Please use this identifier to cite or link to this item:
Title: Análisis y predicción del rendimiento ofensivo de debutantes en las Grandes Ligas de Béisbol
Author: Benavides Esteva, Armando
Director: Casas-Roma, Jordi  
Tutor: Hernández-González, Jerónimo  
Abstract: In the present work a study is made to find knowledge that allows to improve the decision-making process and with more data to select the most prepared players to play in the MLB. For this purpose, using an agile development methodology, a tool that predicts the offensive performance of the players according to their historical statistics in minor leagues is created. To reach the final system, following a quantitative methodology to obtain information, data is obtained through web scrapping, which is analyzed to understand better the info they offer. A study of different grouping and classification models, together with tests and searches of better parameters helps to select the models forming the core of the predictive system. The models allow to group players with similar characteristics and classify new samples. The dataset used on the models is previous processed to facilitate the work of the models and improve the results. The prediction of the new players is based on the historical performance of the players belonging to the same group, which have similar characteristics. The results obtained place the present system in an advantageous position, managing to reduce the errors of the predictions considerably compared to other models in the same field. However, it is essential to consider other approaches especially in predictive calculations, as well as to improve the interpretation of results so that people, regardless of their knowledge of the subject, are able to assimilate them.
Keywords: prediction
data mining
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 9-Jun-2019
Publication license:  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
abenavideseTFM0619memoria.pdfMemoria del TFM2,78 MBAdobe PDFThumbnail