JPH06215035A - Text retrieving device - Google Patents

Abstract

PURPOSE:To provide a text retrieving device for successively outputting texts from a text exactly expressing a retrieval request by applying priority by performing the syntax analysis of the extracted text in the case of text retrieval due to a keyword. CONSTITUTION:This device is provided with a keyword memory 3 for storing the inputted keyword, document memory 4 for storing the retrieved text, text extraction part 5 for retrieving the text based on the keyword, output part 6 for outputting the retrieved result, and syntax analysis decision part 8 for applying the priority to the retrieved text by performing the syntax analysis to the text retrieved by the text extraction part and deciding whether all the kinds of keywords exist in a single sentence or not, whether they are existent in a single clause or not and further whether the keyword has the same modification relation as a question sentence or not.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a question text to be input when searching a text (here, the text means a text composed of one sentence or a plurality of related plural sentences). The present invention relates to a text search device that extracts a keyword, searches for text containing the keyword, and outputs the text.

[0002]

2. Description of the Related Art In recent years, due to advances in computer technology and an increase in the capacity of document storage devices, databases having a large amount of data and sentences have become widespread, and there are many opportunities to handle such databases.

A large amount of data is stored in advance in a storage device such as a hard disk in these databases, and it is common to search the data based on keywords.

As a search based on a keyword in such a text search device, for example, Japanese Patent Laid-Open No. 2-2
As disclosed in Japanese Patent Laid-Open No. 458, the morpheme analysis of a given text is performed to extract a keyword, and the text stored in a storage device is searched based on the extracted keyword, and the keywords match each other. The method of outputting the text to be used is used.

An example of a search method using a conventional text search device will be described with reference to FIGS. 5 and 6.

FIG. 5 is a schematic block diagram of a conventional text search device.

In the figure, 1 is a question input unit for inputting a question sentence including a keyword, and 2 is a keyword extraction unit for performing morphological analysis on the question sentence input in the question input unit 1 to extract keywords. Reference numeral 3 is a keyword memory for storing the keywords extracted by the keyword extracting unit 2, 4 is a document memory in which texts to be searched are already stored, and 5 is based on the keywords stored in the keyword memory 3. A text extraction unit for extracting relevant text from the document memory 4, an output unit 6 for outputting the text extracted by the text extraction unit 5, and a control unit 7 for controlling the entire text search apparatus.

FIG. 6 shows the retrieval result from the output unit 6 in the conventional text retrieval apparatus shown in FIG. 5, and shows the output contents.

As an example of the operation of the text search apparatus shown in FIG. 5, the case where the question sentence "I want to know about Saturn's ring." Is input will be described.

First, when the question sentence "I want to know about Saturn's ring." Is input to the question input unit 1, the keyword extraction unit 2 performs morphological analysis in order to extract the keyword. The question text entered in the question input section 1 is “Saturn / no / ring / about / know / want /.” (Here /
Represents a delimiter by morphological analysis. ) Is decomposed into morphemes. Here, since the noun is used as the keyword, "Saturn" and "ring" are extracted as the noun. Therefore, the keyword extraction unit 2 extracts "Saturn" and "Ring" as keywords and stores them in the keyword memory 3. The content of the search request in this question sentence is about "Saturn's ring", and the one in which the keywords "Saturn" and "Ring" are in a dependent relationship is the most requested text.

When the keywords are stored in the keyword memory 3, the text extracting section 5 sequentially extracts the texts containing the keywords from the texts stored in the document memory 4.

Table 1 shows some of the various texts stored in the document memory 4.

[0013]

[Table 1]

The text extraction unit 5 sequentially searches the contents of the document memory 4 for text. The procedure is shown below.

The text extraction unit 5 determines whether or not the keyword "Saturn" is included in the text of the document memory 4 for all texts. First, the text "As for Saturn, the ring in it is said to be a symbol of Saturn." (Hereinafter referred to as Text 1) is extracted. As you continue to search, the text "Saturn's ring is one of the most beautiful in the solar system." (Hereinafter referred to as text 3) is searched. Continuing the search, "Saturn is surrounded by satellites, and the ring is deprived of its energy."
The text (hereinafter referred to as text 4) is extracted. Continuing on, "What does the ring do in Saturn?" (Text 6 below.
Say. ) Text is extracted. Furthermore, if you continue, the text "Saturn is one of the planets of the solar system." (Hereinafter referred to as text 7) is extracted.

