استخراج مشخصات کاربر بر اساس جریان کاری و جریان اطلاعات

A collaborative team usually consists of team members with various domains. These members’ demands for knowledge are also different from each other. For recommending potentially useful knowledge to suitable members, their user profiles should be well managed and maintained. User profile can be input by the members, but a more intelligent way should be the automatic extraction of the user profiles. Workflow and information flow are two types of collaborative processes, which exist behind every collaborative team. This paper is mainly concerned with how to extract these team members’ user profile from the two types of contexts: workflow and information flow. This paper defines a model for the user profile. Then some methods are proposed for extracting the profile information on the basis of workflow and information flow. This study on the user profile extraction can pave the way for developing knowledge recommender systems, which can recommend proper knowledge to proper team members with a collaborative team.

Knowledge sharing is important for collaborative teams, in which team members are usually geographically and disciplinary distributed. As the team members often belong to different domains, have different backgrounds, their requirements for knowledge are also different from each other (Zhen, Jiang, & Song, 2011). For realizing efficient knowledge sharing among a collaborative team, an important issue lies in how to accurately obtain team members’ profiles, which describe their demands for knowledge. The user profiles should be extracted from context environments intelligently rather than be input by users manually (Adomavicius, Sankaranarayanan, Sen, & Tuzhilin, 2005). Workflow and information flow are two types of collaborative processes, which actually exist behind every collaborative team (Zhuge, 2002, Zhuge, 2002 and Zhuge, 2006). Workflow and information flow can provide us some contextual information for obtaining the members’ profiles. This paper defines a model for the user profile. Then some methods are proposed for extracting the profile information on the basis of workflow and information flow. This study on the user profile extraction can pave the way for developing knowledge recommender systems, which can recommend proper knowledge to proper team members with a collaborative team (Smirnov, Pashkin, Chilov, & Levashova, 2004). The remainder of this paper is organized as follows. Related works are reviewed in Section 2. A model of user profile is defined in Section 3. Section 4 elaborates on the proposed approached for extracting user profile based on workflow and information flow. Section 5 illustrates the results from some numerical experiments. Closing remarks are given in the last section.