CN107193873A - A kind of network search method based on semantic network technology - Google Patents
A kind of network search method based on semantic network technology Download PDFInfo
- Publication number
- CN107193873A CN107193873A CN201710248027.6A CN201710248027A CN107193873A CN 107193873 A CN107193873 A CN 107193873A CN 201710248027 A CN201710248027 A CN 201710248027A CN 107193873 A CN107193873 A CN 107193873A
- Authority
- CN
- China
- Prior art keywords
- concept
- search
- user
- target
- semantic
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- 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
- G06F16/3332—Query translation
- G06F16/3334—Selection or weighting of terms from queries, including natural language queries
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of network search method based on semantic network technology in the design and shipbuilding field of LanExplorer engine, comprising the following steps that for the network search method based on semantic network technology is somebody's turn to do：S1：Build user context model；S2：Situational model is integrated with search engine；S3：Calculate the similarity of the crucial term vector of keyword vector sum concept；S4：Target concept is matched, target concept collection is obtained；S5：The frequency of candidate keywords is defined；S6：Target concept is clustered, the present invention integrates user context model and search engine, analysis and utilization to user context information in information retrieval service are greatly reinforced, using semantic network technology, it can be good at making computer and Swarm Cooperation completion work, the description that structure and features to user context information is formalized, is translated into the intelligible form of computer, and used by search engine.
Description
Technical field
It is specially that one kind is based on semantic network technology the present invention relates to the design and shipbuilding field of LanExplorer engine
Network search method.
Background technology
The main instrument of Query Information is search engine to current user on the internet.Intuitively say, search engine is one
The software systems run on the internet are covered, it is searched according to certain strategy, with specific computer program from internet
Collect information, tissue is being carried out to information with after processing, specific user interface presentation will be passed through with the relevant information of user search
To user.
Nowadays, information technology is developed rapidly, and provides broader shared platform for people, network retrieval into
The conventional channel of information is obtained for people, people retrieve the information of correlation by information retrieval tool, and this is to a certain extent
The problem of solving the classification and retrieval of resource.But traditional retrieval mode based on keyword, due to have ignored keyword
The semantic information itself contained, and relatively low recall ratio and precision ratio is obtained, become increasingly to be not suitable with the demand of retrieval.For
This, comes into operation we have proposed a kind of network search method based on semantic network technology, to solve the above problems.
The content of the invention
It is an object of the invention to provide a kind of network search method based on semantic network technology, to solve above-mentioned background skill
The traditional retrieval mode based on keyword proposed in art, by have ignored the semantic information that keyword is contained in itself, and
Obtain relatively low recall ratio and precision ratio, the problem of becoming increasingly to be not suitable with the demand of retrieval.
To achieve the above object, the present invention provides following technical scheme：A kind of web search side based on semantic network technology
Method, is somebody's turn to do comprising the following steps that for the network search method based on semantic network technology：
S1：Collect the base in terms of user's search preferences, custom, target, psychology, individual character, knowledge, behavior, specialty and creation
Plinth information, builds user context model；
S2：User context module is integrated with existing search engine, it is determined that search point to, search output item,
The entered taxonomy database of output interface, search, the property of search define the quantitative and qualitative with search result；
S3：The searching request that user inputs is transferred in the neighbor node of agent node by network agent node, and
The similarity between the crucial term vector of keyword vector sum concept that user inputs is calculated by Ontology Mapping；
S4：The crucial term vector that user inputs is sent in Ontology by search engine, using each domain body mould
The mode of block parallel inference, is matched to target concept, obtains target concept collection；
S5：Each concept is concentrated to target concept, if some keywords of user's input do not appear in the key of the concept
In term vector, then these keywords will add 1 as the candidate keywords of the concept or by the frequency of correspondence candidate keywords, when certain
When the frequency of candidate keywords reaches boundary value, it will be added into the crucial term vector of the concept；
S6：Target concept is clustered, cluster result is uploaded in semantic base, and feeds back to user, is easy to user
Quickly find file interested.
It is preferred that, in the step S1, user context model can analyze the short-term interest of user, Long-term Interest and its dynamic
The interests change of state, and it is stored, represented and described.
