Towards socio-economic and culturally-aware recommender systems

Recommender systems have gained increasing attention to personalize consumer preferences. While these systems have primarily focused on applications such as advertisement recommendations (e.g., Google), personalized suggestions (e.g., Netflix and Spotify), and retail selection (e.g., Amazon), there is potential for these systems to benefit from a more global, socio-economic, and culturally aware approach, particularly as companies seek to expand into diverse markets. This research investigates the potential of a recommender system that considers cultural identity and socio-economic factors. We review the most recent developments in context-aware recommender systems and explore the impact of cultural identity and socio-economic factors on consumer preferences. We propose an ontology and approach for incorporating these factors into recommender systems. We present a scenario in consumer subscription plan selection within the entertainment industry. We argue that existing recommender systems cannot precisely understand user preferences due to a lack of awareness of socio-economic factors and cultural identity. They also fail to update recommendations in response to changing socio-economic conditions. We explore various machine learning models and develop a final artificial neural network model (ANN) that addresses this gap. We evaluate the effectiveness of socio-economic and culturally aware recommender systems across four dimensions: Precision, Accuracy, F1, and Recall. We find that a highly tuned ANN model incorporating domain-specific data, select cultural indices and relevant socio-economic factors predicts user preference in subscriptions with an accuracy of 95%, a precision of 94%, a F1 Score of 92%, and a Recall of 90%.