart of FIG. 9, a case will be described in which technical terms common to input words and / or output words are classified into the same classification.

  (Step S401) The classification means 34 extracts the input words from the results accumulated by the accumulation means 33 so as not to overlap. If there is no input word, that is, only an output word, it is extracted that there is no input word, that is, the input word is null. For example, the classification unit 34 acquires all input words from the recording medium in which the storage unit 33 has stored the input words and the like, sorts the input words, and deletes the duplicate input words. Word extraction can be performed. The extracted input word may be temporarily stored in a recording medium (not shown).

  (Step S402) The classification means 34 extracts the output words from the results accumulated by the accumulation means 33 so that there are no duplicates as in step S401. The extracted output word may be temporarily stored in a recording medium (not shown).

(Step S403) The classification unit 34 sets the counter m to 1.
(Step S404) The classification unit 34 sets the counter n to 1.

  (Step S405) The classification unit 34 searches whether the storage unit 33 stores the technical terms corresponding to the mth input word extracted in step S401 and the nth output word extracted in step S402. . Specifically, the combination of the mth input word and the nth output word is searched for in the result stored by the storage means 33.

  (Step S406) As a result of the search, the classification unit 34 proceeds to step S407 if there is a combination of the mth input word and the nth output word, otherwise proceeds to step S408. move on. Note that it is determined whether or not the search has been hit a predetermined number of times or more. If the search has been hit a predetermined number of times or more, the process proceeds to step S407, and if not, the process may proceed to step S408. For example, by setting the predetermined number to 2 times, it is possible to avoid including only one technical term in the classification.

  (Step S407) The classification unit 34 associates the technical term corresponding to the searched combination of the mth input word and the nth output word with the mth input word and the nth output word. Classify. Specifically, the technical term is stored in the classification result information in association with the mth input word and the nth output word. When a combination of a plurality of m-th input words and an n-th output word is hit, classification processing is performed so that a plurality of technical terms corresponding to each combination have the same classification.

(Step S408) The classification unit 34 increments the counter n by 1.
(Step S409) The classification unit 34 determines whether or not the nth output word exists in the output words extracted in step S402. And when it exists, it returns to step S405, and when that is not right, it progresses to step S410.

(Step S410) The classification unit 34 increments the counter m by 1.
(Step S411) The classification unit 34 determines whether or not the mth input word exists in the input words extracted in step S401. If it exists, the process returns to step S404, and if not, the process returns to the flowchart of FIG.

  In the flowchart of FIG. 9, nulls are also extracted as input words and output words. Therefore, combinations in which the input word is null and the output word is not null, and combinations in which the input word is not null and the output word is null. Will exist. In these cases, the specific results are only output words, and the specific results are only input words. If both the input word and the output word are null, the classification means 34 does not need to search for a combination in which both the input word and the output word are null in step S405. .

Next, the operation of the technical term classification device 2 according to the present embodiment will be described using a specific example.
Also in this specific example, similarly to the specific example of the first embodiment, the technical term classification device 2 is assumed to be a personal computer. Further, it is assumed that the technical term storage unit 11 stores the technical term shown in FIG. In this specific example, it is assumed that the adjacent phrase storage unit 12 stores the adjacent phrase as a suffix shown in FIG.

  Note that the processes (steps S101 to S108) from the start of the classification process to the search are the same as those in the specific example of the first embodiment except that the adjacent words are different, and the description thereof is omitted.

  Next, a process (step S201) for acquiring a sentence including a search term and specifying an input word or the like will be described. Assume that the search unit 14 composes the search term “machine translation” from the technical term “machine translation” and the adjacent word “do”, and as a result of the search, there is a predetermined number of hits. Then, the acquisition unit 31 acquires a sentence corresponding to the first search result from the document information storage unit 13 (steps S301 and S302). The sentence is "First of all, machine translation from Japanese to English." Then, the specifying means 32 specifies “Japanese”, which is a case corresponding to the search term “machine translation”, as an input word, and “English”, which is a case knowledge corresponding to the search term “machine translation”. It is specified as an output word (step S303). The storage unit 33 determines whether or not the combination of the input word “Japanese”, the output word “English”, and the technical term “machine translation” included in the search term has already been stored (step S304). In this case, if not stored, the storage means 33 stores the input word and the output word in association with the technical term “machine translation” (step S305). The first record in FIG. 11 shows correspondence between input words, output words, and technical terms accumulated in this way.

