CN101086750A - A liverish expert system based on instant message - Google Patents
A liverish expert system based on instant message Download PDFInfo
- Publication number
- CN101086750A CN101086750A CNA2006100518637A CN200610051863A CN101086750A CN 101086750 A CN101086750 A CN 101086750A CN A2006100518637 A CNA2006100518637 A CN A2006100518637A CN 200610051863 A CN200610051863 A CN 200610051863A CN 101086750 A CN101086750 A CN 101086750A
- Authority
- CN
- China
- Prior art keywords
- instant message
- expert
- expert system
- unit
- liverish
- 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
Abstract
The invention relates to a instant liver medical expert system. Patient can get timely information from the long distant service of the expert system, with the system being able to update the expert rules and field knowledge base intelligently.
Description
Technical field
The present invention relates to medical diagnosis, treatment, consulting field and computer information processing field, relate in particular to a kind of liverish expert system based on instant message.
Background technology
Liver is the vitals of human body, and liver diseases (abbreviation hepatopathy) comprises common virus hepatitis, fatty liver, alcoholic liver etc., and especially China is hepatitis B big country, and the patient is numerous.Because hepatopathy is chronic disease often, the patient has a lot of " why " to want the consulting Xiang the doctor, and the progress in diagnosing hepatism and treatment every year is very big clinically, needs the doctor constantly to refresh one's knowledge.Doctor's working time is limited, and a lot of patients can not get the doctor and well serve, and the doctor is also because workload is big and can not well link up with the patient at ordinary times, and the high speed development of modern IT technology is to make the solution of these two problems that better scheme arranged.
Expert system (Expert System) is the computer information processing system of simulation expert decision-making ability, by inference engine of expert system according to information, Expert Rules storehouse and the domain knowledge base of input by analysis reasoning make specialty decision-making, be the most successful application of artificial intelligence.It is not tight that but traditional expert system combines with the internet, can not allow the user pass through the internet and visit expert system anywhere or anytime, can not make full use of internet to go up the knowledge resource of explosive growth.And traditional expert system need make up huge knowledge engineering for maintenance expert's rule base and domain knowledge base, and labor intensive and material resources have influenced promoting the use of of it.
Instant message (Instant Messenger) provides a kind of conveniently exchange way by the internet for the user, and instant message makes online communication that the user can be instant, connects each other fast.Therefore instant message is suitable as the man-machine interface of on-line expert system very much, allows the user visit expert system by the internet quickly and easily.
Summary of the invention
Technical matters to be solved by this invention provides a kind of liverish expert system, can allow the hepatopath use instant message to obtain the remote service of expert system in time by the internet.
Another technical matters to be solved by this invention provides a kind of liverish expert system, and it can upgrade Expert Rules storehouse and domain knowledge base intelligently, thereby reduces the operation expense of expert system, is convenient to penetration and promotion.
In order to solve the problems of the technologies described above, technical scheme of the present invention is:
A kind of liverish expert system based on instant message, comprise instant message processing unit and liverish expert system, it is characterized in that: the patient is undertaken alternately by instant message and liverish expert system, the problem of patient's input passes to the inference machine of liverish expert system through text analyzing and pre-service, inference machine carries out reasoning and provides analysis result under the support of Expert Rules storehouse and domain knowledge base, analysis result passes to the patient by instant message.
Described instant message processing unit comprises the instant message interface unit, and the instant message interface unit has the instant message identification number, and the patient can have access to the instant message processing unit of expert system on the internet by the instant message identification number.Instant message interface unit and patient are undertaken alternately by instant message, patient's problem is passed to question analysis unit, and the answer that the answer extraction unit is provided pass to the patient.
Described instant message processing unit comprises question analysis unit, and patient's problem is carried out text analyzing and pre-service, extracts keyword and passes to liverish expert system.
Described instant message processing unit comprises the answer extraction unit, and the analysis result that liverish expert system provides is analyzed, and extracts optimum answer and passes to the instant message interface unit.
Described expert system comprises inference machine, domain knowledge base, Expert Rules storehouse and meta-knoeledge storehouse.The keyword of inference machine/knowledge conversion unit is converted into domain knowledge by visit meta-knoeledge storehouse with the key to the issue speech that the instant message processing unit transmits.Inference machine visits Expert Rules storehouse and domain knowledge base with domain knowledge, and carries out rational analysis, thereby obtains more effectively analysis result.
Described expert system comprises domain knowledge base, and domain knowledge base is the rational analysis basis of inference machine for inference machine provides the domain knowledge relevant with liver disease.
