时间：2018年6月25日上午8:30

地点：1A502

报告人：韩晓旭副教授

报告题目：Massive data analytics in modern data science

报告摘要：

Massive data analytics in modern data science.

Analytics has been experiencing unprecedented changes in the era of big data. Traditional model-driven analytics are moving to data-driven analytics to handle challenges from all kinds of massive data in modern data science. In this talk, we mainly focus on two topics. The first is complex disease diagnosis with big data in translational bioinformatics.  It plans to answer the query: how to achieve rivaling-clinical and reproducible disease diagnosis using big omics data?

The second is option implied volatility pricing in finance. It wants to answer the question: how to overcome the weakness of traditional model-driven implied volatility pricing via machine learning?

 

For the first topic, we have proposed a profile marker diagnosis approach based on our proposed derivative component analysis (DCA) by using big omics data. We further demonstrated the correctness of our rivaling-clinical diagnosis via a novel data visualization approach, which avoids time-consuming clinical tests, but still verify the correctness of diagnosis.

 

For the second topic, we have proposed a novel integrative learning that optimizes different learning machines to achieve more general and better implied volatility prediction for different types of options by solving a nonlinear programming problem. Technically, the proposed integrative learning is a genetic optimized machine learning algorithm to handle complex learning tasks in data science.

 

韩晓旭个人简介：

韩晓旭：美国福特汉姆大学计算机系副教授，博士生导师，现任大数据研究室主任和生物信息研究室，哥伦比亚大学量化蛋白组研究中心兼职研究员。2004年毕业于美国爱荷华大学应用数学和计算科学博士主要研究领域包括：生物信息学, 大数据, 数据科学,机器学习，商业分析， 信息安全, 云计算, 高性能计算等。担任International Journal of Knowledge Discovery in Bioinformatics期刊编委，担任Cancer informatics，Journal of Computational Biology and Chemistry，ournal of bioinformatics and computational biology，Information system，BioData mining等期刊客座主编。在BMC Medical Genomics，BMC Systems Biology ，BMC Bioinformatics， Journal of Bioinformatics and Computational Biology (JBCB)，Cancer Informatics，Applied Soft Computing ，Computational Knowledge Discovery for Bioinformatics Research等及其它会议发表论文50余篇，出版信息安全和大数据方面的专著二部。