（一）赵岩
1、个人介绍
赵岩， 男，1974年生，博士生，主要研究方向：汉语词法分析，音字转换
2、导师介绍
王晓龙，男，1955年，博士生导师，主要研究方向：人工智能，自然语言处理
3、论坛内容摘要
题目：汉语词性标注的研究
摘要：首先给出汉语词性标注的目的和意义，介绍了常用的词性标注方法，如：HMM模型，最大熵模型，支持向量机模型等。对以上几种方法作了对比和分析。针对词性标注中的长距离约束问题，提出了转换触发对的概念，并将其应用到最大熵词性标注模型中，与传统的HMM模型相比，标注错误减少了34%。
4、报告时间
4月15日 星期五 晚6：30 地点 新技术楼618室
（二）
Speaker: xiao,jinghui
Title: The study of key to pinyin conversion problem for Chinese phonetic input method on handheld devices.
Abstract: Recently it becomes a hotspot for the research on Chinese phonetic input method on handheld devices. In these slides, we propose the key to pinyin conversion problem for sentence-level Chinese phonetic input technique on digit keyboard. We first formulize the problem. And then, according to Bayes' rule, two models are deduced by different hypotheses to solve the problem. And two set of algorithms are provided accordingly. Experiments and discussions are presented later. From the results, two conclusions can be drawn: Firstly, key to pinyin conversion is a bottleneck problem for entence-level Chinese phonetic input technique. Secondly, the methods we proposed are effective to solve the problem.
(三）
speaker: Zhang, Shiwei
Title: CLUSTERING OF GENE EXPRESSION DATA BASED ON SELF-GROWTH TREE
Abstract: A novel approach used in gene expression data is proposed in this paper. Unlike traditional hierarchical clustering, this method has a lower computational complexity and more rational tree structure, and compared with K-means method, it is influenced less by people. It is also applied in the cell cycle data set reported by Cho, and obtains some good results.