Multi-Source Pointer Network for Product Title Summarization
This work addresses the need for accurate and concise product title summarization in E-commerce applications, particularly for mobile users, though it is incremental as it builds on existing pointer network methods.
The paper tackled product title summarization for E-commerce mobile displays by introducing a multi-source pointer network to avoid irrelevant information and retain key details like brand and commodity names, achieving significant performance improvements over baselines and boosting click-through rates in online deployment.
In this paper, we study the product title summarization problem in E-commerce applications for display on mobile devices. Comparing with conventional sentence summarization, product title summarization has some extra and essential constraints. For example, factual errors or loss of the key information are intolerable for E-commerce applications. Therefore, we abstract two more constraints for product title summarization: (i) do not introduce irrelevant information; (ii) retain the key information (e.g., brand name and commodity name). To address these issues, we propose a novel multi-source pointer network by adding a new knowledge encoder for pointer network. The first constraint is handled by pointer mechanism. For the second constraint, we restore the key information by copying words from the knowledge encoder with the help of the soft gating mechanism. For evaluation, we build a large collection of real-world product titles along with human-written short titles. Experimental results demonstrate that our model significantly outperforms the other baselines. Finally, online deployment of our proposed model has yielded a significant business impact, as measured by the click-through rate.