电子商务中基于贝叶斯网络的货源信息检索模型

Bayesian Network Based Business Information Retrieval Model in Electronic Commerce

针对电子商务环境下货源信息检索问题进行了分析研究,应用信息检索和贝叶斯网络的相关理论,提出一种基于贝叶斯网络的货源信息检索模型.该模型采用一种四层节点的贝叶斯网络检索结构,利用货源信息样本文献描述查询要求,并通过网络节点间的条件概率计算出查询与文献之间的相似度.实验表明,该模型取得了较好的检索效果,为企业提供了有价值的货源相关信息.

This paper analyses the problem of business information retrieual in Electronic Commerce （EC） environment and proposes a business information retrieval model by means of Bayesian Network （BN） and Information Retrieval theory. In this model, the BN has four layers of nodes and user query is expressed in terms of the predefined information sample documents. The similarities between documents and the query can be calculated with the conditional probabilities among the nodes in the Bayesian Network. Experiments show that the proposed model has a good performance in searching business information.