Welcome to my comprehensive data science, machine learning, computer vision, and deep learning project collection! This repository contains academic projects showcasing practical applications of AI/ML algorithms and techniques.
Maintained by: Balaji
Repository: A collection of college-level data science projects demonstrating practical applications of statistical analysis, machine learning algorithms, computer vision techniques, and deep learning architectures.
collage-projects/
├── 📊 Root Projects (ML & Statistics)
├── 📷 Computer Vision/
├── 🧠 Deep Learning/
├── 🖼️ Fundamentals of Image Processing/
└── 🤖 Machine Learning/
1. 📈 Naive Bayes Job Prediction (BAyes_theroem.ipynb)
Predicting job placement using Naive Bayes classification
| Feature | Description |
|---|---|
| Algorithm | Gaussian Naive Bayes Classifier |
| Dataset | Student academic data (CGPA, internship, programming knowledge, communication skills) |
| Purpose | Predict whether a student will get a job based on academic and skill features |
| Techniques | Label Encoding, Train-Test Split, Classification Report, Confusion Matrix |
Key Learning: Demonstrates Bayes' theorem application in real-world classification problems.
2. ❤️ Heart Disease Prediction System (HEART_FINAL.ipynb)
Comprehensive ML pipeline for heart disease diagnosis
| Feature | Description |
|---|---|
| Algorithms | Logistic Regression, SVM, Random Forest, XGBoost, LightGBM, KNN, MLP, Stacking Classifier |
| Optimization | Optuna hyperparameter tuning with 50 trials |
| Data Balancing | ADASYN oversampling for class imbalance |
| Explainability | SHAP values for model interpretation |
| Evaluation | ROC-AUC, Precision-Recall curves, Calibration curves, Confusion matrices |
| Features | Input validation for production API, model persistence with joblib |
Key Learning: End-to-end ML pipeline with advanced techniques including ensemble methods and explainable AI.
3. 🏠 House Price Prediction (houseproce.ipynb)
Regression analysis for real estate pricing
| Feature | Description |
|---|---|
| Algorithms | Linear Regression, Ridge Regression, Lasso Regression |
| Techniques | StandardScaler, VIF for multicollinearity, 3D visualization |
| Analysis | Correlation heatmaps, Pairplots, MSE/R² evaluation |
| Visualization | 3D scatter plots with regression planes |
Key Learning: Multiple regression techniques with feature scaling and regularization.
4. 🛒 Market Basket Analysis (store_data.ipynb)
Association rule mining for retail analytics
| Feature | Description |
|---|---|
| Algorithm | Apriori Algorithm |
| Metrics | Support, Confidence, Lift |
| Dataset | 7,500 retail transactions with 20 items each |
| Output | Structured DataFrame of association rules |
Key Learning: Discovering purchasing patterns using association rule mining.
| File | Description | Techniques |
|---|---|---|
| CV_Lab-5_Smoothing.ipynb | Image Smoothing & Filtering - Gaussian blur, Median blur, Bilateral filter, Linear vs Non-linear filters | OpenCV filters, Sharpening kernels |
| CV_Lab-7_edge_dection.ipynb | Noise Reduction & Edge Detection - Salt & pepper noise, Gaussian noise, Median/Gaussian filtering | Noise simulation, Filter comparison |
| CV_Lab-7.ipynb | Edge Detection Algorithms - Canny, Sobel, Prewitt, Roberts edge detectors | Gradient-based edge detection |
| Cv_T2-Review_Module2.ipynb | Image Similarity Search - VGG16 feature extraction, Fashion MNIST similarity search | Transfer learning, Feature matching |
| CV3_Bit_Plane_Sclicing_Lab-4.ipynb | Bit Plane Slicing - Decomposing images into 8 bit planes | Binary image analysis |
| CV3_Lab_Histogram_Lab-3.ipynb | Histogram Equalization - Contrast enhancement, histogram analysis | Image enhancement |
| cv3_Lab-LBP1.ipynb | Local Binary Pattern (Manual) - Manual LBP implementation from scratch | Texture analysis |
| cv3_LAB-LBP2.ipynb | LBP & GLCM Features - Skimage LBP, Gray-Level Co-occurrence Matrix | Texture descriptors, Feature extraction |
| File | Description | Techniques |
|---|---|---|
| DL_Lab-1-and-Lab-2.ipynb | Perceptron Learning - AND, OR, XOR gate implementation with perceptron, Logistic Regression on health data | Weight evolution visualization |
| DL_Hyper-parameter Tuning_Lab-4.ipynb | Neural Network Hyperparameter Tuning - MNIST digit classification, Learning rate, Epochs, Batch size, Dropout, Optimizers | Adam, SGD, Keras Sequential |
