A Virtual Community Member’s Referability Determination Model

Abstract

This paper develops a “virtual community member’s referability determination” model applicable to virtual communities by which to measure the referable value of virtual community members. In terms of community dministrators, there are standard and fair member domain specialty management indicators to enhance the substantial effectiveness of knowledge contributors’ sharing behavior control. In terms of knowledge demanders, the cost of evaluating the members' referability is reduced, and the probability of obtaining the required and correct knowledge is increased. In terms of knowledge contributors, there will be precise and rational member referability evaluation scores, and the defects in the existing virtual community’s incentive mechanisms will be remedied. Thus, the circulation of less referable information can be reduced, with more community members being willing to share knowledge, with the accumulation of professional public knowledge becoming enhanced. Finally the utilization of virtual communities can be activated to catalyze continuous comprehensive development of virtual communities. In addition, this paper develops a Web-based system accordingly for case verification to confirm the feasibility of the methodology. The verification results show that when the system uses about 600 training knowledge articles, the performance of the inference indicators of system can be increased to 82%. Generally speaking, the system performance grows continuously with the periods and training load, and eventually reaches stable and good performance level.