KR100434688B1 - Natural Language Question-Answering Search System for Integrated Access to Database, FAQ, and Web Site - Google Patents

Abstract

Unlike a search system that indexes documents on a website and presents the searched results, the present invention distributes a user's query to each area by a search system integrating FAQ list search and interactive DB search as well as document processing on a website. Provides an integrated query-response search method that enables optimal search response processing. The present invention comprises the steps of (i) morphological / partial parsing of the input natural language query; (Ii) determining whether the analyzed query is a query suitable for the interactive DB search method and searching for a response using the interactive DB search method if the query is suitable for the query; (Iii) applying bottom-up natural language information analysis to a query that is not suitable for the interactive DB retrieval method; (Iii) determining whether the query analyzed by the bottom-up natural language information analysis is a query suitable for the FAQ list retrieval method and retrieving the response using the FAQ list retrieval method if it is a suitable query; (v) retrieving the response with the website query-response retrieval method by determining whether it is not suitable for the FAQ list retrieval method or if the user requires a website retrieval; (Iii) an integrated query-response search method comprising the step of integrating the responses obtained from the interactive DB search, FAQ list search, and website search to select the best answer.

Description

Natural Language Question-Answering Search System for Integrated Access to Database, FAQ, and Web Site}

The present invention relates to an integrated query-response search system that searches an interactive DB, a Frequently Asked Question (FAQ) list, and a website. In particular, the present invention can freely search desired information by processing all searches as natural-language question-answers. A question and answer system and method are described.

Recently, an information retrieval system that processes a myriad of web document information on a website, extracts only information corresponding to a user's request, and provides the information to a user is widely used. In addition, an Internet information retrieval system capable of searching the Internet even in everyday natural language rather than a word format is widely provided. However, in general, it is very difficult to accurately extract the information requested by the information requester from a large collection of web documents, and even though a natural language search is possible, there are many shortcomings for the current search system to correctly obtain a response to a specific query desired by the information requester. .

In order to solve the above problems, an object of the present invention is a search system that integrates a FAQ list search and an interactive DB search as well as document processing on a website, unlike a search system that indexes only documents on a website and presents searched results. It is to provide an integrated query-response search system and method capable of processing an optimally searched response by distributing a user's query to each area.

In the present invention, a bottom-up technique such as statistical information retrieval technique and statistical language analysis, and a top-down technique such as language knowledge base and question-response modeling are integrated and used.

1 is a conceptual diagram of an interactive DB, a FAQ list, and a website query-response integrated search system according to the present invention;

Figure 2a is an overall configuration diagram of a search system to which the present invention is applied. Figure 2b is a detailed configuration diagram of the search system of Figure 2a.

3A and 3B are flow charts illustrating the integrated query-response retrieval method of the present invention.

4 is a detailed flowchart of the interactive DB retrieval step of FIG.

5 is a detailed flowchart of the FAQ list retrieval step of FIG.

6 is a detailed flowchart of the website query-response retrieval step of FIG.

In order to achieve the above object, an integrated query-response search system according to an embodiment of the present invention includes (i) an interactive DB search system; (Ii) a FAQ list retrieval system for retrieving a list of answers to frequently asked questions due to queries and user requests not found in the interactive DB retrieval system; (Iii) a website query-response system that executes a query-response search on a website in response to a user's request and a query not found in the interactive DB search system or the FAQ list search system.

