In the realm of professional sports, particularly football, data scouting has become a crucial tool for clubs worldwide. This article introduces a novel approach to data scouting using AI, specifically supervised learning, to assist scouts in identifying talent from a vast pool of players. By employing stacked deep neural networks, the model aims to provide comprehensive insights into player potential by combining multiple network outputs. The dataset utilized consists of over 7,000 players from various leagues globally, focusing on raw statistics and normalized metrics per minute played. The labeling process involves assessing player evolution based on league transfers over specific seasons to assign numerical potential scores. Through network stacking techniques and iterative model training, the study demonstrates improved accuracy in identifying promising players across different positions. The results showcase the efficacy of the model in predicting player potential and its potential utility as a scouting tool for football clubs.
To Another Language
from source content
arxiv.org
Key Insights Distilled From
by Simon Lacan ... at arxiv.org 03-15-2024
https://arxiv.org/pdf/2403.08835.pdfDeeper Inquiries