CN112507089A - Intelligent question-answering engine based on knowledge graph and implementation method thereof - Google Patents
Intelligent question-answering engine based on knowledge graph and implementation method thereof Download PDFInfo
- Publication number
- CN112507089A CN112507089A CN202011346253.6A CN202011346253A CN112507089A CN 112507089 A CN112507089 A CN 112507089A CN 202011346253 A CN202011346253 A CN 202011346253A CN 112507089 A CN112507089 A CN 112507089A
- Authority
- CN
- China
- Prior art keywords
- question
- answering
- answer
- engine
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
Abstract
The disclosure provides an intelligent question-answering engine based on a knowledge graph and an implementation method thereof. Initializing business process data, and implanting the business process data into a question-answering engine; based on the business process data and the question sentence to be searched, calling a question and answer engine of a required type and a question and answer assembly of a required type; based on the called question-answering component, the called question-answering engine matches answers of the question sentences from a cache matched with the question-answering component; if the answer data are matched, extracting corresponding question and answer data; and if the question is not matched with the question, analyzing the question sentence to obtain question and answer data. The method and the system integrate various types of question-answering components, call the required question-answering engine according to the service flow data and the question to be searched, call the corresponding question-answering components by using the question-answering engine, realize uniform scheduling, and automatically call the required question-answering components according to the types of the question and answer, thereby performing the question-answering flow function and adapting the question-answering result.
Description
Technical Field
The present disclosure relates to intelligent question answering, and more particularly, to an intelligent question answering engine based on a knowledge graph and a method for implementing the same.
Background
With the convenience brought to the life of people by search engines, the demand of users for information retrieval is increasing. The search engines Google, Yahoo, heyday, wikipedia and the like which are well known nowadays can immediately search out relevant webpage information only by inputting the content to be searched by a user. However, in the current era of a large network outbreak, every day network data is explosively increased, which exposes some shortcomings of these search engines, when a user searches, the user can often search too much redundant data, and these search engines can return too much to the user, which is not really needed by the user, and it is difficult to quickly meet the user's needs. Secondly, the current search engine searches through keywords, which results in that the searched results are too diverse, the search range is only limited to the keywords which are completely matched, and the search and shallow semantic analysis similar to synonyms cannot be achieved, so that the information search mode is increasingly difficult to meet the requirements of users.
Disclosure of Invention
In order to solve at least one of the above technical problems, the present disclosure provides an intelligent question-answering engine based on a knowledge graph and an implementation method thereof.
The technical scheme of the disclosure is realized as follows:
an intelligent question-answering engine implementation method based on knowledge graph includes:
based on the business process data and the question to be searched, calling a question-answering engine of a required type and a question-answering assembly of a required type, wherein the business process data comprises a database which is configured by a client in advance and is used for searching and question answer data;
assembling business process data, and implanting the business process data into the called question-answering engine;
matching answers to the question sentences from a cache collocated with the invoked question-answering component by the invoked question-answering engine;
if the answer data are matched, extracting corresponding question and answer data;
and if the question is not matched with the question, analyzing the question sentence to obtain question and answer data.
Further, the analyzing the question sentence to obtain question and answer data includes:
analyzing the part of speech of the question based on a natural language analysis technology, and acquiring the meaning of the question through semantic analysis;
further, the parsing the part of speech of the question based on the natural language analysis technology includes: and analyzing the part of speech of the question by using part of speech marks and part of speech-sense clustering.
Based on the meaning of the question, a database of question-answering components is scheduled and answers to the question are matched.
Further, the invoking a question-answering engine of a required type and a question-answering component of a required type based on the business process data and the question sentence to be searched includes:
defining the type of a question-answering engine based on the business process data and the input question;
selecting a desired question-answering component based on the type of the question-answering engine;
and the called question-answering engine loads the business number flow data and calls a question-answering component of a required type.
Further, the types of question answering engines include: single question-answer, multi-question-answer combination and custom question-answer combination.
Further, the types of the question-answer components comprise question-answer pairs, syntax templates, part-of-speech templates, violent combinations, knowledge-graph-based question-answers, multiple rounds of question-answers, reading comprehension and NLSQL.
Further, after analyzing the question and answering data, adding the obtained question answer into the cache.
