Affiliation:
1. Center for Sport and Exercise Sciences, University of Malaya, Kuala Lumpur 50603, Malaysia
2. Institute of Physical Education, Neijiang Normal University, Neijiang 641000, Sichuan, China
Abstract
China unanimously believed that Emei Martial Arts has the essence of self-improvement. It is a spiritual force that actively promotes human progress. It should be protected and continuously innovated to make it a spiritual pillar of people. However, its promotion faces huge challenges. In recent years, neural networks have made great progress in various fields, such as speech recognition, computer vision, and natural language understanding. On this basis, the combination of neural networks and traditional recommendation methods is helpful for the better development of Emei Martial Arts promotion. Neural networks have a direct analog interaction function and perform coordinated filtering directly through interactive data. Due to the effectiveness of the structure, the neural network can mine nonlinear implicit relationships from the data and find the martial arts items that users want to promote. In order to enhance the effectiveness of the standard recommendation algorithm, a deep neural network-based recommendation algorithm is paired with a neural network-based recommendation algorithm that is proposed in this article. The recall rate of the upgraded deep neural network recommendation model is up to 80%, whereas the recall rate of the model without enhancement is up to 40%, according to the experimental results of this article. The upgraded deep neural network’s recommendation model has a recall rate that is 4% greater than the baseline model. It showed that the recommendation algorithm combined with a neural network has a better recommendation effect so as to achieve a better effect of Emei Martial Arts promotion so that Emei Martial Arts culture can be carried forward and economic development can be promoted at the same time.
Funder
Neijiang Normal University
Subject
Electrical and Electronic Engineering,Energy Engineering and Power Technology,Modeling and Simulation
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献