JP4397221B2 - Link setting device and information using information extracted from text sentence - Google Patents

Description

  The present invention relates to an apparatus and method for performing efficient processing in an information processing apparatus using abstract model data generated from numerical information data in a document and its dependency information.

  In an electronic document, a function (so-called link function) that correlates a description when there is a description related to the document is generally used so that a viewer can easily understand the contents of the document. For example, even if there is a word that you do not understand when you are browsing a document, you can easily refer to the explanation by clicking with the mouse if the explanation is related to the word. it can.

  Since it is cumbersome for a person to set such a link, there is a technique for setting a link by causing a computer to perform matching processing. In the prior art, a link is mechanically set by matching a keyword extracted by morphological analysis of a text sentence in an electronic document with an index character string (for example, Patent Document 1 and Patent Document 2). .

Japanese Patent Laid-Open No. 3-95673

JP-A-7-325827

  However, none of these conventional link setting devices are intended to set hyperlinks in numerical information data (for example, 100, one hundred, etc.). This is because a numerical word does not have a specific meaning in itself, and is not suitable for setting a link by matching. Such a problem that only desired numerical information data cannot be extracted simply by matching text sentences also exists in keyword search.

  In order to solve the above problem, an object of the present invention is to generate an abstract model from numerical information data and its dependency information, and to provide the information processing apparatus as effective information.

(1, 2, 14) The link setting device of this invention
A link setting device for setting a link between target data including related sentence element data in document data,
The text data in the given document data is divided into a plurality of text element data, the type of each text element is determined and recorded in the recording unit, and the numerical text data is included in the given text text data And if included, an extraction / analysis means for specifying numerical information data and recording it in the recording unit, and
Referring to the recording unit, the sentence element data including the numerical information data is identified, and at least one sentence element data having a dependency relation with the numerical information data is determined based on the type of the sentence element data. Dependency information extracting means for extracting dependency information data from the sentence element data,
Abstract model recording that records the numerical information data extracted by the numerical information extraction means and the dependency information data extracted by the dependency information extraction means as abstract model data together with position data indicating the position in the document data in the recording unit Means,
Abstraction model selection means for selecting abstract model data having the same numerical information data and the same dependency information data among a plurality of abstract model data recorded in the recording unit;
Link setting means for setting a link between target data of the abstraction model data based on the position information data of the selected abstraction model data;
It is provided with.

  Thereby, it is possible to accurately set a link even for numerical information data that cannot be set by simple matching alone.

(3) In the link setting device according to the present invention, when the link setting target data is a text sentence, the link setting means sets a link so as to extract only additional information from the link destination text sentence. It is characterized by that.

  Thereby, peripheral information (additional information) can be effectively displayed by omitting the obvious information at the link destination.

(4, 5, 15) The document retrieval apparatus of the present invention
A document search device that searches document data based on target data that is a search element,
The text data in the given document data is divided into a plurality of text element data, the type of each text element is determined and recorded in the recording unit, and the numerical text data is included in the given text text data And if included, an extraction / analysis means for specifying numerical information data and recording it in the recording unit, and
Referring to the recording unit, the sentence element data including the numerical information data is identified, and at least one sentence element data having a dependency relation with the numerical information data is determined based on the type of the sentence element data. Dependency information extracting means for extracting dependency information data from the sentence element data,
Abstract model holding that holds the numerical information data extracted by the numerical information extraction means and the dependency information data extracted by the dependency information extraction means as abstract model data together with the position data indicating the position in the document data. means,
By retrieving the abstract model recording means that generated and recorded the abstract model data from the document data and comparing it with the abstract model data received from the abstract model holding means, the same numerical information data and the same dependency information data are obtained. Abstraction model comparison means for extracting only position information data of abstraction model data having,
Search result display means for displaying, as search results, target data having the same abstract model data, based on the positional information data of the extracted abstract model data;
It is provided with.

  Thereby, a desired search process can be performed even for numerical information data that cannot be searched by a simple matching search.

(6) The document search apparatus according to the present invention is characterized in that the abstract model recording means previously generates and records abstract model data before searching for the entire document data to be searched.

  Thereby, the search processing of document data can be performed efficiently.

(7, 8, 16) The document input verification device of the present invention of the present invention is
A document input verification device for verifying document data input from a document editing device based on target data of a verification element,
Text text data in the document data given from the document editing device is divided into a plurality of text element data, the type of each text element is determined and recorded in the recording unit, and numerical information is included in the given text text data It is determined whether or not data is included, and if included, extraction / analysis means for identifying numerical information data and recording it in a recording unit,
Referring to the recording unit, the sentence element data including the numerical information data is identified, and at least one sentence element data having a dependency relation with the numerical information data is determined based on the type of the sentence element data. Dependency information extracting means for extracting dependency information data from the sentence element data,
Abstract model holding that holds the numerical information data extracted by the numerical information extraction means and the dependency information data extracted by the dependency information extraction means as abstract model data together with the position data indicating the position in the document data. means,
Abstraction model recording means that generates and records abstract model data from document data is retrieved, and the abstraction model data received from the abstract model holding means is the same as the dependency model, and the numerical information data is different. Abstraction model discrimination means for discriminating whether there is model data;
An input error output means for outputting input error information to the document editing device based on the determination result obtained from the abstract model determination means;
It is provided with.
Thereby, it is possible to easily check numerical information data that is likely to be erroneously input.

(9) The document search apparatus of the present invention is characterized in that the abstract model data of the abstract model recording means is generated as needed based on the input from the verification element input means and is constantly updated.

  This makes it possible to efficiently perform document input verification processing.

(10) Each device of the present invention
The document data includes other elements to be processed other than text sentence data, and
Before extracting the abstraction model data, an expression format determination means for determining the expression format of the processing target element;
When the representation format of the target data is a text sentence, abstract model generation means for extracting, from the processing target element, a sentence element having a dependency relationship with the numerical information data as abstract model data together with the position information data ,
When the representation format of the target data is another element, an abstract model generation unit that generates abstract model data based on a predetermined rule and supplies them to the abstract model recording unit together with the position information data,
It is provided with.

  Thereby, the abstract model data can be generated from elements other than the text sentence.

(11) Each device of the present invention is characterized in that the other target data is table data or image data.

  Thereby, abstract model data can be generated from a table or an image.

(12) Each device according to the present invention is characterized in that the document data is described in an XML format, and an attribute of an expression format is attached in advance to each tag included in the file content.

  This facilitates the determination of the expression format and the specification of abstraction processing elements.

(13) Each device of the present invention further includes:
When there are a plurality of languages of text sentences included in the document data, an abstract model translating unit that unifies the abstract model data into the same language by referring to a translation dictionary is provided.

  Thereby, even when a plurality of languages are included, the abstract model data can be generated and used.

  The “abstraction processing element (target data)” in this specification refers to a target of abstraction processing by each device, in which document data is divided in a representation format such as a text sentence including numerical information data, a table, and an image. Data.

  The “abstract model data” in this specification refers to data in which dependency information is associated with numerical information data extracted from a text sentence or the like.

1. First Embodiment [Abstract Model Generation Device]
First, the principle of extracting numerical information data and dependency information in order to generate abstract model data and the processing until the generation of abstract model data will be described below with reference to FIGS.

1-1-1. Functional Block Diagram FIG. 1 shows a functional block diagram of the abstract model generation device according to this embodiment. In this figure, the abstract model generation apparatus according to the present invention includes a cutout means 101, numerical information data extraction means 103, text sentence analysis means 105, dependency information extraction means 107, and abstract model output means 109. .

  The cutout unit 101 cuts out text sentence data one by one from the input document and records it in the recording unit. For example, a process of cutting out text sentence data is performed using a delimiter “.” Appearing in the text sentence.

  The numerical information data extraction unit 103 refers to the text sentence data read from the recording unit in order from the top, and determines whether or not predetermined numerical information data is included. When predetermined numerical information data is included, the numerical information data is extracted and the text sentence is given to the text sentence analyzing means.

  The text sentence analysis unit 105 performs morphological analysis on the text sentence data read from the recording unit, and then analyzes each morpheme based on predetermined grammar definition information, and divides the text sentence data into predetermined sentence element data.

  The dependency information extracting means 107 is based on the sentence element data including the extracted numerical information data, and other sentence element data having a dependency relation with the numerical information data from the text sentence data divided into the sentence element data. To extract.

  The abstract model output means 109 associates the numerical information data extracted by the numerical information data extraction means 103 with the dependency information data extracted by the dependency information extraction means 107, and outputs them as abstract model data. For example, the data is output to a display or a database.

1-1-2. Hardware Configuration FIG. 2 shows an example of a hardware configuration in which the abstract model generation device shown in FIG. 1 is realized using a CPU. The abstract model generation apparatus includes a display 201, a CPU 203, a memory 205, a keyboard / mouse 207, a hard disk 209, a CD-ROM drive 211, and a communication circuit 215.

  The hard disk 209 includes an information extraction program 2091 for performing abstract model extraction processing according to the present invention, a morphological analysis dictionary 2093 for performing morphological analysis processing, grammar definition information 2094 for performing syntax analysis processing, and predetermined numerical values. A numerical unit master 2095 for extracting information data, an abstract model DB 2097 for recording abstract model data by abstract model extraction processing, and the like are recorded. These are stored in the CD-ROM 212 via the CD-ROM drive 211. The recorded data is read and installed.

