CN108984778A - A kind of intelligent interaction automatically request-answering system and self-teaching method - Google Patents
A kind of intelligent interaction automatically request-answering system and self-teaching method Download PDFInfo
- Publication number
- CN108984778A CN108984778A CN201810825398.0A CN201810825398A CN108984778A CN 108984778 A CN108984778 A CN 108984778A CN 201810825398 A CN201810825398 A CN 201810825398A CN 108984778 A CN108984778 A CN 108984778A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- analysis
- intelligent interaction
- answering system
- answer
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses a kind of intelligent interaction automatically request-answering system and its self-teaching methods, wherein system includes analysis of data source module, case study module, knowledge retrieval module and answer generation module.Intelligent interaction automatically request-answering system and self-teaching method of the invention, from the level based on text key word, it is promoted to Knowledge based engineering level, for users, natural language is optimal man-machine interaction mode, automatically request-answering system with natural language come the problem of answering user, it is more friendly than search engine, it is more able to satisfy the demand of user's operation simplification and knowledge precision, realizes the Mailbot of intelligent interaction question and answer.
Description
Technical field
The present invention relates to field intelligent interaction automatic question answering technical field more particularly to a kind of intelligent interaction automatic question answering systems
System and self-teaching method.
Background technique
For users, natural language is optimal man-machine interaction mode, intelligent interaction automatically request-answering system nature language
Speech is answered the problem of user, more friendly than search engine, is more able to satisfy the demand of user's operation simplification and knowledge precision.With
Big data information management, intelligent depth analysis, personal carry the relevant technologies such as smart machine, mobile Internet and application
It advances by leaps and bounds, the form and application scenarios of automatic question answering are undergoing profound change, but also the behavior and demand model of user
Great variation has occurred.By intelligent interaction question answering system from the level based on text key word, is promoted and arrive Knowledge based engineering layer
Realize intelligentized Mailbot in face, it has also become the development trend and target in interaction automatically request-answering system future.
Summary of the invention
The purpose of the present invention is to provide a kind of intelligent interaction automatically request-answering system and its self-teaching methods, from based on pass
The level of keyword is promoted to Knowledge based engineering level, realizes personalized, intelligentized Mailbot.
To achieve the above object, technical scheme is as follows:
A kind of intelligent interaction automatically request-answering system, including analysis of data source module, case study module, knowledge retrieval mould
Block, answer generation module;
Described problem analysis module is based on analysis of data source module to being expressed as matching with data source after case study
Knowledge representation for problem;
The knowledge retrieval module is based on knowledge representation for problem, and relevant knowledge subset is retrieved in knowledge base;
The answer generation module carries out similarity calculation to knowledge subset, dependency analysis, multisource data fusion, knows
Know sequence with merge after carry out answer check and assess after, generation answer.
Further, the analysis of data source module using part of speech analysis, syntactic analysis, semantic analysis, dependency analysis and
The various ways of Entity recognition extract useful information and knowledge from structuring, semi-structured or unstructured data, and
It is expressed as representation of knowledge form.
Further, described problem analysis module carries out semantic understanding to problem using natural language processing technique and divides
Analysis determines expectation answer type by carrying out classification of type to problem.
A kind of self-teaching method of intelligent interaction automatically request-answering system, includes the following steps:
Knowledge identification, by knowledge packaged in knowledge base, carries out analysis retrieval to problem, is confirmed whether there is correlation
Knowledge;
Knowledge creating, when retrieval is less than relevant knowledge or not comprehensive relevant knowledge, the touching of intelligent interaction automatically request-answering system
Hair obtains the demand of new knowledge, carries out analysis to data source and knowledge is extracted;
Knowledge store is stored in knowledge base to the new knowledge of extraction, and the analysis for knowledge cognitive phase is retrieved;
Knowledge sharing, intelligent interaction automatically request-answering system generate answer and share to user;
Knowledge uses, and user solves the problems, such as according to the answer of acquisition, makes decisions, improving efficiency, thinking of promoting innovation;
Knowledge learning is used the feedback of knowledge by user, and is based on automatic question answering appraisal procedure, certainly as intelligent interaction
Dynamic question answering system creation new knowledge or the basis for improving existing knowledge will enter knowledge if knowledge is assessed as valuable
Memory phase, if knowledge be assessed as it is imperfect, back to knowledge identification or the knowledge creating stage.
Intelligent interaction automatically request-answering system and its self-teaching method of the invention, from the level based on text key word,
It is promoted to Knowledge based engineering level, for users, natural language is optimal man-machine interaction mode, and automatically request-answering system is used certainly
Right language is more friendly than search engine come the problem of answering user, is more able to satisfy user's operation and simplifies and knowledge precision
Demand realizes the Mailbot of intelligent interaction question and answer.
