EENMF: An End-to-End Neural Matching Framework for E-Commerce Sponsored Search
This work addresses efficiency and effectiveness challenges in e-commerce sponsored search, which is crucial for revenue generation, but it appears incremental as it builds on existing multi-stage architectures with neural enhancements.
The authors tackled the problem of ad matching in e-commerce sponsored search by proposing an end-to-end neural matching framework (EENMF) that integrates vector-based ad retrieval and neural network-based ad pre-ranking, resulting in significant performance improvements over baselines in real traffic of a large-scale system.
E-commerce sponsored search contributes an important part of revenue for the e-commerce company. In consideration of effectiveness and efficiency, a large-scale sponsored search system commonly adopts a multi-stage architecture. We name these stages as ad retrieval, ad pre-ranking and ad ranking. Ad retrieval and ad pre-ranking are collectively referred to as ad matching in this paper. We propose an end-to-end neural matching framework (EENMF) to model two tasks---vector-based ad retrieval and neural networks based ad pre-ranking. Under the deep matching framework, vector-based ad retrieval harnesses user recent behavior sequence to retrieve relevant ad candidates without the constraint of keyword bidding. Simultaneously, the deep model is employed to perform the global pre-ranking of ad candidates from multiple retrieval paths effectively and efficiently. Besides, the proposed model tries to optimize the pointwise cross-entropy loss which is consistent with the objective of predict models in the ranking stage. We conduct extensive evaluation to validate the performance of the proposed framework. In the real traffic of a large-scale e-commerce sponsored search, the proposed approach significantly outperforms the baseline.