  Thereafter, the acquisition unit 31 acquires a sentence corresponding to the next search result from the document information storage unit 13 (steps S306, S307, and S302). It is assumed that the sentence is “translate Japanese into English using a newly developed algorithm”. Then, in the same manner as described above, the specifying unit 32 specifies the input word “Japanese” and the output word “English” (step S303). In this case, since the combination of the input word “Japanese”, the output word “English”, and the technical term “machine translation” is accumulated, the accumulation process is not performed (step S304). In this way, retrieval result sentences, input words and / or output words are specified and stored in sequence. As a result, as shown in FIG. 11, it is assumed that input words, output words, and technical terms are stored in association with each other. The process is performed on the search results for each search term.

  When the process of specifying and storing input words for all search terms is completed, the classification unit 34 starts the process of classifying the technical terms (step S202).

  First, the classification means 34 acquires and sorts all input words from the accumulation result shown in FIG. Then, the classification means 34 deletes the same input word that exists two or more in the sorting result, leaving only one. In this way, the classification means 34 can extract the input words “Japanese”, “English”, etc. without duplication (step S401). The extracted input word is temporarily stored in a recording medium (not shown).

  Similarly, the classification unit 34 extracts the output words “English”, “Japanese”, “Chinese”, and the like without duplication from the accumulation result shown in FIG. 11 (step S402). The extracted output word is temporarily stored in a recording medium (not shown).

  After that, the classification unit 34 reads the first input word “Japanese” and the first output word “English” from a recording medium (not shown), and records that have those input words and output words as shown in FIG. A search is performed using the accumulation result indicated by (steps S403 to S405). Here, it is assumed that a plurality of records are hit. Then, the classification means 34 determines that there is a search result (step S406), reads each searched record, and stores it in the classification result information (step S407). In the accumulation, the input words and the output words are accumulated as classification names. The records corresponding to the classification “Japanese / English” shown in FIG. 12 are accumulated in this way. By performing such processing for each combination of input words and output words, new classifications are sequentially added to the classification result information. Finally, the classification result information as shown in FIG. 12 is stored in the classification result information storage unit 16. The process of outputting the classification result information is the same as the specific example of the first embodiment.

  Although the case where the classification process is performed is described with reference to the flowchart of FIG. 9, as described above, this is an example, and the present invention is not limited to this classification method. For example, the accumulation results may be sorted by input words, sorted by output words within the same input word range, and classified by collecting technical terms that have the same input words and output words. .

  Further, in this specific example, the case where classification is performed so that technical terms common to input words and / or output words are classified is the same, but as described above, the concept of input words and / or output words is Classification may be performed so that common technical terms are classified into the same classification. In that case, for example, in the flowchart of FIG. 9, the extracted input words and output words are grouped into groups having a common concept using a concept system dictionary, a higher-order lower thesaurus, or the like. In step S405, technical terms corresponding to the input words belonging to the mth group and the output words belonging to the nth group may be searched. For example, when “Japanese”, “English”, “Chinese”, etc. belong to the group “language”, the input word belongs to the group “language” and the output word also has one technical term belonging to the group “language”. It will be classified into classification.

  As described above, according to the technical term classification device 2 according to the present embodiment, it is possible to classify technical terms by using adjacent phrases. Therefore, by using the classification result information, for example, it is possible to classify various language processing results. In addition, by automatically classifying the technical terms, it does not require much time and labor as in the case of manual classification, and it is also necessary to secure people who are familiar with specialized fields. There is also a merit that there is no.

  Further, since classification is performed using input words and output words, technical terms can be classified for each piece of information to be processed and information to be processed. It should be noted that the classification range can be determined according to whether classification is based on common input words or the like, or classification is based on input words having a common concept.

  Also in the present embodiment, when the search term searched by the search unit 14 is included in the document information a predetermined number of times or more, the classification of the technical terms constituting the search term may be performed. The point which may not be so is the same as that of Embodiment 1.

