期刊:
PROCEEDINGS OF THE 2010 INTERNATIONAL CONFERENCE ON MECHANICAL, INDUSTRIAL, AND MANUFACTURING TECHNOLOGIES (MIMT 2010),2010年:125-130
通讯作者:
Wang, Jinqiao
作者机构:
[Wang, Jinqiao; Zhu, Li; Yang, Qing; Sun, JunLi] Huazhong Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.
通讯机构:
[Wang, Jinqiao] H;Huazhong Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.
会议名称:
International Conference on Mechanical, Industrial, and Manufacturing Technologies
会议时间:
JAN 22-24, 2010
会议地点:
Sanya, PEOPLES R CHINA
会议主办单位:
[Wang, Jinqiao;Yang, Qing;Sun, JunLi;Zhu, Li] Huazhong Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.
关键词:
Reinforcement learning;Association rules;Topological graph;Recommendation Systems data mining
摘要:
Reinforcement learning is an important method of machine learning. This paper using the graph theory to express varieties of knowledge points, which their's relationship is expressed by the graph of topological graph. Applied the Technology of association rule Recommendation to deal with the relationship between these knowledge points, give the corresponding of the recommendation work flow chart. In the paper data tables used to store the knowledge points, the algorithm to demonstrate the technical of association rule Recommendation feasibility and rationality.
期刊:
2010 International Conference of Information Science and Management Engineering,2010年1:230-232
作者机构:
Department of Computer Science, Hua Zhong Normal University, Wuhan, China;Overseas Chinese College, Capital University of Economics and Business, Beijing, China
会议名称:
2010 International Conference of Informationa Science and Management Engineering(2010年信息科学与管理工程国际学术会议 ISME 2010)
会议时间:
2010-08-07
会议地点:
西安
会议论文集名称:
2010 International Conference of Informationa Science and Management Engineering(2010年信息科学与管理工程国际学术会议 ISME 2010)论文集
关键词:
area concept extraction;association rule;sememe
摘要:
Ontology learning is from a given area document sets automatic or semi-automatic extraction terms to construct a domain ontology. Area concept extraction is one of the most important aspects in building ontology. In this paper, we proposed an improved area concept extraction algorithm. In the algorithm, we firstly employed association rule algorithm to obtain the similarity between the sememes, and then used the similarity between the sememes to find the similarity between area concepts. Finally our paper achieves the whole area concepts extraction process. By analyzing the experimental results shows the effectiveness and correctness of the algorithm.
摘要:
Ontology learning is a technology. Ontology learning can be used to establish ontology automatically or semi-automatically by introducing the ontology engineering and machine learning technology and many other sciences and technologies. The ontology learning technology which is proposed in our paper is to reduce the time of building an entire ontology. Our paper presents an Ontology Learning model which will enhance the efficiency of extraction concept, and enhance the efficiency of ontology building. It includes several aspects, and area concept extraction is the main aspect of all. The model combines personalized recommendation with concept extraction and realizes a more accurate and stable domain concept extraction method. We describe these techniques and report the results of the experiment examining its effectiveness and efficiency.
摘要:
Nowadays the utility of domain ontologies is widely acknowledged in many area, such as information systems, software engineer. natrue language processing, artificial intelligence, electronic commerce and so on. Ontologies are enable to fulfill knowledge appearance, information retrieval and search. However, there still exists several drawbacks that must be resolved before ontologies become practical and useful tools. A critical issue is the ontology mapping. This matching process will directly influence the precision and recall of information retrieval among several different ontologies. In this paper, we propose an approach of ontology matching and an ontology matching algorithm based on description logic system through analyzing the characteristic of description logics, including definitions of syntax, semantics and basic reasoning services. and finally demonstrate the practicality of this ontology matching method between two concrete Book-System ontologies.
期刊:
2009 FIFTH INTERNATIONAL CONFERENCE ON SEMANTICS, KNOWLEDGE AND GRID (SKG 2009),2009年:392-395
通讯作者:
Wang, Jinqiao
作者机构:
[Wang, Jinqiao; Zhu, Li; Yang, Qing; Sun, JunLi] Huazhong Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.
