Research On Business Email Classification Based On User Knowledge

With the rapped grow of the Internet and E-business technology; enterprises are going into the Internet age. The application of new technology has made the enterprises changing their business model. Communication is becoming more and more important as the enterprises are more focus on the area of Customer Relationship Management and Business Intelligent. New technologies are widely used in order to improve the process of enterprise decision making.Email is widely used in business communication. As one of the most important business communication media, Email contains much useful information. However, most of the enterprises don't take email management serious. Email management is out of order. Under this situation, this paper undertakes the study of email management model and technology in order to improve the Email message analysis ability of the enterprises.Firstly, this thesis performs a thorough structural analysis of the email management problem from three different views which are data management, technology and application. It then proposes a DTA(Data-Technology-Application) model for email management. On top of this, a framework for email management system centered on email knowledge management is proposed.Secondly, this thesis studies the email data mining technology especially the email classification method. It points out that currently available email classification methods are not adequate when being directly applied to business email management because these methods only consider the knowledge from email body, but the knowledge of users. To mend those inadequacies, a user knowledge based na?ve bayes classifier is proposed. The application of this new method can increase the reliability of the email classification process.Lastly, based on the work outlined above, a prototype system is designed for email management. The prototype system's main functions include data management, email data mining and model application. A read example is used to analyze how to apply the prototype system to solve the email auto dispatch problem in customer service center.