  In addition, the number of search results may be included in the classification result information, or it may not be necessary, as in the first embodiment. In the present embodiment, since the number of searches must be counted for each input word and / or output word, for example, when it is determined No in step S304 in the flowchart of FIG. The hit count of the corresponding record may be incremented by one.

  In this embodiment, the case where “Yes” and “Yes” are used as the adjacent words has been described, but other adjacent words may be used. For example, “process”, “execute” or the like may be used, and adjacent phrases other than the final form in the use of the sub-variable may be used. For example, in the use of sub-variables, the adjacent forms of words “shi”, “se”, “sa”, consecutive words “shi”, hypothetical words “sore”, imperative words “shiro” “seyo” Or the like may be used. In addition, the adjacent phrase may include a phrase. For example, “Yes”, “Yes”, etc. may be used as the adjacent phrase. As described above, the adjacent phrase used in the present embodiment is not limited to the above-described specific examples as long as it is a suffix that can constitute a term indicating an operation by being adjacent to the end of the technical term.

(Embodiment 3)
A technical term classification device according to Embodiment 3 of the present invention will be described with reference to the drawings. The terminology classification apparatus according to the present embodiment adds a terminology to classification result information and extends the classification result information by using a dictionary for determining whether or not a plurality of technical terms belong to the same classification. is there.

  FIG. 13 is a block diagram showing the configuration of the technical term classification device 3 according to the present embodiment. In FIG. 13, the technical term classification device 3 according to the present embodiment includes a technical term storage unit 11, an adjacent phrase storage unit 12, a document information storage unit 13, a search unit 14, a classification unit 15, and classification result information. A storage unit 16, an output unit 17, a dictionary information storage unit 41, a determination unit 42, and an additional classification unit 43 are provided. The configurations and operations other than the dictionary information storage unit 41, the determination unit 42, and the additional classification unit 43 are the same as those in the first embodiment, and a description thereof is omitted. In the present embodiment, a configuration in which the technical term classification device 1 according to the first embodiment is further provided with a dictionary information storage unit 41, a determination unit 42, and an additional classification unit 43 will be described. The classification device 2 may further include a dictionary information storage unit 41, a determination unit 42, and an additional classification unit 43. In this case, the same effect as described below can be obtained.

  In the dictionary information storage unit 41, dictionary information is stored. Here, the dictionary information is information indicating a dictionary for determining whether or not a plurality of technical terms belong to the same category. This dictionary information may be, for example, information indicating a synonym dictionary, information indicating a correspondence between a broader word and a lower word, or information indicating a dictionary of synonyms.

  A synonym is a term that has a different form but has almost the same meaning, and is sometimes called a synonym. Where two terms are synonymous, the terms are generally interchangeable. For example, “kitchen” and “kitchen” are synonymous. “Kitchen” and “kitchen” are also synonymous. Expressions of the same object in different languages, such as “Kitchen” and “Kitchen” are also synonymous. Also, the abbreviation and the original term are synonymous, such as “word processor” and “word processor”. The technical terms “word processor” and patent terms “document editing device” and “document creation editing device” are also synonymous. “Floppy (registered trademark) disk” and “FD” are also synonymous. In addition, terms such as “cutter” and “cutter” are also synonymous. A synonym dictionary is a dictionary in which synonyms are grouped together.

Broader terms and narrower terms have the following relationship. A term encompassing the concept of a group of a plurality of narrower terms is a broader term. For example, the broad term “animal” corresponds to the narrow terms “cow”, “sheep”, “lion”, “human”, and the like. Further, the broader terms “metal” correspond to the lower terms “copper”, “gold”, “aluminum”, and the like.
Synonyms are terms with similar meanings. For example, “game” and “game” are synonyms.

  The dictionary information storage unit 41 can be realized by a predetermined recording medium (for example, a semiconductor memory, a magnetic disk, an optical disk, etc.). The process of storing dictionary information in the dictionary information storage unit 41 does not matter. For example, the dictionary information may be stored in the dictionary information storage unit 41 via a recording medium, and the dictionary information transmitted via a communication line or the like is stored in the dictionary information storage unit 41. Alternatively, the dictionary information input via the input device may be stored in the dictionary information storage unit 41. Storage in the dictionary information storage unit 41 may be temporary storage in RAM or the like of performance data read from an external storage device or the like, or may be long-term storage.

