Our Goal Read data correctly, and see into the future
With the advance of information and communication technology, more and more things around us are connected to the Internet, and it has become easier to grasp the information (value) of things and the demand for producing them. In this laboratory, we interpret the meaning of the data generated by things and by the Internet from the viewpoint of statistics. We aim to educate people who can find the value in data by modeling the structure behind it mathematically and reading it correctly.
Connection with Society Predicting with data and statistics
Our society is facing many phenomena that could be called extraordinary. Natural phenomena such as global warming, large earthquakes and pandemics, and technologies such as faster computers, the Internet of Things and algorithms that resemble artificial intelligence, have changed greatly in the past few years. It is important to understand and interpret these changes as well as we can, and then to envision what the future will look like. This laboratory takes up the theme of predicting the future state of various kinds of data, and tries to link science, engineering and society through statistics and information technology.
Product lifetime can be estimated by accelerated life tests
Our research on statistical modeling of accelerated life tests helps predict the lifetime of the electrical devices around us. Predicting lifetime is used to demonstrate the reliability of products and to set warranty periods.
Evaluating the test itself to measure ability accurately
The theory behind our research on estimating ability and optimizing how questions are given is used in widely taken tests such as the TOEIC. As this research progresses, it may support not only certification exams but also everyday learning.
See, Know, Show Processing, analyzing and visualizing data
Data are collected in many ways: from sensors, from web browsing histories, from the results of sports games. First, we make the data visible to people, by drawing graphs or by recoding wins and losses as 1 and 0, and explore how the data can be seen. Next, we think about the mechanism that produced the data. This is called statistical modeling. Observed data contain a lot of noise. A web browsing history, for example, reflects people's preferences, but also mistaken clicks and unintended page transitions. In statistical modeling we separate such noise, asking whether it comes from the person, from the web page, or from pure chance. Finally, we present the results in an understandable form and return them to society.
After Graduation Data science is a tool for every business
This laboratory carries out research based on statistics, the foundation of data science. Our graduates have gone on to a wide range of careers, including systems engineering, consulting, securities, transportation and real estate. Statistics, which deals with data, is needed and applicable in every industry. As information technology and globalization advance, systems of all kinds (companies, information, production) are becoming more complex, and people who can analyze data collected from them and approach them scientifically will be needed in every field. We hope that you will develop an eye for reading data correctly.
Contact
- E-mail: sakumura (at) hosei.ac.jp
- Address: Room W6004, 6th floor, West Building, Koganei Campus, Hosei University, 3-7-2 Kajino-cho, Koganei, Tokyo 184-8584, Japan (map)