Customer Service Automatic Answering System Based on Natural Language Processing

With the rapid development of Internet, information grows explosively, and traditional search engine have failed to meet the needs of users. This paper proposes a customer service automatic answering system with a high-quality knowledge base. First of all, based on unsupervised learning algorithm, this system extracts the question and answer pairs from documents and store them in the knowledge base. Then employing semantic analysis module and the method of Natural Language Processing (NLP), this system gains the meaning of the customers' question accurately, then retrieve the knowledge base and return a high-resolution answer to the user. Furthermore, we construct a dialog management module, which makes reasonable guesses on issues that cannot be matched, and records the dialogue history so that the question-answering system can give more intelligent responses. Finally, due to the diversity of the document structure and the complexity of Chinese natural language, this system adds an edifying function that can add, delete, and modify the question and answer pair in the knowledge. Therefore, our customer service automatic answering system can be more intelligent and efficient than the existing question and answer system.