Leading the data science efforts of our sport analytics startup. Modeling the complex problem of the movements of football team and the roles of players within. Identifying recurrent patterns in team movements using machine learning techniques. Delivering actionable, data-driven insights that can exploit the weaknesses of the upcoming opponent.
Carrying out research in the area of sport analytics. Focusing on the quantitative evaluation of the performance of football players. Designing and developing a quantitative player scouting system with cutting-edge metrics to evaluate the performance of the players, and identify similarities among them.
Involved in a joint research project with FC Barcelona on football analytics. Investigating what kind of factors contribute to the success of football teams and players, from a quantitative viewpoint. Studying problems of telecommunication networks with a special focus on the financial implications of the solutions.
Focusing on network economics related problems using metdologies including game theory, network science, linear programming.
Joint degree program with Semmelweis Medical University.
Major in Telecommunication.
Decachord is able to analyze the football players' trajectories in a game and extract recurring patterns of team-wide movements. This reveals the the playbook of a team (e.g., the upcoming opponent). Decachord has been selected for the Startup Competition of the MIT Sloan Sport Analytics Conference 2017.
The project analyzed the passing network of football teams, i.e., what kind of patterns arise based on the way players are interacting with each others using passes. We managed to identify a key factor that differentiates the passing strategy of FC Barcelona from other top-division football teams. Our work received extensive attention from the media; among others The Economist and BBC.
We derive the movement traits of football players and then identify the similarity among them using advanced machine learning techniques.
We are able to extract the movements of the players related to specific events of the game. This opens up a new line of analyses related to the movements creating scoring opportunities at scale.
Technology offers new ways to measure the locations of the players and of the ball in sports. This translates to the trajectories the ball takes on the field as a result of the tactics the team applies. We reveal the tactics of a team through the analysis of repeating series of passes. We derive insights such as passing strategies for maintaining ball possession or counter attacks, and passing styles with a focus on the team or on the capabilities of the individual players.
I focused on detecting price discrimination on the Internet. Price discrimination, charging customers differently for the same product based on their individual valuations of the products, could benefit from extensive information collected online on the customers and thus contribute to the profitability of e-commerce services. We setup a distributed measurement framework to demonstrate empirically the existence of price differences on the Internet, and uncover what information vectors are used to facilitate the method of charging different prices. Our work received extensive attention from the media; among others The Wall Street Journal picked up our project.
I studied problems of telecommunication networks with a special focus on the financial implications of the solutions. First, I investigated the sustainability of the Internet ecosystem by analyzing the balance of power among the main economic actors in the Internet. Our model applies tools of cooperative game theory and runs with real market data to give actionable, quantitative insights for business negotiations. Second, I analyzed quantitatively the problem of how to share the cost of a backbone network among its customers. We used extensive traffic, routing, and cost data from a tier-1 backbone network and methodologies such as numerical evaluation and game theory. Third, I focused on pricing strategies of Internet access providers considering the loyal intensions of end users, dynamic markets, and uncertainties.