<html><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"></head><body ><div>Hi all,</div><div><br></div><div>Pls go to this talk. Much of our work is related to Daisys work.</div><div><br></div><div>Kenny</div><div><br></div><div><br></div><div><div style="font-size:9px;color:#575757">Sent from Samsung Mobile</div></div><br><br><div>-------- Original message --------</div><div>From: chen-zx <chen-zx@cs.sjtu.edu.cn> </div><div>Date:2014/04/22 09:50 (GMT+08:00) </div><div>To: all@cs.sjtu.edu.cn </div><div>Subject: 23日10点学术报告 </div><div><br></div>各位老师好!<br><br> 来自佛罗里达大学计算机信息科学与工程系的一位助理教授Wang Zhe将于23日（明天）上午10点在404教室做一个题为“ Large <br> Probabilistic Knowledge Base <br> Systems”的学术报告，诚邀有兴趣的老师和同学前来参加！（麻烦老师们转发给学生，谢谢！）以下为讲座具体内容：<br><br> Title: Large Probabilistic Knowledge Base Systems<br> Speaker: Zhe Wang<br> Time: 10:00-11:30, 23rd April<br> Venue: SEIEE 3-404<br> Abstract: Keyword search engines have been the state-of-the-art <br> information retrieval tool over large text corpora for two decades. To <br> date, most search engines have little understanding that keywords and <br> documents refer to entities and relations in real-life. Better search <br> results and experience can be achieved by understanding entities and <br> relations in documents as well as in queries. A knowledge base (KB) <br> containing relevant entities and relations should be the backbone of any <br> application that is fueled by text. Given a large amount of text data, a <br> system is needed that can automatically construct a knowledge base using <br> statistical machine learning (SML) methods, manage the uncertainty <br> inherent in the extracted knowledge, and maintain them over time.<br> In this talk, I first summarize the major results from BayesStore, a <br> probabilistic database system that natively supports SML models and <br> various inference algorithms to perform query-driven knowledge <br> extraction from text and probabilistic query processing over uncertain <br> extractions. Results show that BayesStore can significantly improve <br> performance and answer quality for queries over unstructured text.<br> With BayesStore as a foundation, I propose to build a probabilistic <br> knowledge base (ProbKB) system with a deep integration of the SML <br> methods with scalable data processing frameworks. A ProbKB system should <br> be designed to support various aspects in the life of a knowledge base <br> (KB) including KB extraction, expansion, </body>