The Data Mining Laboratory (DMLab) has produced a wealth of tangible results through many years of collaborative research with industry, and through these efforts has forged strong trusting relationships. In the field of social science, DMLab is home to the research group that has not only amassed extensive research results through industry-academia collaboration but has also built and maintained relationships of mutual trust.

One of the principal avenues of international cooperation pursued by the Data Mining Laboratory is the collaborative research it conducts with the Marketing Department of the Graduate School of Business, Columbia University. Aside from our group conducting the research in multidimensional time-series and stream data analysis, there are no precedents for this kind of work in the data mining field. Moreover, a particularly significant contribution can be expected from DMLab on the theoretical side of customer movement analysis and advertising effect models.
By conducting collaborative research with a research team from Columbia University on ways to introduce these frontier technologies, DMLab is in a position to produce high-caliber results in the field of consumer behavior models. Moreover, the DMLab is looking to become a focal point of exchange with Columbia University and researchers active in applicable fields, and by so doing add a further spur to the Laboratory's research activities.

The DMLab will form a strategic research hub at the core of a network linking social science researchers and computer scientists in Japan. The DMLab has previously collaborated with computer scientists affiliated to Tokyo University, The National Institute of Informatics, Osaka University, and other institutions, and mandates itself to develop the fruits of this work into top-level marketing research.

The DMLab will construct a unique 'Theory& Practice Feedback Loop' through collaboration between academia and industry. To date, DMLab has validated theoretical models on the basis of various simulations and trials (experiments, questionnaire surveys) in the fields of computer science and social science. Yet, between theoretical models such as these and the real world there still remains an unavoidable gap. As a means to traverse this gap, the DMLab aims to promptly introduce theoretical models developed through industry-academia collaboration and to construct feedback loops which can validate their effectiveness in real-life situations.