Knowledge Recommendation Based on Social Network Theory

Abstract

In knowledge management systems (KMS), the overabundance of information has forced users to spend more time searching for information that they need. If KMS can discover users' interests and initiatively recommend knowledge which users are interested in, it will greatly improve users' satisfaction. In this paper, based on social network theory, we propose a new method for establishing a knowledge network by analyzing users' retrieval records. With the network, we introduce an algorithm which is based on centralization analysis and cohesive subgroup analysis in social network theory to meet the requirements of recommendation. The success in a practical project demonstrates the effectiveness of the method.