It is preferred that, in the step S4, target concept is the Similarity value between the crucial term vector of keyword vector sum concept
Maximum concept, and be extended search by other keywords of target concept or utilize nearer with target concept semantic distance
The keyword of concept be extended search.
It is preferred that, in the step S5, the frequency boundary value of keyword is designated as 3.
It is preferred that, in the step S6, in the cluster process of target concept, each ancestor concept is found by target concept,
Document under identical concept is polymerized to a major class, and common ancestor's concept according to concept or belongs to identical concept jointly and is polymerized to
One bigger classification, forms the result of multi-level clustering.
Compared with prior art, the beneficial effects of the invention are as follows：The present invention is integrated by user context model and search engine
Together, analysis and utilization to user context information in information retrieval service have been greatly reinforced, so that search engine
Search result makes search starting point from shared large-scale crawl database to again from the stereotyped to variation transformation of current output item
Miscellaneous taxonomy database and the transformation of search procedure database, using semantic network technology, can be good at making computer and Swarm Cooperation
Work is completed, application semantics net and ontology are handled collected user context information, to user context information
The description that is formalized of structure and features, be translated into the intelligible form of computer, and used by search engine,
Also allowing for each inter-entity reaches common semantic understanding to contextual information simultaneously, so that using already present body to contextual information
Make inferences.
Brief description of the drawings
Fig. 1 is workflow diagram of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
Referring to Fig. 1, the present invention provides a kind of technical scheme：A kind of network search method based on semantic network technology, should
Network search method based on semantic network technology is comprised the following steps that：
S1：Collect the base in terms of user's search preferences, custom, target, psychology, individual character, knowledge, behavior, specialty and creation
Plinth information, builds user context model, and user context model can analyze the short-term interest of user, Long-term Interest and its dynamic
Interests change, and it is stored, represented and described；
S2：User context module is integrated with existing search engine, it is determined that search point to, search output item,
The entered taxonomy database of output interface, search, the property of search define the quantitative and qualitative with search result；
S3：The searching request that user inputs is transferred in the neighbor node of agent node by network agent node, and
The similarity between the crucial term vector of keyword vector sum concept that user inputs is calculated by Ontology Mapping；
S4：The crucial term vector that user inputs is sent in Ontology by search engine, using each domain body mould
The mode of block parallel inference, is matched to target concept, obtains target concept collection, and target concept is keyword vector sum concept
The concept of Similarity value maximum between crucial term vector, and be extended search by other keywords of target concept or utilize
The keyword of nearer concept is extended search with target concept semantic distance；
S5：Each concept is concentrated to target concept, if some keywords of user's input do not appear in the key of the concept
In term vector, then these keywords will add 1 as the candidate keywords of the concept or by the frequency of correspondence candidate keywords, when certain
When the frequency of candidate keywords reaches boundary value, it will be added into the crucial term vector of the concept, the frequency boundary of keyword
Value is designated as 3；
S6：Target concept is clustered, cluster result is uploaded in semantic base, and feeds back to user, is easy to user
File interested is quickly found, in the cluster process of target concept, each ancestor concept is found by target concept, will be same
Document under concept is polymerized to a major class, and common ancestor's concept according to concept or belongs to identical concept jointly and be polymerized to one more
Big classification, forms the result of multi-level clustering.
General thinks, the process of an Ontology Mapping should include following several parts：(1) body is standardized；(2) it is similar
The extraction of degree；(3) Semantic mapping；(4) mapping is performed；(5) mapping post processing, during the extraction of wherein similarity is Ontology Mapping
One most important step, is exactly mainly the calculating for carrying out similarity, in the present invention by the method and profit of Case-based Reasoning
It is combined together and is learnt from other's strong points to offset one's weaknesses with the method for heuristic rule, and corresponding weights is set to two methods.Example calculation phase
Like the method for degree, to be the Joint Distribution probability that occurs using a certain amount of example in two concepts calculate two concepts
Similarity, for an example, similarity is calculated using Jaccard coefficients, and the calculation formula of the coefficient isConcept C is calculated with instantiation1And C2
Similarity, be designated as Siminstance(C1,C2), then calculation formula is
Concept in body is layering, and body can also regard one concept of each node on behalf in a conceptional tree, tree as, heuristic
Rule can be made with some rules defined in the properties of conceptional tree, some field axioms and domain expert for foundation
Fixed, the similarity obtained by calculating is designated as Simrule(C1,C2).The similarity of Case-based Reasoning and rule-based similarity are pressed into phase
The weight answered is merged, then concept C1With concept C2Between semantic similarity can represent Sim with equation belowsemantic(C1,C2)
=WinstanceSiminstance(C1,C2)+WruleSimrule(C1,C2), wherein Winstance+Wrule=1.Due to the diversity of body
With easy structure, the bridge of Semantic mapping must be erected between the bodies by wanting to complete the task of information interchange, and Ontology Mapping is present
It has been a major issue present in semantic net evolution, the method for the Ontology Mapping proposed in the present invention, which is integrated, to be changed
A variety of methods in terms of mapping similarity measure are entered, have there is good efficiency and the degree of accuracy, its weight is usually according to existing warp
Test and provide.Search Requirement is navigated in suitable data source on semantic network platform and performs retrieval service by the present invention, so that
It is further to improve precision ratio and recall ratio, to realize Internet resources precise search, effectively using network information resource, eliminate
Internet resources isolated island.