According to the present invention, the interactive DB retrieval system comprises: (i) a morpheme analyzer for analyzing the user's natural language query; (Ii) a parser that determines a mathematical relationship between the analyzed morphemes; (Iii) a granter for assigning a PLO tagger and semantic code to the parsed phrase; (Iii) a Lexico-Syntick pattern matcher for mapping the PLO tagger and the semantic coded distinctions into SQL statements; (Iii) a searcher that accesses a suitable DB for the query described by the SQL statement; (Iii) a response generator for finding an optimal response in the accessed DB. In an embodiment of the present invention, the bottom-up natural language information searcher is based on (i) a morphological syntax dictionary and statistical language information. A morpheme / syntax analyzer that analyzes the user's natural language query in morphemes and phrases; (ii) a complex noun, a compound noun dictionary, and a compound noun based on the air information between nouns A noun processor; (ⅲ) an information searcher for performing a search using the weights of the stopword dictionary and the indexed noun; (포함) a semantic distinguishing matcher that performs keyword matching using Korean wordnet semantic information on a user query In addition, the top-down information filter is a Lexico-Shin that expresses the vocabulary, parts of speech, syntax, and semantic information in a regular expression form. Question form filter that classifies 30 kinds of text patterns and question forms to identify question intent; (ⅵ) Words that construct weighted dictionaries containing words that depend on specific search areas and adjust weights according to the characteristics of the areas. In addition, the FAQ list search system updates the index used in the statistical natural language analyzer, the relational information of nouns and verbs used in the semantic filter, and the specialized domain dictionary used in the word refined filter. It has a FAQ editor that includes a knowledge base updater that updates the Lexico-Syntic patterns and question type definitions that are used semi-automatically in the question type filter. According to the present invention, the website query-response system is a bottom-up natural language information searcher. And a top-down correct sentence extractor, wherein the bottom-up natural language information searcher includes (i) a morphological phrase A morpheme / syntax analyzer that analyzes a user's natural language query into morphemes and phrases based on dictionaries and statistical language information; (Ii) a compound noun processor for processing morphemes and phrases into queries based on single words, compound noun dictionaries, and air information between nouns; (Iii) an information searcher for performing a search using a stopword dictionary and a weighted index of nouns; And (iii) a semantic discriminating matcher that performs keyword matching using Korean WordNet semantic information on the user query, wherein the top-down correct sentence extractor comprises: (i) a natural language of the user based on the morphological syntax dictionary and statistical language information; Morpheme / syntax analyzer to parse queries into morphemes and phrases; (Iii) a PLO recognition and semantic code assigner that assigns a PLO tagger and semantic code to the parsed phrase; (Iii) a pattern matcher for determining question types using Lexico-Syntectic patterns with PLO information and semantic code assigned syntaxes; And (iii) a top-down information filter using heuristics such as semantic similarity, syntax structure in ISA format, clue words, and position within a document.

According to the integrated query-response search method of an embodiment of the present invention, the web server of the search site, the stemming / partial parsing of the user's natural language query input from the terminal; Determining whether the query is suitable for an interactive DB search, and if the query is suitable for the query, (a) determining, by the web server of the search site, a mathematical relationship between the analyzed morphemes; (b) the web server of the search site assigning a PLO tagger and semantic code to the parsed phrase; (c) the web server of the site mapping the division to which the PLO tagger and the semantic code are assigned to an SQL statement; (d) performing an interactive DB search comprising the web server of the search site accessing a predetermined DB and generating a response to the query described in the SQL statement; (E) analyzing, by the web server of the search site, the natural language query of the user in morpheme and syntax based on a morpheme phrase dictionary and statistical language information when it is determined that the analyzed query is not suitable for an interactive DB search; (f) processing, by the web server of the search site, the query analyzed with the morpheme and syntax into a compound noun based on a single word, a compound noun dictionary, and air information between nouns; (g) a web server of the search site performing a search using a stopword dictionary and weights of indexed nouns; And (h) performing bottom-up natural language information analysis, wherein the web server of the search site performs keyword matching on the user query using Korean wordnet semantic information. If the query analyzed in the bottom-up natural language information analysis step is a query suitable for the FAQ list search and the query is suitable for the FAQ list search, (i) the web site of the search site normalizes the vocabulary, parts of speech, syntax, and semantic information. The Lexicon-Syntectic pattern and the question form expressed in the form of expression are classified into a plurality of types to determine the intention of the question, and (j) a dictionary including words dependent on the search area is constructed and weighted according to the characteristics of the area. A top-down information filtering step of extracting correct answer sentences by adjusting a value; the web server of the search site is the FAQ lease. Determining whether it is not suitable for the search method or if the user requests a website search, and after performing step (e), (k) assigns the PLO tagger and semantic code to the analyzed phrase, and (l) the PLO information and semantic code. Is used to determine the question type using Lexico-syntactic pattern and (m) correct answer sentence through heuristic top-down filtering using semantic similarity, ISA format syntax structure, clue words and position in the document. There is provided an integrated query-response search method comprising performing a website query-response search for extracting the information from the user. Summarizing the steps for generating a response from the user's query, The analysis information is passed to the DB search system, and if there is a search result, the result is provided to the user. Otherwise, the FAQ list search system is requested to be searched. If there is a search result of the FAQ search system, the search result is provided to the user. Otherwise, the website query-response system is requested to search and the user is provided the result. Depending on the area in which the system is applied, the above three systems may be requested at the same time to provide all the results to the user, or some may be provided in order of priority.

