JP3013388B2 - Linguistic information retrieval device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a linguistic information search apparatus, and more particularly, to a part and a whole of linguistic information using a characteristic characteristic of the linguistic information, that is, various attributes of the linguistic information as a search instruction key (search key). The present invention relates to a language information search device configured to arbitrarily search for. [Prior Art] The linguistic information constituting a document is digitized, that is, replaced as electronic linguistic information in a storage means of an information processing apparatus and input, and the linguistic information is used for generating a new document, or is used for public communication. 2. Description of the Related Art Electronic document processing technology for transmitting and receiving data via an information processing network such as a network or a LAN has come into practical use. Word processors (word processors) and personal computers (personal computers), which have become very popular in ordinary households, are common examples of this type of electronic document processing device.
This is a workstation as one means of office automation in a company or the like. In a conventional document processing apparatus of this type, the document is simply digitized, and the reusability of the electronic document is poor, such as creating a new document using the digitized document. Therefore, as an approach for taking advantage of the digitization such as reusability as described above, it has been considered that linguistic information is decomposed into detailed attributes and structured. Construct a database that converts the digitized linguistic information into records that are decomposed into various toxicities that are unique characteristics,
When linguistic information is searched using the above various attributes as search keys, conventionally, for example, as in an electronic translation device disclosed in Japanese Patent Publication No. 59-865, the word of the word as an attribute of the first language is used. A database in which a set of a document to be read or an example sentence and a word or an example sentence having a second language having a corresponding meaning as an attribute is stored is stored, and a database of a first language or a second language is stored. Generally, one of the words is input as a search key for translation, and the corresponding second language or first language word or example sentence is displayed on the display means. [Problems to be solved by the invention] In order to handle a structured electronic document as described above,
A different user interface is required. For example, "Smalltalk-80" (trade name) known as an object-oriented programming environment (Adele Goldber
g, “Smalltalk-80, The interactive programming envi
The inspector in ronment ", Addison Wesley, 1984) is a two-part interface, limited to linguistic information, a list menu showing a list of instance variable names and a text view showing the value of the selected variable. This is not a user interface, but it is an indispensable user interface for referencing and changing the structure of an object, etc. However, one inspector can only obtain information on one variable or the object itself, and refers to the relationship between all language information. Like changing one variable with
It is not envisaged to use the “whole” information as a text such as the language information and the “parts” constituting the text at the same time. The present invention realizes a user interface that simultaneously provides linguistic information by combining the advantages of both, by capturing linguistic information from structured “part” information and “whole” information as text. Another object of the present invention is to make it possible to use digitized language information effectively. [Means for Solving the Problems] The object is to provide a storage unit that stores linguistic information as a record decomposed for each attribute, and a part of the record stored in the storage unit that displays details about the specified attribute. Display means, an entire display means for displaying a plurality of attributes relating to the records stored in the storage means as flat text as a whole, and a partial search means for searching the records stored in the storage means for each attribute unit; An overall search means for performing a search based on the relation of the entire record stored in the storage means; a partial search instruction means for instructing the partial search means and the overall search means to perform a search; The search instruction means, the whole display means, the partial search instruction means and the partial display means constitute a simultaneously displayed user interface. Achieved by [Operation] At the same time, the entire display of the search result obtained by the search key input from the search instruction means as flat text similar to a normal dictionary and the partial display showing the hierarchical structure of items related to the search key are simultaneously performed. By displaying, the subsequent search based on both displays can be executed efficiently. [Example] Hereinafter, an example of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an embodiment of a linguistic information search apparatus according to the present invention, wherein 1 is a storage means, 2 is a character input means, 3 is a whole search instruction means, 4 is a whole display means, 5 Is a partial search instructing means, and 6 is a partial display means. In FIG. 1, a character input means 2 has a record creation / registration function when constructing a database, in addition to a search key instruction input function described later. First, when constructing a database, the input linguistic information is decomposed into various attributes such as reading, pronunciation, notation, meaning, part