An ACO-Based Trust Inference Algorithm

The trust value between two nodes is important for their interactions in social network . In this paper we propose an ACO-based trust inference algorithm. First, we use the Ant Colony Optimization algorithm to find two trust trains with high trust value between any two nodes, since the most accurate...

Full description

Saved in:
Bibliographic Details
Published inIIAI-AAI 2014 : proceedings : 2014 IIAI 3rd International Conference on Advanced Applied Informatics : Kitakyushu, Japan, 31 August - 4 September 2014 pp. 216 - 220
Main Authors Leyi Shi, Yao Wang, Xin Liu
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.11.2014
Subjects
Accuracy
Algorithm design and analysis
Ant colony optimization
Clustering algorithms
Heuristic algorithms
Inference algorithms
Social network services
social network; social trust; ant colony optimization algorithm; trust trains
Online AccessGet full text
DOI10.1109/BWCCA.2014.70

Cover

More Information
Summary:The trust value between two nodes is important for their interactions in social network . In this paper we propose an ACO-based trust inference algorithm. First, we use the Ant Colony Optimization algorithm to find two trust trains with high trust value between any two nodes, since the most accurate information comes from the node with the highest trust value. We optimized the corresponding parameters of the ACO algorithm to make it adaptive to our problem. Indirect trust value between two nodes is calculated according to direct trust value and trust chains got by ACO. The statistical analysis indicates that our algorithm gains more accuracy of trust value computing compared with a classical trust algorithm.
DOI:10.1109/BWCCA.2014.70