A Sequence to Sequence Model for Extracting Multiple Product Name Entities from Dialog
This work addresses a domain-specific problem for e-commerce voice ordering systems by enabling customers to order multiple items with one utterance, improving shopping efficiency and experience, though it is incremental as it builds on existing transformer-based methods.
The paper tackles the problem of recognizing multiple product name entities from voice ordering utterances in e-commerce, where existing systems like Amazon Alexa only capture single entities, and proposes Entity Transformer (ET) architectures that achieve a 12% performance improvement over non-neural models on their test set.
E-commerce voice ordering systems need to recognize multiple product name entities from ordering utterances. Existing voice ordering systems such as Amazon Alexa can capture only a single product name entity. This restrains users from ordering multiple items with one utterance. In recent years, pre-trained language models, e.g., BERT and GPT-2, have shown promising results on NLP benchmarks like Super-GLUE. However, they can't perfectly generalize to this Multiple Product Name Entity Recognition (MPNER) task due to the ambiguity in voice ordering utterances. To fill this research gap, we propose Entity Transformer (ET) neural network architectures which recognize up to 10 items in an utterance. In our evaluation, the best ET model (conveRT + ngram + ET) has a performance improvement of 12% on our test set compared to the non-neural model, and outperforms BERT with ET as well. This helps customers finalize their shopping cart via voice dialog, which improves shopping efficiency and experience.