通讯机构:
[Wang, Jinqiao] H;Huazhong Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.
会议名称:
Fifth International Conference on Semantics,Knowledge and Grid(第五届语义、知识与网格国际会议 SKG 2009)
会议时间:
2009-10-12
会议地点:
珠海
会议论文集名称:
Fifth International Conference on Semantics,Knowledge and Grid(第五届语义、知识与网格国际会议 SKG 2009)论文集
关键词:
reinforcement learning;association rules;topological graph;Recommendation Systems data mining
摘要:
Reinforcement learning is an important method of machine learning. This paper using the graph theory to express varieties of knowledge points, which their's relationship is expressed by the graph of topological graph. Applied the Technology of association rule Recommendation to deal with the relationship between these knowledge points, give the corresponding of the recommendation work flow chart. In the paper data tables used to store the knowledge points, the algorithm to demonstrate the technical of association rule Recommendation feasibility and rationality.
摘要:
In this paper, we have integrated fuzzy logic in the formal context, developing fuzzy formal concept analysis to support the automatic generation of primitive fuzzy ontology that can deal with uncertain information. The proposed automatic fuzzy ontology generation approach consists of the following steps: first, fuzzy formal concepts are created through fuzzy formal context; then, the decision rules from the generated formal concept are obtained using the implicative relations between two fuzzy concept lattices; and next, Fuzzy clustering is adopted to promote the fuzzy concept hierarchy of fuzzy concept lattice; finally, a primitive fuzzy ontology is constructed according to the concept hierarchy structure. A complete fuzzy ontology can be generated through the expanding of primitive ontology. Since the fuzzy concept lattice should be from specific fuzzy formal context, the proposed automatic generation approach is basis on the concept hierarchy structure which supporting concept learning, thus, for each different application, it is suitable to pursue the own fuzzy domain ontology model automatically.
期刊:
2009 IEEE INTERNATIONAL CONFERENCE ON GRANULAR COMPUTING ( GRC 2009),2009年:682-685
通讯作者:
Yang, Qing
作者机构:
[Chen, Wei; Yang, Qing; Wen, Bin] Huazhong Normal Univ, Dept Comp Sci, Wuhan 430079, Peoples R China.
通讯机构:
[Yang, Qing] H;Huazhong Normal Univ, Dept Comp Sci, Wuhan 430079, Peoples R China.
关键词:
fuzzy ontology;fuzzy clustering;semantic information retrieval
摘要:
For expressing the fuzziness and uncertainty of domain knowledge, realizing the semantic retrieval of fuzzy information, this paper produces an extended fuzzy ontology model and proposes a kind of semantic query expansion technology which can implement semantic information query based on the property values and the relationships of fuzzy concepts. The extended fuzzy ontology provides appropriate support for Learning Evaluation. To access the effect of the proposed model, many experiments have been given for the performance evaluation. The results show that this system can improve retrieval accuracy and promote intelligent semantic query.
期刊:
2009 INTERNATIONAL CONFERENCE ON INDUSTRIAL MECHATRONICS AND AUTOMATION,2009年:297-300
通讯作者:
Yang, Qing
作者机构:
[Chen, Wei; Wang, Jinqiao; Zhu, Li; Yang, Qing] Huazhong Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.
通讯机构:
[Yang, Qing] H;Huazhong Normal Univ, Dept Comp Sci, Wuhan, Peoples R China.
关键词:
semantic web;description logics;reasoning mechanism;ontology construction
摘要:
Ontology is becoming one of the important research point in the area of the semantic web, which is enable to fulfill knowledge appearance, information retrieval and search. And as the basis of ontology description language, description logics are the formal tool of knowledge representation and reasoning, providing strong capbilities of representation and effective reasoning service mechanism, which have been extensively applied to various fields of computer science. This paper detailedly studies the theory of description logic and its reasoning mechanism using in Book -System ontology construction, and finally testifies the practical reasoning in the process of the concrete ontology construction.