Detailed description of the invention
Fig. 1 is the structural block diagram of the intelligent interaction automatically request-answering system of one embodiment of the invention;
Fig. 2 is the recurrent neural networks model figure of the intelligent interaction automatically request-answering system of one embodiment of the invention;
Fig. 3 is the attention model figure of the intelligent interaction automatically request-answering system of one embodiment of the invention.
Specific embodiment
Technical solution of the present invention is described in further detail with reference to the accompanying drawings and examples.
As shown in Figure 1, a kind of intelligent interaction automatically request-answering system of the invention, including analysis of data source module, problem point
Analyse module, knowledge retrieval module and answer generation module.
Wherein, case study module is based on analysis of data source module to being expressed as matching with data source after case study
Knowledge representation for problem, analysis of data source module use part of speech analysis, syntactic analysis, semantic analysis, dependency analysis and Entity recognition
Various ways extract useful information and knowledge, and be expressed as knowing from structuring, semi-structured or unstructured data
Know representation, it includes but not limited to the natural languages such as part of speech analysis, the identification of problem focus that case study module, which mainly uses,
Processing method carries out semantic understanding and analysis to problem, in general, problem can be divided into factoid questions, list problem, assume to ask
Topic, confirmation problem and cause and effect problem etc., problem types will affect the judgement to desired answer type, meanwhile, by the expection of problem
Answer type is divided into such as: people, mechanism, place, quantity, time fundamental type, it is expected that answer type can for knowledge retrieval and
Answer extracting provides reference frame.
Knowledge retrieval module be based on knowledge representation for problem, in knowledge base retrieve relevant knowledge subset, be related to data mining,
A variety of methods such as information retrieval, knowledge retrieval and discovery and technology.
Answer generation module carries out similarity calculation, dependency analysis, multisource data fusion, knowledge row to knowledge subset
Sequence with merge after carry out answer check and assessment after, generate answer, answer may be extraction text or multi-medium data segment.
A kind of self-teaching method of intelligent interaction automatically request-answering system, includes the following steps:
Knowledge identification, by knowledge packaged in knowledge base, carries out analysis retrieval to problem, is confirmed whether there is correlation
Knowledge;
Knowledge creating, when retrieval is less than relevant knowledge or not comprehensive relevant knowledge, the touching of intelligent interaction automatically request-answering system
Hair obtains the demand of new knowledge, carries out analysis to data source and knowledge is extracted;
Knowledge store is stored in knowledge base to the new knowledge of extraction, and the analysis for knowledge cognitive phase is retrieved;
Knowledge sharing, intelligent interaction automatically request-answering system generate answer and share to user;
Knowledge uses, and user solves the problems, such as according to the answer of acquisition, makes decisions, improving efficiency, thinking of promoting innovation;
Knowledge learning is used the feedback of knowledge by user, and is based on automatic question answering appraisal procedure, certainly as intelligent interaction
Dynamic question answering system creation new knowledge or the basis for improving existing knowledge will enter knowledge if knowledge is assessed as valuable
Memory phase, if knowledge be assessed as it is imperfect, back to knowledge identification or the knowledge creating stage.
Intelligent interaction automatically request-answering system of the invention is based on recurrent neural network, the interdependent parsing of Utilizing question sentence
Tree is indicated from word and the phrase layer sentence that faces the problems, and mainly establishes entity topic collection and entity using semantic similarity
The matching relationship of relationship, for the automatic question answering to single relations problems.The model hashes layer, one three layers of neural network by word
It is constituted with semantic layer, as shown in Fig. 2, being each word (such as " cat "), addition head and the tail separator (such as " # in hash layer
Cat# "), then according to triple alphabetical in word (as " and #ca, cat, at# "), word hash is turned into term vector as neural network
Input.Neural network is made of convolutional layer, pooling layers of Max and feedforward layer.Firstly, using word on the part in sentence
Context information obtains one group of local context feature vector by the processing of convolutional layer, partially in following traits space, language
The similar N member phrase (Word-n-grams) of justice will be mapped as similar vector.In view of the meaning of sentence is often depending on
A few keyword extracts most significant local feature vectors using pooling layers of Max, to obtain the complete of regular length
Office's feature vector.Finally, extracting non-linear semantic feature from global characteristics vector by Feedforward Neural Networks network layers.Semantic layer
Actually and a neural network, activation primitive Softmax calculate semantic feature similarity, to obtain problem reality
The similarity of body reference (Entity mentions) and knowledge base entity similarity and relation schema and knowledge base relation.