  Further, the technical term storage unit 11, the adjacent phrase storage unit 12, the document information storage unit 13, and the dictionary information storage unit 41 may be realized by the same recording medium or by different recording media. Also good.

  The determination unit 42 determines whether the technical term to be classified belongs to the same classification as the technical term that has already been classified. This determination is performed with reference to the dictionary information stored in the dictionary information storage unit 41. Further, the technical terms to be classified are technical terms stored in the technical term storage unit 11. The technical terms to be classified may be all technical terms stored in the technical term storage unit 11 or some technical terms. In the latter case, technical terms that are not included in the classification result information may be technical terms that are subject to classification, or for certain categories, technical terms that are not included in the classification are subject to classification. A certain technical term may be used (in this case, the technical term to be classified differs for each classification).

  For example, when the determination unit 42 performs a determination with reference to dictionary information indicating a dictionary of synonyms, it may determine that the two technical terms belong to the same category when the two technical terms have a synonym relationship. . For example, when the determination unit 42 makes a determination with reference to dictionary information indicating correspondence between a broader word and a lower word, the technical terms of a plurality of lower words corresponding to the broader terminology are the same. When belonging to a category, it may be determined that the technical term of the broader term belongs to the category to which the technical terms of the plurality of lower terms belong. For example, a lower limit value may be set for the number of lower-order words in the “plurality of lower-order words”. That is, when a predetermined number or more of low-level technical terms corresponding to the broader terminology belong to the same category, it is determined that the broader terminology belongs to the category to which the low-level terminology belongs. May be.

  When the determination unit 42 determines that the technical term that is the classification target belongs to the same classification as the technical term that has already been classified, the additional classification unit 43 has already classified the technical term that is the classification target. Categorize in the same category as the technical term By adding the technical terms to the classification by the additional classification unit 43, the classification result information is expanded.

  Next, operation | movement of the technical term classification device 3 by this Embodiment is demonstrated using the flowchart of FIG. In this flowchart, only the process for expanding the classification will be described, and the description regarding the process for creating the classification result information by using the adjacent words and the like described in the first embodiment will be omitted.

  (Step S501) The determination unit 42 determines whether it is time to start adding a classification. If it is time to start adding a class, the process proceeds to step S502. If not, the process in step S501 is repeatedly executed until it is determined that it is time to start adding a class.

(Step S502) The determination unit 42 sets the counter i to 1.
(Step S503) The determination unit 42 sets the counter j to 1.
(Step S504) The determination unit 42 reads the i-th technical term stored in the technical term storage unit 11.

  (Step S505) The determination unit 42 refers to the classification result information stored in the classification result information storage unit 16, and determines whether or not the i-th technical term is included in the j-th classification in the classification result information. . If it is included, the process proceeds to step S508. If it is not included, the process proceeds to step S506.

  (Step S506) The determination unit 42 refers to the dictionary information stored in the dictionary information storage unit 41, and the i-th technical term and one of the technical terms included in the j-th classification are in the same classification. Determine if there is. And when it is the same classification, it progresses to Step S507, and when that is not right, it progresses to Step S508.

(Step S507) The additional classification unit 43 adds the i-th technical term to the j-th classification in the classification result information.
(Step S508) The determination unit 42 increments the counter j by 1.

(Step S509) The determination unit 42 determines whether or not the jth classification exists in the classification result information. And when it exists, it returns to step S505, and when that is not right, it progresses to step S510.
(Step S510) The determination unit 42 increments the counter i by 1.

  (Step S511) The determination unit 42 determines whether or not the i-th technical term exists in the technical term storage unit 11. And when it exists, it returns to step S503, and when that is not right, it returns to step S501.

  In the flowchart of FIG. 14, the process is terminated by powering off or a process termination interrupt. In the flowchart of FIG. 14, the classification result information may be output when returning from step S511 to step S501. Further, the output of the classification result information may be executed only when a technical term is added to the classification result information.

