基于晶体增长分类算法的个性化在线服务

摘要/Abstract

摘要： 为给用户提供精确的个性化信息服务，提出了一种晶体增长分类算法，以此对网络上的信息资源进行分类。晶体增长分类算法利用反复提取原型向量的方式来聚合信息。其中，原型向量的提取根据输入的搜索关键词及其上下文关系，通过径向基函数神经网络完成。原型向量反复提取过程反映了一种树型关系，而每个阶段的结果形成树型目录的一个节点。信息分类结束后，形成一个树型目录。在此目录的基础上，结合在线服务技术和用户需求，以向导的方式对信息进行二次匹配，从而保证个性化信息服务的精确度。另外，所提算法也具有合理的计算复杂度，且实验结果验证了算法的有效性和可行性。

关键词: 径向基函数神经网络, 概念模糊集, 在线服务, 搜索引擎

Abstract: In order to provide accurate and personalized information service, a classification algorithm of crystalline increment was proposed to classify the information resource in the network. To aggregate information, the classification algorithm of crystalline increment adopted reiterative extraction of original vector, which was created by Radial Basis Function neural network according to the input search key words and their context. The process of reiterative extraction of original vector reflected a tree type relationship, and the result of each step formed a node of the tree type catalog. If the process of information classification was accomplished, a tree type catalog would be established. Based on this catalog, integrated with on-line service technology and customer demand, the information could be piloted to perform second match, so as to guaranteed the accuracy of personalized information service. Besides, this algorithm had reasonable computing complexity. Its effectiveness and feasibility has been verified by experiment results.

Key words: radial basis function neural network, concept fuzzy set, on-line service, search engine

中图分类号:

G354.4

丁振国,王静. 基于晶体增长分类算法的个性化在线服务[J]. .

DING Zhen-guo, WANG Jing. Personalized on-line service based on classification algorithm of crystalline increment[J]. .

参考文献

[1]YANG K H, PAN C C, LEE T L. Approximate search engine optimization for directory service[A].Proceedings of the 2003 International Parallel and Distributed Processing Symposium[C]. CA, USA: IEEE Computer Society Press, 2003. 242-249.

[2]FU X J,WANG L P. Data dimensiona lity reduction with application to simplifying RBF network structure and improving classification performance[J]. IEEE Transactions on Systems, Man and Cybernetics, Part B:Cybernetics, 2003,18(6):399-409.

[3]LIN W M, ZHAN T S, YANG C D. Distribution system reliability worth analysis with the customer cost model based on RBF neural network[J].IEEE Transactions on Power Delivery, 2003,18(7):1015-1021.

[4]KARAYIANNIS N B,M.M R. On the construction and training of reformulated radial basis function neural networks[J]. IEEE Transactions on Neural Networks, 2003,18(7): 835-846.

[5]DOBSA J, BASIC B D. Concept decomposition by fuzzy k-means algorithm[A].Proceedings of the 2003 IEEE/WIC International Conference on Web Intelligence[C]. CA, USA: IEEE Computer Society Press, 2003. 684-688.

[6]TOMIYAMA T, OHGAYA R, SHINMURA A, et al. Concept-based web communities for googleTM search engine[A]. Proceedings of the 12th IEEE International Conference on Fuzzy Systems[C].CA,USA:IEEE Computer Society Press, 2003. 1122-1128.

[7]LIU J G, HUANG Z H, WU W P. Web mining for electronic business application[A]. Proceedings of the Fourth International Conference on Parallel and Distributed Computing, Applications and Technologies[C]. CA, USA: ISIF/IEEE, 2003. 872-876.

[8]KAWANO H. Web archiving strategies by using web mining techniques[A]. Processing of the 2003 IEEE Pacific Rim Conference on Communications, Computers and Signal[C]. CA, USA: ISIF/IEEE,2003. 915-918.

[9]LIN S D, KNOBLOCK C. Exploiting a search engine to develop more flexible web agent[A].Proceedings of the 2003 IEEE/WIC International Conference on Web Intelligence[C]. CA, USA: IEEE Computer Society Press,2003.54-60.

[10]SATO N, UEHARA M, SAKAI Y. Temporal information retrieval in cooperative search engine[A]. Proceedings of the 14th International Workshop on Database and Expert Systems Applications[C]. CA,USA: IEEE Computer Society Press, 2003. 215-220.