Sportanalytics from social media data

OData support
Supervisor:
Nagy István
Department of Telecommunications and Media Informatics

Nowadays, the sports analytics can be used for several purposes. It is able to improve the athlete’s performance in training time, helps in preparation against opponents, informs visitors about matches, gives information about recent condition of teams, and it can be the base of professional sports betting. The scope of my study is the specification and the development of a predictive model for sport betting purposes. It is able to predict the likelihood of three possible match outcome and making a suggestion for the stake. I chose football as the analyzed sport, and Facebook as a social media source, due to its popularity in Europe. The analyzed matches were played in leagues with the highest prestige.

The main thesis of my work is the bookmaker’s odds based on the non-professional gamblers. These gamblers often have emotional bonding with their favorites, which also decreases the efficiency of sport betting market. In contrast, the predictive models and robots have an objective viewpoint, and they are able to calculate with hundreds of statistics. We can observe sport betting robots in the market, but a number of robots are dwarfed in comparison to the stock exchange.

During the thesis, I analyzed the statistics, odds of bookmakers and social media content. Based on this substance, I created predictive models, which are able to estimate the match outcomes in a profitable way.

Downloads

Please sign in to download the files of this thesis.