In this repository, you can find my fall semester graduation project, which predicts the winning team from a live game using data visualization and machine learning of the LoL game.
League of Legends (LoL) Live Match Outcomes Predicts With Machine Learning,
League of Legends (LoL) Champion Dynamics,
League of Legends (LoL) Global Server Distribution,
League of Legends (LoL) Data Visualization,
League of Legends (LoL) Champions Gender Distribution,
Using Riot Api For Live Game Data
This project uses Python programming to visually analyze the dynamics of League of Legends (LoL) and predict which team will win in a live match, focusing on champion dynamics and global server distribution. We provide a brief but comprehensive overview of the various elements in the game through pie charts, bar graphs and a world map and predict the winning team with machine learning technology. A notable feature of our project is the inclusion of a world map showing LoL's global server distribution. We provide a visual representation of the widespread global LoL community by highlighting various server locations via color-coded markers. In another aspect of our project, we offer artificial intelligence that predicts which team will win 10 minutes after the start of a live game, based on in-game dynamics, with our model that we train based on certain sources. In summary, this project offers a brief review of League of Legends dynamics by combining Python programming with data visualization techniques, as well as using machine learning fundamentals to predict who will win a live match using Python programming and machine learning techniques. Summarizing complex data with visually appealing graphs and maps, this project is a valuable resource for both casual players and dedicated enthusiasts who want to grasp the nuances of the League of Legends gaming experience, including how game dynamics are affected by gender and play time. For enthusiasts, it is a resource that appeals to players and those interested, with its mechanism that predicts which team will win.
In this project, we established the web application infrastructure using the Flask framework. We used various python libraries and preferred the pandas library for visualization. We accessed the data sets using Riot API and arranged them according to our needs. On the machine learning side, we used the logistic regression approach using the scikit - learn library. We created a world map using the Plotly library and here we can see the regions where LoL has servers in the world.
If you are curious about the appearance of our website, you can look at the screenshots in the pdf file.