Please use this identifier to cite or link to this item:
Title: Learning to rank aplicado al análisis avanzado del desempeño de jugadores en la NBA
Author: Iturbe Azcorra, Iker
Tutor: Hernández-González, Jerónimo  
Others: Casas-Roma, Jordi  
Abstract: Historically, data gathering and processing in sports has been based on yearly cumulative statistics to compare players' performance. In the current Big Data Era, we have seen a large increase in the amount of data available. This enables to perform much more advanced statistical analyses by weighing the information about, for example, the defensive capabilities and performances of the opponents or the importance of the game. There are two inter-related hypotheses on which this advanced study of the NBA player´s offensive performance is based. The first one is that the advanced knowledge of the game is what helps the player to make proper decisions when the opponent team poses a new problem by means of their defensive strategy. And the second one is the importance of the game. In the particular case of the NBA, there is no relegation at the end of the league and the regular season is far too long compared to European leagues. All this means that the players' physical performance is much more important than the knowledge of the game, so that yearly cumulative data loses its significance in the comparison of players' performance. The main aim of this study is providing a good statistical analysis taking into account these two factors to compare players' weak and strong points. The result will help coaches and managers when making quick and accurate decisions about recruiting their future staff in a very volatile market.
Keywords: analytics
data mining
learning to rank
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Jun-2019
Publication license:  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
iiturbeaTFM0619memoria.pdfMemoria del TFM1,79 MBAdobe PDFThumbnail