主講人：樊衛(wèi)國 美國弗吉尼亞理工大學(xué)教授

時(shí)  間：2013年6月25日（星期二）下午4:00-5:30

地  點(diǎn)：bwin必贏唯一官網(wǎng)313教室

Abstract：The recent surge in the usage of social media, such as Facebook, Twitter, and online communities and forums has created an enormous amount of user-generated content (UGC). UGC serves as a huge gold mine that is yet to be tapped for various business and consumer intelligence applications.  While there are streams of research that seek to mine this UGC, these research studies, however, are often done in a piecemeal fashion without a coherent and systematic framework.  In this paper, we synthesize existing research studies and propose an integrated text analytic framework based on statistical language processing and computational linguistics.  The framework effectively leverages the rich social media contents and quantifies the words in the text using various automatically extracted signal cues.  These extracted signal cues can then be used as modeling inputs for business value discovery.  We show case the usefulness of the framework by performing product defect discovery using UGC from popular online discussion forums. Implications of such a framework for both research and practice are provided.