A smart question-answering engine for knowledge-graph based applications, comprising:
the question-answer component module comprises various question-answer components and is used for analyzing the question and sentence and acquiring related question-answer data;
the question answering engine module comprises a plurality of types of question answering engines and is used for scheduling the required question answering components;
and the application service module is used for starting a question-answering function, initializing the business process data and implanting the business process data into the question-answering engine.
Further, the types of question answering engines include: single question-answer, multi-question-answer combination and custom question-answer combination.
Further, the types of the question-answer components comprise question-answer pairs, syntax templates, part-of-speech templates, violent combinations, knowledge-graph-based question-answers, multiple rounds of question-answers, reading comprehension and NLSQL.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the disclosure and together with the description serve to explain the principles of the disclosure.
FIG. 1 is a schematic workflow diagram of a question-answering engine of the present disclosure;
Detailed Description
The present disclosure will be described in further detail with reference to the drawings and embodiments. It is to be understood that the specific embodiments described herein are for purposes of illustration only and are not to be construed as limitations of the present disclosure. It should be further noted that, for the convenience of description, only the portions relevant to the present disclosure are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
It should be noted that, the step numbers in the text are only for convenience of explanation of the specific embodiments, and do not serve to limit the execution sequence of the steps. The method provided by the embodiment can be executed by a related server, and the following description takes an electronic device such as a server or a computer as an example of an execution subject.
Example one
Referring to fig. 1, a method for implementing an intelligent question-answering engine based on a knowledge graph includes:
based on the business process data and the question to be searched, a question-answering engine of a required type and a question-answering component of a required type are called, and a user enables the question-answering engine to call the question-answering component appointed by the user by configuring matching rules of the question-answering engine and the question-answering component;
initializing the business process data, including: assembling service flow data, implanting the service flow data into the called question-answer engine, and leading the question-answer data, question sentences, answers, templates and the like in a service range in batches by a user;
matching answers to the question sentences from a cache collocated with the invoked question-answering component by the invoked question-answering engine;
if the answer data are matched, extracting corresponding question and answer data;
and if the question is not matched with the question, analyzing the question sentence to obtain question and answer data.
And analyzing the question, and adding the obtained question answer into a cache after obtaining the question-answer data so as to be convenient for next calling.
As a preferred implementation of this embodiment, analyzing the question sentence to obtain question-answer data includes:
analyzing the part of speech of the question by using part of speech tagging and part of speech/word meaning clustering based on a natural language analysis technology, and acquiring the meaning of the question through semantic analysis;
based on the meaning of the question, a database of question-answering components is scheduled and answers to the question are matched.
In the embodiment, in the initialization stage, a user firstly assembles and implants business process data to wait for the scheduling of a subsequent question-answering engine; then, determining the type of a question-answering engine according to the demand of the question-answering reference, wherein the determination of the type of the question-answering engine can further determine the scheduling demand of the question-answering component; types of question-answering engines include: single question-answer, multi-question-answer combination and custom question-answer combination. After determining the type of the question-answering engine, the question-answering engine calls a required question-answering component through a question-answering engine context algorithm and carries out question-answering search, wherein the type of the question-answering component comprises: question-answer pairs, syntactic templates, part-of-speech templates, violent combinations, knowledge-graph-based question-answers, multiple rounds of question-answers, reading comprehension, NLSQL (Nature Language SQL). In the question and answer searching process, the system firstly queries whether relevant question answers of the questions exist in the cache region, if yes, the data are taken out from the cache and returned to the application front end, the question and answer data are presented to the user, if no relevant question information exists in the cache region, the answers of the questions are retrieved from the database through relevant question and answer searching functions, question and answer results are adapted, and meanwhile the questions and the answers are returned to the application front end and added into the cache.
As a preferred embodiment of this embodiment, based on the business process data and the question sentence to be searched, the method for invoking the question-answering engine of the required type and the question-answering component of the required type includes:
defining the type of a question-answering engine based on the business process data and the input question;
selecting a desired question-answering component based on the type of the question-answering engine; the matching rules between the types of the question answering engines and the selected question answering components are manually configured in the background by the user at first, and the question answering engines can automatically select the matching rules according to precedent after enough running times in the later period;
and the called question-answering engine loads the business flow data and calls the question-answering assembly of the required type. In the embodiment, various types of question-answering components are integrated, a required question-answering engine is called according to business process data and question sentences to be searched, the corresponding question-answering components are called by using the question-answering engine, unified scheduling is realized, the required question-answering components can be automatically called according to the types of the question and answer, and therefore the question-answering process function is performed and question-answering results are adapted.