of speech, and example sentences, and these are stored in the storage means 1 as records having a hierarchical structure such as a tree structure. register. In this way, a database is constructed by registering the language information in the storage unit 1 as a record decomposed into its attributes. The whole search instruction means 3 and the whole display means 4 and the partial search instruction means 5 and the partial display means 6 constitute a user interface, and are presented on a common display means (view). Using this user interface, the user inputs a desired inspection key from the entire search instruction means 3 or the partial search instruction means 5. The result searched according to the input of the search key is simultaneously displayed on the whole display means 4 and the partial display means 6 on the display means and presented to the user. The user performs subsequent search work based on this presentation. FIG. 2 is an explanatory diagram of a structured record of a language using an English-Japanese electronic dictionary search system as an example, and shows a record in which dictionary items (language attributes) corresponding to headwords are structured. In the figure, 21 is a word, 22 is a headword, 23 is a heading notation, 24 and 25 are two different notations (Color and Color) registered as heading notation 23, 26 is a word importance rank, and This indicates that the importance is “2” 27. 28 is pronunciation, 29 is phonetic symbol, 30 is phonetic notation registered as phonetic symbol 29, 31 is phonetic symbol katakana notation, 32 is katakana phonetic notation (color), 33 is the part of speech of the word 21, 34 is a first part of speech, 35 is a part of speech name, and indicates that the first part of speech is a noun 36. 37 indicates the meaning of the first part of speech 34, and indicates that the meaning 37 has a first meaning 38 and a second meaning 45, and the first meaning 38 has a translation 39 as “color” 40 and “color”. 41, and an English sentence 42 and its translation 44 are shown as an example sentence 42 of the first definition 38. These example sentences
In the text 42 and the translation 44, the actual example sentences and the actual translation are structured in the following. In the second definition 45, there is a translation 46, and "paint" 47 is provided as an actual word. "Other" (transitive verb) 50 as the part of speech name 49 in the second part of speech 48
However, a first meaning 52 as a meaning 51, a translation 53 thereof, and a "coloring" 54 as an actual example sentence of the translation 53 are structured.
If there is a third part-of-speech, a fourth part-of-speech,..., It is structured in the same manner as above, but illustration and description are omitted. The record structured as described above is stored in the storage unit 1 shown in FIG.
Is stored sequentially to build a database. Generally, since each item of the electronic foreign language dictionary is described simply according to simple rules, the structure of the item is easy. In addition, a thesaurus is registered in the storage means 1 in addition to the above. This thesaurus is originally a dictionary created to limit keywords, but in a search system, in order to improve search efficiency, in addition to the limited keywords, its superordinate concepts and subordinate concepts, related words, or synonyms, antonyms, etc. And can be created using the above structured database. FIG. 3 is an explanatory view of an example of the above-mentioned electronic dictionary database registration procedure. FIG. 3 (a) is a flowchart of the processing operation, and FIG. 3 (b) is an explanatory view of a one-word format. The one-word format is composed of an array such as a headword, a blank, a phonetic symbol, a blank, a part of speech, a blank,... Hereinafter, the database registration processing operation will be described with reference to this flowchart. First, the character input means 2 shown in FIG. 1 reads word by word from an original dictionary file on which a database is built, and outputs a word-by-word data stream of the format shown in FIG. Is generated (S1),
This is stored in the word record area of the storage means 1 from the head of ST to its blank (NULL) as a headword (S2). Then, the next character after ST is the phonetic symbol notation '['
Is determined (S3). If YES, up to "]" shown in (b) is stored as phonetic symbols (S7), and the procedure goes to S4. If NO in S3, the next character after ST is the part-of-speech symbol ('

[','] ') Is determined (S4), and YE
If it is S, the part-of-speech symbol or line feed symbol following ST is stored as a part-of-speech item (S8), and the process returns to S4. If the answer is NO, a line feed symbol comes and the ST of the word ends, so that it is registered in the storage means as a record having an attribute decomposed using the above-mentioned heading as a key (S5). By repeating this (Yes in S6 → S1 →...), A database consisting of all ST-based records generated from the original dictionary is constructed. When performing dictionary lookup (dictionary search) using the database generated as described above, the user gives a search key from the character input means 2 and sequentially searches for a desired word. 4 and 5 are explanatory views of a specific example of the user interface according to the present invention. In the view as the user interface, reference numeral 33 denotes an entire search instruction unit, reference numeral 34 denotes an entire display unit, and reference numeral 35 denotes a partial search. An instruction unit, 36 is a partial display unit, 37 is an input display unit, 38 is a front and rear search button, and 39 is a cursor. The whole search instructing unit 33, the whole display unit 34, the partial search instructing unit 35, and the partial display unit 36 are the whole search instructing unit 3, the whole display unit 4, the partial search instructing unit 5, and the partial displaying unit 6 in FIG.