| DL_Lab-7_Transfer_Learning.ipynb | Transfer Learning & Model Comparison - VGG16/19, ResNet50/101/152, InceptionV3, MobileNet, DenseNet, NASNet | Pre-trained models, Feature extraction |
| DL_T2_Review_Module-2.ipynb | Advanced Architectures - Bidirectional LSTM, Stacked LSTM, Conv2D, Depthwise Separable Conv2D | Parameter calculation, Architecture design |
| File | Description | Techniques |
|---|---|---|
| FIP_Lab-1-Basic_Functions.ipynb | Basic Image Operations - Add, Subtract, Multiply, Divide, Bitwise operations (AND, OR, NOT) | OpenCV arithmetic operations |
| FIP-Lab-2-Split_Channels.ipynb | Color Channel Separation - RGB channel visualization, Color space manipulation | Channel extraction |
| Fip_Lab-3-Gray_Scale.ipynb | Gray Level Slicing - Intensity-based image segmentation, with/without background preservation | Thresholding |
| Fip_Lab-4-Histogram_Equalization.ipynb | Histogram Equalization - Contrast enhancement, Before/after histogram comparison | Image enhancement |
| Fip_Lab-5-Spatial_filtering.ipynb | Spatial Filtering - Averaging, Gaussian, Min/Max filters, Salt & pepper noise removal | Linear/Non-linear filters |
| Fip_T2_Review_Module-2.ipynb | Morphological Operations - Opening, Closing, Hole filling, Skeletonization, Thickening | Binary morphology |
| FIP-Lab-LZW-Coding.ipynb | LZW Compression - Lempel-Ziv-Welch encoding/decoding with step-by-step visualization | Data compression |
| File | Description | Techniques |
|---|---|---|
| decision tree.ipynb | Decision Tree Classification - Job prediction using entropy criterion, Tree visualization | CGPA, skills-based classification |
| diabetes.ipynb | Iris Dataset Analysis - K-Fold cross-validation, MinMaxScaler, Data preprocessing | Exploratory data analysis |
| head brain.ipynb | Simple Linear Regression - Head size vs Brain weight prediction, Scatter plots | MSE, R² score evaluation |
| houseproce.ipynb | Multiple Linear Regression - House price prediction with Ridge/Lasso regularization | 3D visualization, VIF analysis |
| hyper parameter.ipynb | Neural Network Hyperparameters - Activation functions, Optimizers, Weight initializers | Keras/TensorFlow configuration |
| knn_svm_decision_random.ipynb | Classifier Comparison - KNN, SVM, Decision Tree, Random Forest on Iris dataset | Accuracy comparison, Confusion matrices |
| mini project.ipynb | Calorie Burn Prediction - Exercise data analysis, Age groups, Gender-based insights | Plotly visualizations, Seaborn |
| pca_lda.ipynb | Dimensionality Reduction - PCA vs LDA comparison, Gaussian Naive Bayes classification | Accuracy comparison, Feature reduction |
| reg_logestic.ipynb | Logistic Regression - K-Fold, Stratified K-Fold, Leave-One-Out cross-validation | Validation strategies |
| store data.ipynb | Association Rule Mining - Apriori algorithm, Support/Confidence/Lift analysis | Market basket analysis |
| svm.ipynb | Support Vector Machine - RBF kernel classification on Iris dataset | SVM prediction |
| Category | Technologies |
|---|---|
| Languages | Python 3.x |
| Environment | Jupyter Notebooks, Google Colab |
| Data Processing | Pandas, NumPy |
| Machine Learning | Scikit-learn, XGBoost, LightGBM |
| Deep Learning | TensorFlow, Keras |
| Computer Vision | OpenCV, Skimage |
| Visualization | Matplotlib, Seaborn, Plotly |
| Optimization | Optuna |
| Explainability | SHAP |
| Association Mining | Apyori |
-
Clone this repository:
git clone https://github.com/9046balaji/collage-projects.git cd collage-projects -
Open in Google Colab (Recommended):
- Click the "Open in Colab" badge on any notebook
- Or upload notebooks directly to Google Colab
-
Run Locally:
pip install jupyter pandas numpy scikit-learn matplotlib seaborn opencv-python tensorflow keras jupyter notebook
| Category | Count | Key Topics |
|---|---|---|
| Root Projects | 4 | Naive Bayes, Heart Disease ML, Regression, Market Basket |
| Computer Vision | 8 | Edge Detection, LBP, Histogram, Image Similarity |
| Deep Learning | 4 | Perceptron, CNN, LSTM, Transfer Learning |
| Image Processing | 7 | Morphology, Filtering, Compression, Enhancement |
| Machine Learning | 11 | Classification, Regression, Clustering, Dimensionality Reduction |
| Total | 34 Notebooks |
This project is for educational purposes. Feel free to use and modify for learning.
GitHub: @9046balaji
Last Updated: January 2026