As a preferred implementation manner, the question-answer component of this embodiment integrates question-answer pairs, and the question-answer pairs form individual question-answer pairs in the database by defining a plurality of pairs of questions and answers in advance. When the user searches for a question, the answer to the question-answer pair is returned by exact retrieval, fuzzy retrieval or approximate matching to the relative cause or associated question. The improved accuracy and relevance of the answers to the question sentences, and the searched results are only limited in the database, so that the results are not cluttered and redundant.
As a preferred implementation, the question-answering component of this embodiment integrates a knowledge graph question-answering, and the knowledge graph forms a graph-based data format by extracting unstructured, semi-structured, and structured data, and forms a huge relational network graph by points and edges, which greatly optimizes search performance. The problem of traditional search can be solved through knowledge-graph question-answering. Meanwhile, the data with the highest similarity is searched in the database by combining a natural language processing technology and through approximate matching, semantic understanding, relation judgment and the like.
As a preferred implementation, the question-answering component of this embodiment integrates NLSQL, which is also called natural language database technology, and can convert the natural language input by the user into executable SQL statements. The purpose of converting the natural language into the SQL statement is to serve as an intelligent interface of the database, reduce the threshold of database query and enable a non-professional user to freely query data as required, such as speaking, on the premise of not learning and mastering the database programming language. In the technical scope, NLSQL essentially converts a natural language statement of a user into a semantic representation that is readable and understandable by a computer, can run and conforms to computer rules, and needs a computing mechanism to solve the natural language statement and generate an executable procedural language for accurately expressing the statement semantic.
The embodiment calls the service data of the user, the required engine type and the question-answering component to realize the question-answering function through the scheduling of the question-answering engine; meanwhile, the problems are understood through semantic recognition and semantic analysis, a database is scheduled, and a search result is returned, wherein the question-answer engine is used as a hub for mutual calling among question-answer components such as atlas question-answer, question-answer pair search, NLSQL and the like, so that the operations of atlas search and question-answer search are more visual, direct and simple, and are more efficient.
Example two
A smart question-answering engine for knowledge-graph based applications, comprising:
the question-answer component module comprises various question-answer components and is used for analyzing the question and sentence and acquiring related question-answer data; the types of the question-answer components comprise question-answer pairs, syntax templates, part-of-speech templates, violent combinations, NLP KBQA, multiple rounds of question-answers, reading comprehension and NLSQL.
The question answering engine module comprises a plurality of types of question answering engines and is used for scheduling the required question answering components; types of question-answering engines include: single question-answer, multi-question-answer combination and custom question-answer combination.
And the application service module is used for starting a question-answering function, initializing the business process data and implanting the business process data into a question-answering engine.
The principle and effect of the present embodiment are the same as those of the first embodiment, and the description of the present embodiment is not repeated.
It will be understood by those skilled in the art that the foregoing embodiments are merely for clarity of illustration of the disclosure and are not intended to limit the scope of the disclosure. Other variations or modifications may occur to those skilled in the art, based on the foregoing disclosure, and are still within the scope of the present disclosure.
Claims (9)
1. An intelligent question-answering engine implementation method based on knowledge graph is characterized by comprising the following steps:
based on the business process data and the question to be searched, calling a question-answering engine of a required type and a question-answering assembly of a required type, wherein the business process data comprises a database which is configured by a client in advance and is used for searching and question answer data;
assembling business process data, and implanting the business process data into the called question-answering engine;
matching answers to the question sentences from a cache collocated with the invoked question-answering component by the invoked question-answering engine;
if the answer data are matched, extracting corresponding question and answer data;
and if the question is not matched with the question, analyzing the question sentence to obtain question and answer data.
2. The method of claim 1, wherein analyzing the question sentences to obtain question-answer data comprises:
analyzing the part of speech of the question based on a natural language analysis technology, and acquiring the meaning of the question through semantic analysis;
and scheduling the database of the question-answer component based on the meaning of the question, and matching the answer of the question.
3. The method of claim 1, wherein invoking a question-and-answer engine of a desired type and a question-and-answer component of a desired type based on the business process data and a question to be searched comprises:
defining the type of a question-answering engine based on the business process data and the input question;
selecting a desired question-answering component based on the type of the question-answering engine;
and the called question-answering engine loads the business number flow data and calls a question-answering component of a required type.