In the intelligent interaction question answering system for being mostly based on neural network, for from the similarity calculation of different answers,
Problem is represented as identical vector.However, should be calculated and be asked based on attention model dynamic according to the focus of candidate answers
The vector of topic indicates, as shown in figure 3, each vocabulary in problem is shown as term vector (Word embedding) by system, and passes through
Too long short-term memory model (Long short-term memory, LSTM) obtains the sentence vector of problem;Answer is real from answer
Body ae, answer relationship ar, answer type at, answer context ac etc. knowledge based library be expressed as answer vector ee,
er,et,ec;It is finally input with sentence vector sum answer vector, is obtained by attention model.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that the foregoing is merely a specific embodiment of the invention, the guarantor that is not intended to limit the present invention
Range is protected, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all be contained in this hair
Within bright protection scope.
Claims (4)
1. a kind of intelligent interaction automatically request-answering system, which is characterized in that including analysis of data source module, case study module, know
Know retrieval module, answer generation module;
Described problem analysis module is based on analysis of data source module to being expressed as the problem of matching with data source after case study
The representation of knowledge;
The knowledge retrieval module is based on knowledge representation for problem, and relevant knowledge subset is retrieved in knowledge base;
The answer generation module carries out similarity calculation, dependency analysis, multisource data fusion, knowledge row to knowledge subset
Sequence with merge after carry out answer check and assessment after, generation answer.
2. intelligent interaction automatically request-answering system according to claim 1, which is characterized in that the analysis of data source module is adopted
With part of speech analysis, syntactic analysis, semantic analysis, dependency analysis and Entity recognition various ways, from structuring, it is semi-structured or
In unstructured data, useful information and knowledge are extracted, and be expressed as representation of knowledge form.
3. intelligent interaction automatically request-answering system according to claim 1, which is characterized in that described problem analysis module uses
Natural language processing technique carries out semantic understanding and analysis to problem, determines expectation answer class by carrying out classification of type to problem
Type.
4. according to claim 1 to the self-teaching method of intelligent interaction automatically request-answering system described in 3 any one, feature exists
In including the following steps:
Knowledge identification, by knowledge packaged in knowledge base, carries out analysis retrieval to problem, is confirmed whether that there are correlations to know
Know;
Knowledge creating, when retrieval is less than relevant knowledge or not comprehensive relevant knowledge, the triggering of intelligent interaction automatically request-answering system is obtained
The demand for taking new knowledge carries out analysis to data source and knowledge is extracted;
Knowledge store is stored in knowledge base to the new knowledge of extraction, and the analysis for knowledge cognitive phase is retrieved;
Knowledge sharing, intelligent interaction automatically request-answering system generate answer and share to user;
Knowledge uses, and user solves the problems, such as according to the answer of acquisition, makes decisions, improving efficiency, thinking of promoting innovation;
Knowledge learning is used the feedback of knowledge by user, and is based on automatic question answering appraisal procedure, is asked automatically as intelligent interaction
Knowledge store will be entered if knowledge is assessed as valuable by answering system creation new knowledge or improving the basis of existing knowledge
Stage, if knowledge be assessed as it is imperfect, back to knowledge identification or the knowledge creating stage.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810825398.0A CN108984778A (en) | 2018-07-25 | 2018-07-25 | A kind of intelligent interaction automatically request-answering system and self-teaching method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810825398.0A CN108984778A (en) | 2018-07-25 | 2018-07-25 | A kind of intelligent interaction automatically request-answering system and self-teaching method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108984778A true CN108984778A (en) | 2018-12-11 |
Family
ID=64551061
