Abstract
In recent years, recommendation systems based on collaborative filtering (CF) have achieved a high performance. Most of the existing recommendation systems use similarity measures to determine the suitability of items for users based on latent factor models (LFM). However, these recommendation systems reduce the explainability of recommendations and hide the reasons for recommending specific items. As a result, users tend to distrust the recommendation results. To address this problem, we propose the neural explicit factor model (NEFM). Based on the user–item rating matrix, we propose adding both user–feature attention matrix and an item–feature quality matrix to improve the explainability of user and item vectors. In addition, a feedforward neural network and a one-dimensional convolutional neural network extract features from user, item and the item–feature vector. Finally, a prediction layer performs the inner product of user data, item data, and item features. Experiments on the MovieLens and Yahoo Movies datasets validate the proposed model, and comparisons with similar recommendation models show the higher accuracy and explainability of our proposal.
Highlights
With the accelerated growth of information availability, people face more choices over time
The main contributions of this work can be summarized as follows: 1. We propose a novel neural explicit factor model (NEFM) based on collaborative filtering (CF) to extract information about users, items, and feature vectors through nonlinear projection of a neural network
We propose NEFM to construct an explainable recommendation system based on CF
Summary
With the accelerated growth of information availability, people face more choices over time. Recommender systems play a pivotal role in alleviating information overload [1] and are widely used in applications such as Web search, e-commerce, and entertainment. Matrix factorization (MF) [5], a widely used CF method, embeds user/item IDs into a vector and models the user–item interaction by applying the inner product [6], [7]. MF is a latent factor model (LFM) [8]–[10] in which the latent factor represents the underlying reason of a user’s preference for an item (e.g., a user likes a movie because it combines action and comedy). LFMs like MF provide a high efficiency, the latent factors hide from the users the reasons regarding the generated recommendations
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have