Emotional Recognition of Crisis Groups in Sports Events Based on Cloud Computing

In sports events, a large amount of data closely related to the event is generated due to the involvement of a wide range of audiences, media and participants. These data contain rich emotional information, reflecting the emotional tendencies of the group toward the event. Especially in the crisis of sports events, changes in group emotions have a significant impact on the direction and outcome of the event. Therefore, in order to effectively monitor and analyze the emotional dynamics in sports events, a cloud computing based method for identifying group emotions in sports event crises is proposed. By collecting data related to the event, including audience feedback, media reports and social media comments, and integrating these data into a unified cloud computing platform. Subsequently, the data are cleaned and denoised to eliminate irrelevant information and noise. In order to unify data formats and standardize processing, the optimal segmentation function is adopted, which adaptively segments data based on its characteristics and distribution, thereby improving the accuracy and efficiency of data processing. On this basis, extract features related to crisis emotion recognition in sports events and complete data preprocessing work. Next, based on the dynamic evolution structure of the development law of crisis public opinion in sports events, the weights of each indicator were calculated by constructing group emotion recognition indicators. These indicators and weights together constitute the group emotion recognition model in sports event crises. Through cloud computing technology, real-time monitoring and recognition of group emotions during sports event crises have been achieved. The experimental results show that the recognition trend using this method is highly consistent with the actual development trend of the event, demonstrating excellent stability and convergence. By comparing the average fitness and mean square error of different methods, it was found that the recognition performance of this method is significantly better than other methods in the reference literature. In the recognition results of specific competition events, the recognition results of this method are completely consistent with the actual situation, accurately identifying positive and negative emotions in the actual situation. In addition, this method can effectively identify athletes’ emotional tendencies in different types of competitions, and has good robustness and practical application value. Therefore, this method provides a more reliable and effective tool for research and application in related fields, with broad application prospects.