Information filtering via biased random walk on coupled social network

Area of Science:

• Computer Science
• Artificial Intelligence
• Social Network Analysis

Background:

• Recommender systems have advanced significantly, yet often neglect social influence on user behavior.
• Existing methods primarily focus on user or item similarities, overlooking crucial social network dynamics.

Purpose of the Study:

• To develop a novel recommender system algorithm that integrates social influence with user preferences.
• To enhance recommendation accuracy and address the user cold-start problem.

Main Methods:

• A biased random walk algorithm was designed and implemented on coupled social networks.
• The algorithm considers both social interests and individual user preferences for generating recommendations.

Main Results:

• Numerical analyses on Epinions and Friendfeed datasets confirmed improved recommendation performance.
• Experimental results demonstrated superior effectiveness in alleviating the user cold-start problem compared to existing methods.

Conclusions:

• Integrating social influence significantly enhances recommender system performance.
• The proposed biased random walk algorithm offers a more effective solution for the user cold-start problem in recommender systems.