Last update: May 22, 2025
Welcome to our website!
We study information-theoretic frameworks and their applications to unravel the valuable information from the data collected and generated from complex and diverse natural phenomena, social phenomena and human activities, especially cases that are transient and difficult to process routinely.
Center for Data-driven Science and Artificial Intelligence, Director, Professor
Keyword: Complex Systems, Behavior Analysis
Research: Collective dynamics and information theory of complex systems
Looking at a flock of birds and fish, they behave in a complex way, as if they were a single multicellular organism, exhibiting a well-organized movement or sometimes changing the state disorderly. We are trying to reveal the information-theoretic structure and the control mechanisms of collective motion in the group of animals based on the field measurement data. We are also interested in other examples of complex phenomena including biological systems as the research subjects.
Center for Data-driven Science and Artificial Intelligence, Professor
Keyword: Combinatorial reconfiguration, Computational complexity, Graph algorithms
Research: Structural analysis for solution space in combinatorial reconfiguration problems
In constraint satisfaction problems and graph problems, the solution space can be structured as a graph by defining natural adjacency relations between solutions. Theoretical analysis of the structural properties of such solution space graphs is conducted, focusing on investigating their computational complexity and possibilities for efficient estimation.
Center for Data-driven Science and Artificial Intelligence, Associate Professor
Keyword: Learning Analytics, Behavior Analysis
Research: Learning analytics in IT education and e-learning
To practice effective education with IT or e-learning, it is necessary to improve instruction and educational contents based on evidence. So, we investigate every different learning activities and model them through developing applications and tools for actual educational fields. We also study on methodology of analytics for these learning activities.
Center for Data-driven Science and Artificial Intelligence, Assistant Professor
Keyword: SRL Learning Analytics, SSI Data Analytics, Distributed Confidential Analytics
Research: Construction of a Visualization and Behavioral Analysis Framework for Supporting Self-Regulated Learning (SRL)
This study aims to promote learners’ autonomous learning by supporting Self-Regulated Learning (SRL) through the visualization of learning histories and the development of tools tailored to actual educational settings. It further analyzes and models behavioral data derived from classroom practice to construct a data-driven framework for understanding and enhancing SRL.
Center for Data-driven Science and Artificial Intelligence, Assistant Professor
Keyword: Statistical Science, Sparse Estimation, Model Selection, Real Estate Data Analysis
Research: Development of a data analysis method and its application
With sparse estimation, nonparametric regression, varying coefficient model, etc., we try to develop a new data analysis method and its estimation algorithm. For example, prediction for the prices of real estate and ecology investigation of marine lives are our targets, and real data analyses for them are conducted.
Akihito Takeuchi
Graduate School of Information Sciences, D1
Research: Classification of performer expertise based on acoustic features in Manzai Comedy
This research analyzes acoustic features such as speech timing and prosody in manzai comedy to classify the expertise of performers. By comparing highly trained professionals and amateurs, it aims to reveal structural differences in conversational rhythm ("ma") and contribute to applications in dialogue robots and improved human-to-human communication.
Simon Adrien
COLABS exchange student
Research: Learning analytics in blended learning
Blended learning is an emerging way of learning combining classroom and online learning. Furthermore, online resource store many logs about the user activity describing his learning behavior. Using this, we can investigate learning practices that lead to academic success, and try to predict students at risk of dropping out.
Center for Innovation in Education Research and Practice, Graduate School of Education, Assistant Professor
Concurrent position at the Center for Data-driven Science and Artificial Intelligence
Keyword: Educational Technology, Learning Support Systems, Speech Recognition
Research: Development of Smartphone-Based Learning Materials for Introductory Foreign Language Education
Learning a second foreign language for the first time at university requires continuous and repeated study. We are developing educational applications for smartphones—devices that most university students carry—to facilitate easy and accessible language learning.
Address: Multimedia Education and Research Complex, Kawauchi 41, Aoba-ku, Sendai, Japan, 980-8576
E-mail: mineaki.ohishi.a4 at tohoku.ac.jp (Mineaki Ohishi)