A Review-based Context-Aware Recommender Systems: Using Custom NER and Factorization Machines

Recommender Systems depend fundamentally on user feedback to provide recommendation. Classical Recom-menders are based only historical data and also suffer from several problems linked the lack of such as sparsity. Users’ reviews represent a massive amount valuable rich knowledge information, but they still ignored by most current recommender systems. Information users’ preferences contextual could be extracted integrated into more accurate recommendations. In this paper, we present Context Aware System model, Bidirectional Encoder Representations Transformers (BERT) pretrained model customize Named Entity Recognition (NER). The allows automatically extract information then insert Contextual Machine Factorization compte predict ratings. Empirical results show that our improves quality recommendation outperforms existing Systems.