Enhanced Collaborative Filtering for Personalized E-Government Recommendation

The problems with the information overload of e-government websites have been a big obstacle for users to make decisions. One promising approach solve this problem is deploy an intelligent recommendation system on platforms. Collaborative filtering (CF) has shown its superiority by characterizing both items and latent features inferred from user–item interaction matrix. A fundamental challenge enhance expression user or/and item embedding implicit feedback. This negatively affected performance in e-government. In paper, we firstly propose learn positive items’ leveraging negative original features. We present mixed collaborative (NMCF) method CF-based recommender system. Such mixing beneficial extending expressiveness Comprehensive experimentation real-world dataset showed that our improved significantly compared state-of-the-art baseline algorithms.