如何将数据转换成信息

（转自2006年国际数学家大会官方网站）

Plenary Session: Iain Johnstone

How to Convert Data into Information
For some time now the following wording can be found in many job advertisements: “Minimum requirements: Experience in the implantation of Information Management Systems (Business Intelligence, Data Warehousing, Data Mining)". Given the growing amount of data handled in many sectors, experience in information management is becoming increasingly necessary. The mathematician Iain Johnstone will give a plenary lecture at the ICM2006 on “High Dimensional Statistical Inference and Random Matrices”, which will deal with the management of massive amounts of data.

Until recently, statistics was centered on the study of one- and multi-dimensional random variables. However, the development of computation has led to the era of “data mining", and all organizations – banks, hospitals, research centres – handle enormous quantities of data which must often be constantly available, such as financial assets. Statistics in high dimensions are essential for analysing this data, and this will be the central theme of Johnstone’s lecture. This branch of mathematics shows how to organize and summarize data, whether it involves an electrocardiogram, Internet traffic or a stocks and shares, in such a way that they provide useful information.

Iain Johnstone was born in Melbourne in 1956. In 1977 he graduated in mathematics at the Australian National University, specializing in pure mathematics and statistics. He obtained his doctorate in statistics from Cornell University in 1981. Since then he has been associated with Stanford University in California, where in 1992 he became professor of statistics and biostatistics. In addition to his work on biostatistics, in the field of statistics he has received much recognition.

Lecturer: Iain Johnstone

Title: High Dimensional Statistical Inference and Random Matrices

Date: Friday, August 25th: 10:15-11:15

ICM2006 Scientific Programme: http://www.icm2006.org/scientificprogram/plenarylectures/