Improving women football tactics analysis by using extreme learning and accumulated optimization algorithm.

The research will fill the existing gaps in the current approaches to the analysis of the strategies of women football by offering a new deep-learning framework and creating the Accumulated Chimp Optimization Algorithm (ACHOA). Traditional approaches like Deep Convolutional Neural Networks (DCNN) are very processing-intensive and thus they are not effective in real-time applications. Moreover, the random factor selection that is typical of the existing approaches involving the use of Deep Extreme Learning Machines (DELM) makes the reliability and accuracy of such models in terms of forecasting less than satisfactory. In order to overcome these deficiencies, we propose a new ACHOA-enhanced DELM model that enhances the performance of the model by making the parameters stable. In a way our method is a supplement to video recordings, as it makes use of the continuous frames to determine how effective a player is, how many skills may be estimated to be under the hood, and where he/she makes positional errors, which enables a more dramatically tactical goal that could not be reached before. One of the main contributions of this study is the ACHOA itself that enhances the reliability of the DELM model. The model has been trained and tested on video data from the 2021-2022 UEFA Women's Champions League and the model has been validated using this real-world data. It achieved a classification accuracy of over 95%, and there was better agreement between the factors and the evaluations made by the experts. Further, the possibility of real-time monitoring helps to make tactical shifts, which enhances pre-game measures and develops performance metrics in women's soccer.