Key Techniques of Automatic Question-Answering Customer Service System in College Informatization Domain

Abstract

To improve the efficiency of customer service for college information, an automatic question-answering customer service system was designed and implemented. It could collect the domain-specific corpus was collected in combination with the general corpus, train word vectors, represent the semantics of words, and implement the question similarity calculation algorithm based on word co-occurrence and syntactic analysis to achieve question matching. To minimize the range of question matching and thus make the matching more efficient, a question classification module based on support vector machine (SVM) was implemented. Finally, the validity and usability of the proposed method were verified by experiments. The proposed method shows a certain universality. Given that some colleges have not established their automatic question-answering customer service system, this paper is of great research significance.