Hereinafter, with reference to the accompanying drawings will be described in detail the present invention.

1 illustrates a structure of an integrated natural language query-response search system that analyzes a user's natural language query and searches for an interactive DB, a FAQ list, and a web site to generate a response. As shown, the integrated natural language query-response system includes an interactive DB search system 10; A FAQ list retrieval system 20 for retrieving a list of answers to frequently asked questions due to queries and user requests not found in the interactive DB retrieval system 10; A web site query-response system 30 that executes a query-response search on a web site in response to a user's request and a query not found in the interactive DB search system 10 or the FAQ list search system 20.

Example of Zone Dictionary Dictionary headings Meaning Category Gardner (@city | @person) teacher (@position) lecture (% lecture) New York (@city) dollar (@unit_money) Sogang University (@organization | @building) Securities Pacific National Bank (@company) human (% person) KG (% unit_weight) CM (% unit_length) Of the semantic categories of Table 1, '@X' means that the heading belongs to the semantic category 'X', and '% X' means that the heading is the semantic category 'X'. '@X | @Y' means that the heading belongs to the semantic category 'X' and also to 'Y'. The semantic category can be arbitrarily determined as shown in Table 1, or the sense code shown in WordNet can be given as it is.

In addition, the FAQ list retrieval system 20 includes a bottom-up natural language information searcher 21, a top-down information filter 22, and a FAQ editor 23. The bottom-up natural language information searcher 21 includes a morpheme syntax dictionary and statistical language information. A morpheme / syntax analyzer 210 for analyzing the user's natural language query based on the morpheme and syntax; A compound noun processor 211 that processes a morpheme and phrase-queried query into a compound noun based on a single word, a compound noun dictionary, and air information between two nouns (the probability that two target nouns appear adjacent in the document). ; An information searcher 212 that performs a search using a stopword dictionary and a weighted index of nouns; And a semantic discriminating matcher 213 that performs keyword matching using Korean wordnet semantic information with respect to a user query. The compound noun processor reduces the lexical discrepancy problem between the words in the input query and the words in the search target sentence. For example, if the input query is "information search" which is a compound noun and the sentence to be searched is "searching for information" or vice versa, the adjacent information "information" and "search" are combined as one compound noun. Synthesized by 'information search' to make the word 'information search' coincide between two sentences, solving the problem of lexical discrepancy and improving search performance.

The top-down information filter 22 is a question-type filter for classifying question intent by classifying about 30 Lexicon-syntectic patterns and question types that express a sentence of a vocabulary, part-of-speech, syntax, and semantic information in a regular expression form. 220); A word refiner 221 may be configured to construct a dictionary including words dependent on a specific search region and adjust weights according to characteristics of the region using the dictionary. The question form filter analyzes the question form of the input question sentence and the question form of the FAQ to be analyzed and removes only the same question form. For example, if your query sentence is "How do I change my password?" And the FAQ question is "How do I change my password?" And "Where do I report my password forgot?" The first FAQ question asks how and the second FAQ question asks about the place, so only the first question remains and the second question is removed. The word refining filter plays a role of manually defining and weighting words that are considered to be important by manually analyzing a target FAQ for each FAQ sentence. For example, if the user's query sentence is "Change Password" and the FAQ to be analyzed is "Change Password" with the word "Password" as the main word and "Password" with "ID" as the main word, "And" ID "with key words as" keyword ", the first FAQ question is higher than the second FAQ question because the word" secret "in the user's query is the key word in the first FAQ question. Has a weight.