As described above, the text 1, the text 3,
Text 4, text 6, and text 7 are extracted. Text 2 and text 5 are not extracted because the keyword "Saturn" is not included.

Subsequently, the text extraction unit 5 determines whether or not the extracted keyword 1, text 3, text 4, text 6, and text 7 include the next keyword "ring". Text 1, Text 3, Text 4, and Text 6 include the keyword “ring”, but Text 7 does not include the keyword “ring”. Therefore, the text 1, the text 3, the text 4, and the text 6 are extracted from the text extraction unit 5 and transmitted to the output unit 6, and the text 1, the text 3, the text are output at the output unit 6 as shown in FIG. 4 and the text 6 are output in this order.

As described above, conventionally, as a result of the keyword search, the texts including all kinds of keywords are output from the output unit 6 in the order of being stored in the document memory 4.

[0019]

With the above-mentioned configuration, although it is possible to extract the text including the keyword, the texts are output in the order stored in the document memory 4, so that the question text It was not always output from the text that accurately represents the content of the intended search request.

The present invention has been made in view of the above problems, and based on the result of parsing the extracted text,
(EN) A text search device that gives priority to the output order of texts and outputs in order from texts that accurately represent the contents of a search request.

[0021]

In order to solve the above problems, a text search device of the present invention is a morpheme from a question input section in which a question text including a keyword is input, and a question text input in the question input section. A keyword extraction unit that extracts keywords by analysis, a keyword memory that stores the keywords extracted by the keyword extraction unit, a document memory that stores a large amount of data such as text to be searched, and a keyword memory. A text extraction unit that extracts the corresponding text from the document memory based on the keyword, an output unit that outputs the text extracted by the text extraction unit, and a control unit that controls the entire
In addition to performing a syntactic analysis of the extracted text, a syntactic analysis determination unit that determines the priority order for outputting the text from the output unit based on the syntactic information of the keyword extracted by the keyword extraction unit Is.

[0022]

According to the present invention, with the above-described configuration, the extracted text is parsed using the keyword of the question sentence, and the extracted text is given a priority order to accurately represent the search request. It is possible to provide a text search device that automatically prioritizes and outputs text.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION A text search device according to an embodiment of the present invention will be described below with reference to FIGS.

FIG. 1 is a schematic configuration diagram of a text search device according to the present invention, FIGS. 2 and 3 are flowcharts showing a flow of processing of a parsing judgment unit 8 which is a feature of the present invention, and FIG.
It is an output result when this invention is implemented.

The present invention is different from the conventional example in that a syntactic analysis determination unit 8 is provided between the text extraction unit 5 and the output unit 6, and this syntactic analysis determination unit 8 is input to the question input unit 1. The parsing is performed on the extracted question sentence and the text extracted by the text extraction unit 5, and the texts extracted by the text extraction unit 5 are rearranged based on the results of the parsing. It has the function of determining the priority of the output text. An outline of processing in the syntax analysis determination unit 8 will be described based on the flowcharts shown in FIGS.

In S1, the question sentence input to the question input unit 1 is syntactically analyzed, and the syntactic information of the keyword extracted by the keyword extracting unit 2, that is, which sentence and which clause the keyword is included in, Investigate the relationship of the relationships between keywords. In S2, one of the texts extracted by the text extraction unit 5 is added to the syntax analysis determination unit 8
To read. In S3, morpheme analysis is performed from the text read in S2 to extract keywords, and at the same time, the syntax information of the keywords in the read text (which sentence, which clause is included in the keyword, or between the keywords). Check the relationship).

In S4, S6, and S8, the priority order in the output unit 6 is determined based on the syntax information of the keyword extracted in S3. First, in S4, the priority is extracted by the text extraction unit 5. In the text
It is determined whether or not all types of keywords are present in one sentence. If all types of keywords are present in one sentence, the process proceeds to S6, and if all types of keywords are not present in one sentence. To proceed to S5. In S5, the priority order when all kinds of keywords are not present in one sentence is determined, and when all kinds of keywords are not present in one sentence, it is determined that there is almost no relationship between keywords, and the priority order is determined. (Hereinafter, referred to as ID) is determined to be the fourth (ID = 4) (hereinafter, if ID = n is written,
It indicates that the priority is n. ).

Next, in S6, the text extraction unit 5
It is determined whether or not all types of keywords are present in one phrase in the text extracted in step S7. If all types of keywords are present in one phrase, the process proceeds to S8, and all types of keywords are 1 If not included in one clause, the process proceeds to S7. In S7, the priority order when all types of keywords are not present in one phrase is determined, and when all types of keywords are not present in one phrase, it is determined that there is not much relation between keywords, ID
= 3 is determined.

In S8, it is determined whether the relationship between the keywords in the text extracted by the text extraction unit 5 and the relationship between the keywords in the question sentence input to the question input unit 1 are the same, If the relationships are the same, the process proceeds to S10. If the relationships are different,
Proceed to S9.

In S9, the priority order is determined when the relationship between the keywords in the text extracted by the text extraction unit 5 and the relationship between the keywords in the question sentence input to the question input unit 1 are different. If it is close to the content of the search request, it is determined that there is no match, and the ID
= 2 is determined.

In S10, the priority order is determined when the relationship between the keywords in the text extracted by the text extracting unit 5 and the relationship between the keywords in the question sentence input to the question input unit 1 are the same. In this case, it is determined that the contents match the content of the search request, and ID = 1 is determined.

In S11, all the texts extracted by the text extraction unit 5 are processed in the above S2 to S10.
It is determined whether or not the processing of step S1 has been completed, and if all the extracted texts have been completed, S1
Go to 2. In S12, the texts are rearranged in descending order of priority (smaller ID) based on the IDs determined in S2 to S10.

In S13, it is determined whether or not the rearranged texts have the same priority order. If there is the same priority order, the process proceeds to S14 and there is no one having the same priority order. In that case, the process proceeds to S17.

In steps S14 to S16, if there are those having the same priority, the priority is determined by another method.
In S14, it is determined whether or not there is a keyword not used in the determination of the priority order in S4 to S10 among the keywords extracted in S3, and if the keyword exists, this keyword is used. , S3 are again performed on the text read in S3 to determine the priority order.

In S15 and S16, the above S2 to S1
If the priorities are the same even after the process of 4, the priorities are determined according to the distance between the keywords (the number of characters existing between a certain keyword and another keyword).

In S17, based on the priorities determined in the processes of S2 to S16, the texts with higher priorities are transmitted to the output unit 6 in order.

As an example, the search for the question sentence "I want to know about Saturn's ring." Will be described with reference to Tables 1, 2 and 3.

Since the flow from the question sentence input to the question input unit 1 to the text extraction of the text extraction unit 5 is the same as the flow of the text extraction of the conventional example, the description thereof will be omitted here. A method of prioritizing the texts extracted by the text extraction unit 5 in the syntax analysis determination unit 8 will be described with reference to FIGS. 2 and 3.

First, in S1 of FIG. 2, the question input unit 1
The question sentence "I want to know about Saturn's ring." Entered from is morphologically analyzed, "Saturn" and "ring" are extracted as keywords, and further syntactic analysis is performed. The keywords "Saturn" and "Ring" are the same sentence,
It is determined that the keywords are in the same phrase and have a dependency relationship. This result is stored in the syntactic analysis determination unit 8 as the syntactic analysis result of the input question sentence.

Next, the texts extracted by the text extraction unit 5 in S2 are sequentially read by the syntactic analysis determination unit 8 and subjected to morphological analysis in S3.

In S3, first, the morphological analysis of the text 1 is performed, and "Saturn / to / related / to say / ba /, / the / in / ring / wa / saturn / of / symbol / to / speak / . "Is analyzed. Next, the ID is determined by syntactic analysis based on the result of this morphological analysis.
For text 1, it is determined whether the keywords "Saturn" and "ring" are in the same sentence. Since the keyword of text 1 exists in the same sentence, the process proceeds to S6. In S6, it is determined whether a keyword exists in the same phrase. Since the keywords "Saturn" and "Ring" of the text 1 do not exist in the same clause, the process proceeds to S7, where ID =
After it is determined to be 3, the process proceeds to S11. In S11,
For all the texts extracted by the text extraction unit 5, it is checked whether the ID has been determined by the syntactic analysis. If all the texts extracted by the text extraction unit 5 are completed, the process proceeds to S12, and if not completed, the process returns to S2 and is repeated.

An example of parsing text 3, text 4, and text 6 will be described below.

Subsequently, the morphological analysis of the text 3 is performed in S3. Text 3 is "Saturn / no / ring / ha /,
/ Solar system / Of / most / beautiful / thing / of / one / The analyzed text 3 is also subjected to S4, S6, S8 and S10 similarly to the text 1. Since the syntactic analysis result of the text 3 and the syntactic analysis result of the question sentence are the same. , ID = 1 is determined.

