A0443
Title: Using aggregated cluster validity indexes to cluster football players performance data
Authors: Christian Hennig - University of Bologna (Italy) [presenting]
Serhat Akhanli - Mugla Sitki Kocman University (Turkey)
Abstract: In cluster analysis applications, it is often difficult to decide between the many available clustering methods and to choose an appropriate number of clusters. We provide a case study to apply an approach based on several validation criteria that refer to different desirable characteristics of clustering, including stability. These characteristics are chosen based on the aim of clustering, and this allows the definition of a suitable validation index as a weighted average of calibrated individual indexes measuring the desirable features. We analyse football (soccer) player performance data with mixed type variables from the 2014-15 season of eight European major leagues. We cluster these data based on a tailor-made dissimilarity measure. We derive two different clusterings, namely a partition of the data set into major groups of essentially different players, and a second one that divides the data set into many small clusters, which can be used for finding players with a very similar profile to a given player. It is discussed what characteristics are desirable for these clusterings. Weighting the criteria for the second clustering is informed by a survey of football experts.