  The installation may be performed using data downloaded from the Internet 216 or the like using the communication circuit 215.

1-1-3. Flowchart Processing based on the information extraction program 2091 will be described using the flowcharts of FIGS. In the following, an example will be described in which a text sentence including the text sentence “Operating profit is 32,763 million yen, the same level as the previous year” is input to the clipping means 101.

  In the abstract model extraction process flowchart shown in FIG. 3, the CPU 203 reads a text sentence and inputs it to the memory 205 (step S301).

  Text sentence data is cut out from the head of the input text sentence (step S303). In the present embodiment, the punctuation mark “.” Is used as a text sentence delimiter. For example, when the text sentence data is read, the first sentence, “Operating income is 32,976 million yen, which is the same level as the previous year.” Is extracted.

  The CPU 203 performs numerical information data extraction processing based on the extracted text sentence data (step S305).

1-1-3-1. Numerical Information Data Extraction Processing In the flowchart of the numerical information data extraction processing shown in FIG. 4, the CPU 203 extracts unit characters of numerical information data based on the numerical unit master 2095 (step S401). For example, in this embodiment, unit characters relating to amounts such as “million yen” and “100 million yen” are recorded in the numerical unit master.

  In particular, the numerical unit master may be set according to the contents of a text sentence from which numerical information data is extracted. For example, in the case of securities reports and financial statements, “Yen”, “Billions of yen”, “Millions of yen”, “Thousands of yen”, etc., which are unit characters relating to monetary amounts, are set, such as historical chronology and newspaper articles In this case, “year”, “month”, “day”, etc., which are unit characters relating to the date, may be set. Thereby, only necessary information can be extracted according to the contents of the document.

  If the numerical unit master is not used, a predetermined unit character for the numerical information data may be set in the information extraction program 2091.

  The CPU 203 searches whether or not the extracted unit character exists from the head of the text sentence data (step S403). For example, it is searched from the head of the text sentence “Operating income was 32,763 million yen, which is the same level as the previous year”, whether or not the unit characters “million yen” or “100 million yen” exist.

  If a predetermined unit character exists in the text sentence data (step S405, YES), numbers consecutive in front of the unit character are extracted as constituent elements of the numerical information data (step S407). In this case, if a plurality of numbers are continuous, a character string of numbers obtained by concatenating all the continuous numbers from the head direction is extracted as a constituent element of the numerical information data. Note that even if symbols such as a comma “,” and a decimal point “.” Are consecutive, character strings are concatenated and extracted without being separated there. Further, the CPU 203 concatenates the constituent elements of the extracted numerical information data and the unit characters to obtain numerical information data.

  For example, in the case of the above text sentence data, “32776” is extracted as a constituent element of the numerical information data, concatenated with the unit character “million yen”, and “32776 million yen” is extracted as the numerical information data. 205 or a predetermined area of the hard disk 209.

  If there is another unit character in the text sentence data (step S409, NO), the CPU 203 repeats step S407. If there is no other unit character, the process ends (YES in step S409). In addition, if no unit character exists in step S405, the process is ended (NO in step S405).

1-1-3-2. Text sentence analysis processing If numerical information data is extracted in the numerical information data extraction process (FIG. 3, step S305) (step S307, YES), the CPU 203 performs a text sentence analysis process (step S309).

  In the flowchart of the text sentence analysis process shown in FIG. 5, the CPU 203 performs a morphological analysis of the text sentence data based on the morphological analysis dictionary 2093 (step S501). For example, for the morphological analysis process, a “tea bowl” of Matsumoto Laboratory of Nara Institute of Science and Technology may be used.

  FIG. 7, FIG. 7a and FIG. 7b show an example in which the abstract model extraction processing is performed on the text sentence data “Operating profit was 32,976 million yen, which is the same level as the previous year” 701.

  The CPU 203 divides the text sentence data 701 (FIG. 7) into morphemes, which are the smallest meaningful language units, through the morphological analysis processing in step S501.

  For example, in FIG. 7, the text sentence data 701 includes “operating profit (noun) / ha (particle) / previous year (noun) / and (particle) / same level (noun) / ) / 32776 million yen (noun) / and (particle) / na (verb) /. (End of sentence). In the text sentence data 703 shown in FIG. 7 and FIG. 7, the text sentence data is displayed by inserting “/” for each morpheme for the sake of explanation, but actually, as shown in A of FIG. It is recorded in a predetermined area of the memory 205 or the hard disk 209 as morphological analysis data.

Based on the grammar definition information 2094, the CPU 203 performs a syntax analysis of the text sentence data subjected to the morphological analysis in the above (step S503). For example, the syntax analysis process may use YACC, which is an LR grammar parser for ascending syntax analysis. The parser refers to a syntax analysis program, which is a program that analyzes an input character string and creates a syntax analysis tree.
The LR grammar parser performs syntax analysis based on preset grammar definition information 2094. FIG. 8 shows an example of grammar definition information 2094 set in this embodiment. In the grammar definition information 2094 shown in FIG. 8, grammar information is defined by BNF notation.

  In step S503, the CPU 203 inputs the text sentence data subjected to morphological analysis to YACC, which is an LR grammar parser, and performs syntax analysis. For example, if the text sentence data 701 shown in FIG. 7 is parsed using the YACC and the grammar definition information 2094, the result is as shown in 705.

  In 705, based on the grammar definition information 2094, syntax analysis is performed step by step as in levels 1-8. Thereby, the whole text sentence data can be divided | segmented into predetermined sentence element data in each level. For example, at level 2, the text sentence data 701 is “operating profit is (subordinate phrase) / previous year (subordinate phrase) / same level (subordinate phrase) / 32976 million yen (subordinate phrase) / Can be divided into each sentence element data of “verb phrase)”.

  As shown in B to D of FIG. 7a and E to I of FIG. 7b, the CPU 203 records the parsed data in a predetermined area of the memory 205 or the hard disk 209 for each level.

  The CPU 203 acquires the text sentence data divided into the sentence element data at the “˜phrase” level from the text sentence data that has been parsed (step S505).

  For example, the text sentence data 707 divided into level 2 of the text sentence data 705 in FIG. 7, specifically, the syntax analysis data (level 2) shown in C of FIG. 7a is acquired.

  The CPU 203 performs a process for determining a main phrase based on the acquired syntax analysis data (level 2) (step S507). Here, the main phrase is sentence element data including the subject, and it is determined whether or not it is a main phrase by the particle at the end of the sentence element data. For example, if the end of the sentence element data is any one of “to”, “to is”, and “to”, the sentence element data is determined to be a main phrase.

  FIG. 9 shows an example in which a main phrase is determined. In the text sentence data 901 shown in FIG. 9, the particle at the end of the sentence element data “Operating profit is (subordinate phrase)” is “ha” from the text sentence data 703 immediately after the morphological analysis in FIG. As a result, it is possible to determine that the sentence element data “operating profit is (subordinate phrase)” ends with “˜ha” and is a main phrase.

  If there are subordinate clauses of the combination modification before the subordinate phrase determined to be the main phrase here, the subordinate phrase string including them may be determined as the main phrase, and such a subordination phrase may be determined. Dependent phrases need not be included in the main phrase. For example, if “(Subordinate clause) of the current fiscal year” precedes “Operating income (Subordinate clause)”, it is determined that “Operating income of the current fiscal year” (Subordinate clause column) is the main phrase. Alternatively, only “operating profit (subordinate phrase)” may be determined as the main phrase.

  The CPU 203 performs processing for determining a predicate in the acquired text sentence data 707 (step S509). Here, the predicate is statement element data including a predicate. The CPU 203 determines whether the predicate is based on the part of speech at the end of the sentence element data.

  For example, when the part of speech at the end of the sentence element data is “noun”, “adjective”, or “verb”, the sentence element data is determined to be a predicate or a component of the predicate. In particular, in the case of “auxiliary verbs” and the above “nouns” and “adjectives” whose meaning is low even if they are “verbs”, the predicates are connected with the immediately preceding sentence element data. Shall be composed.

  In the present embodiment, when the end of sentence element data is any one of “to”, “becomes”, “was to”, and “becomes to”, The part of speech is determined to be an “auxiliary verb”. Note that the determination as to whether the part of speech at the end of the sentence element data is an “auxiliary verb” may be made based on morphological analysis. In this case, it is only necessary to analyze whether the verb is a main verb (independent) or an auxiliary verb (non-independent) by morphological analysis.

  FIG. 9 shows an example of determining a predicate. In the text sentence data 903 shown in FIG. 9, the sentence element data “Nata (verb phrase)” is a “verb” based on the result of the morphological analysis shown in FIG. Thereby, the sentence element data “Nana (verb phrase)” can be determined to be a predicate or a component of the predicate. Further, since the sentence element data “Nana (verb phrase)” is classified as an “auxiliary verb”, it is concatenated with the immediately preceding sentence element data to form a predicate.

  As a result, the sentence element data “Nata (verb phrase)” and the immediately preceding sentence element data “32776 million yen and (subordinate phrase)” are concatenated to become “32976 million yen” as a predicate. Is done.

