Multi-data mining for understanding leadership behavior

Abstract

In this chapter, we propose an approach for understanding leadership behavior in dot-jp, a non-profit organizations in Japan, by analyzing heterogeneous multi-data composed of questionnaire and e-mail archives. From questionnaire, we obtain qualitative relations between staff members. From e-mail archives, we obtain quantitative relations between staff members. By integrating these different kind of relations, we discovered that a leader should give supportive messages as well as directive messages to staff members to constract his/her reliable relation with staff members. © 2008 Springer-Verlag Berlin Heidelberg.