Next, operation | movement of the technical term classification device 3 by this Embodiment is demonstrated using a specific example.
In this specific example, as in the specific example of the first embodiment, the technical term classification device 3 is assumed to be a personal computer. Further, it is assumed that the technical term storage unit 11 stores the technical term shown in FIG. Further, it is assumed that the classification result information shown in FIG. 16 is configured and stored in the classification result information storage unit 16 by the same method as in the first embodiment. Further, it is assumed that the dictionary information storage unit 41 stores the synonym dictionary shown in FIG. In FIG. 17, a plurality of synonyms are associated with one ID.

  First, when the process of adding classification is started (step S501), the determination unit 42 reads the first technical term “document editing device” (steps S502 to S504), and the technical term is the first classification, That is, it is determined whether it is included in the classification of ID = 1 (step S505). Here, it is assumed that it is determined not to be included. Then, the determination unit 42 uses the synonym dictionary shown in FIG. 17 to determine whether or not the first technical term “document editing device” and the technical term included in the first category are the same category. Judgment is made (step S506).

  Specifically, the determination unit 42 refers to the synonym dictionary and acquires the synonyms “word processor”, “word processor”, and “document creation / editing device” of the technical term “document editing device”. Then, it is determined whether or not the acquired synonym is included in the first classification. In this case, since “word processor” and “word processor” are included, the determination unit 42 determines that the technical term “document editing device” and the technical term included in the first classification are the same classification. Then, the technical term “document editing device” and the classification identification information “ID = 1” are passed to the additional classification unit 43. Then, the additional classification unit 43 adds the technical term “document editing device” to the classification of ID = 1 (step S507). In this way, technical terms are added and classification result information is expanded.

  Next, a case where the dictionary information storage unit 41 stores the upper and lower thesaurus shown in FIG. 18 will be described. When three or more subordinate words are included in the same field, a broader word corresponding to the subordinate word is added to the field.

  Here, it is assumed that the counter i is 2 and the counter j is 2. Then, the determination unit 42 reads the second technical term “natural language processing” and determines whether or not the technical term is included in the second classification (step S505). Here, it is assumed that it is determined not to be included. Then, the determination unit 42 determines whether or not the second technical term “natural language processing” and the technical term included in the second classification are the same classification using the upper and lower thesaurus in FIG. (Step S506).

  Specifically, the determination unit 42 refers to the upper and lower thesaurus, and the lower terms “machine translation”, “syntactic analysis”, “semantic analysis”, “translation processing”, “kana-kanji conversion”, etc. of the technical term “natural language processing”. To get. Then, it is determined whether the acquired low-order word is included in the second classification. In this case, since the acquired three low-level words “machine translation”, “syntax analysis”, and “translation processing” are included in the second classification, the determination unit 42 determines that the technical term “natural language processing” and the second It is determined that the technical terms included in the classification are the same. Then, the technical term “natural language processing” and classification identification information “ID = 2” are passed to the additional classification unit 43. Then, the additional classification unit 43 adds the technical term “natural language processing” to the classification of ID = 2 (step S507). In this way, technical terms are added and classification result information is expanded.

  As described above, according to the technical term classification device 3 according to the present embodiment, it is possible to classify technical terms by using dictionary information, and as a result, it is possible to expand classification result information. For example, when the document information includes only information indicating the technical document and does not include the information indicating the patent document, it is often used in the patent document as in the above specific example. There is a possibility that the “document editing device” is not classified. However, by extending the classification using dictionary information, it may be possible to add terms that are not included in such literature information to the classification.

  In the present embodiment, the configuration in which the dictionary information storage unit 41, the determination unit 42, and the additional classification unit 43 are added to the technical term classification devices 1 and 2 of the first and second embodiments has been described. The terminology classification device 3 according to 1 extends the classification result information using dictionary information, so the adjacent phrase storage unit 12, the search unit 14, and the like are not essential components. Therefore, as shown in FIG. 19, the technical term classification device 3 includes a technical term storage unit 11, a classification result information storage unit 16, an output unit 17, a dictionary information storage unit 41, a determination unit 42, and an addition. The classification unit 43 is included, and the adjacent phrase storage unit 12 and the search unit 14 may not be included.

  In each of the above embodiments, each processing or each function may be realized by centralized processing by a single device or a single system, or distributed processing by a plurality of devices or a plurality of systems. May be realized.

