Research on English Corpus Tagging Based on BiLSTM model
Juanjuan Guo · 2021 5th Asian Conference on Artificial Intelligence Technology (ACAIT) · 2021
The recognition and tagging of special words in English corpus can effectively improve students' learning efficiency. Based on BiLSTM model and CRF model, a BiLSTM-CRF model model is constructed to recognize and automatically label special words in English corpus. The results show that the average accuracy of BiLSTM-CRF model is 95.35% and the average recall rate is 94.83%, which are much higher than other models. We can know from the above that BiLSTM-CRF model can label English professional corpora well and is a practical method.