Treffer: Novel data-driven approach for enhanced prediction of sports outcomes.

Creating a successful sports performance analysis procedure is a desirable task for sports team managers. Throughout the past few decades, the sports market has experienced remarkable growth. Because it offers practical data for sports market activities, sports outcome prediction is an appealing sports analytical task. This research proposed using Machine Learning (ML) algorithms to predict cricket game score outcomes, which would enhance the sports outcome prediction process. We proposed a novel crayfish optimizer with fine-tuned XG Boost (CO-FXG Boost) for the prediction of sports outcomes. The dataset information were gathered from Kaggle which contained information from cricket events. Min-max normalization was used as a phase of preprocessing to obtain the data. To predict sports results, the performance assessment measures include F1-measure (84%), accuracy (75.62%), recall (81%), and precision (86.21%). When compared to other conventional algorithms, the suggested sports outcome prediction technique can deliver a promising prediction result. [ABSTRACT FROM AUTHOR]