  That is, as shown in the text sentence data 905 of FIG. 9, the text sentence data is “Operating profit was (main phrase) / previous year and (subordinate phrase) / same level (subordinate phrase) / 32976 million yen. (Predicate) "is divided into each sentence element data. In practice, the CPU 203 records analysis result data 910 as shown in FIG. 9 a in a predetermined area of the memory 205 or the hard disk 209.

1-1-3-3. Dependency abstraction model extraction processing
In the text sentence analysis process (FIG. 3, step S309), when the text sentence data is divided into predetermined types of sentence element data, the CPU 203 further performs a dependency abstract model extraction process (step S313).

  In the flowchart of the text sentence analysis process shown in FIG. 6, the CPU 203 acquires sentence element data to which the numerical information data belongs (step S601). For example, since the numerical information data extracted in the numerical information data extraction process is “32776 million yen”, the text element data 905 shown in FIG. Is obtained (predicate).

  If the CPU 203 determines that the sentence element data including the numerical information data is a main phrase (step S603, YES), the CPU 203 proceeds to step S609 described later, and if it is determined that the sentence element data is not a main phrase (step S603, NO), further It is determined whether or not it is a predicate (step S605).

  If the CPU 203 determines that the sentence element data to which the numerical information data belongs is a predicate (step S605, YES), the CPU 203 extracts the noun in the main phrase of the original text sentence data as dependency information data of the numerical information data. If it is determined that it is not a predicate (step S613) (step S605, NO), it is further determined that the sentence element data to which the numerical information data belongs is a subordinate phrase, and whether or not the sentence element data is a combination modification. Is determined (step S607).

  If the CPU 203 determines that the sentence element data including the numerical information data is the combination modification (YES in step S607), the noun in the sentence element data immediately after the sentence element data is changed to the dependency information of the numerical information data. If it is extracted as data (step S615) and it is determined that the modification is not a combination modification (NO in step S607), sentence element data other than the sentence element data to which the numerical information data belongs are extracted as dependency information data of the numerical information data ( Step S609).

  In the following description, the sentence element data to which the numerical information data belongs are divided into “predicate”, “main phrase”, and “subordinate phrase”.

1-1-3-3-1. In the case of “predicate” For example, since the sentence element data “numerical information data that has become 32767 million yen (predicate)” is determined to be a predicate, the text sentence data 905 shown in FIG. The dependency information data is extracted on the basis of “operating profit (main phrase)”. That is, “business profit (noun)” is extracted as dependency information data with reference to the text sentence data 703 shown in FIG.

  Accordingly, the numerical information data “32776 million yen” and the dependency information data “operating profit” become the abstract model data.

1-1-3-3-3. In the case of “main phrase” For example, let us consider a case where “Ieyasu Tokugawa was entrusted to the great conqueror in 1603” as text data.

  According to steps S301 to S309 shown in FIG. 3, the above text sentence data is “in the year 1603 (main phrase) / Ieyasu Tokugawa (main phrase) / conquered general (subordinate phrase) / assigned (predicate) ) "Sentence element data. In this case, “year” is used as a unit character, and “1603” is extracted as numerical information data.

  In this case, since the sentence element data to which the numerical information data belongs is “1603 (prominent phrase)”, the CPU 203 relates sentence element data other than “1603 (prominent phrase)” from the text sentence data. Extracted as received information data. That is, “Ieyasu Tokugawa (main phrase) / subordinate phrase (subordinate phrase) / assigned (predicate)” is extracted as dependency information data.

  As a result, the numerical information data “1603” and the dependency information data “Ieyasu Tokugawa has been appointed as a great general of conquest” become the abstract model data.

1-1-3-3-3. In the case of “subordinate phrase” When the sentence element data including the numerical information data is a subordinate phrase, the CPU 203 determines whether or not the sentence element data is a combination modification (step S607). That is, it is determined whether or not a statement is included in the subsequent sentence element data.

  For example, if the end of sentence element data including numerical information data is “to” (step S607, YES), it is determined that the word to be modified is a body word, and the sentence element data is determined to be a combined modification. To do.

  For example, consider a case where “I bought a note of 100 yen yesterday” is input as text data. In accordance with steps S301 to S309 shown in FIG. 3, the text data is “I am (main phrase) / yesterday (subordinate phrase) / 100 yen (subordinate phrase) / note (subordinate phrase) / buy (description). Part) "sentence element data. Note that “yen” is used as the unit character in this case, and “100 yen” is extracted as numerical information data.

  In this case, since the sentence element data to which the numerical information data belongs is “100 yen (subordinate phrase)” and the end of the sentence element data is “to”, the CPU 203 determines that the subordinate phrase is a combination modification. judge. That is, it is determined that the word modified by the subordinate phrase is a body word, and the noun of the sentence element data immediately after the sentence element data in the text sentence data is extracted as dependency information data.

  For example, “note (noun)”, which is sentence element data immediately after, is extracted as dependency information data from “note (subordinate phrase)”.

  As a result, the numerical information data “100 yen” and the dependency information data “note” become the abstract model data.

  On the other hand, when the CPU 203 determines that the modification is not a combination modification (NO in step S607), that is, it is determined that the word to be modified is a predicate, and sentence element data other than the sentence element data to which the numerical information data belongs is determined as dependency information data. (Step S609).

  For example, consider a case in which “Yori Yoromo opened the Kamakura Shogunate in 1192” is input as text data.

  According to steps S301 to S309 shown in FIG. 3, the text sentence data is “Gen Yoromo (primary phrase) / 1192 (subordinate phrase) / Kamakura shogunate (subordinate phrase) / opened (predicate)”. Divided into sentence element data. In this case, “year” is used as a unit character, and “1192” is extracted as numerical information data.

  In this case, since the sentence element data to which the numerical information data belongs is “in 1192 (subordinate phrase)” and the end of the sentence element data is other than “to”, the CPU 203 determines that the subordinate phrase is not a combination modification. judge. That is, it is determined that the word modified by the subordinate phrase is a predicate, and sentence element data other than the sentence element data to which the numerical information data belongs in the text sentence data is extracted as dependency information data.

  For example, sentence element data other than the sentence element data “1192 (subordinate phrase)” is “source Yoromo (main phrase) / kamakura shogunate (subordinate phrase) / opened (predicate)” as dependency information data Extracted.

  As a result, the numerical information data “1192” and the dependency information data “Gen Yoromo opened the Kamakura Shogunate” become the abstract model data.

  The CPU 203 repeats the processing steps S601 to S609 as many times as the numerical information data extracted by the numerical information data extraction process (FIG. 4) (step S611, YES). If there is no other numerical information data, the process ends. (Step S611, NO).

1-1-3-4. Conclusion When the dependency abstract model extraction process is completed, the CPU 203 records the abstract model data in the abstract model DB 2097 (FIG. 3, step S315). For example, when the text sentence data “Operating profit was 32,763 million yen, the same level as the previous year” shown in FIG. 7, the numerical information data “32776 million yen is stored in the abstract model DB 2097. ”, The dependency information data“ operating profit ”is recorded.

  In this embodiment, the abstract model data extracted by the abstract model generation device is recorded in the abstract model DB 2097. However, the present invention is not limited to this, and the abstract model data is stored in other models. You may make it hand over to an application.

  Further, in order to perform processing for extracting numerical information data and dependency information data from text sentence data, it may be used by being incorporated in another application. Furthermore, the abstract model data may be displayed on the display 201 of the abstract model generation device for presentation to the user.

  The CPU 203 repeats the above processing (FIG. 3, steps S303 to S315 (step S317, NO)) as many times as the number of text sentence data cut out from the text sentence. If there is no text sentence data to be processed, the abstract model extraction is performed. The process ends (step S317, YES).

  As described above, according to the present invention, nouns having a dependency relationship can be accurately extracted even when nouns corresponding to numerical information data are not in the vicinity. In addition, information extracted as information corresponding to numerical information data is not limited to a specific noun, and a noun or sentence having a dependency relationship can be accurately extracted.

1-2. 1-1 embodiment In the said embodiment, it demonstrated on the assumption that the text-text data input is a single sentence. In the present embodiment, a case where the text sentence data is a compound sentence or a double sentence including two sentences will be described.

  Note that a compound sentence is a sentence in which a subject / predicate relationship is established, and a subject / predicate relationship is further found in its constituent parts. A heavy sentence is a sentence formed by combining two or more independent sentences with equal qualifications.

1-2-1. Functional Block Diagram FIG. 10 is a functional block diagram of the abstract model generation device according to this embodiment. In this figure, the abstract model generation apparatus according to the present invention includes a cutout means 101, numerical information data extraction means 103, text sentence analysis means 105, extraction target sentence determination means 106, dependency information extraction means 107, abstract model. Output means 109 is provided.

  The extraction target sentence determination unit 106 determines a sentence from which numerical information data is to be extracted from the text sentence data divided into sentence element data by the text sentence analysis unit, and supplies the sentence to the dependency information extraction unit 107. For example, in the case of a compound sentence, after dividing into a main sentence and a sub sentence, it is determined whether or not each sentence is an extraction target.

  Note that the cutout unit 101, the numerical information data extraction unit 103, the text sentence analysis unit 105, the dependency information extraction unit 107, and the abstract model output unit 109 are the same as those in the first embodiment.

1-2-2. Hardware configuration The hardware configuration is the same as in the first embodiment.

