管理学院上海管理论坛学术讲座第136期
讲座题目：Machine learning methods for protein structure prediction and ambulatory assessment of alcohol craving机器学习在蛋白质结构预测和嗜酒评测中的应用研究
时间：2015年6月12日（周五），上午10:00
主讲人：尚奕教授（美国密苏里大学）
主讲人简介：
尚奕，美国密苏里大学教授，计算机系研究生主管；
1997年博士毕业于伊利诺伊大学香槟分校；
在非线性优化、人工智能、移动计算等领域发表论文160余篇，获得6项专利；研究已被NSF, NIH, Army, DARPA, Microsoft, and Raytheon等资助；
ACM终生会员和IEEE高级会员。
内容简介：
Protein folding is one of biggest unsolved problems in science. In recent years, the gap between the number of known sequences and known structures is rapidly increasing. Computational protein structure prediction provides a cheap, fast, and large scale approach to address the challenge. In this talk, I will present our system Mufold and machine methods for quality assessment and selection of predicted models. Our methods are developed based on a comprehensive set of model features and outperform the best existing methods on some benchmark datasets.
Currently, most methods in clinical psychology research primarily rely on questionnaires and interviews with examiners, which could not provide real-life subject behavioral and psychology data monitoring and collecting services. In this talk, I will present a smart phone-based mobile Ambulatory Assessment System (mAAS) for psychology research, especially alcohol craving studies. Body sensor and survey data collected are used by machine learning methods to build models to assess and predict alcohol or other substance use. The system has been deployed in a field study and initial data collected from real subjects show promising results.
欢迎感兴趣的教师、学者和广大同学踊跃参加！