MAIN PROJECTS
Green AI: Which Programming Language Consumes the Most?
An empirical experiment was conducted to assess the impact of programming languages on AI's environmental sustainability, comparing five languages (C++, Java, Python, MATLAB, R) across seven AI algorithms and multiple datasets.
GamePlayStyle-Classifier
Binary Anomaly Detection - By employing a random forest model (ML), the gameplay style of a player in a video game is analyzed at runtime based on mouse and keyboard usage. Through binary classification, the algorithm identifies which of the two players is on the computer during the game [Python]
QuizWebApp
An interactive multiplayer quiz with login authentication. Storage of questions, scores, and authentication information in the database [GoLang]
Real-time victory prediction
This project predicts real-time victory probabilities for two teams in a League of Legend game using machine learning, specifically Long Short-Term Memory (LSTM) networks, to process temporal game data and provide minute-by-minute predictions on which team is winning. [Python]
Human Activity Recognition - Federated Learning
This project focuses on classifying human activities—such as sitting, running, and standing—using a Long Short-Term Memory (LSTM) model. The classification is based on data collected from accelerometers, gyroscopes, and other sensors embedded in both smartphones and smartwatches. [Python]