Predicting the probability that a user will click on specific advertisement has been prevalent issue in online advertising, attracting much research attention past decades. As hot frontier driven by industrial needs, recent years have witnessed more and novel learning models employed to improve advertising CTR prediction. Although extant provides necessary details algorithmic design for address...

Modeling feature interactions is of crucial importance to predict click-through rate (CTR) in industrial recommender systems. Because great performance and efficiency, the factorization machine (FM) has been a popular approach learn interaction. Recently, several variants FM are proposed improve its performance, they have proven field information play an important role. However, feature-length ...

Aiming at the defect that click-through rate of marketing advertisements cannot provide accurate prediction results for company in time strategy Internet companies, this paper uses a deep learning algorithm to establish model advertisements. The suggested is called high-order cross-feature network (HCN). Furthermore, also introduces combination feature vectors into graph structure and as nodes ...