  In each of the above embodiments, each component may be configured by dedicated hardware, or a component that can be realized by software may be realized by executing a program. For example, each component can be realized by a program execution unit such as a CPU reading and executing a software program recorded on a recording medium such as a hard disk or a semiconductor memory. The software for realizing the technical term classification device in each of the above embodiments is a program as follows. In other words, the program includes an adjacent phrase storage unit in which an adjacent phrase that is a phrase that can be adjacent to the technical term is stored in the technical term stored in the technical term storage unit in which two or more technical terms are stored in the computer. A search step for searching for the search term in a plurality of document information that constitutes a search term in which the adjacent words and phrases stored in are adjacent to each other, and a search result in the search step And a classification result stored in a classification result information storage unit that stores classification result information that is information indicating a result classified in the classification step. And an output step for outputting information.

  Further, another program determines whether the technical term to be classified, which is stored in the technical term storage unit in which the technical term is stored, belongs to the same classification as the technical term that has already been classified. In the determination step that refers to dictionary information that is information indicating a dictionary for determining whether or not a plurality of technical terms belong to the same classification, in the determination step, the technical terms that are the classification targets are already classified. If it is determined that the technical term belongs to the same classification as the technical term in question, the technical term that is the classification target is stored in the classification result information storage unit, and the classification result is information indicating the result of classification of the technical term In the information, an additional classification step for classifying into the same classification as the technical terms already classified, and the classification result information stored in the classification result information storage unit An output step of outputting, is intended for causing the execution.

  In the above program, the output step for outputting information does not include at least processing performed only by hardware, for example, processing performed by a modem or an interface card in the output step.

  Further, this program may be executed by being downloaded from a server or the like, and a program recorded on a predetermined recording medium (for example, an optical disk such as a CD-ROM, a magnetic disk, a semiconductor memory, or the like) is read out. May be executed by

  Further, the computer that executes this program may be singular or plural. That is, centralized processing may be performed, or distributed processing may be performed.

  FIG. 20 is a schematic diagram showing an example of the appearance of a computer that executes the program and realizes the technical term classification device according to the embodiment. The above-described embodiment is realized by computer hardware and a computer program executed on the computer hardware.

  20, a computer system 100 includes a computer 101 including a CD-ROM (Compact Disk Read Only Memory) drive 105 and an FD (Flexible Disk) drive 106, a keyboard 102, a mouse 103, and a monitor 104.

  FIG. 21 is a diagram illustrating a computer system. In FIG. 21, in addition to the CD-ROM drive 105 and the FD drive 106, a computer 101 includes a CPU (Central Processing Unit) 111, a ROM (Read Only Memory) 112 for storing a program such as a bootup program, A CPU (Random Access Memory) 113 that is connected to the CPU 111 and temporarily stores application program instructions and provides a temporary storage space, a hard disk 114 that stores application programs, system programs, and data, a CPU 111 and a ROM 112. Etc. to each other. The computer 101 may include a network card (not shown) that provides connection to the LAN.

  A program that causes the computer system 100 to execute the function of the technical term classification device according to the above embodiment is stored in the CD-ROM 121 or the FD 122, inserted into the CD-ROM drive 105 or the FD drive 106, and stored in the hard disk 114. May be forwarded. Instead, the program may be transmitted to the computer 101 via a network (not shown) and stored in the hard disk 114. The program is loaded into the RAM 113 at the time of execution. The program may be loaded directly from the CD-ROM 121, the FD 122, or the network.

The program does not necessarily include an operating system (OS) or a third-party program that causes the computer 101 to execute the function of the technical term classification device according to the above-described embodiment. The program may include only a part of an instruction that calls an appropriate function (module) in a controlled manner and obtains a desired result. How the computer system 100 operates is well known and will not be described in detail.
Further, the present invention is not limited to the above-described embodiment, and various modifications are possible, and it goes without saying that these are also included in the scope of the present invention.

  As described above, according to the technical term classification device and the like according to the present invention, technical terms can be appropriately classified, which is useful as a system for classifying technical terms.

