A Matrix Factorization Technique with Word Embedding for Recommendation
Proceedings of 2020 the 10th International Workshop on Computer Science and Engineering · 2020
In recent years, machine learning has achieved great success in computer vision and natural language processing.However, little attention has been paid to how to apply these explorations in the field of natural language processing to recommendation systems.In this paper, inspired by word embedding techniques, we take advantage of word embedding techniques and propose a word embedding based recommendation method.Specifically, word2vec is first used to train a corpus composed of item name and item attributes, the implicit item information is obtained, and then incorporate them into a matrix factorization model, which combines both latent and pre-learned features for recommendation.We have performed experiments on the MovieLens dataset, and the results show that our algorithm performs better than traditional matrix factorization algorithms and has a higher prediction rate.