Detection of FAQs Matching Inquiry E-mails by Automatic Generation of Characteristic Word Groups from Past Inquiry E-mails

摘要

This paper discusses how to detect inquiry e-mails corresponding to predefined FAQs (frequently asked questions). Web-based interactions such as ordering and registration forms on a Web page are usually provided with FAQ pages to help users. However, most users submit their inquiry e-mails without checking such pages. This causes help desk operators to process large numbers of e-mails even if some contents match FAQs. Automatic detection of such e-mails is proposed based on an SVM (support vector machine) and a specific Jaccard coefficient based on positive and negative already-received inquiry e-mails. Experimental results show its effectiveness, and we also discuss future work to improve our method.