摘要:
Recently ontologies are playing very important part in many areas,such as intelligent information retrieve, knowledge management and organization,electronic commerce and so on,however, several drawbacks must be overcome before ontologies become useful and practical tools. As the number of ontologies are made publicly available and accessible on the web increases steadily,a single ontology is no longer enough to support the tasks envisaged by a distributed environment like the semantic web.Multiple ontologies need to be accessed for several applications. A critical issue is ontology integration,which can largely improve the efficiency to enrich such a domain ontology with less time and lower cost for obtaining related knowledge.This paper has deeply studied the principles of ontology integration,then proposes a procedure model for ontology construction and a new framework for ontology integration based on machine learning through analyzing the characteristics and problems in the process of ontology integration.
期刊:
2009 INTERNATIONAL FORUM ON INFORMATION TECHNOLOGY AND APPLICATIONS, VOL 2, PROCEEDINGS,2009年:377-381
通讯作者:
Wei, Chen
作者机构:
[Wei, Chen; Li, Zhu; Qing, Yang; Bin, Wen] Cent China Normal Univ, Dept Comp Sci, Wuhan 430079, Peoples R China.
通讯机构:
[Wei, Chen] C;Cent China Normal Univ, Dept Comp Sci, Wuhan 430079, Peoples R China.
关键词:
fuzzy ontology;semantic reasoning;fuzzy matching rule base;semantic web
摘要:
In this paper, we construct a fuzzy ontology model and propose using a fuzzy matching rule base to promote a fuzzy ontology generation frame which supports rough concept descriptions on intrinsic semantic level. To consider the rule base and express the fuzziness, the formal analysis of concept vector is developed for the generation of fuzzy ontology that can deal with uncertain information. The proposed fuzzy semantic extension technique taking advantage of the fuzzy matching rules consists of the following steps: vectorization of fuzzy concept, establishment of basic concept pair, and synthesis of semantic information. The formal descriptions of fuzzy approximations of concepts and valid fuzzy semantic reasoning can be obtained through semantic matching of concepts. As such, applying the provided frame to subjective credit reporting management, the SCRM ontology model will potentially improve the fuzzy concepts reasoning and effectively facilitate the semantic expansion. It is, especially, suitable for acquiring acceptable degree of subjects through the intelligent reasoning of trust relationship.
摘要:
Different doctor may give different diagnosis descriptions on the same result in the medical management system, some even not standardized and lots of the diagnosis are same., doctor's repeated input is inefficiency and can not solve the customer's queuing problem. The traditional K-nearest neighbor algorithm has the drawback of slow recommend while the sample space is large. This paper adopt two-level K-nearest neighbor algorithm to reduce time complexity of the recommendation algorithm in order to improve the efficiency of doctors input and the standardize description of the item, implement the initiative recommend of the medical item diagnosis.
期刊:
SEVENTH WUHAN INTERNATIONAL CONFERENCE ON E-BUSINESS, VOLS I-III,2008年:839-844
作者机构:
[Wang, Xueping; Huang, Zhufeng; Yang, Qing; Zhang, Lianfa] Cent China Normal Univ, Dept Comp, Wuhan 430079, Hubei, Peoples R China.
关键词:
self-learning;Bayesian Network;e-learning
摘要:
E-learning is becoming one of the most important educational means. As more and more organizations and institutions are moving towards the e-learning strategy, self-learning model becomes a big challenge. Background knowledge and learning objectives of various groups of students on the network are very different. Self-learning system, which uses different learning programs for different students, can enhance the efficiency of learning process. In a self-learning system, the algorithm dealing with uncertainty factors of Self-learning model is very important. Bayesian network artifice is a very effective one within various methods dealing with uncertainty. In this paper, we applied Bayesian network method to self-learning model; designed Bayesian network structure in a self-learning model; assigned the local probability distribution and discussed the way to acquire and propagate related evidences. The practice has proven Bayesian network approach for self-learning model is a very effective method.