Graph Intention Network for Click-through Rate Prediction in Sponsored Search

Abstract

Estimating click-through rate (CTR) accurately has an essential impact on improving user experience and revenue in sponsored search. For CTR prediction model, it is necessary to make out user's real-time search intention. Most of the current work is to mine their intentions based on users' real-time behaviors. However, it is difficult to capture the intention when user behaviors are sparse, causing the behavior sparsity problem. Moreover, it is difficult for user to jump out of their specific historical behaviors for possible interest exploration, namely weak generalization problem. We propose a new approach Graph Intention Network (GIN) based on co-occurrence commodity graph to mine user intention. By adopting multi-layered graph diffusion, GIN enriches user behaviors to solve the behavior sparsity problem. By introducing co-occurrence relationship of commodities to explore the potential preferences, the weak generalization problem is also alleviated. To the best of our knowledge, the GIN method is the first to introduce graph learning for user intention mining in CTR prediction and propose end-to-end joint training of graph learning and CTR prediction tasks in sponsored search. At present, GIN has achieved excellent offline results on the real-world data of the e-commerce platform outperforming existing deep learning models, and has been running stable tests online and achieved significant CTR improvements. CCS CONCEPTS• Information systems → Sponsored search advertising; Recommender systems.

Social media intention mining for sustainable information systems: categories, taxonomy, datasets and challenges

“…Li et al [25] presented graph intention network-based model to detect behavioral intentions in click through rate (CTR). Real-world data of e-commerce platform has been used to assess the proposed model, and it delivers promising results.…”

“…Real-world data of e-commerce platform has been used to assess the proposed model, and it delivers promising results. Giannopoulos et al [26] proposed a clientcentered intent-aware query framework to shield user data privacy in personalized web search [25]. Hashemi et al [27] proposed a multiple intent model to infer users' behavioral intention from America Online (AOL) search query log.…”

Order-aware Embedding Neural Network for CTR Prediction

“…FiGNN models feature interactions via graph propagation on the fully-connected fields graph [16]. GIN utilizes user behaviors to construct a co-occurrence commodity graph to mine user intention [15]. GCMC [2] treats the recommendation task as a link prediction problem and employs a graph auto-encoder framework on the user-item bipartite graph to learn user and item embeddings.…”

GraphSAIL

“…Recently, with the developments and improvements of graph neural networks (GNNs) [5,10,32], graphs have proven to be among the best performing mechanisms for modeling the abundant relationship information present in recommendation datasets. Useritem interaction graphs [23,24,29] and item-item co-occurrence graphs [12] have been employed and shown to alleviate the data sparsity and cold start problems and improve recommendation relevance [12,22,24,29]. However, as is the case for deep recommendation models, graph neural network (GNN) based recommendation systems usually suffer from a low training efficiency.…”