In addition, the FAQ editor 23 updates the index used in the statistical natural language analyzer, the relational information of nouns and verbs used in the semantic filter, and the specialized domain dictionary used in the word refinement filter, and is used semi-automatically in the question type filter. It includes a knowledge base updater 230 for updating the Lexico-syntactic pattern and question type definition.

In addition, the website query-response system 30 includes a bottom-up natural language information searcher 31 and a top-down natural language information searcher 32 for receiving a search result from the bottom-up natural language information searcher 31 and searching. The natural language information searcher 31 may include a morpheme / syntax analyzer 310 for analyzing a user's natural language query into a morpheme and a phrase based on a morpheme phrase dictionary and statistical language information; A compound noun processor 311 for processing the morpheme and syntax analyzed query into a compound noun based on a single word, a compound noun dictionary, and air information between nouns; An information searcher 312 that performs a search using a stopword dictionary and weighted nouns of the index; And a semantic discriminating matcher 313 for performing keyword matching using Korean WordNet semantic information on the user query, and the top-down correct sentence extractor 32 includes a natural language of the user based on the morpheme syntax dictionary and statistical language information. A morpheme / syntax analyzer 320 that parses the query into morphemes and phrases; A pattern matcher 321 for assigning a PLO tagger and semantic code to the parsed phrase and determining a question type using a Lexico-syntactic pattern to which the PLO information and semantic code are assigned; Top-down information filter 322 using heuristics, such as semantic similarity, ISA format syntax structure, clue words, and position within a document. The top-down information filter assigns scores to predefined features, such as semantic similarity, ISA-type syntax structure, clue words, and positions within sentences, and the following weights when those features appear in the correct answer sentence: Calculate the score of each sentence using the sum. In the above formula ts _i is the score of sentence i , f _ij is the score of the j- th quality of the sentence i . That is, f _i1 is the score for the first feature of sentence i and f _i2 is the score for the second feature. A, B, C, and D are weight constants used to add each feature score and are determined through experiments for each applied area. Based on the scores of the sentences thus determined, the top n items are extracted as correct answer sentences. FIG. 2A schematically shows the overall configuration of a search system to which the present invention is applied. In FIG. 2A, a user terminal composed of a plurality of computers ( U _{1, ...} U _n ) is connected to any one search site web server 6 via the Internet 5. The web site 6 of this search site is provided with a user interface 7 and a document refinement and automatic classification system 8, a database (DB1) supporting a web server of a plurality of clients and a relational database integrated search, and a web. It has a database (DB2) for site query-response search, a database (DB3) for interactive DB search, and a database for FAQ list search (DB4). The search site web server 6 includes an operator terminal M. FIG. 2B shows the internal configuration of the search site web server 6, where the user interface 7 is a natural language interactive analyzer 7 ₁ . And the SQL generator 7 ₂ . The document refinement and automatic classification system 8 also includes a natural language information search engine 8 ₂ . The web server 8 includes an interactive DB search system 10 for analyzing a user's natural language query type to obtain an appropriate search result, a FAQ list search system 20 for providing a related inquiry FAQ, and a correct answer to a question. Or a website query-response search system 30 for recommending links. Details of the interactive DB retrieval system 10, the FAQ retrieval system 20, and the website query-response retrieval system 30 are the same as the description of the retrieval system of the present invention as described above. The web server 8 includes an integrated response generator 9 that finds the optimal response from the accessed DB according to the user's natural language query type.

