Mining Of E-commerce Product Quality Feature Relationship Based On Pattern Recognition And Its Application

With the coming of the Web2.0 era,the rapid development of e-commerce has changed people's consumption pattern and enterprise's production and management mode.Massive reviews on the e-commerce platform,published by user,contain lots of valuable information.These information,mined and utilized rationally,will provide a new way of product quality supervision for manufacturing enterprises.The main contents of this paper are as follows:(1)Extraction of explicit product features.In this paper,a method based on pattern matching is proposed.By analyzing lots of reviews,based on the theory of dependency grammar and syntactic dependency relationship between words,summarizes five criteria that explicit product features need to meet.The experimental results show that the method proposed in this paper is feasible and effective.(2)Extraction of implicit product features.A method of considering context weight is proposed in this paper.The co-occurrence matrix of product feature cluster and opinion is constructed,by calculating feature similarity and feature clustering.And evaluates the credibility of context information by calculating the context weight,when there are several contextual features in the comment.Then,with comprehensive consideration opinion and contextual feature,the implicit product feature is extracted.The experiment shows that the method proposed in this paper reduces the complexity of the algorithm and improves the accuracy and recall of the implicit product feature extraction.(3)Construction of the feature-semantic association structure tree and the statistics of the network's key parameters.In this paper,overall considered the explicit product features and the implicit product features,and based on pattern recognition,the feature-semantic association structure tree is constructed.The statistic of the product feature's emotional score are realized by the integration of emotional intensity dictionary,which can generate the connotation of product quality management.(4)Realization of the early warning system.Based on the research results of this paper,a product quality early-warning system based on Chinese reviews is designed and implemented.The system can help the production enterprises to find product quality problems and product quality early warning,also can improve the quality management level of e-commerce product and reduce the risk of the quality of ecommerce product.