Although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of changes, modification can be carried out to these embodiments, replace without departing from the principles and spirit of the present invention by understanding
And modification, the scope of the present invention is defined by the appended.
Claims (5)
1. a kind of network search method based on semantic network technology, it is characterised in that：The web search based on semantic network technology
Method is comprised the following steps that：
S1：Collect the basis letter in terms of user's search preferences, custom, target, psychology, individual character, knowledge, behavior, specialty and creation
Breath, builds user context model；
S2：User context module is integrated with existing search engine, it is determined that search is pointed to, searches for output item, output
The entered taxonomy database of interface, search, the property of search define the quantitative and qualitative with search result；
S3：The searching request that user inputs is transferred in the neighbor node of agent node by network agent node, and passed through
Ontology Mapping calculates the similarity between the crucial term vector of keyword vector sum concept of user's input；
S4：The crucial term vector that user inputs is sent in Ontology by search engine, using each domain body module simultaneously
The mode of row reasoning, is matched to target concept, obtains target concept collection；
S5：Each concept is concentrated to target concept, if some keywords of user's input do not appear in the keyword of the concept to
In amount, then these keywords will add 1 as the candidate keywords of the concept or by the frequency of correspondence candidate keywords, as certain candidate
When the frequency of keyword reaches boundary value, it will be added into the crucial term vector of the concept；
S6：Target concept is clustered, cluster result is uploaded in semantic base, and feeds back to user, is easy to user quick
Find file interested.
2. a kind of network search method based on semantic network technology according to claim 1, it is characterised in that：The step
In S1, user context model can analyze the short-term interest, Long-term Interest and its dynamic interests change of user, and it is carried out
Storage, expression and description.
3. a kind of network search method based on semantic network technology according to claim 1, it is characterised in that：The step
In S4, target concept is the maximum concept of Similarity value between the crucial term vector of keyword vector sum concept, and general by target
Other keywords read are extended search or are extended using the keyword with the nearer concept of target concept semantic distance
Search.
4. a kind of network search method based on semantic network technology according to claim 1, it is characterised in that：The step
In S5, the frequency boundary value of keyword is designated as 3.