1 is a block diagram showing the configuration of a technical term classification device according to Embodiment 1 of the present invention. The flowchart which shows operation | movement of the technical term classification device by the embodiment The figure which shows an example of the technical vocabulary in the embodiment The figure which shows an example of the adjacent phrase in the embodiment The figure which shows an example of the classification result information in the same embodiment The block diagram which shows the structure of the technical term classification device by Embodiment 2 of this invention. The flowchart which shows operation | movement of the technical term classification device by the embodiment The flowchart which shows operation | movement of the technical term classification device by the embodiment The flowchart which shows operation | movement of the technical term classification device by the embodiment The figure which shows an example of the adjacent phrase in the embodiment The figure which shows an example of the accumulation | storage result, such as an input word, in the embodiment The figure which shows an example of the classification result information in the same embodiment The block diagram which shows the structure of the technical term classification device by Embodiment 3 of this invention. The flowchart which shows operation | movement of the technical term classification device by the embodiment The figure which shows an example of the technical vocabulary in the embodiment The figure which shows an example of the classification result information in the same embodiment The figure which shows an example of the synonym dictionary in the embodiment The figure which shows an example of the high-order low-order thesaurus in the embodiment The block diagram which shows the structure of the technical vocabulary classification apparatus by another form. The schematic diagram which shows an example of the external appearance of the computer system in each said embodiment The figure which shows an example of a structure of the computer system in each said embodiment

Explanation of symbols

1, 2, 3 Technical term classification device 11 Technical term storage unit 12 Adjacent word storage unit 13 Literature information storage unit 14 Search unit 15, 21 Classification unit 16 Classification result information storage unit 17 Output unit 31 Acquisition unit 32 Identification unit 33 Storage unit 34 Classification means 41 Dictionary information storage section 42 Judging section 43 Additional classification section

Claims (17)