1-2-3. Flowchart Processing based on the information extraction program 2091 will be described using the flowchart of FIG. In the following, an example will be described in which a text sentence including text sentence data “This month's food expense is 45400 yen because my father bought 7000 yen of wine” is input to the cutting means 101.

  In the abstract model extraction process flowchart shown in FIG. 11, the processes of steps S301 to S309 performed by the CPU 203 are basically the same as those in the first embodiment.

  For example, when the text sentence data is read, the first sentence, “This month's food expense is 45400 yen because my father bought 7000 yen of wine” is cut out as text sentence data.

  The CPU 203 performs numerical information data extraction processing based on the extracted text sentence data (step S305).

1-2-3-1. Numerical Information Data Extraction Processing For example, in this embodiment, “Yen” is recorded as a unit character in the numerical value unit master, and the CPU 203 performs “7000 yen” and “45400” by the numerical information data extraction processing shown in FIG. Each “circle” is extracted as numerical information data.

1-2-3-2. Text sentence analysis processing If numerical information data is extracted in the numerical information data extraction process (FIG. 11, step S305) (YES in step S307), the CPU 203 performs text sentence analysis processing (step S309). As shown in the first embodiment, the CPU 203 divides the text sentence data into predetermined sentence element data according to the flowchart of the text sentence analysis process shown in FIG.

  FIG. 13 shows the text sentence data after the text sentence analysis processing is performed. As shown in the text sentence data 1301, the text sentence data is: “This month's food cost is (main phrase) / father's (main phrase) / 7000 yen (subordinate phrase) / wine (subordinate phrase) / buy. (Predicate) / 45400 Yen (Predicate) ”is divided into each sentence element data.

1-2-3-3. Extraction target sentence decision processing
When the text sentence analysis process is completed, the CPU 203 performs an extraction target sentence determination process (step S310). FIG. 12 shows a flowchart of the extraction target sentence determination process.

  The CPU 203 determines the type of sentence based on the text sentence data subjected to the syntax analysis process. That is, the CPU 203 determines whether the input text sentence data is a compound sentence, a double sentence, or a single sentence based on the number and arrangement of combinations of the main phrase and predicate (FIG. 12, step S1201).

  If the combination of the main phrase and the predicate is one, the CPU 203 determines that the text sentence data is a single sentence (step S1201, simple sentence). If the combination of the main phrase and predicate is two and the arrangement is “main phrase-main phrase-predicate-predicate”, it is determined that the text sentence data is a compound sentence (step S1201, compound sentence). Further, if there are two combinations of the main phrase and the predicate and the arrangement is “main phrase-predicate-main phrase-predicate”, it is determined that the text sentence data is a heavy sentence (step S1201, heavy sentence).

  If the CPU 203 determines that the input text sentence data is a compound sentence, the CPU 203 divides the text sentence data into a main sentence and a sub sentence (step S1203). For example, in the case of the text sentence data 1301 shown in FIG. 13, it is determined that the sentence structure is “main case phrase-main case phrase-predicate-predicate”, so that it is a compound sentence. Text sentence data 1303 “This month's food cost was (main phrase) / 45400 yen (predicate)” and text sentence data 1305 as a sub sentence (father was (main phrase) / 7000 yen (subordinate phrase) / Wine (subordinate phrase) / Bought (predicate) ".

  On the other hand, if the CPU 203 determines that the input text sentence data is a heavy sentence, the CPU 203 divides the text sentence data into a first half sentence and a second half sentence (step S1205). For example, in the case of the text sentence data 1401 shown in FIG. 14, it is determined that the sentence structure is “major phrase-predicate-prominent phrase-predicate”, so that it is a heavy sentence. Text sentence data 1403 “Food expenses for this month are (primary phrases) / 45400 yen (predicate)” and text sentence data 1305 as sub-sentence is “Father is (primary phrases) / 7000 yen (subordinate phrase) ) / Wine (subordinate clause) / bought (predicate) ”.

CPU203 acquires the divided simple sentence or input sentence (step S1207), determines whether or not contain numerical information data to the simple sentence (step S1209).

If the acquired single sentence includes numerical information data (step S1209, YES), the CPU 203 determines the single sentence as an extraction target sentence and stores it in the memory 205 (step S1211). If numerical information data is not included, step S1211 is skipped.

  If there is another simple sentence (YES in step S1213), the CPU 203 returns to step S1207, acquires the next simple sentence, and performs the same processing as in steps S1209 to S1213. If there is no single sentence to be processed, the process ends (NO in step S1213).

  For example, when the text sentence data 1303 as the main sentence in FIG. 13 is acquired as a single sentence, the numerical information data “45400 yen” is included, so the text sentence data 1303 is determined as an extraction target sentence. Further, the text sentence data 1305 as the sub sentence is acquired as a single sentence, and the numerical information data “7000 yen” is included, so that the text sentence data 1305 is determined as the extraction target sentence. In the case of the heavy text text data 1401 shown in FIG.

1-2-3-4. Dependency abstraction model extraction processing
In the extraction target sentence determination process (FIG. 11, step S310), when the input text sentence data is divided into predetermined types of sentence element data and divided into simple sentences, the CPU 203 further performs a dependency abstract model extraction process (step S313) is performed. Note that the dependency abstraction model extraction processing is the same as the flowchart of FIG. 6 shown in the first embodiment.

  The CPU 203 repeats the processing steps S313 to S315 as many as the number of simple sentences determined by the extraction target sentence determination process (FIG. 12) (step S316, YES), and if there is no other single sentence, the process ends (step S316). , NO).

  For example, numerical information data “45400 yen” and dependency information data “food expenses” are extracted from the text sentence data 1303 shown in FIG. 13 and recorded in the abstract model DB 2097. Similarly, numerical information data “7000 yen” and dependency information data “wine” are extracted from the text sentence data 1305 shown in FIG. 13 and recorded in the abstract model DB 2097.

2-3-5. Summary As described above, according to the present invention, it is possible to accurately extract a noun or sentence that is dependent on numerical information data regardless of whether the text sentence data is a single sentence, a compound sentence, or a heavy sentence. it can.

  In the above embodiment, the text sentence data has been described on the premise of a compound sentence or a duplicate sentence in which the combination of the main phrase and the predicate is two, but a combination of two or more may be used. In this case, the arrangement of the main phrase and the predicate may be pre-patterned, and it may be configured to determine whether it is a compound sentence or a heavy sentence based on which pattern it belongs to, and to divide it into simple sentences. .

3. 1-2 embodiment In the above embodiment, the input text sentence data is divided into sentence element data, and based on the type of sentence element data to which the numerical information data belongs, the numerical information data and the numerical information data It is configured to extract only dependency information data that is in a dependency relationship.

  In the present embodiment, a case will be described in which numerical information data and dependency information data are extracted as main information based on predetermined sentence element data, and additional information is extracted based on other sentence element data.

1-3-1. Functional Block Diagram FIG. 15 is a functional block diagram of the abstract model generation device according to this embodiment. In this figure, the abstract model generation apparatus according to the present invention includes a cutout means 101, numerical information data extraction means 103, text sentence analysis means 105, extraction target sentence determination means 106, dependency information extraction means 107, and additional information extraction. Means 108 and abstract model output means 109 are provided.

  The additional information extraction unit 108 extracts sentence element data other than the sentence element data including the numerical information data and the dependency information data based on the text sentence data divided into the sentence element data, and the numerical information data and the dependency information are extracted. It is given to the abstract model output means 109 as additional information related to information data.

  Note that the cutout unit 101, the numerical information data extraction unit 103, the text sentence analysis unit 105, the dependency information extraction unit 107, and the abstract model output unit 109 are the same as those in the first embodiment.

1-3-2. Hardware configuration The hardware configuration is the same as in the first embodiment.

1-3-3. Flowchart Processing based on the information extraction program 2091 will be described using the flowchart of FIG. In the following, as in the first embodiment, an example in which a text sentence including the text sentence data “Operating income was 32,976 million yen, the same level as the previous year” is input to the clipping unit 101 is described as an example. Will be described.

  In the abstract model extraction process flowchart shown in FIG. 11, the processes of steps S301 to S315 performed by the CPU 203 are basically the same as those in the first embodiment.

  For example, when the text sentence data is read, the first sentence “Operating profit was 32,976 million yen, the same level as the previous year” is extracted as the text sentence data.

  The CPU 203 performs numerical information data extraction processing based on the extracted text sentence data (step S305).

  In the numerical information data extraction process (FIG. 16, step S305), if numerical information data is extracted (step S307, YES), the CPU 203 performs a text sentence analysis process (step S309). As shown in the first embodiment, the CPU 203 divides the text sentence data into predetermined sentence element data according to the flowchart of the text sentence analysis process shown in FIG.

  FIG. 18 shows the text sentence data after the text sentence analysis processing is performed. As shown in the text sentence data 1801, the text sentence data is “Operating profit was (main phrase) / previous year (subordinate phrase) / same level (subordinate phrase) / 32776 million yen (predicate)” Is divided into each sentence element data.

  In the text sentence analysis process (FIG. 16, step S309), when the input text sentence data is divided into predetermined types of sentence element data, the CPU 203 further performs a dependency abstract model extraction process (step S313). Note that the dependency abstraction model extraction processing is the same as the flowchart of FIG. 6 shown in the first embodiment.

