An approach to expert recommendation based on fuzzy linguistic method and fuzzy text classification in knowledge management systems

Since organizational tacit knowledge such as know-how and experiences usually resides in the owner’s brain, consulting the expert is an effective and efficient way to utilize this type of knowledge. However, users are no longer able to effectively find the appropriate experts in the knowledge management system due to the complexity and diversity of the expertise and the knowledge needs. In this paper, an approach to expert recommendation is proposed to assist the user to find the required experts. The method adopts the fuzzy linguistic method to construct the expert profile, that is, to model expert’s expertise. In addition, the fuzzy text classifier is used to get the relevant degree of the document to each knowledge area when the document is registered, which is the base of the following user profile construction. Then, the user profile consisting of the time and the relevance factors of the rated documents is constructed to derive the overall knowledge needs level of the user. Consequently, the expert that fulfills the knowledge needs most is recommended based on the similarity between the derived expert profile and the user profile. The developed prototype system, “knowledge management system in aircraft industry company”, is introduced and the experimental results show the proposed approach is feasible and effective.

Research highlights
► We propose an expert recommendation approach to foster tacit knowledge sharing.
► We adopt Fuzzy linguistic group decision making to construct the expert profile.
► We construct the fuzzy text classifier to analyze the registered documents.
► We analyze the rated documents to identify the users knowledge needs.
► We use LWA operator to calculate the matching degree of the expert and the user.