Described expert system comprises the Expert Rules storehouse, and the Expert Rules storehouse provides the expertise and the dependency rule of rational analysis for inference machine.
Described expert system comprises the meta-knoeledge storehouse, deposits the knowledge about domain knowledge and Expert Rules in the meta-knoeledge storehouse.Expert system is converted into domain knowledge by visit meta-knoeledge storehouse with the key to the issue speech, and inference machine carries out reasoning according to domain knowledge, thereby obtains intelligence, effective analysis result more.
A kind of liverish expert system is characterized in that: expert system has reduced the maintenance workload and the operation expense of expert system by knowledge maintenance unit and regular maintenance unit intelligence, dynamically upgrade domain knowledge base and Expert Rules storehouse.
Described expert system comprises domain knowledge base and knowledge maintenance unit.Vertical search engine is visited intelligently according to the domain knowledge that the meta-knoeledge storehouse provides in the knowledge maintenance unit, and the latest fields knowledge that will retrieve from the internet is updated to the domain knowledge base in time by the domain knowledge bank interface.
Described expert system comprises Expert Rules storehouse and regular maintenance unit.Problem for there not being respective rule in the Expert Rules storehouse manually provides corresponding answer by the hepatopathy expert.Expert's answer extracts Expert Rules by the Expert Rules extraction unit of regular maintenance unit, and is updated in the Expert Rules storehouse by Expert Rules storehouse maintenance interface.
Described expert system comprises the meta-knoeledge storehouse, and meta-knoeledge is extracted intelligently by visit Expert Rules storehouse and domain knowledge base in the meta-knoeledge storehouse.
Technique effect of the present invention is:
(1) this liverish expert system can allow the patient obtain the remote service of expert system easily and efficiently.Its principle is as follows: thus the patient is connected to the instant message identification number visit liverish expert system of liverish expert system by instant message client.The patient imports problem by instant message client, by the instant message processing unit instant message is carried out text analyzing and pre-service, extracts keyword, and liverish expert system is given in transmission.Inference engine of expert system is converted to domain knowledge with keyword, thereby and visits the Expert Rules storehouse and domain knowledge base carries out rational analysis with domain knowledge.Analysis result returns to the patient by the instant message processing unit by instant message.
(2) this liverish expert system can upgrade Expert Rules storehouse and domain knowledge base intelligently. and its principle is as follows: vertical search engine is visited intelligently according to the domain knowledge that the meta-knoeledge storehouse provides in the knowledge maintenance unit, and the latest fields knowledge that will retrieve from the internet is updated to the domain knowledge base in time by the domain knowledge bank interface.Problem for there not being respective rule in the Expert Rules storehouse manually provides corresponding answer by the hepatopathy expert.Expert's answer extracts Expert Rules by the Expert Rules extraction unit of regular maintenance unit, and is updated in the Expert Rules storehouse by Expert Rules storehouse maintenance interface.
Description of drawings
The invention will be further described below in conjunction with the drawings and specific embodiments.
Fig. 1 is based on the synoptic diagram of the liverish expert system of instant message
Fig. 2 is the structural representation of question analysis unit
Fig. 3 is the structural representation of answer extraction unit
Fig. 4 is the structural representation of liverish expert system
Fig. 5 is the structural representation of Expert Rules extraction unit
Embodiment
Fig. 1 is the synoptic diagram that the present invention is based on the liverish expert system of instant message, and this invention comprises that instant message processing unit 1 and expert system 2. instant message processing units 1 comprise instant message interface unit 11, question analysis unit 12 and answer extraction unit 13.Expert system 2 comprises inference engine of expert system 21, knowledge maintenance unit 22, domain knowledge base 23, meta-knoeledge storehouse 24, Expert Rules storehouse 25 and regular maintenance unit 26.Instant message interface unit 11 has the instant message identification number that can be provided with, and user's instant message client just can be connected to instant message interface unit 11 by visiting this instant message identification number.The user imports problem at instant message client, passes to instant message interface unit 11 by instant message.Instant message interface unit 11 passes to question analysis unit 12 with the instant message that receives, carry out text analyzing and pre-service by 12 pairs of instant messages of question analysis unit, pre-service comprises to be carried out the problem classification and extracts keyword instant message, and question analysis unit 12 passes to inference engine of expert system 21 with pretreated result.Inference engine of expert system 21 is converted to relevant domain knowledge by visit meta-knoeledge storehouse 24 with keyword, and visits domain knowledge base 23 and Expert Rules storehouse 25 with these domain knowledges and problem category.Inference engine of expert system 21 carries out rational analysis under the driving of the relevant knowledge in domain knowledge base 23 and Expert Rules storehouse 25, provide analysis result and analysis result is passed to answer extraction unit 13.13 pairs of analysis results of answer extraction unit carry out the problem relevancy ranking and select optimum answer passing to instant message interface unit 11, return to the user by instant message.
