The 57th In Silico Megabank Research Seminar will be held on Friday, February 27. This Time, we will be welcoming Dr. Mahito Sugiyama, The Institute of Scientific and Industrial Research, Osaka University as our lecturer, and he will be speaking on “Finding Statistically Significant Structure from Big Data ”.

・Date: Friday, February 27, 2015
・Time: 5:00 pm‐6:30 pm
・Venue: Small Conference Room 2(3rd Floor), Tohoku Medical Megabank Building
・Title: Finding Statistically Significant Structure from Big Data
・Lecturer: Mahito　（The Institute of Scientific and Industrial Research,  Osaka University）

＊This lecture is transferable as a class in the medical research-related lecture course.

・Abstract: Data mining, whose purpose is knowledge discovery from the big data, is utilized in various fields from basic science such as chemistry and biology to the application to management and marketing. Especially, the development of approaches to find combinational structure hiding in data such as gene pairs co-occurred and expressed and the structure shared by compounds with specific activation is the center of the topic concerning data mining. However, although securing the statistical significance on the discovered knowledge, in other words, the calculation of P value is a requirement in major application domain for data mining including biological science, there was no adequate focus for a long time. In this lecture, initiatives for this rapidly advancing research topic in recent year, triggered by the literature of Terada et al. presented at PNAS in 2013 are introduced. Furthermore, a special attention is paid to the recent outcomes that enable resolution to this matter by combining superfast algorithm developed in the field of data mining and multiple testing procedures developed in the field of statistics.

・Organizer: Takahiro Mimori, Masao Nagasaki