This is a display related to the operation of. As described above, in the present embodiment, all the parts that are visible to the user, that is, the “display unit / instruction unit” are integrated in the same display (the same view), and the search result is a flat text like an ordinary dictionary, and The partial display reflecting the hierarchical structure is presented at the same time. In the figure, the partial display presented on the partial display unit 36 is a headword (choos) displayed on the input display unit 37.
Regarding e), this is an example of a second part of speech (part of speech 2) indicated by hatching on the partial display section 36, and a display example of the English sentence. That is, the display content presented on the partial display unit 36 is the result of a search for the second part of speech, that is, the transitive verb (other) presented on the overall display unit 34. In this way, since the linguistic information about a headword is displayed so as to present both the entire text and the item-by-item portion at the same time, the user can use the entire search result and the portion in accordance with the dictionary use situation, This electronic English-Japanese dictionary can be used efficiently. First, if you want to see visually at the overall notation level,
A general search is performed after a quick look at the entire display of the presented display content. The overall search here is a function that searches for related words without being aware of the attribute that the user focuses on.
There are “concept search” and “notation search”. The “concept search” is a function of searching for a synonym, a synonym, or a word having the same synonym one after another, and is performed with reference to a thesaurus stored in the storage unit. This thesaurus is originally a dictionary created to limit keywords, but in a search system, in order to improve search efficiency, in addition to the limited keywords, the superordinate concept, subordinate concept, related words, or synonyms,
It describes antonyms. For example, in the case of “choose”, synonyms “select” “el
ect "" desire "and part-of-speech form" choice "Notation search is a function that searches for similar spellings, pseudophonic words, idioms, etc. at the notation level.
In the case of "e", similar spelling is searched for "cheese" and similar pronunciation is "Tuesday" etc. The actual search is performed by the search system by automatically combining pronunciation and meaning attributes. In addition, since displaying all information can be complicated, specifying the "display level" in such a case allows simple display such as only part of speech and bilingual display. It is also possible to switch to. If the user's intention to the search result is clear and it is known which attribute is desired to be used, a partial search is performed. For example, as shown in FIG. 4, it is suitable for searching for an example sentence related to a specific part of speech, or for viewing only a phrase as a whole. Further, a grandchild can be subtracted from the obtained attribute.
In FIG. 4, if the meaning of “carefully” in the example sentence is not understood, “carefully” in the example sentence of the partial display unit 36 in FIG.
Is selected by the cursor 39 and the search is instructed by the partial search instructing section 35, "carefully" is searched as shown in FIG. 5, and the search result is presented on the entire display section 34. The same applies to the case of the whole search. However, even if the search for the display word in the display unit (the whole view of the display means) is performed, the original word previously input to the input display unit 37 is always stored. A word as a search key that has been input, including candidates for a plurality of items, can be searched instantaneously by using the front and rear search buttons 38, so that there is no need to uselessly search for the same word many times. Furthermore, the combination of the search procedure of performing a partial search after grasping the meaning of the whole display can effectively execute the use efficiency of the electronic dictionary, and the concept of words can be easily grasped. Also improve. FIG. 6 is a flowchart illustrating the flow of a dictionary search process according to the present invention. Hereinafter, the dictionary search procedure described above will be described with reference to FIG. The search is performed by first determining whether the type of search desired by the user is a full search or a partial search (S11). This is executed by judging which of the whole search instructing unit 33 and the partial search instructing unit 35 has been input in the user interface shown in FIG. If the search instruction is a general search, it is further determined whether the instruction input is a concept search or a notation search (S1).