The operation of the search system configured as described above and the integrated natural language query-response search method for the interactive DB, FAQ list, and website of the present invention will be described in detail with reference to FIG. 3A. As shown in FIG. 3A, in step S11, the natural language query input from the user is analyzed into a morpheme and a partial syntax structure which is a minimum unit having one meaning in Korean grammar using a morpheme / syntax analysis dictionary and statistical language information. . This analyzed query is then distributed to a suitable search area, i.e. interactive DB search, FAQ list search, website search system 10, 20, 30, in step S12.

The query distributed to the interactive DB search system 10 is assigned a semantic code including schema information of the DB to the Korean WordNet information / domain dictionary in step 13, and the query with the semantic code is assigned to the query sentence. Structured Query, an interactive database query statement that is pre-stored through a Lexico-syntactic pattern determiner 130 that includes lexical, part-of-speech, PLO information, syntax information, and semantic code, and is described by regular expressions. Language). The SQL query thus determined finds the most optimal response in the database through the interactive database in step S14. When the interactive DB search system 10 is used as a separate system, it has a structure as shown in FIG.

On the other hand, the query list distributed to the FAQ list and the website query system 20, 30 processes the compound nouns using the single word and the compound noun dictionary in step S15, and then searches using the weights of the stopword dictionary and the indexed nouns. Using the bottom-up natural language information retrieval method, it is classified into FAQ query and website query.

Here, the query analyzed by the FAQ list query is given the semantic code of the Korean WordNet information / domain dictionary. The query word to which this semantic code is assigned includes a lexical, part-of-speech, PLO information, syntax information, and semantic code of the query sentence, and passes through a top-down FAQ filter 22 through a Lexico-Syntectic pattern determiner described in regular expression. Find the answer (S16). When using this FAQ list retrieval system 20 as a separate system, it has the structure as shown in FIG.

On the other hand, the query analyzed by the website correct answer document is given a PLO recognition and semantic code in step S17, and determines the pattern for determining the question type using the Lexico-Syntectic pattern, and the semantic similarity, ISA format. The correct answer sentence is found by passing through a top-down correct sentence extractor 32 including a top-down information filter 322 using heuristics such as a syntax structure, a word, and a position in a document. When using this website query-response system as a separate system 30, it has a structure as shown in FIG.

Then, in step S18, the best response is found by integrating the interactive DB response, the FAQ response obtained from the top-down FAQ filter 21, and the correct answer sentence obtained from the top-down correct sentence extractor 32.

In the integrated natural language question-answering system described above, a procedural sequence for searching an interactive DB, a FAQ list, and a website is shown in a flowchart in FIG. 3B. As shown in FIG. 3, a user query is input into the integrated natural language question-response system according to the present invention (S21). The integrated natural language question-answering system performs morpheme and partial parsing on the query input in step S21. Then, it is determined whether the user query analyzed in step S21 is suitable for the interactive DB search.

If the query is suitable for the interactive DB, the interactive DB search is performed (S22). Then, it is passed to the FAQ list retrieval module, and in step S23 it is determined whether the user requires a FAQ list retrieval to obtain more information than the response obtained in the interactive DB retrieval, and if so, the user query in step S25 Perform a list search. On the other hand, if not required, it is passed to step S28, which is an integrated response selection step.

In step S26, it is determined whether the user requests a website search, and if so, the website is searched to extract an appropriate response (S27). If the user does not require a website search, the search results are presented to the user through the integrated response selection step S28.

On the other hand, if the user query is not suitable for the interactive DB in step S21, bottom-up natural language information analysis is performed (S24). It is determined whether the analyzed user query is suitable for the FAQ list search (S24). If appropriate, the FAQ list is retrieved in step S25. However, if it is not suitable for the user's FAQ list search, the website is searched and an appropriate response is extracted in step S27. Even when the user requests a web site search, the web site may be searched to extract an appropriate response (S26). In this way, the integrated search result for the user's query is properly presented to the user through the selection of the integrated response (S28).

For example, if the user query of the integrated natural language query-response retrieval system is "What is Kim Jung-min's phone number?", The relevant interactive DB is searched to provide the desired information. How do I get after-sales service? ”If the answer is a question frequently asked by users, the FAQ list is searched for an appropriate response. In addition, if a user query is a query that can be answered by a specific document on the website, such as "Where is the nearest agency in Sillim-dong?", The website is searched and an appropriate document is provided as an answer.

Hereinafter, each of the above-described interactive DB search method, FAQ list search method and website query-response search method will be described in more detail.

First, the interactive DB search method will be described in detail with reference to FIG. According to the present invention, the query of the user input to the interactive DB search system 10 is subjected to morphological analysis and partial syntax analysis to generate a predetermined syntax unit (eg, noun phrase) (S110). The generated syntactic unit includes PLO information, which is information about which noun phrases belong to proper nouns, categories, people, places, and institutions, and extended semantic codes by including DB schema information in Korean WordNet and domain dictionaries. It is given (S120). After the PLO recognition and semantic code assignment process, Lexic-Syntic pattern matcher including the lexical, part-of-speech, PLO information, syntax information, and semantic code of the query sentence is described in the form of regular expression. The user query is mapped to a predefined SQL statement (S130). When the SQL statement is selected, a response is extracted through DB access (S140), and a proper response sentence is generated by using a predefined response frame (S150).

Next, a frequently asked question, the FAQ list search method that can easily find the answer of the FAQ will be described based on FIG. Since the structure of the FAQ list retrieval system 20 has been described above, a detailed description thereof will be omitted. The FAQ list retrieval system obtains a suitable query-response pair via a bottom-up natural language information retriever 21 and a top-down information filter 22.

Bottom-up information retrieval analyzes the query through morpheme / syntax analysis (S210), and processes a compound noun using a single word and a compound noun dictionary (S211). After that, the search is performed using the stopword dictionary and the weight of the indexed noun (S212). A keyword matching search is performed by semantic discriminating matching using Korean WordNet semantic information (S213).

Therefore, in the case of using another keyword of similar meaning, which is a problem in a system using only the keyword matching search method by semantic discrimination matching using wordnet semantic information, the FAQ list search system 20 uses semantic classification and abstraction. Inference can be used to resolve errors from using other keywords.

On the other hand, the top-down information filter 22 improves the performance of the search by resolving errors resulting from information omissions other than nouns, and by refining words based on the dictionary considering the characteristics of the search target area. This top-down information filtering system accurately identifies user's question intention through lexico-syntectic pattern that expresses the sentence sentence, part-of-speech, syntax, and semantic information in the form of regular expression and the question type filtration that classifies 30 kinds of question forms. Identify and improve search performance (S220). In addition, the word reconstruction using a dictionary that includes words dependent on a specific search area is adjusted according to the characteristics of the search area (S221). This FAQ list retrieval system 20 has a FAQ editor, which constructs a robust FAQ list retrieval system 20 using a collection of entered user queries and creates a new question-answer pair in the FAQ list. Add In addition, the FAQ editor 23 automatically updates the relational information of nouns and verbs used in indexing and semantic filtering used in statistical information retrieval systems, and specialized domain dictionaries used in word refinement. The Lexico-syntactic pattern and the question type definition used are updated (S230).

Next, a website query-response retrieval method will be described with reference to FIG. As shown in FIG. 6, the user's natural language query performs top-down filtering 32 on the top N correct answer documents retrieved by the bottom-up natural language information searchers 21 and 31 shown in FIG. 5.

In the bottom-up natural language information searcher (21, 31), first, the user's natural language query is grouped into a certain syntax unit through morphological analysis and partial syntax analysis. Each syntax unit is given PLO information and a semantic code. The user query assigned the PLO information and the semantic code is passed to the pattern matcher module for determining the question type (S321). To determine the question type, the module uses the Lexico-Syntic pattern, which is described in regular expression form, including the lexical, part-of-speech, PLO information, syntax information, and semantic code of the query sentence in the Lexico-Syntax pattern S321). The pattern matcher reads the Lexico-syntactic pattern described in text form and filters the sentence that is correct according to the question type of the current query (S322). Table 1 is an example of Lexico-Syntectic pattern. In Table 1, the vocabulary starting with $ means Synset and its lower term of Korean WordNet to which the vocabulary belongs, and% means synonym.