Further, subsequently, the third text extracted by the text extracting unit 5 is similarly repeated. Text 4 says "Saturn / ha /, / satellite / in / surrounding / retaining /./ and // ring / ha /, / satellite / from / energy / deprived / retained / retained /." Is morphologically analyzed, and the ID is determined through the processes of S4 and S5, and the text 4 has ID = 4. Further, the text 6 is also subjected to morphological analysis such as “Saturn / in / ring / wa / what / working / doing / of / wondering /?”, And S4, S6, S8 and S9. ID = 2 is determined through the process of. As described above, the IDs are determined by parsing all the texts extracted by the text extraction unit 5.

When the ID is finally determined by the syntax analysis, the output order is determined in step S12 of FIG. 3 from the highest priority order by the syntax analysis. In S12, if there are those having the same priority, the process proceeds to S13, and if they do not exist, the output order is designated to the output unit 6 in S17. In this embodiment, the text 1 has ID = 3, the text 3 has ID = 1, the text 4 has ID = 4, and the text 6 has ID = 2. Therefore, in S17, the text 3, the text 6, the text 1, and the text 1, The text 4 is output in order from the output unit 6.
The result is shown in FIG.

In the present embodiment, the case where there is no item having the same priority has been described. However, when there are a plurality of items having the same priority, as shown in S14 of FIG. It is also possible to make a determination by using a keyword other than the above keywords. Also, S1
If the same is true after step 4, it is possible to determine the priority order based on the distance between keywords in the extracted text (such as the number of characters between keywords) as shown in S15 and S16.

In the present embodiment, the case where the number of keywords extracted by the keyword extracting unit 2 is two has been described as an example, but the case where there are three or more keywords is basically the same. When there are three, it is also possible to use the syntactic analysis result by three keywords, or the syntactic analysis result regarding the keyword with a high appearance frequency in the question sentence.

[0048]

As is apparent from the above description, according to the present invention, the text extracted by the keyword is syntactically analyzed to give priority to the order of outputting the extracted text from the output section. Since the ranking is determined, it is possible to sequentially output the search requests from the texts that accurately represent the search requests. Furthermore, in the extracted text, whether all types of keywords are present in a single sentence, or in a single phrase, and whether the keywords have the same dependency relationship as the question sentence. By determining whether or not, it is possible to output the text representing the content of the search request more simply and in order.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a text search device including a syntax analysis determination unit of the present invention.

FIG. 2 is a flowchart showing a flow of processing of a parsing determination unit according to the present invention.

FIG. 3 is a flowchart showing a flow of processing of a parsing determination unit according to the present invention.

FIG. 4 is a diagram showing an output result when the present invention is implemented.

FIG. 5 is a schematic configuration diagram of a conventional text search device.

FIG. 6 is a diagram showing an output result according to a conventional example.

[Explanation of symbols]

 1 question input unit 2 keyword extraction unit 3 keyword memory 4 document memory 5 text extraction unit 6 output unit 7 control unit 8 syntax analysis determination unit

Claims (4)

[Claims] 1. A document memory in which a plurality of texts are stored, a question input section for inputting a question sentence required to extract the texts stored in the document memory, and a question input section A keyword extraction unit that extracts a plurality of types of keywords from the question sentence, a text extraction unit that extracts a text including the keyword from the document memory based on the keywords extracted by the keyword extraction unit, and the text extraction unit An output unit that outputs the text extracted from the text extraction unit; and a syntactic analysis determination unit that performs syntactic analysis on the text extracted by the text extraction unit, wherein the syntactic analysis determination unit is the keyword extraction unit. Based on the syntactic information of the keyword extracted by the above, the priority order for outputting the text from the output unit is determined. Search device. 2. The syntactic information is information as to whether or not all types of keywords extracted by the keyword extracting unit are included in one sentence, and the syntactic analysis determining unit includes all types of keywords. 2. The text search device according to claim 1, wherein the priority order of the texts included in the single sentence is set higher than the priority orders of the other texts. 3. The syntax information is information as to whether or not all types of keywords extracted by the keyword extraction unit are included in one clause, and the syntax analysis determination unit is used for all types of the keywords. 2. The text search device according to claim 1, wherein the priority included in the text included in the single clause is higher than the priority included in the other texts. 4. The syntactic analysis determination unit determines the relationship between the keywords in the question sentence input to the question input unit and the relationship between the keywords in the text extracted by the text extraction unit. 2. The text search device according to claim 1, wherein the priority order of the texts is set to be the highest when both have the same dependency relationship.