2) If it is determined that a concept search has been instructed, a similar concept is searched for by referring to a thesaurus stored in the storage means (S13), and filtering based on the notation level is performed (S13).
16). This filtering is a process of extracting the items of the similar concept searched in S13 at the level of the notation. The filtered search results are presented to the user by displaying them on the entire display section 34 of the user interface. If it is determined in S12 that the search instruction is a notation, the spelling checker searches the input search key for a similar notation (S14). This spelling checker is a function that verifies the input spelling and is originally an algorithm for detecting spelling mistakes, but here it is used as a function to extract notations with similar spelling (color / colour, etc.) I have. The similar notation searched by the spelling checker is then searched for a similar pronunciation item using a pronunciation dictionary (items classified into pronunciation attributes) (S15), and the similar pronunciation search result is filtered by the notation level. Apply (S1
6). The search result is displayed on the entire display unit 34 and presented to the user (S17). On the other hand, if it is determined in S11 that the input instruction is a partial search, the instruction attribute is checked to identify the instructed attribute (S18), and then a search for the attribute is executed (S19). The search result is displayed on the partial display unit 36 (S20) to present it to the user. Note that the present invention is not limited to the foreign-language electronic dictionary search system such as the English-Japanese electronic dictionary described in the above-described embodiment. For example, by applying the present invention as a dictionary editing system, the entire text can be visually observed. Therefore, it is possible to provide a dictionary editing environment in which an error can be grasped and then edited in attribute units by partial search, so that a high-quality electronic dictionary can be created. Also, on the view of the user interface,
By combining the whole display part and the partial display,
As described above, it is possible to use words for constructing a thesaurus based on a created electronic dictionary instead of digitizing an existing thesaurus because words can be captured in units of concepts. [Effects of the Invention] As described above, according to the present invention, in using a language database such as an electronic dictionary, a word is grasped from the viewpoint of the whole attribute and the viewpoint of a part decomposed into individual attributes. In addition, it is easy to arrange concepts and to grasp the spatial meaning between words having common attributes, and it is possible to make optimal use according to the situation.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of a language information search apparatus according to the present invention, FIG. 2 is an explanatory view of a structured record of a language using an English-Japanese electronic dictionary search system as an example, and FIG. FIGS. 4 and 5 are explanatory diagrams of a specific example of a user interface according to the present invention, and FIG. 6 is a flowchart illustrating the flow of a dictionary search process according to the present invention. It is a chart. 1 ... storage means, 2 ... character input means, 3 ... overall search instruction means, 4 ... overall display means, 5 ... partial search instruction means, 6 ... partial display means.

Continuation of front page (56) References JP-A-63-101967 (JP, A) JP-A-62-287336 (JP, A) JP-A-63-101968 (JP, A) JP-A-62-229467 (JP) , A) Ogawa, Sahara, Kaneko, "Hypermedia Creation Support System Including Audio and Video," Proc. Of the 1989 Informatics Symposium, p. 43-52 (Heisei 1-1-17) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06F 17/21-17/30 JICST Scientific and Technical Reference File

Claims (1)

(57) [Claims] 1. A storage means for storing linguistic information as a record decomposed for each attribute, a partial display means for displaying details of a specified attribute among the records stored in the storage means, a storage means Whole display means for displaying a plurality of attributes relating to the record stored in the memory as flat text as a whole, partial search means for searching the record stored in the storage means for each attribute, and storage in the storage means An entire search means for searching the obtained record on the basis of the entire association, a partial search instruction means for instructing the partial search means and the entire search means to perform a search, and an overall search instruction means, Means, the whole display means, the partial search instruction means and the partial display means constitute a simultaneously displayed user interface Language information retrieval apparatus according to claim.