Evaluation of sports public service under fuzzy integral and deep neural network

摘要

This study aims to explore the evaluation value of deep neural network and fuzzy integral (FI) technology on sports public services under the background of big data, so as to provide the public with more efficient and convenient services. A fuzzy integral-convolutional neural network (FI-CNN) algorithm model is constructed based on convolutional neural network (CNN) and FI, and is compared with the mixed National Institute of Standards and Technology, ImageNet, and Stanford sports event data sets. In addition, it is applied to the evaluation of public sports services in the eastern, central, western, and northeastern regions. The results show that the evaluation accuracy of the FI-CNN model based on the three data sets (98.8%, 85.9%, and 87.1%, respectively) are significantly higher than that of the not optimized FI-CNN model (96.3%, 82.8%, 83.2%), and the running time (633.5 s, 8169.4 s, and 1291.7 s, respectively) is much lower than that of the noise fusion convolutional neural network (NF-CNN) model (683.8 s, 8749.6 s, and 1673.2 s, respectively). The constructed FI-CNN evaluation model shows higher evaluation accuracy and shorter running time. The eastern region shows the highest investment in sports public services (4.26), the highest output of sports public services (3.98), and the highest effect of sports public services (4.23). In short, the overall performance of sports public service input, sports public service output, and sports public service effect in the eastern region is relatively good.

关键词