报告人：孙浩军 教授

报告时间：2016年7月19日 15:00

报告地点：计算机与通信工程学院会议室（基础楼217）

邀 请 人：王翠荣、刘杰民 教授

报告人简介：孙浩军，博士，教授， 1982年毕业于河北大学数学系， 1982-2000年在河北大学任教，2000-2005年在加拿大舍布克大学留学，并于2005年获博士学位。同年回国工作，现任汕头大学工学院计算机系教授。其研究涉及了多个科学领域，其中包括模式识别、数据挖掘、图像处理和理解、信息系统、神经网络、信号处理等。近年来主持国家自然基金、省基金以及横向课题的研究，在国际国内期刊、国际会议发表论文多篇。目前兼任中国计算机学会信息系统专委会委员，中国人工智能学会机器学习专业委员会通讯委员。

报告内容摘要：Clustering high-dimensional data is a challenging task in data mining, and clustering high-dimensional categorical data is even more challenging because it is more difficult to measure the similarity between categorical objects. We propose a hierarchical algorithm with attribute weighting for clustering high-dimensional categorical data, based on a recently proposed information-theoretical concept named holo-entropy. The algorithm proposes new ways of exploring entropy, holo-entropy and attribute weighting in order to determine the feature subspace of a cluster and to merge clusters even though their feature subspaces differ.High-dimensional mixture data clustering is a new issue of data mining in rescent years. Focused on the problem, this reseach study the information amalgamation of mixture data. We proposed a data discretization method based on the fuzzy-set which preserves the sequence data after discretization of numerical data， and proposed a hierarchical clustering method based on the information entropy.