The Research on information recommendation model based on an improved deep learning Model

Abstract

In recent years, information recommendation system has been widely used in various fields. Deep learning is increasingly combined with information recommendation system to capture user preference or item interaction evolution over time. In this paper, we improve the deep learning model applied to information recommendation system, and use LDA (Linear Discriminant Analysis) classifier instead of the Softmax. Experimental results show that our method has better accuracy, recall rate and F1-Score. It has certain significance for the development of recommendation information service for follow-up research.