A technical term storage unit for storing two or more technical terms;
An adjacent phrase storage unit that stores adjacent phrases that can be adjacent to technical terms;
A plurality of document information that is information indicating a document in a specific field by configuring a search term in which the adjacent word stored in the adjacent word storage unit is adjacent to the technical term stored in the technical word storage unit A search unit for searching for the search term;
According to the search result by the search unit, a classification unit that classifies the technical terms included in the search term;
A classification result information storage unit in which classification result information, which is information indicating a result classified by the classification unit, is stored;
And an output unit that outputs the classification result information stored in the classification result information storage unit. A document information storage unit for storing a plurality of document information;
The technical term classification device according to claim 1, wherein the search unit searches for a search term in a plurality of document information stored in the document information storage unit. The adjacent phrase is a prefix that can be adjacent to the beginning of the technical term, or a suffix that can be adjacent to the end of the technical term,
The classification unit, when the search term searched by the search unit is included in the literature information, classifies the technical terms constituting the search term according to the adjacent words constituting the search term. The technical term classification device according to claim 1 or claim 2. The terminology classification device according to claim 3, wherein the classification unit classifies the technical terms constituting the search term when the search term searched by the search unit is included in the document information a predetermined number of times or more. . The adjacent words are divided into a plurality of groups,
When the search term searched by the search unit is included in the literature information, the classification unit classifies the technical terms constituting the search term according to a group of adjacent phrases constituting the search term. The technical term classification device according to claim 3 or 4. The technical term classification device according to claim 3, wherein the classification unit classifies the technical terms in association with the number of search results of the search terms searched by the search unit. The adjacent phrase is a suffix that can constitute a term indicating an action by being adjacent to the end of a technical term,
The classification unit includes:
Obtaining means for obtaining one or more sentences including the search term from literature information;
Specifying means for specifying an input word and / or an output word for the search term in one or more sentences acquired by the acquiring means;
Storing means for storing the input words and / or output words specified by the specifying means in association with technical terms constituting the search terms;
The classification means which classify | categorizes the technical vocabulary corresponding to the said input word and / or output word according to the input word and / or output word which the said accumulation | storage means accumulate | stored was provided. Technical term classification device. The technical term classification device according to claim 7, wherein the classification unit classifies technical terms common to input words and / or output words into the same classification. 8. The technical term classification device according to claim 7, wherein the classification means classifies technical terms having a common concept of input words and / or output words into the same classification. A dictionary information storage unit that stores dictionary information that is information indicating a dictionary for determining whether a plurality of technical terms belong to the same classification;
A determination unit that determines whether or not the technical term to be classified, which is stored in the technical term storage unit, belongs to the same classification as the technical term that has already been classified, with reference to the dictionary information;
When the judgment unit determines that the technical term that is the classification target belongs to the same classification as the technical term that has already been classified, the technical term that is the classification target is changed to the technical term that has already been classified. The technical term classification device according to any one of claims 1 to 9, further comprising an additional classification unit for classifying into the same classification. The dictionary information is information indicating a dictionary of synonyms,
The technical term classification device according to claim 10, wherein the determination unit determines that the two technical terms belong to the same classification when they have a synonym relationship. The dictionary information is information indicating correspondence between broader terms and narrower terms,
The determination unit determines that the technical term of the broader term belongs to the classification to which the technical term of the lower term belongs when a plurality of the technical terms of the lower term corresponding to the technical term of the broader term belong to the same classification. The technical term classification device according to claim 10. A technical term storage unit for storing technical terms;
A classification result information storage unit in which classification result information, which is information indicating a result of classification of technical terms, is stored;
A dictionary information storage unit that stores dictionary information that is information indicating a dictionary for determining whether a plurality of technical terms belong to the same classification;
A determination unit that determines whether or not the technical term that is the classification target stored in the technical term storage unit belongs to the same classification as the technical term that has already been classified, with reference to the dictionary information;
When the judgment unit determines that the technical term that is the classification target belongs to the same classification as the technical term that has already been classified, the technical term that is the classification target is changed to the technical term that has already been classified. An additional classifier that classifies the same class;
And an output unit that outputs the classification result information stored in the classification result information storage unit. The terminology storage unit stores two or more technical terms, the adjacent term storage unit stores adjacent words that can be adjacent to the technical terms, the search unit, the classification unit, and the classification unit. A classification result information storage unit for storing classification result information, which is information indicating a result obtained, and a technical term classification method processed using an output unit,
The search unit constitutes a search term in which the adjacent term stored in the adjacent term storage unit is adjacent to the technical term stored in the technical term storage unit, and indicates information in a specific field. A search step for searching for the search term in a plurality of document information,
A classification step in which the classification unit classifies technical terms included in a search term according to a search result in the search step;
An output step in which the output unit outputs the classification result information stored in the classification result information storage unit. A technical term storage unit that stores technical terms, a classification result information storage unit that stores classification result information, which is information indicating the result of classification of technical terms, and a judgment as to whether multiple technical terms belong to the same category A terminology classification method that is processed using a dictionary information storage unit that stores dictionary information that is information indicating a dictionary for performing, a determination unit, an additional classification unit, and an output unit,
Judgment in which the judgment unit judges whether or not the technical term to be classified, which is stored in the technical term storage unit, belongs to the same classification as the technical term already classified, with reference to the dictionary information Steps,
When the additional classification unit determines in the determination step that the technical term that is the classification target belongs to the same classification as the technical term that has already been classified, the technical term that is the classification target is already classified. An additional categorization step to classify the same categorization as the terminology used,
An output step in which the output unit outputs the classification result information stored in the classification result information storage unit. On the computer,
Adjacent to the technical terms stored in the technical term storage unit storing two or more technical terms, the adjacent terms stored in the adjacent phrase storage unit storing adjacent phrases that can be adjacent to the technical terms A search step for searching the search term in a plurality of document information that is information indicating a document in a specific field,
A classification step of classifying technical terms included in the search terms according to the search results in the search step;
A program for executing an output step of outputting classification result information stored in a classification result information storage unit in which classification result information, which is information indicating a result classified in the classification step, is stored. On the computer,
Whether the technical terms to be classified that are stored in the technical term storage unit that stores the technical terms belong to the same category as the technical terms that have already been classified, or whether multiple technical terms belong to the same category A determination step of determining with reference to dictionary information that is information indicating a dictionary for determining whether or not;
In the determination step, when it is determined that the technical term that is the classification target belongs to the same classification as the technical term that has already been classified, the technical term that is the classification target is stored in the classification result information storage unit. In the classification result information, which is information indicating the result of classification of the technical terms, an additional classification step for classifying into the same classification as the technical terms already classified,
An output step of outputting the classification result information stored in the classification result information storage unit.

Images