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810825398.0A Pending CN108984778A (en) | 2018-07-25 | 2018-07-25 | A kind of intelligent interaction automatically request-answering system and self-teaching method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108984778A (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109658271A (en) * | 2018-12-19 | 2019-04-19 | 前海企保科技(深圳)有限公司 | A kind of intelligent customer service system and method based on the professional scene of insurance |
CN109885672A (en) * | 2019-03-04 | 2019-06-14 | 中国科学院软件研究所 | A kind of question and answer mode intelligent retrieval system and method towards online education |
CN110276403A (en) * | 2019-06-25 | 2019-09-24 | 北京百度网讯科技有限公司 | Method for establishing model and device |
CN110619042A (en) * | 2019-03-13 | 2019-12-27 | 北京航空航天大学 | Neural network-based teaching question and answer system and method |
CN110727781A (en) * | 2019-10-21 | 2020-01-24 | 国网江苏省电力有限公司电力科学研究院 | Power multi-source knowledge retrieval result fusion method and device |
CN111200737A (en) * | 2019-12-29 | 2020-05-26 | 航天信息股份有限公司企业服务分公司 | Intelligent robot-assisted question answering system and method for live video platform |
CN111813911A (en) * | 2020-06-30 | 2020-10-23 | 神思电子技术股份有限公司 | Knowledge automatic acquisition and updating system based on user supervision feedback and working method thereof |
CN111886601A (en) * | 2019-03-01 | 2020-11-03 | 卡德乐人工智能私人有限公司 | System and method for adaptive question answering |
CN111985238A (en) * | 2020-06-30 | 2020-11-24 | 联想(北京)有限公司 | Answer generation method and equipment |
CN112035627A (en) * | 2020-07-27 | 2020-12-04 | 深圳技术大学 | Automatic question answering method, device, equipment and storage medium |
CN112115275A (en) * | 2020-09-18 | 2020-12-22 | 湖南工学院 | Knowledge graph construction method and system for math tutoring question-answering system |
CN113139036A (en) * | 2020-01-20 | 2021-07-20 | 海信视像科技股份有限公司 | Information interaction method and equipment |
CN113609830A (en) * | 2021-04-07 | 2021-11-05 | 新大陆数字技术股份有限公司 | Literature question-answering method, system and storage medium based on NLP technology |
CN113742462A (en) * | 2020-07-20 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Answer monitoring method and device |
CN116881426A (en) * | 2023-08-30 | 2023-10-13 | 环球数科集团有限公司 | AIGC-based self-explanatory question-answering system |
CN117874208A (en) * | 2024-03-11 | 2024-04-12 | 羚羊工业互联网股份有限公司 | Method for realizing large model memory sharing, knowledge question-answering method and related equipment thereof |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101086843A (en) * | 2006-06-07 | 2007-12-12 | 中国科学院自动化研究所 | A sentence similarity recognition method for voice answer system |
CN101118554A (en) * | 2007-09-14 | 2008-02-06 | 中兴通讯股份有限公司 | Intelligent interactive request-answering system and processing method thereof |
CN101373532A (en) * | 2008-07-10 | 2009-02-25 | 昆明理工大学 | FAQ Chinese request-answering system implementing method in tourism field |
CN104809197A (en) * | 2015-04-24 | 2015-07-29 | 同程网络科技股份有限公司 | On-line question and answer method based on intelligent robot |
CN105893476A (en) * | 2016-03-29 | 2016-08-24 | 上海智臻智能网络科技股份有限公司 | Intelligent questioning and answering method, knowledge base optimization method and device, and intelligent knowledge base |
-
2018
- 2018-07-25 CN CN201810825398.0A patent/CN108984778A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101086843A (en) * | 2006-06-07 | 2007-12-12 | 中国科学院自动化研究所 | A sentence similarity recognition method for voice answer system |
CN101118554A (en) * | 2007-09-14 | 2008-02-06 | 中兴通讯股份有限公司 | Intelligent interactive request-answering system and processing method thereof |
CN101373532A (en) * | 2008-07-10 | 2009-02-25 | 昆明理工大学 | FAQ Chinese request-answering system implementing method in tourism field |
CN104809197A (en) * | 2015-04-24 | 2015-07-29 | 同程网络科技股份有限公司 | On-line question and answer method based on intelligent robot |
CN105893476A (en) * | 2016-03-29 | 2016-08-24 | 上海智臻智能网络科技股份有限公司 | Intelligent questioning and answering method, knowledge base optimization method and device, and intelligent knowledge base |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109658271A (en) * | 2018-12-19 | 2019-04-19 | 前海企保科技(深圳)有限公司 | A kind of intelligent customer service system and method based on the professional scene of insurance |
CN111886601B (en) * | 2019-03-01 | 2024-03-01 | 卡德乐人工智能私人有限公司 | System and method for adaptive question-answering |
CN111886601A (en) * | 2019-03-01 | 2020-11-03 | 卡德乐人工智能私人有限公司 | System and method for adaptive question answering |
CN109885672A (en) * | 2019-03-04 | 2019-06-14 | 中国科学院软件研究所 | A kind of question and answer mode intelligent retrieval system and method towards online education |
CN109885672B (en) * | 2019-03-04 | 2020-10-30 | 中国科学院软件研究所 | Question-answering type intelligent retrieval system and method for online education |
CN110619042A (en) * | 2019-03-13 | 2019-12-27 | 北京航空航天大学 | Neural network-based teaching question and answer system and method |