4. The method of any of claims 1-3, wherein the types of question-answering engines include: single question-answer, multi-question-answer combination and custom question-answer combination.
5. The method of any of claims 1-3, wherein the types of question-answer components include question-answer pairs, syntactic templates, part-of-speech templates, violent combinations, knowledge-graph-based question-answers, multiple rounds of question-answers, reading comprehension, NLSQL.
6. A method according to any one of claims 1 to 3, wherein said analyzing said question and after obtaining question-answer data, adding said obtained question-answer to said cache.
7. An intelligent knowledge-graph-based question-answering engine, comprising:
the question-answer component module comprises various question-answer components and is used for analyzing the question and sentence and acquiring related question-answer data;
the question answering engine module comprises a plurality of types of question answering engines and is used for scheduling the required question answering components;
and the application service module is used for starting a question-answering function, initializing the business process data and implanting the business process data into the question-answering engine.
8. The intellectual property graph based question-answering engine according to claim 7 wherein the types of question-answering engines include: single question-answer, multi-question-answer combination and custom question-answer combination.
9. The intellectual property map based question answering engine according to claim 7 wherein the types of question answering components include question answering pairs, syntactic templates, part of speech templates, violent combinations, intellectual property map based question answering, multiple rounds of question answering, reading comprehension, NLSQL.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011346253.6A CN112507089A (en) | 2020-11-25 | 2020-11-25 | Intelligent question-answering engine based on knowledge graph and implementation method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011346253.6A CN112507089A (en) | 2020-11-25 | 2020-11-25 | Intelligent question-answering engine based on knowledge graph and implementation method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112507089A true CN112507089A (en) | 2021-03-16 |
Family
ID=74966105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011346253.6A Pending CN112507089A (en) | 2020-11-25 | 2020-11-25 | Intelligent question-answering engine based on knowledge graph and implementation method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112507089A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113127619A (en) * | 2021-04-19 | 2021-07-16 | 厦门渊亭信息科技有限公司 | Intelligent question answering method and device |
| CN113138637A (en) * | 2021-04-16 | 2021-07-20 | 北京沃东天骏信息技术有限公司 | Computer configuration information processing method, device and storage medium thereof |
| CN113779263A (en) * | 2021-08-24 | 2021-12-10 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Detection method, device and equipment of decision engine |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103229162A (en) * | 2010-09-28 | 2013-07-31 | 国际商业机器公司 | Providing answers to questions using logical synthesis of candidate answers |
| CN105159996A (en) * | 2015-09-07 | 2015-12-16 | 百度在线网络技术（北京）有限公司 | Deep question-and-answer service providing method and device based on artificial intelligence |
| CN107368548A (en) * | 2017-06-30 | 2017-11-21 | 华中科技大学鄂州工业技术研究院 | Intelligent government affairs service interaction method and system |
| CN108052605A (en) * | 2017-12-13 | 2018-05-18 | 广州佰聆数据股份有限公司 | A kind of intelligent Answer System based on client feature library |
| CN108446286A (en) * | 2017-02-16 | 2018-08-24 | 阿里巴巴集团控股有限公司 | A kind of generation method, device and the server of the answer of natural language question sentence |
| CN108491394A (en) * | 2018-06-27 | 2018-09-04 | 杭州贝店科技有限公司 | A kind of semantic analysis, device, computer equipment and storage medium |
| CN109033229A (en) * | 2018-06-29 | 2018-12-18 | 北京百度网讯科技有限公司 | Question and answer treating method and apparatus |
| CN110046227A (en) * | 2019-04-17 | 2019-07-23 | 腾讯科技（深圳）有限公司 | Configuration method, exchange method, device, equipment and the storage medium of conversational system |
| CN110717019A (en) * | 2019-08-28 | 2020-01-21 | 厦门快商通科技股份有限公司 | Question-answering processing method, question-answering system, electronic device and medium |
| WO2020178626A1 (en) * | 2019-03-01 | 2020-09-10 | Cuddle Artificial Intelligence Private Limited | Systems and methods for adaptive question answering |
-
2020
- 2020-11-25 CN CN202011346253.6A patent/CN112507089A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103229162A (en) * | 2010-09-28 | 2013-07-31 | 国际商业机器公司 | Providing answers to questions using logical synthesis of candidate answers |