  Therefore, as in the first embodiment, the numerical information data “32776 million yen” and the dependency information data “operating profit” are the abstract model data.

1-3-3-1. Additional abstraction model extraction processing
When the dependency abstract model extraction process is completed, the CPU 203 performs an additional abstract model extraction process (step S331). FIG. 17 shows a flowchart of the additional abstract model extraction process.

  The CPU 203 acquires the numerical information data extracted in the numerical information data extraction processing and the sentence element data to which the dependency information data extracted in the dependency abstract model extraction processing belongs (FIG. 17, step S1701).

  Next, the CPU 203 reads the text sentence data 1801 divided into the sentence element data in the text sentence analysis process, and adds the sentence element data other than the sentence element data to which the numerical information data and the dependency information data acquired above belong to the additional information. (Step S1703).

  For example, in the text sentence data 1801 in FIG. 18, the sentence element data to which the numerical information data belongs is “32776 million yen (predicate)”, and the sentence element data to which the dependency information data belongs is “operating profit is (major Phrase). Therefore, “same year (subordinate phrase)” and “same level (subordinate phrase)” are extracted as sentence element data other than the sentence element data to which the numerical information data and the dependency information data belong.

  The CPU 203 repeats the processing steps S1701 to S1703 as many times as the numerical information data extracted by the numerical information data extraction process (FIG. 4) (step S1705, YES), and terminates the process if there is no other numerical information data. (Step S1705, NO).

  When the additional abstract model extraction processing is completed, the CPU 203 records the abstract model data in the abstract model DB 2097 (FIG. 16, step S333).

  For example, as shown by 1803 in FIG. 18, in the abstract model DB 2097, in addition to the numerical information data “32776 million yen” and the dependency information data “operating income”, the additional information “same level as the previous year”. Is recorded.

  The CPU 203 repeats the above processing (FIG. 11, steps S303 to S333 (step S317, NO)) for the number of text sentence data cut out from the text sentence, and when there is no text sentence data to be processed, the abstract model extraction is performed. The process ends (step S317, YES).

1-3-4. Summary As described above, according to the present invention, in addition to numerical information data and dependency information data that are main information of text sentence data, additional information related to the main information can be extracted.

  As a result, not only the dependency information directly related to the numerical information data but also the additional information indirectly related to the numerical information data or the dependency information data can be extracted.

  For example, the additional information “same level as last year” regarding the numerical information data “32776 million yen” and the dependency information data “operating profit” is related to the situation where the “operating profit” is “32976 million yen”. -It can be used as additional information as a situation.

  In the above embodiment, the case where the text sentence data is a single sentence has been described. However, the text sentence data may include a compound sentence or a duplicate sentence. In this case, as shown in the 1-1 embodiment, text sentence analysis processing (step S309). And an extraction target sentence determination process may be inserted between the dependency abstract model extraction process (step S313).

1-4. Other Embodiments In the above-described embodiment, the division processing is performed after the numerical information data extraction processing. However, the numerical information data extraction processing may be performed during the text sentence analysis processing. Specifically, after the text sentence data is divided into sentence element data in the text sentence analysis process, it is determined whether or not numerical information data is included for each sentence element data, and if included, the syntax of the text sentence data What is necessary is just to comprise so that an analysis may be performed.

  In the above embodiment, when the numerical information data is extracted, it is configured to extract, as the numerical information data, numbers consecutive in front of a predetermined unit character existing in the text sentence data. Alternatively, numbers that are continuous in the back may be extracted as numerical information data. For example, this is the case when a currency symbol such as “¥” or “$” is used as a unit character.

  In the above embodiment, YACC, which is an LR grammar parser for ascending syntax analysis, is exemplified in the syntax analysis processing, but other types of parsers may be used. For example, an LL grammar parser for descending parsing corresponds to this.

  In the above embodiment, a text sentence written in Japanese has been described. However, if numerical information data can be extracted and can be divided into sentence element data in morphological analysis processing and syntax analysis processing, It may be a written text. For example, in the case of English, numerical information data may be extracted based on English numeric units and configured using an English morphological analysis program and a syntax analysis program.

2. Second Embodiment [Link Setting Device]
An apparatus for setting a link to an abstraction processing element such as a text sentence including numerical information data using the abstraction model generation apparatus of the first embodiment will be described below with reference to FIGS. To do.

2-1. Functional Block Diagram FIG. 19 is a functional block diagram of the link setting device according to the present embodiment. In this figure, a link setting device according to the present invention includes a document data input means 1201, an element division means 1203, an abstract model data extraction means 1204, an abstract model recording means 1211, an abstract model selection means 1213, and a link setting means 1215. It has. The abstract model data extraction unit 1204 includes an expression format determination unit 1205, an abstract model generation unit (text sentence) 1207, and an abstract model generation unit (table) 1209.

  Document data is input from the document data input unit 1201 and is divided for each abstraction processing element by the element dividing unit. The divided abstraction processing elements are given to the abstraction model data extraction unit 1204.

  The expression format discrimination means 1205 of the abstract model data extraction means 1204 determines the expression format of the acquired abstraction processing element. For example, it is determined whether the abstraction processing element is a text sentence (sentence) or a table (table). Thus, the element is given to the abstract model generation means (text sentence) 1207 when the abstract processing element expression format is a text sentence, and the element is abstract when the element expression form is a table. This is given to a modelized model generating means (table) 1209. When the document data includes only a text sentence, the above-described expression format determination unit 1205 and the abstract model generation unit (table) 1209 are not necessary.

  The abstract model generation means (text sentence) 1207 has the same function as the abstract model generation apparatus (shown in FIG. 1) in the first embodiment. That is, other sentence elements having a dependency relationship with the numerical information data are extracted from the input text sentence, and these are provided to the abstract model recording means 1211 together with the position information data (specifically, the flowchart of FIG. 3 and the like). The processing shown in steps S301 to S317 of FIG. The abstract model generation means (table) 1209 performs processing for generating abstract model data based on a predetermined rule (described later) defined for the table data, and abstract model recording means 1211 together with the position information data. To give.

  The abstract model recording unit 1211 records the abstract model data given from the abstract model generation unit (text sentence) 1207 and the abstract model generation unit (table) 1209. The abstract model selection means 1213 searches the abstract model recording means 1211 recorded, selects all the same abstract model data, and provides each position information data to the link setting means 1215. The link setting unit 1215 sets a link (for example, a hyperlink) for the abstraction processing elements having the same abstraction model data based on the position information data.

2-2. Hardware Configuration FIG. 20 shows an example of a hardware configuration in which the link setting device shown in FIG. 19 is realized using a CPU. 20 includes a display 201, a CPU 203, a memory 205, a keyboard / mouse 207, a hard disk 209, a CD-ROM drive 211, and a communication circuit 215.

  A link setting program 221 is recorded on the hard disk 209. The link setting program 221 includes an abstract model data generation script 223 for performing abstract model data generation processing (shown in FIG. 3) for a text sentence. Further, the link setting program 221 also includes an expression format correspondence table 225 for determining the expression format (text sentence, table, etc.) of the abstraction processing element, and an abstract model DB 227 for recording the generated abstract model data. I have.

  These programs and the like are installed by reading data recorded on the CD-ROM 212 via the CD-ROM drive 211. The installation may be performed using data downloaded from the Internet 216 or the like using the communication circuit 215.

2-3. Specific Example of Processing for Setting Link Based on Abstraction Model Data Processing for generating abstract model data from a text sentence including numerical information data and setting a link using the flowchart shown in FIG. explain. The following processing is an example in the case where a link is set for a text sentence or a table in an XML format file (XML file).

  First, document data to be linked is input (step S2001). For example, the securities report file data as shown in FIG. 22A is read from the CD-ROM drive via the CD-ROM 212 and recorded in the memory 205. Note that the data of the securities report file shown in FIG. 22A is described in XML as shown in FIG. 22B. For example, the text sentence a1 of the file shown in FIG. 22A corresponds to the α part in FIG. 22B.

  The input document data is divided into each abstract processing element such as a text sentence and a table (step S2003). Specifically, a portion surrounded by a predetermined start tag and end tag is extracted from the XML data shown in FIG. 22B. For example, in the BusinessResults and Item fields of the representation format correspondence table 225 shown in FIG. 23, a symbol “◯” indicating that the subject is an abstraction process is entered in the column of the representation format (text sentence or table). Therefore, the α to γ portions (shown in FIG. 22B) surrounded by these start tags and end tags are extracted. In the expression format correspondence table 225 of FIG. 23, “-” indicates that the abstraction processing element is not extracted.

  Next, based on the expression format correspondence table 225 shown in FIG. 23, it is determined whether the content of the extracted part is a text sentence or a table (step S2005). For example, from the expression format correspondence table 225 of FIG. 23, it is determined that the content A1 of the α portion that is the BusinessResults tag of FIG. 22 is a text sentence, and the content B1 to B3 and the content of the β portion and the γ portion that are the Item tag C1 to C3 are determined to be tables.

  If it is determined that it is a text sentence or a table, as described in (i) and (ii) below, abstract model data generation processing is performed in a different manner depending on the expression format (steps S2007 and S2009).

