报告题目：Structured Prediction in Natural Language Processing

报告人：吴苑斌博士

主持人：孙仕亮

时间：2013年4月17日上午9：00

地点：信息楼629

报告摘要：Structured prediction is one of the most important machine learning methods in natural language processing (NLP). Different from traditional classification problems, the output spaces of structured prediction are sets of combinatorial objects, such as sequences, trees and sub-graphs. On one hand, structured prediction provides a unified framework for many NLP tasks; on the other hand, research objects from computational linguistics introduce many challenging inference and learning problems for statistical machine learning. These new problems attract much attention in both NLP and machine learning fields.

报告人简介：Yuanbin Wu is a research fellow at National University of Singapore (NUS). He got his B.S. and Ph.D degree from Fudan University. His research interests include natural language processing and machine learning, especially, opinion mining, information extraction, structured prediction and online learning.