Application Research of Intention Recognition and Semantic Slot Filling Combined Model in Electric Power Customer Service.

The development of artificial intelligence technology is changing with each passing day, intelligent voice technology has been applied in more and more industries and scenarios. In the human-machine dialogue system, the natural language understanding module is responsible for converting the natural language text input by the user into a structured semantic representation that is convenient for machine understanding and calculation. This paper studies the construction of a joint model of intention recognition and slot filling in Natural Language Understanding (NLU) in a human-machine dialogue system and conducts an application experiment on the performance of the existing model in a laboratory environment. The experimental reveals the traditional single model is combined. The model has a better effect on the understanding of interactive information, hierarchical information and contextual information.