(I) When it is determined that the expression format is a text sentence, the abstract model data generation script 223 performs the abstract model data generation process for the text sentence described in the first embodiment, and “numerical information data = "Dependency information" is extracted as abstract model data (step S2007). Specifically, the abstract model data generation process is a process shown in steps S301 to S317 in the flowchart of FIG. 3, and sentence elements having a dependency relationship with the numerical information data are extracted by this process. For example, “sales = 8,191,752 million yen” is extracted from the content A in FIG. 22B as abstract model data. Further, along with the abstract model data, the position ID (for example, “br2003” A0 shown in FIG. 22) is given to the abstract model DB 227.

(Ii) When it is determined that the expression format is a table, abstract model data is generated based on a rule predetermined for the table (step S2009). In this embodiment, it is stipulated that data satisfying “numerical information data = dependency information” is generated as abstraction model data by combining contents belonging to each part determined as a table. For example, “8,191,752 million yen = sales” and “100 = sales” are extracted as abstract model data from the contents B1 to B3 of the β portion shown in FIG. 22B, and the contents C1 to C1 of the γ portion are extracted. From C3, abstract model data “152,967 million yen = operating profit” and “1.9 = operating profit” are extracted. Further, the position ID (for example, “sales_2003” B0 for the β portion and “ifo_2003” C0 for the γ portion) is given to the abstract model DB 227 together with the abstract model data.

  In the abstract model DB 227, abstract model data and its position information data are recorded (step S2011). FIG. 24 shows a specific example of data recorded in the abstract model DB 227.

  Next, the link setting program 221 searches the abstract model DB 227 to select all of the same abstract model data, and outputs the ID of each position information data to the link setting program 221 (step S2013). For example, IDs “br2003” and “ifo_2003” having the same abstract model data shown in FIG. 24 are given to the link setting program 221.

  The link setting program 221 sets a link based on the position ID of these abstraction model data (step S2015). For example, a link is set based on IDs “br2003” and “ifo_2003” between the α portion and the β portion shown in FIG. 22B using XLink. XLink is a link language that expresses links by using URIs.

  As a result, as shown in FIG. 25, when a part of the table (shaded portion X) is clicked while browsing a document, a related text sentence can be displayed in the pop-up window Y.

  In FIG. 25, when a text sentence is displayed in the pop-up window Y (that is, when the link destination is a text sentence), only “additional information” is extracted so that “current fiscal year” and “previous fiscal year” are displayed. Only the part of “increased by 2% compared to the year” can be displayed. The process of extracting the additional information can be executed by the process already described in FIG.

  As described above, using abstract model data generated from a document including numerical information data, it is possible to easily set a link for each part such as a text sentence in the document data.

  In the above embodiment, a link is set between the text sentence and the table, but a link may be set between the text sentence and the image. However, the image needs to have a data structure that can generate at least abstract model data based on attribute data such as headings. Moreover, you may make it set a link between text sentences.

  In the above embodiment, a case where there are two abstract models has been described. However, when three or more abstract models are found, a one-to-many link is set for each target data. For example, it is possible to set a function to display a plurality of link destinations on the screen by using XLink, or display a list of a plurality of link destinations and select a desired link destination in a general language such as JAVA. Write and set the link.

  In the above embodiment, the expression format attribute is not attached to the tag included in the XML document (file contents), but it is an expression format attribute such as a text sentence or a table or an abstract processing element. Attributes may be added and described in advance. Accordingly, even if there is no representation format correspondence table 225 shown in FIG. 23, it is possible to determine the representation format of the abstract model data and extract the abstract processing element.

3. Third Embodiment [Document Search Device]
In the above embodiment, the abstract model data generated based on the numerical information data is used for setting the link. However, the abstract model data may be used in the document search apparatus. An apparatus for searching a document file for an element such as a text sentence including numerical information data using the abstract model generation apparatus of the first embodiment will be described below with reference to FIGS. . In the present embodiment, a case where only text text is included in the document data will be described.

3-1. Functional Block Diagram FIG. 26 shows a functional block diagram of the document search apparatus according to this embodiment. In this figure, a document search apparatus according to the present invention includes a search element input means 2501, abstract model data extraction means (table) 2503, abstract model holding means 2505, abstract model recording means 2507, abstract model comparison means 2509. A search result output unit 2511 is provided. The abstract model recording means 2507 records each abstract model data generated in advance for the abstract processing elements in the entire document data.

  A search element (text sentence) is input to the search element input means for each abstract processing element, and is supplied to the abstract model generation means (text sentence) 2503. Upon receiving the abstraction processing element, the abstraction model generation means (text sentence) 2503 extracts the sentence element having a dependency relation with the numerical information data as abstraction model data, and stores it in the abstraction model holding means 2505 together with the position information data. give. The generation process of the abstract model data is performed by the process shown in the abstract model data generation flowchart of FIG.

  The abstract model comparison unit 2509 searches the abstract model recording unit 2507 to search the abstract model data generated from the entire document data in advance and the search target received from the abstract model generation unit (text sentence) 2503. By comparing with the abstraction model data, the position information data of the same abstraction model data is acquired and provided to the search result output means 2511.

  The search result output unit 2511 outputs an abstraction processing element having the same abstract model data as a search result based on the position information data.

3-2. Hardware Configuration FIG. 27 shows an example of a hardware configuration in which the document search apparatus shown in FIG. 26 is realized using a CPU. 27 includes a display 201, a CPU 203, a memory 205, a keyboard / mouse 207, a hard disk 209, a CD-ROM drive 211, and a communication circuit 215.

  A document search program 231 is recorded on the hard disk 209. The document search program 231 records an abstract model data generation script 223 for performing abstract model data generation processing (shown in FIG. 3) for a text sentence including numerical information data, and the generated abstract model data. An abstraction model DB 227 is provided. In this embodiment, since only the text sentence is included in the document data, the expression format correspondence table 225 (FIG. 20) for determining the expression format (text sentence, table, etc.) of the abstraction processing element is Although not provided, it may be provided when a plurality of expression formats exist in the text sentence data.

  These programs and the like are installed by reading data recorded on the CD-ROM 212 via the CD-ROM drive 211. The installation may be performed using data downloaded from the Internet 216 or the like using the communication circuit 215.

3-3. Specific Example of Document Search Processing A file search processing using abstract model data generated from document data will be described using the flowchart shown in FIG. In order to simplify the explanation, it is assumed that the XML format file (XML file) to be searched in the following example includes only a text sentence. However, the present invention can be implemented even when a table or the like is included. .

  Before performing the document search process, the abstract model DB 227 records all the abstract model data for the entire document data to be searched. The recording timing in the abstract model DB 227 may be real time or document verification. Specifically, abstraction model data for document data to be searched (for example, an XML file shown in FIG. 22) is generated and recorded in advance by the same processing as steps S301 to S317 shown in FIG.

  First, an input of a text sentence to be searched is received (step S2601). Further, in response to a search start input (such as pressing a search button), XML data of the input text sentence is output to the abstract model data generation script 223 (step S2603).

  The abstract model data generation script 223 extracts the abstract model data from the input text sentence, and outputs the abstract model data and its position ID to the memory 205 (step S2605). The memory 205 holds the abstract model data and its position ID (step S2607).

  The document search program 231 searches the abstract model DB 227 in which the abstract model data is extracted and recorded from the document data, and compares it with the abstract model data read from the memory 205, so that the abstract model data has the same data. ID is extracted (step S2609).

  The document search program displays a text sentence corresponding to the ID of the abstract model data as a search result (step S2611). For example, the highlighting is performed for the part specified as the search result.

  As described above, by generating abstract model data from a large amount of documents to be searched and comparing them with abstract model data generated from input text sentences, etc., a document search device that differs from simple matching Can be realized.

  In the above embodiment, the abstract model data is generated in advance for the entire document and recorded in the abstract model DB 227. However, it may be extracted each time during the search process.

  In the above-described embodiment, document data including only a text sentence is set as a search target. However, document data including a table or an image may be searched.

4). Fourth Embodiment [Document Editing Device with Verification Function]
Further, the abstract model data generated based on the numerical information data described above may be used as a verification device in the document editing program. An apparatus for verifying a document for an element such as a text sentence including numerical information data using the abstract model generation apparatus according to the first embodiment will be described below with reference to FIGS. 29 and 31. FIG. In the present embodiment, a case where only text text is included in the document data will be described.

4-1. Functional Block Diagram FIG. 29 is a functional block diagram of the document input verification apparatus according to this embodiment. In this figure, a document input verification apparatus according to the present invention includes verification element input means 2701, element extraction means 2702, abstract model generation means (text sentence) 2703, abstract model holding means 2705, abstract model recording means 2707, An abstract model discriminating unit 2709 and an input error output unit 2711 are provided. In the abstract model recording unit 2707, abstract model data is always generated and stored for document data input to the document editing apparatus.

  The verification element input unit 2701 receives input of document data (text sentence) from a document editing apparatus such as a word processor / editor, and provides the input to the element extraction unit 2703. The element extraction unit 2703 extracts an abstract processing element from the input document data. The abstract model generation means (text sentence) 2703 extracts the abstract processing elements as the dependency model data as the abstract model data, and supplies them to the abstract model holding means 2705 together with the position information data. The generation process of the abstract model data is performed by the process shown in the abstract model data generation flowchart of FIG.

  The abstract model discriminating means 2709 searches the abstract model recording means 2707 and extracts abstract model data having the same dependency information and different numerical information data from the abstract model data received from the abstract model holding means 2705. It is determined whether or not there is an abstraction processing element to have, and the determination result is given to the input error output means 2711.

  The input error output unit 2711 outputs input error information to the document editing apparatus based on the determination result acquired from the abstract model determination unit 2709.

4-2. Hardware Configuration FIG. 30 shows an example of a hardware configuration in which the document input verification apparatus shown in FIG. 29 is realized using a CPU. The document input verification apparatus shown in FIG. 30 includes a display 201, a CPU 203, a memory 205, a keyboard / mouse 207, a hard disk 209, a CD-ROM drive 211, and a communication circuit 215.

  A document editing application 241 and a document verification program 243 are recorded on the hard disk 209. The document verification program 231 has a function of receiving text sentence data input from the document verification application 241 and outputting the verification result. As shown in FIG. 30, the document verification program 231 includes an abstract model data generation script 223 for performing abstract model data generation processing (shown in FIG. 3) for a text sentence including numerical information data, and the generated abstract. An abstract model DB 227 for recording the model data is provided. In this embodiment, since only the text sentence is included in the document data, the expression format correspondence table 225 (FIG. 20) for determining the expression format (text sentence, table, etc.) of the abstraction processing element is Although not provided, it may be provided when a plurality of expression formats exist in the text sentence data.

  These programs and the like are installed by reading data recorded on the CD-ROM 212 via the CD-ROM drive 211. The installation may be performed using data downloaded from the Internet 216 or the like using the communication circuit 215.

4-3. Specific examples of document verification processing
A process for outputting an input error when numerical information data is erroneously input while editing a document will be described using the flowchart shown in FIG. In order to simplify the explanation, it is assumed that the XML format file (XML file) to be verified in the following example includes only a text sentence. However, the present invention can be implemented even when a table or the like is included. .

  Before performing the document verification process, the abstract model DB 227 records all abstract model data for the document data input to the document editing apparatus. The recording timing in the abstract model DB 227 may be real time or document verification. Specifically, abstraction model data for document data to be searched (for example, an XML file shown in FIG. 22) is generated and constantly updated by the same processing as steps S301 to S317 shown in FIG.

  First, information input from a document editing apparatus such as an XML document editing word processor is acquired (step S2801). Further, upon receiving a verification start input (such as pressing a search button), an abstract processing element is extracted from the input text sentence group and output to the abstract model data generation script 223 (step S2803).

  The abstract model data generation script 223 extracts the abstract model data from the received abstract processing element, and outputs the abstract model data and its position ID to the memory 205 (step S2805). The memory 205 holds the abstract model data and its position ID (step S2807).

  The abstract model data discriminating program 241 searches the abstract model DB 227 to extract abstract model data (similar abstraction) having the same dependency information as the abstract model data read from the memory 205 and different numerical information data. And the result is output to the document verification program 241 (step S2809).

  If there is similar abstraction model data, the document verification program 241 outputs input picture error information to the document editing device (step S2811). As a result, for example, a warning window that alerts the user to an incorrect input is displayed on the screen of a word processor or the like. When it is determined that there is similar abstract model data, if the position ID and numerical information data are acquired from the abstract model DB 227, it is possible to simultaneously display the corresponding part and the numerical value to be corrected.

  As described above, it is possible to easily recognize erroneous input of numerical information data by generating abstract model data from the contents of a document already input and using it for document verification.

  In the above embodiment, the abstract model data is extracted in advance for the entire document and recorded in the abstract model DB 227. However, the abstract model data may be extracted from the document already input when the search is executed.

  In the above embodiment, document data including only a text sentence is set as a verification target. However, document data including other elements such as a table or an image may be verified.

5). Other Embodiments In the above embodiment, even when there are a plurality of languages of text sentences included in the document file, abstract model data is generated by a processing method suitable for the language, and dependency information is stored in a dictionary or the like. It is possible to unify the language of the abstract model data by translating using.

  In the above-described embodiment, an XML format file is used as a document for performing an abstraction processing element. However, other format files such as an HTML format file may be used.

  In the above embodiment, the CPU 203 is used to realize processing by software. However, some or all of them may be realized by hardware such as a logic circuit. In addition, you may make it make an operating system (OS) process a part of program further.

It is a figure which shows the example of the functional block diagram of the information extraction apparatus in embodiment of this invention. It is an example which shows the example of the hardware block diagram of the abstraction model production | generation apparatus of this invention. It is a figure which shows the example of the flowchart in the "abstraction model extraction process" of this invention. It is a figure which shows the example of the flowchart in the "numerical information data extraction process" of this invention. It is a figure which shows the example of the flowchart in the "text sentence division | segmentation process" of this invention. It is a figure which shows the example of the flowchart in the "dependency information extraction process" of this invention. It is a figure which shows the example of "morphological analysis" or "syntax analysis" of this invention. It is a figure which shows the example of "morpheme analysis data" or "syntax analysis data" of this invention. It is a figure which shows the example of the "syntactic analysis data" of this invention. It is a figure which shows the example of the "grammar definition information" of this invention. It is a figure which shows the example which determines the "main phrase" or "predicate" of this invention. It is a figure which shows the example of the "analysis result data" of this invention. It is a figure which shows the example of the functional block diagram of the abstraction model production | generation apparatus in embodiment of this invention. It is a figure which shows the example of the flowchart in the "abstraction model extraction process" of this invention. It is a figure which shows the example of the flowchart in the "extraction target sentence determination process" of this invention. It is a figure which shows the example in the case of "dividing a compound sentence into a main sentence and a sub sentence" of this invention. It is a figure which shows the example in the case of "dividing a heavy sentence into the first half sentence and the latter half sentence" of this invention. It is a figure which shows the example of the functional block diagram of the abstraction model production | generation apparatus in embodiment of this invention. It is a figure which shows the example of the flowchart in the "abstraction model extraction process" of this invention. It is a figure which shows the example of the flowchart in the "additional information extraction process" of this invention. It is a figure which shows the example in the case of extracting additional information from the text sentence of this invention. It is a figure which shows the example of the functional block diagram of the link setting apparatus in embodiment of this invention. It is an example which shows the example of the hardware block diagram of the link setting apparatus of this invention. It is a figure which shows the example of the flowchart in the "link setting process" of this invention. It is a figure which shows the example of the XML file of this invention. It is a figure which shows the example of the expression format correspondence table | surface of this invention. It is a figure which shows the example of data of abstraction model DB of this invention. It is a figure which shows the example using the link set by this invention. It is a figure which shows the example of the functional block diagram of the document search apparatus in embodiment of this invention. It is an example which shows the example of the hardware block diagram of the document search device of this invention. It is a figure which shows the example of the flowchart in the "document search process" of this invention. It is a figure which shows the example of the functional block diagram of the document input verification apparatus in embodiment of this invention. It is an example which shows the example of the hardware block diagram of the document input verification apparatus of this invention. It is a figure which shows the example of the flowchart in the "document input verification process" of this invention.

Explanation of symbols

101 ... Extraction means 103 ... Numerical information data extraction means 105 ... Text sentence division means 106 ... Extraction target sentence determination means 107 ... Dependency information extraction means 108 ... ..Additional information extraction means 109... Abstraction model output means

Claims (16)

