Design of Virtual Assistant and Human Machine Dialogue System Based on Natural Language Processing Algorithms

Virtual assistants are already quite common in current mobile devices. Through functions similar to human secretaries, they bring people closer to cold machines, improve user experience, and also facilitate users' work needs. This article designs a virtual assistant software based on the NLP (Natural Language Processing) algorithm and a human-machine dialogue system. The experimental results show that the average coverage rates of DL (Deep Learning) and GA (Genetic Algorithm) are 72.1% and 63.9%, respectively. The coverage rate of the method in this article is 82.3%. In order to further understand the impact of proposition coverage on answer reasoning, this paper continued the experiment and obtained the experimental results as shown in the figure. Compared with the method proposed in this paper, the accuracy of DL and GA decreased by 4.7% and 12.4%, respectively. In a sense, this is somewhat counterintuitive because many examples use DL and GA methods without proposition and context relevance. But from the results obtained, we can infer that the method proposed in this article does indeed contribute to the inference of the answer. It can be seen that the system designed in this article increases the software's fun and functional diversity, which can better improve the interaction experience between virtual assistants and users.