Followee recommendation in microblog using matrix factorization model with structural regularization.

Area of Science:

• Computer Science
• Social Network Analysis
• Information Retrieval

Background:

• Microblogging platforms have rapidly grown as primary communication and information-sharing tools.
• Recommending high-quality information sources (followees) is a critical and competitive service for microblog users.

Purpose of the Study:

• To develop an accurate and effective model for followee recommendation in microblogging environments.
• To address the challenge of identifying valuable information sources within large social networks.

Main Methods:

• A matrix factorization model, adapted from traditional item recommender systems, was employed.
• Structural regularization was incorporated to utilize the inherent social network structure and constrain the model.
• The model was evaluated using a real-world microblogging dataset.

Main Results:

• The proposed matrix factorization model with structural regularization demonstrated promising results in improving followee recommendation accuracy.
• Exploiting social network structure significantly enhanced the performance of the recommendation system.
• The model proved effective in identifying users who are high-quality information sources.

Conclusions:

• The developed model offers a promising approach for enhancing followee recommendation services on microblogs.
• Integrating social network structural information is key to improving the accuracy of these recommendations.
• This research contributes to the field of recommender systems within the context of social media platforms.