HybridBERT4Rec: A Hybrid (Content-Based Filtering and Collaborative Filtering) Recommender System Based on BERT

Because a user’s behavior depends mainly on the user’s current interests, which can change over time, the sequential recommendation approach has become more prevalent in recommender systems. Many methods have been proposed that aim to model sequential recommendations. One of these is BERT4Rec, which applies the bidirectional-encoder-representations-from-transformers (BERT) technique to model user behavior sequences by considering the target user’s historical data, i.e., a content-based filtering (CBF) approach. Despite BERT4Rec’s effectiveness, we argue that considering only this historical data is insufficient to provide the most accurate recommendation. We believe that if BERT were to consider other users’ interactions in its analysis, it would increase the model accuracy. Therefore, we propose a new method called HybridBERT4Rec, which applies BERT to both CBF and collaborative filtering (CF). For CBF, we want to extract the characteristics of the target user’s interactions with purchased items. (We implement this in the same way as in BERT4Rec, with our model generating a target user profile.) For CF, we want to find neighboring users who are similar to the target user. Here, we extract the target item’s characteristics using all other users who rated the target item as a second input to BERT. This generates a target item profile. After obtaining both profiles, we use them to predict a rating score. We experimented with three datasets, finding that our model was more accurate than the original BERT4Rec.