Name
Title | ||
|---|---|---|
A reliability and link analysis based method for mining domain experts in dynamic social networks. |
Abstract | ||
|---|---|---|
People express opinions or convey some emotion in a form of communities in a specific social network such as Twitter, Facebook, and Google Plus and so on. Researches have applied link analysis to capture clusters or detect communities, as well as mine and analyze sentiments published on theWeb. Most previous approaches are lack of evaluating the reliability of the information and exploring the specialty in specific areas. Besides, the user possessing lowauthority value does not mean he/she still has lower authority in his/her own community. Motivated by that, a synthetic method is proposed to extract domain experts through analyzing the information on the Web and in-degree and out-degree of the set of nodes in the large social networks. In addition, we consider the temporal factor in the process of optimizing the final objective function. Experimental results indicate that our proposed method DEM-RLA, focused on the reliability of information and authority of users in a small community of a complex social network, is very useful for the prediction of domain experts. According to this research, we offer a more comprehensive insight for the task of mining domain experts in a complex network. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.3233/JIFS-161205 | JOURNAL OF INTELLIGENT & FUZZY SYSTEMS |
Keywords | Field | DocType |
Social network,information reliability,link analysis,temporal trend,domain experts | Social network,Link analysis,Artificial intelligence,Mathematics,Machine learning | Journal |
Volume | Issue | ISSN |
34 | 4 | 1064-1246 |
Citations | PageRank | References |
0 | 0.34 | 20 |
Authors | ||
3 | ||