| CN105159996A (en) * | 2015-09-07 | 2015-12-16 | 百度在线网络技术（北京）有限公司 | Deep question-and-answer service providing method and device based on artificial intelligence |
| CN108446286A (en) * | 2017-02-16 | 2018-08-24 | 阿里巴巴集团控股有限公司 | A kind of generation method, device and the server of the answer of natural language question sentence |
| CN107368548A (en) * | 2017-06-30 | 2017-11-21 | 华中科技大学鄂州工业技术研究院 | Intelligent government affairs service interaction method and system |
| CN108052605A (en) * | 2017-12-13 | 2018-05-18 | 广州佰聆数据股份有限公司 | A kind of intelligent Answer System based on client feature library |
| CN108491394A (en) * | 2018-06-27 | 2018-09-04 | 杭州贝店科技有限公司 | A kind of semantic analysis, device, computer equipment and storage medium |
| CN109033229A (en) * | 2018-06-29 | 2018-12-18 | 北京百度网讯科技有限公司 | Question and answer treating method and apparatus |
| WO2020178626A1 (en) * | 2019-03-01 | 2020-09-10 | Cuddle Artificial Intelligence Private Limited | Systems and methods for adaptive question answering |
| CN110046227A (en) * | 2019-04-17 | 2019-07-23 | 腾讯科技（深圳）有限公司 | Configuration method, exchange method, device, equipment and the storage medium of conversational system |
| CN110717019A (en) * | 2019-08-28 | 2020-01-21 | 厦门快商通科技股份有限公司 | Question-answering processing method, question-answering system, electronic device and medium |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113138637A (en) * | 2021-04-16 | 2021-07-20 | 北京沃东天骏信息技术有限公司 | Computer configuration information processing method, device and storage medium thereof |
| CN113138637B (en) * | 2021-04-16 | 2024-04-05 | 北京沃东天骏信息技术有限公司 | Computer configuration information processing method, device and storage medium thereof |
| CN113127619A (en) * | 2021-04-19 | 2021-07-16 | 厦门渊亭信息科技有限公司 | Intelligent question answering method and device |
| CN113779263A (en) * | 2021-08-24 | 2021-12-10 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Detection method, device and equipment of decision engine |
| CN113779263B (en) * | 2021-08-24 | 2024-01-05 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Decision engine detection method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109284363B (en) | Question answering method and device, electronic equipment and storage medium | |
| CN110399457B (en) | Intelligent question answering method and system | |
| US9448995B2 (en) | Method and device for performing natural language searches | |
| CN109684448B (en) | Intelligent question and answer method | |
| US10146751B1 (en) | Methods for information extraction, search, and structured representation of text data | |
| US10262062B2 (en) | Natural language system question classifier, semantic representations, and logical form templates | |
| US20180341871A1 (en) | Utilizing deep learning with an information retrieval mechanism to provide question answering in restricted domains | |
| US10339453B2 (en) | Automatically generating test/training questions and answers through pattern based analysis and natural language processing techniques on the given corpus for quick domain adaptation | |
| KR102491172B1 (en) | Natural language question-answering system and learning method | |
| CN112507089A (en) | Intelligent question-answering engine based on knowledge graph and implementation method thereof | |
| US20210374168A1 (en) | Semantic cluster formation in deep learning intelligent assistants | |
| CN112035506A (en) | Semantic recognition method and equipment | |
| US20230030086A1 (en) | System and method for generating ontologies and retrieving information using the same | |
| CN114218472A (en) | Intelligent search system based on knowledge graph | |
| CN112579733A (en) | Rule matching method, rule matching device, storage medium and electronic equipment | |
| Steinmetz et al. | From natural language questions to SPARQL queries: a pattern-based approach | |
| US11487795B2 (en) | Template-based automatic software bug question and answer method | |
| CN111159381A (en) | Data searching method and device | |
| CN114117000A (en) | Response method, device, equipment and storage medium | |
| Rashid et al. | A study on semantic searching, semantic search engines and technologies used for semantic search engines | |
| CN117251455A (en) | Intelligent report generation method and system based on large model | |
| CN113157887A (en) | Knowledge question-answering intention identification method and device and computer equipment | |
| CN109684357B (en) | Information processing method and device, storage medium and terminal | |
| CN111898024A (en) | Intelligent question and answer method and device, readable storage medium and computing equipment | |
| CN115982322A (en) | Water conservancy industry design field knowledge graph retrieval method and retrieval system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |