主 题：Large Scale Computing for Data Mining Applications in Insurance
主讲人：Prof. Yuying Li
主持人：鲁万波 教授
时 间：2014年9月18日下午16:00—17:00
地 点：通博楼B座212学术会议室
主办单位：统计学院 科研处
主讲人简介：
Prof. Yuying Li received her B.S. in Mathematics of Sichuan University and her M.S. and Ph. D. in Computer Science fromUniversityofWaterloo. Her main research interests include design novel predicative analytics for business and finance applications. Currently Prof. Li research focuses on optimization methods for rare class classification and detection, feature selection, semi-supervised, and unsupervised predications. She has published many articles which have appeared in top field journals, such as Journal of Banking and Finance, Journal of Computational Optimization and Applications, Journal of Quantitative Finance.
内容提要：
Large amounts of data are continually accumulated every day. Extracting useful information from data presents exciting challenges to information technology, computational mathematics, and statistical analysis. Potential benefits of success are unimaginable.
In this talk, I will discuss two real-world data mining applications in insurance, in health care and auto-insurance fraud detection, to illustrate some of these challenges and potentials. I will describe our participation and solution sin achieving the fourth place in the 2013 Heritage Health Provider Competition for hospital admission prediction. In addition, I will discuss difficulties in Auto Insurance Fraud Detection and present a new method to tackle this difficult problem.
