Analyzing World Cup Impact Through an Evolutionary Optimization Approach Based on Sentiment Polarity with Pre-trained Word Embeddings.

The 2022 Qatar World Cup created massive global attention and generated widespread discussions on different social media platforms, including the X platform. The event was the subject of intense debate after Qatar was announced as the host. Opinions were divided, with supporters and critics weighing in based on political, ethical, cultural, and social considerations. Public sentiments evolved throughout three key phases-before, during, and after the event-shaped by numerous factors. This study aims to analyze these sentiments during these three stages based on a novel hybrid evolutionary approach. Three versions for each stage were produced by applying pre-trained word embeddings with 100 and 400 features and sentiment features combined with word embeddings. In total, nine different versions of datasets were employed to examine the proposed approach. Furthermore, five different metaheuristic algorithms were applied: the multi-verse optimizer (MVO), the genetic algorithm (GA), the particle swarm optimization (PSO), the salp swarm algorithm (SSA), and the whale optimization algorithm (WOA). The five metaheuristic algorithms were combined with the feature selection-support vector machine (FS-SVM) and weighting-support vector machine (WSVM) to examine the newly created dataset versions. The results reveal that people's perspectives shifted from negative before the event to positive during and after the event. Moreover, a comparison of the proposed MVO-WSVM and MVO-SVM-Fs approaches with other metaheuristic algorithms showed the superior accuracy of the proposed approaches in sentiment prediction. [ABSTRACT FROM AUTHOR]