CN110276403A (en) * | 2019-06-25 | 2019-09-24 | 北京百度网讯科技有限公司 | Method for establishing model and device |
CN110727781A (en) * | 2019-10-21 | 2020-01-24 | 国网江苏省电力有限公司电力科学研究院 | Power multi-source knowledge retrieval result fusion method and device |
CN110727781B (en) * | 2019-10-21 | 2022-11-01 | 国网江苏省电力有限公司电力科学研究院 | Power multi-source knowledge retrieval result fusion method and device |
CN111200737A (en) * | 2019-12-29 | 2020-05-26 | 航天信息股份有限公司企业服务分公司 | Intelligent robot-assisted question answering system and method for live video platform |
CN113139036B (en) * | 2020-01-20 | 2022-07-01 | 海信视像科技股份有限公司 | Information interaction method and equipment |
CN113139036A (en) * | 2020-01-20 | 2021-07-20 | 海信视像科技股份有限公司 | Information interaction method and equipment |
CN111985238A (en) * | 2020-06-30 | 2020-11-24 | 联想(北京)有限公司 | Answer generation method and equipment |
CN111813911A (en) * | 2020-06-30 | 2020-10-23 | 神思电子技术股份有限公司 | Knowledge automatic acquisition and updating system based on user supervision feedback and working method thereof |
CN113742462A (en) * | 2020-07-20 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Answer monitoring method and device |
CN112035627A (en) * | 2020-07-27 | 2020-12-04 | 深圳技术大学 | Automatic question answering method, device, equipment and storage medium |
CN112035627B (en) * | 2020-07-27 | 2023-11-17 | 深圳技术大学 | Automatic question and answer method, device, equipment and storage medium |
CN112115275A (en) * | 2020-09-18 | 2020-12-22 | 湖南工学院 | Knowledge graph construction method and system for math tutoring question-answering system |
CN113609830A (en) * | 2021-04-07 | 2021-11-05 | 新大陆数字技术股份有限公司 | Literature question-answering method, system and storage medium based on NLP technology |
CN116881426A (en) * | 2023-08-30 | 2023-10-13 | 环球数科集团有限公司 | AIGC-based self-explanatory question-answering system |
CN116881426B (en) * | 2023-08-30 | 2023-11-10 | 环球数科集团有限公司 | AIGC-based self-explanatory question-answering system |
CN117874208A (en) * | 2024-03-11 | 2024-04-12 | 羚羊工业互联网股份有限公司 | Method for realizing large model memory sharing, knowledge question-answering method and related equipment thereof |
CN117874208B (en) * | 2024-03-11 | 2024-06-07 | 羚羊工业互联网股份有限公司 | Method for realizing large model memory sharing, knowledge question-answering method and related equipment thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108984778A (en) | A kind of intelligent interaction automatically request-answering system and self-teaching method | |
CN107943998B (en) | Man-machine conversation control system and method based on knowledge graph | |
CN109829039B (en) | Intelligent chat method, intelligent chat device, computer equipment and storage medium | |
CN109635273B (en) | Text keyword extraction method, device, equipment and storage medium | |
CN106934012B (en) | Natural language question-answering implementation method and system based on knowledge graph | |
CN108038234B (en) | Automatic question template generating method and device | |
KR102640564B1 (en) | Question and answer system training device and computer program therefor | |
CN109460459B (en) | Log learning-based dialogue system automatic optimization method | |
CN104679910A (en) | Intelligent answering method and system | |
US20230350929A1 (en) | Method and system for generating intent responses through virtual agents | |
CN110347802B (en) | Text analysis method and device | |
CN111709223B (en) | Sentence vector generation method and device based on bert and electronic equipment | |
CN111353026A (en) | Intelligent law attorney assistant customer service system | |
CN112836030A (en) | Intelligent dialogue system and method | |
CN116010581A (en) | Knowledge graph question-answering method and system based on power grid hidden trouble shooting scene | |
CN116910220A (en) | Multi-round dialogue interaction processing method, device, equipment and storage medium | |
CN116226338A (en) | Multi-round dialogue system and method based on searching and generating fusion | |
Alam et al. | Comparative study of speaker personality traits recognition in conversational and broadcast news speech. | |
CN105631032B (en) | Question and answer Knowledge Base, the apparatus and system recommended based on abstract semantics | |
CN113869040A (en) | Voice recognition method for power grid dispatching | |
CN110931002B (en) | Man-machine interaction method, device, computer equipment and storage medium | |
CN117473034A (en) | Interactive text processing method and device, electronic equipment and storage medium | |
CN115905187B (en) | Intelligent proposition system oriented to cloud computing engineering technician authentication | |
CN111552816A (en) | Dynamic cognitive semantic matching method for big data text mining | |
CN116974554A (en) | Code data processing method, apparatus, computer device and storage medium |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181211 |