Abstract
Association football has been the subject of many research studies. In this work we present a study on player similarity using passing sequences extracted from games from the top-5 European football leagues during the 2017/2018 season. We present two different approaches: first, we only count the motifs a player is involved in; then we also take into consideration the specific position a player occupies in each motif. We also present a new way to objectively judge the quality of the generated models in football analytics. Our results show that the study of passing sequences can be used to study player similarity with relative success.

Abstract
Convolutional Neural Networks (CNN) have been successfully applied to images, text and audio, but their performance are not so good when applied to feature-based tabular data. Exceptions are works such as TabNet and DeepInsight, which employ end-to-end approaches. In this work, we propose an alternative way of using CNNs to model tabular data where knowledge is extracted from the feature space before being introduced to the network. Our strategy, Map-Optimize-Learn (MOL), changes the shape representation of samples in order to produce suitable input data for the CNN architecture. The strategy is applied to a real-world scenario of children and teenagers with cardiac pathology and compared against baseline and state of the art Machine Learning (ML) algorithms for tabular datasets. Preliminary results suggest that the strategy has potential to improve prediction quality of tabular data over end-to-end CNN methods and classical ML methods.

Abstract

Abstract

Abstract

Abstract