Expert Finding Considering Dynamic Profiles and Trust in Social Networks

Abstract

Recently, social network services that express individual opinions and thoughts have been significantly developed. As unreliable information is generated and shared by arbitrary users in social network services, many studies have been conducted to find users who provide reliable and professional information. In this paper, we propose an expert finding scheme to discover users who can answer users’ questions professionally in social network services. We use a dynamic profile to extract the user’s latest interest through an analysis of the user’s recent activity. To improve the accuracy of the expert finding results, we consider the user trust and response quality. We conduct a performance evaluation with the existing schemes through various experiments to verify the superiority of the proposed scheme.

Expert Recommendation for Answering Questions on Social Media

“…Ref. [29] proposes a method of searching an expert by creating user profiles based on the social activity analysis of users and using human relations on social media. A dynamic user profile is generated through the periodic analysis of recent activity by a user.…”

“…Consequently, the trust in the recommendation result is degraded. The method of recommending an expert by generating dynamic profiles of users exhibits higher precision than the profile-based expert recommendation method [29]. However, such an expert recommendation method cannot recommend an expert precisely when the number of users who have formed human relations with the user is insufficient or when the users are actively engaged in the field of interest.…”

“…The superiority of the proposed method is verified through the performance comparison between the proposed expert recommendation method and the existing methods. First, a performance evaluation is conducted using the existing method developed by [29], which recommends an expert by analyzing the activity information of users. After generating dynamic profiles by collecting and analyzing the recent activities of each user, the method developed by [29] generates a question cluster using the ontology when a question is requested.…”

“…To ameliorate this drawback of the profile-based expert recommendation method, studies have been conducted on methods of recommending an expert by analyzing user activity details [29][30][31][32]. This method determines the similarity of a user's question by considering activities such as the user's wr recommended.…”

“…Such a method recommends an expert more precisely compared to the profile-based expert recommendation method. In [29], an expert recommendation method was proposed using dynamic profiles, which are generated by analyzing the social activities of users. This method considers the generated dynamic profile of a user with a question, and a user with the highest similarity is recommended as an expert.…”