AI/ML Projects
Explore my machine learning projects that demonstrate practical implementations of various ML algorithms and techniques. Each project showcases different aspects of machine learning, from decision trees to neural networks.
Decision Tree Implementation
Developed a decision tree classifier from scratch to understand the fundamentals of tree-based learning algorithms. Implemented the algorithm to handle both categorical and numerical data, with features including information gain calculation, tree pruning, and visualization of the decision boundaries.
KNN Regression Analysis
Implemented K-Nearest Neighbors regression algorithm to predict continuous values based on feature similarity. The project includes distance metric calculations, neighbor weighting schemes, and cross-validation for optimal k-value selection.
Neural Network Implementation
Built a neural network from scratch to understand the inner workings of deep learning. Implemented forward propagation, backpropagation, and various activation functions. The network was tested on classic machine learning datasets to demonstrate its learning capabilities.
Naive Bayes Classifier
Created a Naive Bayes classifier implementation for text classification tasks. The project includes preprocessing steps like tokenization and feature extraction, probability calculations, and Laplace smoothing for handling zero probabilities.
Reinforcement Learning Implementation
Developed a reinforcement learning algorithm to solve simple game environments. Implemented Q-learning with features including exploration vs exploitation strategies, reward function design, and policy optimization for optimal decision-making.