Finding a partner for sports such as tennis or badminton is often time-consuming and inefficient, as it can be difficult to arrange a game and even harder to find someone of a similar skill level to ensure the match is enjoyable for both players. Existing solutions, such as Facebook groups and online forums, rely on manual searching, which discourages many people from engaging in the sport. As part of this thesis, we developed a system that recommends the most suitable partners for tennis and badminton using machine learning, based on user characteristics and previously played matches. The system takes into account factors such as gender, age, ELO rating, and match history, as well as match outcomes and how much the players reported enjoying the game. The solution is implemented as a web application built using modern web technologies and a hybrid recommender system. The final product improves the experience of finding suitable partners compared to existing solutions, enabling more players to participate in the sport and helping to build a stronger sports community.
|
