At the last Annual Conference of Japan Society of Educational Technology (JSET) in 2005 and the Third WebCT Workshop in Japan, we have proposed a new mathematical analysis method of learning process via e-Learning which applied Complex Networks Theory. There, we have indicated that it is possible to identify some characteristics of communication network structure from log data of Bulletin Board System (BBS). The purpose of this paper is to develop the former studies and to clarify the following matters that is: 1) Uneven distribution in terms of network centrality and clustering coefficient is found in the learners' network structure which is formed on a virtual learning environment such as BBS: 2) The centrality and the coefficient are correlated; 3) Some interesting characteristics are observed among Knowledge Transforming Process, Knowledge Telling Process, and the learners' communication network structure.
