Joint Intention Detection and Semantic Slot Filling Based on BLSTM and Attention
Tingting Chen, Lin Min, Yanling Li · 2019
Spoken language understanding (SLU) of the dialogue system usually involves two tasks: intent detection and semantic slot filling. The current Joint intention detection and semantic slot filling has become the mainstream method of SLU research. A Bidirectional long short-term memory (BLSTM)model based on the attention mechanism is used to jointly identify the intent and semantic slot filling of the Hohhot bus query. The experimental results show that the model achieves a good performance in the intent detection and semantic slot filling, and the result based on the character mark is better than the one based on word mark. The F1 score is better than the others based on the LSTM model.