Fig. 2 is the structural representation of question analysis unit 12, and question analysis unit 12 comprises Chinese word segmentation unit 31, part-of-speech tagging unit 32, problem taxon 33 and keyword extracting unit 34.Question analysis unit is to carrying out text analyzing and pre-service by the user by the problem of instant message input.Chinese word segmentation unit 31 carries out Chinese word segmentation to problem under the driving of specialized dictionary and general dictionary handles, and question text is decomposed into Chinese word and phrase.Chinese word segmentation unit 31 can adopt latent Marko husband (HHMM) model of multilayer or shortest path model.The Chinese word segmentation result passes to part-of-speech tagging unit 32 and carries out part-of-speech tagging, and part-of-speech tagging is followed " modern Chinese texts cutting and the part-of-speech tagging standard V1.0 " that Peking University formulated in 1994.Keyword extracting unit 34 is according to the result of part-of-speech tagging, with technical term and predicate as keyword.Problem taxon 33 is classified problem by interrogative according to the result of part-of-speech tagging, can be divided into following a few class:
(1) how/how How
(2) Why why
(3) How many how much
(4) what What is
(5) When
(6) Where where
(7) Who who
(8) problem that can not sort out of Other
Fig. 3 is the structural representation of answer extraction unit 13, comprises answer modular unit 41 and answer selected cell 42.The analysis result of liverish expert system passes to answer extraction unit 13, selects corresponding answer template by answer modular unit 41 according to problem category, and analysis result is organized into answer.Answer selected cell 42 sorts answer according to the degree of correlation with problem, select optimum answer.
Fig. 4 is the structural representation of expert system 2, comprises inference engine of expert system 21, knowledge maintenance unit 22, domain knowledge base 23, meta-knoeledge storehouse 24, Expert Rules storehouse 25 and regular maintenance unit 26.The keyword of inference engine of expert system 21/knowledge converting unit 211 is visited meta-knoeledge storehouse 24 according to the pre-service result that instant message processing unit 1 transmits, keyword is converted to the domain knowledge relevant with hepatopathy, then with domain knowledge visit domain knowledge base 23 and Expert Rules storehouse 25.Inference engine of expert system 21 carries out rational analysis under the driving in domain knowledge base 23 and Expert Rules storehouse 25, provide analysis result.Meta-knoeledge storehouse 24 comprises meta-knoeledge extraction unit 241, and meta-knoeledge extraction unit 241 is by visit domain knowledge base 23 and Expert Rules storehouse 25, extracts about the meta-knoeledge of domain knowledge and Expert Rules and upgrades meta-knoeledge storehouse 24.Rule maintenance unit 26 comprises Expert Rules storehouse maintenance interface 261 and Expert Rules extraction unit 262.Problem for there not being respective rule in the Expert Rules storehouse 25 manually provides corresponding answer by the hepatopathy expert.Expert's answer extracts Expert Rules by the Expert Rules extraction unit 262 of regular maintenance unit 26, and is updated in the Expert Rules storehouse 25 by Expert Rules storehouse maintenance interface 261.Knowledge maintenance unit 22 comprises domain knowledge base maintenance interface 221 and vertical search engine 222.Knowledge maintenance unit 22 drives vertical search engine 222 according to the domain knowledge in the meta-knoeledge storehouse 24, the network of relation resource in vertical search engine 222 search hepatopathy fields also extracts domain knowledge from Search Results, be updated in the domain knowledge base 23 by domain knowledge base maintenance interface 221.
Fig. 5 is the structural representation of Expert Rules extraction unit 262, for the problem that does not have the respective handling rule in the Expert Rules storehouse 25, manually provide corresponding answer by the hepatopathy expert, problem and expert's answer passes to regular maintenance unit 26, also therefrom extracts Expert Rules by 262 problem analyses of Expert Rules extraction unit and expert's answer.Expert Rules extraction unit 262 comprises Chinese word segmentation unit 51, part-of-speech tagging unit 52, problem taxon 53, keyword extracting unit 54, rule template 55 and regular generation unit 56.Problem and expert's answer is decomposed into Chinese word and phrase by Chinese word segmentation unit 51, mark out the grammatical item of sentence then by part-of-speech tagging unit 52, keyword extracting unit 54 is come out technical term and predicate according to the result of part-of-speech tagging as keyword extraction.Problem taxon 53 is classified problem by interrogative according to the result of part-of-speech tagging.Rule template 55 provides corresponding Expert Rules template according to problem category, and regular generation unit 56 generates Expert Rules under the driving of Expert Rules template and keyword, and is updated in the Expert Rules storehouse 25.