A link setting device for setting a link between target data including related sentence element data in document data,
The text data in the given document data is divided into a plurality of text element data, the type of each text element is determined and recorded in the recording unit, and the numerical text data is included in the given text text data And if included, an extraction / analysis means for specifying numerical information data and recording it in the recording unit, and
Referring to the recording unit, the sentence element data including the numerical information data is identified, and at least one sentence element data having a dependency relation with the numerical information data is determined based on the type of the sentence element data. Dependency information extracting means for extracting dependency information data from the sentence element data,
Abstract model recording that records the numerical information data extracted by the numerical information extraction means and the dependency information data extracted by the dependency information extraction means as abstract model data together with position data indicating the position in the document data in the recording unit Means,
Abstraction model selection means for selecting abstract model data having the same numerical information data and the same dependency information data among a plurality of abstract model data recorded in the recording unit;
Link setting means for setting a link between target data of the abstraction model data based on the position information data of the selected abstraction model data;
A link setting device comprising: A program for realizing, using a computer, a link setting device for setting a link between target data including related sentence element data in document data, the computer comprising the following means: :
A) The text sentence data in the given document data is divided into a plurality of sentence element data, the type of each sentence element is determined and recorded in the recording unit, and the numerical information data is included in the given text sentence data. Determining whether or not to include, if included, extraction / analysis means for identifying numerical information data and recording it in the recording unit;
B) Referring to the recording unit, the sentence element data including the numerical information data is specified, and at least one sentence element data having a dependency relationship with the numerical information data is determined based on the type of the sentence element data. Dependency information extraction means for extracting dependency information data from the determined sentence element data;
C) Abstraction in which the numerical information data extracted by the numerical information extraction means and the dependency information data extracted by the dependency information extraction means are recorded as abstract model data in the recording unit together with position data indicating the position in the document data. Model recording means,
D) Abstract model selection means for selecting abstract model data having the same numerical information data and the same dependency information data from among a plurality of abstract model data recorded in the recording unit;
E) Link setting means for setting a link between target data of the abstraction model data based on the position information data of the selected abstraction model data. In the link setting device or the link setting program according to claim 1 or 2,
The link setting unit sets a link so that only additional information is extracted from the text text of the link destination when the target data of the link destination is a text sentence. A document search device that searches document data based on target data that is a search element,
The text data in the given document data is divided into a plurality of text element data, the type of each text element is determined and recorded in the recording unit, and the numerical text data is included in the given text text data And if included, an extraction / analysis means for specifying numerical information data and recording it in the recording unit, and
Referring to the recording unit, the sentence element data including the numerical information data is identified, and at least one sentence element data having a dependency relation with the numerical information data is determined based on the type of the sentence element data. Dependency information extracting means for extracting dependency information data from the sentence element data,
Abstract model holding that holds the numerical information data extracted by the numerical information extraction means and the dependency information data extracted by the dependency information extraction means as abstract model data together with the position data indicating the position in the document data. means,
By retrieving the abstract model recording means that generated and recorded the abstract model data from the document data and comparing it with the abstract model data received from the abstract model holding means, the same numerical information data and the same dependency information data are obtained. Abstraction model comparison means for extracting only position information data of abstraction model data having,
Search result display means for displaying, as search results, target data having the same abstract model data, based on the positional information data of the extracted abstract model data;
A document retrieval apparatus comprising: A program for realizing, using a computer, a document search apparatus that searches for document data based on target data that is a search element, the computer having the following means:
A) The text sentence data in the given document data is divided into a plurality of sentence element data, the type of each sentence element is determined and recorded in the recording unit, and the numerical information data is included in the given text sentence data. Determining whether or not to include, if included, extraction / analysis means for identifying numerical information data and recording it in the recording unit;
B) Referring to the recording unit, the sentence element data including the numerical information data is specified, and at least one sentence element data having a dependency relationship with the numerical information data is determined based on the type of the sentence element data. Dependency information extraction means for extracting dependency information data from the determined sentence element data;
C) Abstraction in which numerical information data extracted by the numerical information extraction unit and dependency information data extracted by the dependency information extraction unit are stored as abstract model data in the recording unit together with position data indicating a position in the document data Model holding means,
D) The same numerical information data and the same dependency information are obtained by searching the abstract model recording means for generating and recording the abstract model data from the document data and comparing it with the abstract model data received from the abstract model holding means. Abstraction model comparison means for extracting only position information data of abstraction model data having data;
E) Search result display means for displaying target data having the same abstract model data as a search result based on the extracted position information data of the abstract model data. In the document search device or document search program according to any one of claims 4 and 5,
In the abstract model recording means, the abstract model data is generated and recorded before searching for the entire document data to be searched. A document input verification device for verifying document data input from a document editing device based on target data of a verification element,
The text data in the document data given from the document editing device is divided into a plurality of text element data, the type of each text element is determined and recorded in the recording unit, and numerical information is included in the given text text data. It is determined whether or not data is included, and if included, extraction / analysis means for identifying numerical information data and recording it in a recording unit,
Referring to the recording unit, the sentence element data including the numerical information data is identified, and at least one sentence element data having a dependency relation with the numerical information data is determined based on the type of the sentence element data. Dependency information extracting means for extracting dependency information data from the sentence element data,
Abstract model holding that holds the numerical information data extracted by the numerical information extraction means and the dependency information data extracted by the dependency information extraction means as abstract model data together with the position data indicating the position in the document data. means,
Abstraction model recording means that generates and records abstract model data from document data is retrieved, and the abstraction model data received from the abstract model holding means is the same as the dependency model, and the numerical information data is different. Abstraction model discrimination means for discriminating whether there is model data;
An input error output means for outputting input error information to the document editing device based on the determination result obtained from the abstract model determination means;
A document input verification device comprising: A program for realizing, using a computer, a document input verification device for verifying document data input from a document editing device based on target data of a verification element, and comprising the following means on a computer Program to:
A) The text sentence data in the document data given from the document editing device is divided into a plurality of sentence element data, the type of each sentence element is determined and recorded in the recording unit, and the given text sentence data is included in the given text sentence data It is determined whether or not numerical information data is included, and if included, extraction / analysis means for specifying the numerical information data and recording it in the recording unit,
B) Referring to the recording unit, the sentence element data including the numerical information data is specified, and at least one sentence element data having a dependency relationship with the numerical information data is determined based on the type of the sentence element data. Dependency information extraction means for extracting dependency information data from the determined sentence element data;
C) Abstraction in which numerical information data extracted by the numerical information extraction unit and dependency information data extracted by the dependency information extraction unit are stored as abstract model data in the recording unit together with position data indicating a position in the document data Model holding means,
D) The abstract model recording means that generates and records the abstract model data from the document data is searched, the dependency model information is the same as the abstract model data received from the abstract model holding means, and the numerical information data is different. Abstraction model discrimination means for discriminating whether or not there is abstraction model data;
E) Input error output means for outputting input error information to the document editing apparatus based on the determination result acquired from the abstract model determination means. In the document input verification device according to claim 7 or 8,
The abstract model data of the abstract model recording means is generated as needed based on the input from the verification element input means, and is always updated. The apparatus or program according to any one of claims 1 to 9, wherein the document data includes an element to be processed other than text sentence data,
Before extracting the abstraction model data, an expression format determination means for determining the expression format of the processing target element;
When the representation format of the target data is a text sentence, abstract model generation means for extracting, from the processing target element, a sentence element having a dependency relationship with the numerical information data as abstract model data together with the position information data ,
When the representation format of the target data is another element, an abstract model generation unit that generates abstract model data based on a predetermined rule and supplies them to the abstract model recording unit together with the position information data,
Characterized by comprising.   11. The apparatus or program according to claim 10, wherein the other target data is table data or image data.   12. The apparatus or program according to claim 1, wherein the document data is described in an XML format, and an attribute of an expression format is attached in advance to each tag included in the file content. What to do. The apparatus or program according to any one of claims 1 to 12, further comprising:
An abstract model translation means for unifying the abstract model data into the same language by referring to a translation dictionary when there are a plurality of languages of text sentences included in the document data is provided. A link setting method for setting a link between target data including related sentence element data in document data,
The text data in the given document data is divided into a plurality of text element data, the type of each text element is determined and recorded in the recording unit, and the numerical text data is included in the given text text data And if included, an extraction / analysis step for identifying numerical information data and recording it in the recording unit, and
Referring to the recording unit, the sentence element data including the numerical information data is identified, and at least one sentence element data having a dependency relation with the numerical information data is determined based on the type of the sentence element data. A dependency information extraction step of extracting dependency information data from the sentence element data,
Abstraction model data that is recorded in the recording unit together with position data indicating the position in the document data, using the numerical information data extracted in the numerical information extraction step and the dependency information data extracted in the dependency information extraction step as abstract model data Recording step;
An abstract model data selection step for selecting abstract model data having the same numerical information data and the same dependency information data from among a plurality of abstract model data recorded in the recording unit;
A link setting step for setting a link between target data of the abstraction model data based on the positional information data of the selected abstraction model data;
A link setting method characterized by comprising: A document search method for searching document data based on target data that is a search element,
The text data in the given document data is divided into a plurality of text element data, the type of each text element is determined and recorded in the recording unit, and the numerical text data is included in the given text text data And if included, an extraction / analysis step for identifying numerical information data and recording it in the recording unit, and
Referring to the recording unit, the sentence element data including the numerical information data is identified, and at least one sentence element data having a dependency relation with the numerical information data is determined based on the type of the sentence element data. A dependency information extraction step of extracting dependency information data from the sentence element data,
Abstraction model data that is stored in the recording unit together with position data indicating the position in the document data, using the numerical information data extracted in the numerical information extraction step and the dependency information data extracted in the dependency information extraction step as abstract model data Holding step,
The same numerical information data and the same dependency information data are obtained by searching the abstract model recording means that has generated and recorded the abstract model data from the document data and comparing it with the abstract model data held in the abstract model data holding step. Abstraction model data comparison step for extracting only position information data of abstraction model data having
A search result display step for displaying, as search results, target data having the same abstract model data, based on the positional information data of the extracted abstract model data;
A document search method characterized by comprising: A document input verification method for verifying document data input from a document editing device based on target data of a verification element,
The text data in the document data given from the document editing device is divided into a plurality of text element data, the type of each text element is determined and recorded in the recording unit, and numerical information is included in the given text text data. It is determined whether or not the data is included, and if included, the extraction / analysis step for identifying the numerical information data and recording it in the recording unit,
Referring to the recording unit, the sentence element data including the numerical information data is identified, and at least one sentence element data having a dependency relation with the numerical information data is determined based on the type of the sentence element data. A dependency information extraction step of extracting dependency information data from the sentence element data,
Abstraction model data that is stored in the recording unit together with position data indicating the position in the document data, using the numerical information data extracted in the numerical information extraction step and the dependency information data extracted in the dependency information extraction step as abstract model data Holding step,
Abstraction model recording means that generated and recorded abstraction model data from document data is retrieved, and the abstraction model data held in the abstraction model data holding step has the same dependency information and different numerical information data Abstraction model data determination step for determining whether there is abstraction model data;
An input error output step for outputting input error information to the document editing device based on the determination result obtained in the abstraction model data determination step;
A document input verification method characterized by comprising:

Images