Listing of projects I completed related to data mining as part of the Data Mining course, with links to the implementation repositories. Each project includes a notebook with the complete algorithm implementation, covering exploratory data analysis, model training, evaluation, and result interpretation. Additionally, each project is accompanied by a paper in IEEE format, with description of all aspects.
Project: Insights in Wheat Seed and Irish Flower Data Through PCA
Analysis implementing PCA for dimensionality reduction on wheat seed and iris flower datasets. The project demonstrates effective feature extraction and visualization of high-dimensional agricultural data.
Project: Decision Tree Model for Banknote Verification
Development of a robust decision tree classifier for detecting counterfeit banknotes. The model processes various banknote measurements to determine authenticity with high precision.
Project: Predictive Analysis of Heart Disease
Implementation of a Naive Bayes classifier for heart disease prediction. The project processes medical data to assess disease risk factors and provide probabilistic health insights.
Project: Water Quality Prediction Using KNN
An application of the KNN algorithm for water quality assessment. The model analyzes multiple water parameters to predict potability, demonstrating practical use of nearest neighbor classification.
Project: Handwritten Digit Classification
A from-scratch implementation of an Artificial Neural Network for the MNIST dataset. The project includes complete forward and backward propagation mechanisms for digit recognition.
Project: Lake & Granite Image Segmentation
Application of k-means clustering for advanced image processing. The project successfully segments different regions in lake and granite images, showcasing unsupervised learning techniques.
Project: ID3 and C4.5 Implementation
Custom implementation of ID3 and C4.5 decision tree algorithms from scratch. The project includes a comparative analysis of both algorithms using an educational dataset for college admissions.