Question type Person pattern (@Person |% person) (jCs | jOc) {0,1} (\?) {0,1} $ (@ person |% person) (jCs) {0,1} (\?) {0,1 } (. + (jOm) {0,1}) {0,1} who. * $ (@ person |% person) (. + (jOm) {0,1}) {0,1}% name (jCs | jOc) {0,1} (\?) {0,1} $ (@ person |% person) (. + (jOm) {0,1}) {0,1}% name (jCs) {0, 1} (What | How). * $ Who jCsWho jOm Symbol Description $: Hyponym%: Synonym {}: Repeat |: selection +: repeat one or more times *: repeat zero or more times

<Table 1> Lexico-syntatic pattern example

The top N documents retrieved using the bottom-up natural language information searcher (21, 31) also undergo a PLO recognition and semantic code assignment process. After this preprocessing of the user query and the retrieved document, the website query-response retrieval system 30 extracts a sentence estimated to be a correct answer from the retrieved document. To extract the correct answer, we use the question type, PLO information, semantic similarity, and various heuristics found in the pattern matcher module. The question type and PLO information serve as a filter to leave only the sentences that are likely to be correct and to remove the rest from candidates. Heuristics, such as semantic similarity and syntactic structure in ISA format, clue words, and position in a document, are used to score sentences that pass through the filter to select the sentences most likely to be correct. The website query-response retrieval system 30 also ranks the top N documents found by the bottom-up natural language information searcher (21, 31) by giving a low score to documents that do not contain the correct answer sentences. do.

The present invention is a search system that integrates FAQ list search and interactive DB search as well as document processing on the web, unlike a search system that displays only the documents on the web and presents the searched results. Organically integrated top-down natural language processing techniques, such as natural language processing technology, language knowledge base, and question-answer modeling, can provide users with convenient and accurate information.

Claims (11)

