Codes written for studies conducted while assisting research for Chris Snijders in Jheronimus Academy of Data Science
The aim of this study was to estimate soccer players' playing style in a data-driven manner through the use of spatio-temporal data. In this way, a 79 length vector is defined for each player for describing his style of play. These vectors can be later used to determine the similarity between players, allowing to find the players who are most similar to any given player. The preliminary results of the study were also published as a web application at https://manfredi.pythonanywhere.com/.