5. a kind of network search method based on semantic network technology according to claim 1, it is characterised in that：The step
In S6, in the cluster process of target concept, each ancestor concept is found by target concept, the document under identical concept is polymerized to one
Individual major class, and common ancestor's concept according to concept or belong to identical concept jointly and be polymerized to a bigger classification, form multilayer
The result of cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710248027.6A CN107193873A (en) | 2017-04-17 | 2017-04-17 | A kind of network search method based on semantic network technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710248027.6A CN107193873A (en) | 2017-04-17 | 2017-04-17 | A kind of network search method based on semantic network technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107193873A true CN107193873A (en) | 2017-09-22 |
Family
ID=59872003
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710248027.6A Pending CN107193873A (en) | 2017-04-17 | 2017-04-17 | A kind of network search method based on semantic network technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107193873A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109033372A (en) * | 2018-07-27 | 2018-12-18 | 北京未来媒体科技股份有限公司 | A kind of content information retrieval method and system based on artificial intelligence |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007085528A1 (en) * | 2006-01-26 | 2007-08-02 | International Business Machines Corporation | Knowledge management system, program product and method |
| KR20090033149A (en) * | 2007-09-27 | 2009-04-01 | 주식회사 시맨틱스 | Semantic web based index method and search engine using the same |
| CN103064945A (en) * | 2012-12-26 | 2013-04-24 | 吉林大学 | Situation searching method based on body |
| CN104050243A (en) * | 2014-05-28 | 2014-09-17 | 黄斌 | Network searching method and system combined with searching and social contact |
| CN104090958A (en) * | 2014-07-04 | 2014-10-08 | 许昌学院 | Semantic information retrieval system and method based on domain ontology |
-
2017
- 2017-04-17 CN CN201710248027.6A patent/CN107193873A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007085528A1 (en) * | 2006-01-26 | 2007-08-02 | International Business Machines Corporation | Knowledge management system, program product and method |
| KR20090033149A (en) * | 2007-09-27 | 2009-04-01 | 주식회사 시맨틱스 | Semantic web based index method and search engine using the same |
| CN103064945A (en) * | 2012-12-26 | 2013-04-24 | 吉林大学 | Situation searching method based on body |
| CN104050243A (en) * | 2014-05-28 | 2014-09-17 | 黄斌 | Network searching method and system combined with searching and social contact |
| CN104090958A (en) * | 2014-07-04 | 2014-10-08 | 许昌学院 | Semantic information retrieval system and method based on domain ontology |
Non-Patent Citations (3)
| Title |
|---|
| 史斌: "面向语义网的语义搜索引擎关键技术研究", 《中国博士学位论文全文数据库信息科技辑》 * |
| 吴芳: "基于用户情境及语义网技术的个性化搜索引擎", 《江西图书馆学刊》 * |
| 李春苗: "语义网中基于相似度计算的本体映射研究_李春苗", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109033372A (en) * | 2018-07-27 | 2018-12-18 | 北京未来媒体科技股份有限公司 | A kind of content information retrieval method and system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7844592B2 (en) | Ontology-content-based filtering method for personalized newspapers | |
| Abbas et al. | A literature review on the state-of-the-art in patent analysis | |
| US9715493B2 (en) | Method and system for monitoring social media and analyzing text to automate classification of user posts using a facet based relevance assessment model | |
| US8341159B2 (en) | Creating taxonomies and training data for document categorization | |
| Batsakis et al. | Improving the performance of focused web crawlers | |
| CN108846029B (en) | Information correlation analysis method based on knowledge graph | |
| CN103838833A (en) | Full-text retrieval system based on semantic analysis of relevant words | |
| US9971828B2 (en) | Document tagging and retrieval using per-subject dictionaries including subject-determining-power scores for entries | |
| US20030212663A1 (en) | Neural network feedback for enhancing text search | |
| US20110191335A1 (en) | Method and system for conducting legal research using clustering analytics | |
| CN112559684A (en) | Keyword extraction and information retrieval method | |
| Ahmadov et al. | Towards a hybrid imputation approach using web tables | |
| KR20120092756A (en) | Method and system for searching mobile application using human activity knowledge database | |
| Calado et al. | A Bayesian network approach to searching Web databases through keyword-based queries | |
| Vijaya et al. | LionRank: lion algorithm-based metasearch engines for re-ranking of webpages | |
| CN107193873A (en) | A kind of network search method based on semantic network technology | |
| CN111666420B (en) | Method for intensively extracting experts based on subject knowledge graph | |
| Wang et al. | Research on discovering deep web entries | |
| Lu et al. | Influence model of paper citation networks with integrated pagerank and HITS | |
| Ganta et al. | Search engine optimization through spanning forest generation algorithm | |
| Meng et al. | Personalized Semantic Retrieval System based on Statistical Language Model | |
| Liu et al. | Research on learning video recommendation system based on DBSCAN clustering algorithm | |
| Tang et al. | An Agricultural Information Recommendation Method Based on Matrix Decomposition Knowledge Graph Algorithm | |
| Wu et al. | A personalized intelligent web retrieval system based on the knowledge-base concept and latent semantic indexing model | |
| Wei et al. | Product information retrieval based on opinion mining |
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 |
Application publication date: 20170922 |
|
| RJ01 | Rejection of invention patent application after publication |