Application of Artificial Intelligence in Customer Service Field
Xingping Wu, Xusheng Liu, Qianjun Wu, Xiaolong Wang · 2021 3rd International Conference on Applied Machine Learning (ICAML) · 2021
As the demand for customer service continues to increase, more and more companies are trying to apply artificial intelligence technology in the field of customer service, enabling customer service through intelligent customer service, reducing customer service pressure and reducing operating costs. Currently, the existing intelligent customer service has a limited degree of intelligence and can only answer simple user questions, and it is difficult to understand complex user expressions. In order to solve the problem of low accuracy of multi-round dialogue language understanding, this paper proposes a language understanding model based on the fusion of CNN and Attention. The model builds an “intention-slot” joint model based on the “encoding-decoding” framework, and uses hidden semantic information that combines intention recognition and slot filling, avoiding the problem of information loss in traditional isolated tasks, and achieving end-to-end language understanding. At the same time, an improved attention mechanism based on CNN is introduced in the decoding process to reduce the interference of redundant information in the original text, thereby improving the accuracy of language understanding. Finally, by applying the model to intelligent customer service, it is verified through experimental comparison that the proposed fusion model has a good effect in improving the performance of intent recognition and slot filling, and can improve the user experience of intelligent customer service.