NCG-LS: Named Entity Recognition Model Specializing for Analyzing Product Titles

Abstract

Entity recognition of product titles is essential for retrieving and recommending product information, where product title text has the characteristics of high entity density and fine type granularity. Most of the current studies focus on only one of the two features instead of considering the two challenges together. Our approach, named NCG-LS, proposes NEZHA-CNN-GlobalPointer architecture with the addition of label semantic networks, and uses multi-granularity contextual and label semantic information to fully capture the internal structure and category information of words and texts to improve the entity recognition accuracy. Through a series of experiments, we proved the efficiency of NCG-LS over a dataset of Chinese product titles from JD, improving the F1 value by 5.98%, when compared to the BERT-LSTM-CRF model on the product title corpus.