In a most preferred embodiment of the present invention, should comprise instant message processing unit 1 and expert system 2 based on the liverish expert system of instant message.Instant message processing unit 1 is connected with user's instant message client software by the internet, and the user is by the service of instant message visit expert system.
The method of operating of this most preferred embodiment is as follows: thus user's instant message client software is by the instant message identification number visit liverish expert system of Internet connection liverish expert system.The user imports information such as medical history, analysis data, problem at instant message client, pass to liverish expert system by instant message, is handled by instant message processing unit 1.The instant message interface unit 11 of instant message processing unit 1 receives the instant message that users transmit and passes to question analysis unit 12 and carries out pre-service, comprises problem classification and keyword extraction.The Chinese word segmentation unit 31 of question analysis unit 12 carries out Chinese word segmentation to instant message under the driving of general dictionary and specialized dictionary handles, and instant message is resolved into Chinese word and phrase.Part-of-speech tagging unit 32 marks out the grammatical item of speech and phrase, and problem taxon 33 is classified problem according to interrogative, and keyword extracting unit 34 is come out technical term and predicate as keyword extraction.The pre-service result of question analysis unit 12 passes to expert system 2, the keyword of inference engine of expert system 21/knowledge converting unit 211 pre-service results visit meta-knoeledge storehouse 24, keyword is converted to the domain knowledge relevant with hepatopathy, then with domain knowledge visit domain knowledge base 23 and Expert Rules storehouse 25.Inference engine of expert system 21 carries out rational analysis under the driving in domain knowledge base 23 and Expert Rules storehouse 25, provide analysis result.The analysis result of expert system passes to answer extraction unit 13, selects corresponding answer template by answer modular unit 41 according to problem category, and analysis result is organized into answer.Answer selected cell 42 sorts answer according to the degree of correlation with problem, select optimum answer and pass to the user by instant message interface unit 11.
In this most preferred embodiment, the problem for there not being the respective handling rule in the Expert Rules storehouse 25 manually provides corresponding answer by the hepatopathy expert, passes to the user by instant message processing unit 1.Problem and expert's answer passes to regular maintenance unit 26, by 262 problem analyses of Expert Rules extraction unit and expert answer, therefrom extracts Expert Rules and is updated in the Expert Rules storehouse.Knowledge maintenance unit 22 drives vertical search engine 222 according to the domain knowledge in the meta-knoeledge storehouse 24, the network of relation resource in vertical search engine 222 search hepatopathy fields also extracts domain knowledge from Search Results, be updated in the domain knowledge base 23 by domain knowledge base maintenance interface 221.
Claims (6)
1, a kind of liverish expert system based on instant message, comprise instant message processing unit (1) and expert system (2), it is characterized in that: the user visits liverish expert system by instant message, the user passes to the instant message processing unit (1) of liverish expert system through instant message in instant message client input relevant information, instant message processing unit (1) passes to expert system (2) after instant message is carried out pre-service, expert system (2) is made analysis, and analysis result passes to instant message processing unit (1) back and returns to the user by instant message.
2, according to claim 1 described liverish expert system based on instant message, it is characterized in that: instant message is as the interface of visit expert system.
3, instant message processing unit according to claim 1 (1) is characterized in that: instant message is carried out pre-service, with problem classification and extract keyword.
4, according to claim 1 described expert system (2), it is characterized in that: inference engine of expert system (21) is made rational analysis to user's problem under the driving of hepatopathy domain knowledge Kuku (23), meta-knoeledge storehouse (24) and Expert Rules storehouse (25).
5, according to claim 4 described expert systems (2), it is characterized in that: upgrade non-existent Expert Rules in the Expert Rules storehouse (25) by regular maintenance unit (26).
