Exploring the use of Time-Dependent Cross-Network Information for Personalized Recommendations

TL;DR

This paper introduces a cross-network, time-aware recommendation system that leverages auxiliary data from social media to improve the timeliness, accuracy, diversity, and novelty of recommendations, especially when user profiles are incomplete or preferences change over time.

Contribution

It presents a novel method that combines historical user models from multiple networks with time-aware latent factors for improved recommendations.

Findings

Outperforms baseline methods in accuracy, novelty, and diversity.

Utilizes Twitter data to enhance YouTube recommendations.

Effective across different time granularities.

Abstract

The overwhelming volume and complexity of information in online applications make recommendation essential for users to find information of interest. However, two major limitations that coexist in real world applications (1) incomplete user profiles, and (2) the dynamic nature of user preferences continue to degrade recommender quality in aspects such as timeliness, accuracy, diversity and novelty. To address both the above limitations in a single solution, we propose a novel cross-network time aware recommender solution. The solution first learns historical user models in the target network by aggregating user preferences from multiple source networks. Second, user level time aware latent factors are learnt to develop current user models from the historical models and conduct timely recommendations. We illustrate our solution by using auxiliary information from the Twitter source network to improve recommendations for the YouTube target network. Experiments conducted using multiple time aware and cross-network baselines under different time granularities show that the proposed solution achieves superior performance in terms of accuracy, novelty and diversity.