delete delete delete delete Morpheme / partial parsing of a natural language query of a user input from a terminal by a web server of a search site; Determining, by the web server of the search site, whether the analyzed query is a query suitable for an interactive DB search, and if the query is a suitable query, (a) determining, by the web server of the search site, a mathematical relationship between the analyzed morphemes; (b) the web server of the search site assigning a PLO tagger and semantic code to the parsed phrase; (c) the web server of the site mapping the division to which the PLO tagger and the semantic code are assigned to an SQL statement; (d) performing an interactive DB search comprising a web server of the search site accessing a predetermined DB and generating a response to a query described in the SQL statement; When the web server of the search site determines that the analyzed query is not suitable for an interactive DB search, (e) the web server of the search site may stem the user's natural language query based on the morpheme syntax dictionary and statistical language information. Parsing; (f) processing, by the web server of the search site, the query analyzed with the morpheme and syntax into a compound noun based on a single word, a compound noun dictionary, and air information between nouns; (g) a web server of the search site performing a search using a stopword dictionary and weights of indexed nouns; And (h) performing bottom-up natural language information analysis by a web server of the search site performing keyword matching on the user query using Korean wordnet semantic information. The web server of the search site determines whether the query analyzed in the bottom-up natural language information analysis step is a query suitable for the FAQ list search, and if the query is suitable for the FAQ list search, (i) the web site of the search site searches the query sentence. Lexicon-Syntectic patterns and question types that express part-of-speech, syntax, and semantic information in regular expressions are classified into a number of ways to determine questioning intention, and (j) construct a dictionary that includes words that depend on the search domain. A top-down information filtration step of extracting the correct sentence by adjusting the weight according to the characteristics of the region by using; The web server of the search site determines whether it is not suitable for the FAQ list search method or if the user requests a website search, and after performing step (e), (k) assigns a PLO tag and semantic code to the analyzed phrase. (1) determine the question type using Lexico-Syntax pattern, and (m) the semantic similarity, the syntax structure of the ISA format, the clue words and their position in the document. An integrated natural language query-response search method comprising the step of performing a website query-response search for extracting correct sentences through top-down filtration using a heuristic of. delete 6. The method of claim 5, wherein the top-down information filtering step comprises: (n) a web server of the search site, an index used in a statistical natural language analyzer, relationship information of nouns and verbs used in semantic filtering, and a specialized region used in word refinement filtering. And a FAQ editing step of updating the dictionary and updating the lexico-syntactic pattern and question type definitions used in the question type filtration. delete (i) morphologically analyzing the user's natural language query from the terminal by the web server of the search site; (ii) determining a mathematical relationship between the analyzed morphemes by the web server of the search site; A web server of a search site assigning a PLO tagger and a semantic code to the parsed phrase; (i) a web server of the search site mapping a phrase to which a PLO tagger and semantic code is assigned to an SQL sentence; (Iii) the web server of the search site accesses a suitable DB for the query described in the SQL statement; (iii) the web server of the search site extracts a response through the DB access and defines a predefined response framework. By generating a response sentence, The step of determining the mathematical relationship between the analyzed morphemes is: Extracting two representative morphemes consisting of real morphemes and formal morphemes from a word based on the result of the morphological analysis; And Representing the morphemes according to the following rules, The step of assigning the semantic code is: Analyzing the target region to determine a semantic category; Manually collecting words corresponding to each of the semantic categories and storing them in a database; And Interactive DB retrieval method comprising the step of assigning a semantic code to each morpheme, noun phrase or verb phrase of the input sentence based on the database: Where the rule is ● The leftmost morph of a word basically becomes the real morph of that word, ● The rightmost morpheme of a word becomes the formal morpheme of that word, ● Words consisting of only one morpheme serve as real morphemes and formal morphemes. ● Commas, exclamation marks, periods, and special symbols affect the structure of sentences, so they are treated like formal morphemes. ● When a commander or an investigator comes to the beginning of a word, the next statement is treated as a real morpheme. (i) analyzing, by the web server of the search site, a natural language query of the user into a morpheme and a phrase based on a morpheme phrase dictionary and statistical language information; (Ii) processing, by the web server of the search site, the query analyzed with the case and phrase into a compound noun based on a single word, a compound noun dictionary, and air information between nouns; (Iii) the web server of the search site performing a search using a stopword dictionary and a weighted index of nouns; And (Iii) a bottom-up natural language information retrieval step comprising a web server of the search site performing keyword matching on the user query using Korean wordnet semantic information; (Iii) a web server of the search site classifies the Lexic-Syntectic pattern and the question form, which express the vocabulary, the part-of-speech, the phrase, and the semantic information in a regular expression form, to identify the intention of the question ; And (Iii) a top-down information filtering step comprising the web server of the search site constructing a dictionary containing words dependent on the search area and using the same to adjust weights according to the characteristics of the area; and (Iii) The web server of the search site updates the index used in the statistical natural language analyzer, the relational information of nouns and verbs used in semantic filtering, and the specialized domain dictionary used in word refinement filtering, and the Lexi used in question type filtering. A method for retrieving a list of frequently asked questions, which includes editing the co-syntactic pattern and question type definitions. (i) analyzing, by the web server of the search site, a natural language query of the user into a morpheme and a phrase based on a morpheme phrase dictionary and statistical language information; (Ii) the web server of the search site processing the query analyzed in the morpheme and phrase into a compound noun based on a single word, a compound noun dictionary, and air information between nouns; (Iii) the web server of the search site performing a search using a stopword dictionary and a weighted index of nouns; And (Iii) a bottom-up natural language information retrieval step comprising a web server of the search site performing keyword matching on the user query using Korean wordnet semantic information; and (I) assigning a PLO tagger and semantic code to the parsed syntax by the web server of the search site after performing step (i); (Iii) determining, by the web server of the search site, a question type using a Lexico-Synectic pattern with the PLO information and a semantic code; And (Iii) a web server of the search site performing top-down filtering using heuristics such as semantic similarity, ISA format syntax structure, clue words, and position within a document.