Design And Implementation Of Intelligent Customer Service Multi-service Semantic Comprehension System Based On Grammar Rules Matching

With the development of computer science technology and the theory of machine learning,people's consumption level is getting higher and they are demanding more about the quality of service.The demands about the intelligent client service system that they have always been exposed to is getting higher,such as,to enjoy more intelligence,diversification,friendly interaction,facilitation and so on.It is of great necessity to improve the present intelligent client service system,as it has low efficiency,deficient intelligence,and as it requires maintenance with large quantity of human resources and financial resources,besides,it becomes harder to live up to the demands of customers.For the purpose of further improving the quality of intelligent client service system,reducing businesses cost and improving the automation of customer service and intelligent management,this paper puts forward a design scheme,which is based upon grammar rule matching.The scheme is about intelligent client service and multi-businesses semantic understanding system.The plan adds to the design of named entity recognition mode and semantic ranking mode on the basis of traditional grammar rule match mode.While the intelligent client service identification mode helps improving the accuracy of identifying the natural language,the new addition enables the identification efficiency of the whole mode to go better.The plan introduces identifying the natural language network,which is based upon the extended fonn of Backus-Naur Form,which is the expansion of ABNF and finite state automaton.The introduction of the network is used as the basic mode of understanding natural language,and is used to understand the real purposes of customers,and their concrete semantic message.Meanwhile,combined with named entity recognition mode,which is an entity that could identify the expressions of customers.It improves the accuracy of the model identification by using named entity recognition mode.At the same time,the intelligent client service and multi-businesses semantic understanding system which is based on grammar rule matching has semantic ranking mode and SVMRank algorithm which is based on machine learningAt the same time,the intelligent client service and multi-businesses semantic understanding system has semantic ranking model based on the sort of semantic recognition SVMRank algorithm idea of machine learning based on the results of the expressions of customer's words by getting the top 5 semantic information.By reducing the number of semantic information,reduce the service pressure of the data search module,so as to improve the overall efficiency of the system.