Advanced search
Start date

Data Mining for Individual and Collective Analysis in Team Sports


With the enhancement and miniaturization of sensors capable of obtaining and transmitting various types of data, the Internet of Things has been notably gaining space in both scientific research and industry. Sports analysis is among a multitude of applications of such a technology. Specifically, sensors have aided the statistical analysis of the athletes' performances. This type of devices is already capable of transmitting data such as speed, heart rate and player positioning in real time. Those data have been increasingly used by elite teams in various sports, such as soccer, basketball, and rugby. Despite the large and valuable volume of information obtained, the current software tools for data analysis focus on the visual and individualized analysis. In other words, the data obtained is only presented to the technical staff through visualization tools. For the sake of collective analysis, the data is individually presented for each player or projected on the field or court, disregarding any type of relationship between the athletes. In a practical standpoint, the possibilities of analysis are virtually restricted to observing physical factors, which can help to avoid injuries and to develop personalized physical training for each player. On the other hand, support for tactical decisions and collective performance of the teams is practically null. The objective of this project is to research and develop tools that help in the better understanding of athletes' performance, taking into account the collective behavior of the team. To do this, we will use data mining techniques to find typical and atypical patterns in data from sensors used by the athletes, facilitating the analysis of the positioning and interaction of players during matches and training sections. In addition, it will enable pointing out the most similar or different aspects of such patterns for exploratory analysis and extracting collective tactical indicators. In this way, it will be possible to aid technical staffs to improve the individual and collective performance of their teams and make decisions regarding the adoption or disposal of specific strategies and training approaches. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
Articles published in other media outlets (0 total):
More itemsLess items

Scientific publications
(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)
ARAGAO DA SILVA, YURI GABRIEL; SILVA, DIEGO FURTADO; WU, XT; JERMAINE, C; XIONG, L; HU, XH; KOTEVSKA, O; LU, SY; XU, WJ; ALURU, S; et al. On Convolutional Autoencoders to Speed Up Similarity-Based Time Series Mining. 2020 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), v. N/A, p. 10-pg., . (19/06080-2, 17/24340-6)
SILVA, DIEGO F.; BATISTA, GUSTAVO E. A. P. A.; WANI, MA; KANTARDZIC, M; SAYEDMOUCHAWEH, M; GAMA, J; LUGHOFER, E. Elastic Time Series Motifs and Discords. 2018 17TH IEEE INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS (ICMLA), v. N/A, p. 6-pg., . (16/04986-6, 13/26151-5, 17/24340-6)

Please report errors in scientific publications list by writing to: