Click for new scientific resources and news about Corona[COVID-19]

Paper Information

Journal:   JOURNAL OF ADVANCES IN COMPUTER ENGINEERING AND TECHNOLOGY   winter 2018 , Volume 4 , Number 1; Page(s) 21 To 26.

Utilizes the Community Detection for Increase Trust using Multiplex Networks

Author(s):  Habibi Rahimeh*, HAROUNABADI ALI
* ACRCE Khozestan
Today, e-commerce has occupied a large volume of economic exchanges. It is known as one of the most effective business practices. Predicted trust which means trusting an anonymous user is important in online communities. In this paper, the trust was predicted by combining two methods of multiplex network and community detection. In modeling the network in terms of a multiplex network, the relationships between users were different in each layer and each user had a rank in each layer. Then, the ratings of two layers including the weight of each layer were aggregated and four effective features of the Trust were achieved. Then, the network was divided into overlapping groups via community detection’ algorithms, each group representative was considered as the community centers and other features were extracted through similar comments. At the end, 48J decision tree algorithm was used to advance the work. The proposed method was assessed on Epinions data set and accuracy of trust was 96%.
Keyword(s): Trust,Community Detection,Multiplex Network,social Network
  • ندارد
مباني نظري و تجربي ونداليسم: مروري بر يافته هاي يك تحقيق Yearly Visit 29
Latest on Blog
Enter SID Blog