6, according to claim 4 described expert systems (2), it is characterized in that: drive vertical search engine (222) by domain knowledge, the association area on the internet is updated one's knowledge in the domain knowledge base (23).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006100518637A CN101086750A (en) | 2006-06-09 | 2006-06-09 | A liverish expert system based on instant message |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2006100518637A CN101086750A (en) | 2006-06-09 | 2006-06-09 | A liverish expert system based on instant message |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101086750A true CN101086750A (en) | 2007-12-12 |
Family
ID=38937704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2006100518637A Pending CN101086750A (en) | 2006-06-09 | 2006-06-09 | A liverish expert system based on instant message |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101086750A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102083116A (en) * | 2011-01-10 | 2011-06-01 | 浪潮通信信息系统有限公司 | Method for constructing intelligent communication alarm fault positioning expert system based on meta-knowledge |
| CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
| CN105335488A (en) * | 2015-10-16 | 2016-02-17 | 中国南方电网有限责任公司电网技术研究中心 | Knowledge base construction method |
| CN105843962A (en) * | 2016-04-18 | 2016-08-10 | 百度在线网络技术（北京）有限公司 | Information processing and displaying methods, information processing and displaying devices as well as information processing and displaying system |
| CN106339602A (en) * | 2016-08-26 | 2017-01-18 | 丁腊春 | Intelligent consulting robot |
| WO2018157349A1 (en) * | 2017-03-02 | 2018-09-07 | 深圳前海达闼云端智能科技有限公司 | Method for interacting with robot, and interactive robot |
-
2006
- 2006-06-09 CN CNA2006100518637A patent/CN101086750A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102083116A (en) * | 2011-01-10 | 2011-06-01 | 浪潮通信信息系统有限公司 | Method for constructing intelligent communication alarm fault positioning expert system based on meta-knowledge |
| CN102083116B (en) * | 2011-01-10 | 2014-08-20 | 浪潮通信信息系统有限公司 | Method for constructing intelligent communication alarm fault positioning expert system based on meta-knowledge |
| CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
| WO2017041372A1 (en) * | 2015-09-07 | 2017-03-16 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
| CN105068661B (en) * | 2015-09-07 | 2018-09-07 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method based on artificial intelligence and system |
| CN105335488A (en) * | 2015-10-16 | 2016-02-17 | 中国南方电网有限责任公司电网技术研究中心 | Knowledge base construction method |
| CN105843962A (en) * | 2016-04-18 | 2016-08-10 | 百度在线网络技术（北京）有限公司 | Information processing and displaying methods, information processing and displaying devices as well as information processing and displaying system |
| CN106339602A (en) * | 2016-08-26 | 2017-01-18 | 丁腊春 | Intelligent consulting robot |
| CN106339602B (en) * | 2016-08-26 | 2019-02-26 | 丁腊春 | A kind of intelligent consulting robot |
| WO2018157349A1 (en) * | 2017-03-02 | 2018-09-07 | 深圳前海达闼云端智能科技有限公司 | Method for interacting with robot, and interactive robot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Al-Moslmi et al. | Named entity extraction for knowledge graphs: A literature overview | |
| CN105843781B (en) | For improving the method and system of the matrix sort of buffer efficiency | |
| CN103823857B (en) | Space information searching method based on natural language processing | |
| Kawtrakul | Ontology engineering and knowledge services for agriculture domain | |
| CN105138864B (en) | Protein interactive relation data base construction method based on Biomedical literature | |
| CN104573006A (en) | Construction method of public health emergent event domain knowledge base | |
| CN102270212A (en) | User interest feature extraction method based on hidden semi-Markov model | |
| Villaverde et al. | Supporting the discovery and labeling of non-taxonomic relationships in ontology learning | |
| CN101086750A (en) | A liverish expert system based on instant message | |
| CN104281702A (en) | Power keyword segmentation based data retrieval method and device | |
| CN105740227A (en) | Genetic simulated annealing method for solving new words in Chinese segmentation | |
| CN112989208A (en) | Information recommendation method and device, electronic equipment and storage medium | |
| Kanagarajan et al. | Intelligent sentence retrieval using semantic word based answer generation algorithm with cuckoo search optimization | |
| CN107480137A (en) | With semantic iterative extraction network accident and the method that identifies extension event relation | |
| CN104391969A (en) | User query statement syntactic structure determining method and device | |
| Hsu et al. | A hybrid Latent Dirichlet Allocation approach for topic classification | |
| Vossen et al. | KYOTO: an open platform for mining facts | |
| Aracri et al. | Integrated use of KOS and deep learning for data set annotation in tourism domain | |
| An et al. | Exploring pre-trained language models to build knowledge graph for metal-organic frameworks (mofs) | |
| Ciravegna et al. | LODIE: Linked Open Data for Web-scale Information Extraction. | |
| Manna et al. | CookingQA: A question answering system based on cooking ontology | |
| Faralli et al. | DWS at the 2016 open knowledge extraction challenge: a hearst-like pattern-based approach to hypernym extraction and class induction | |
| Talita et al. | Challenges in building domain ontology for minority languages | |
| Ben Mustapha et al. | A dynamic composition of ontology modules approach: application to web query reformulation | |
| Wang et al. | Evolution of the Internet’s Support for Chinese Enterprises’ Innovation Based on Big Data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |