Introduction: The project, titled "Prediction of Student Performance from Gameplay Using Neural Network," focuses on using machine learning techniques, specifically neural networks, to predict student performance based on their gameplay data.
Data Sources Kaggle Competition https://www.kaggle.com/competitions/predict-student-performance-from-game-play
About Dataset The dataset includes information about various in-game events, elapsed time, and other relevant features. The goal is to build a predictive model that can assess and anticipate students' performance levels in different aspects of the game.
Process: The project involves preprocessing the data, reducing memory usage, visualizing missing values, and engineering features. The neural network models are trained and evaluated using a group-based K-fold cross-validation approach.
Results The project assesses the performance of the models and explores different neural network architectures for predicting student outcomes.