AutoClassifier: AI-Powered Car Identification

Deep learning-based car identification model using computer vision for automotive surveillance. Processes 16,185 images across 196 car classes, classifying cars by make, model, and year using CNNs and RCNN for precise vehicle detection and localization.

Deep Learning Computer Vision CNN RCNN Transfer Learning MobileNetV2

Support Ticket Categorization

NLP-based machine learning model using transformers for automatic support ticket classification. Accurately predicts ticket categories, assigns priorities, and estimates resolution times through advanced text classification and prompt engineering.

NLP Transformers Text Classification Prompt Engineering Machine Learning

Advanced Computer Vision: Face Detection & Recognition

Real-time face detection and recognition system for broadcast environments using CNNs. Implements triplet loss and Siamese networks for crew member identification in video streams, enhancing security and personalization.

Computer Vision CNN Triplet Loss Siamese Networks Real-time Processing

Customer Churn Prediction

Ensemble learning model for predicting customer churn in telecom industry. Combines random forests and XGBoost to achieve high prediction accuracy, supporting customer retention strategies through advanced predictive analytics.

Ensemble Learning Random Forest XGBoost Predictive Analytics Classification

Manufacturing Yield Prediction

Predictive model for semiconductor manufacturing yield optimization. Employs PCA for feature engineering and grid search for model tuning to identify factors affecting production yield and reduce operational costs.

Feature Engineering PCA Grid Search Predictive Modeling Regression

Neural Networks for Predictive Modeling

Dual-purpose neural network project: regression model for signal quality prediction and image classifier for street-level photography. Focuses on optimization techniques, overfitting prevention, and hyperparameter tuning.

Neural Networks Regression Image Classification Hyperparameter Tuning Deep Learning

Unsupervised Learning: Vehicle Clustering

Clustering and classification system for vehicle segmentation based on fuel consumption and silhouette analysis. Utilizes K-means clustering and SVM to discover hidden patterns in unlabeled automotive data.

Unsupervised Learning K-means Clustering SVM Dimensionality Reduction Pattern Recognition

Healthcare & Marketing Predictive Models

Supervised learning models for healthcare patient condition prediction and marketing customer conversion. Handles imbalanced datasets using logistic regression and SVM for accurate classification in critical domains.

Supervised Learning Logistic Regression SVM Classification Imbalanced Data

New Wheels: SQL & Data Visualization

Business analytics case study analyzing sales data to address declining revenue. Uses SQL queries, aggregation, and visualization to generate actionable insights for customer retention and product optimization strategies.

SQL Data Visualization Business Analytics Data Aggregation Decision Support

Applied Statistics for Decision-Making

Statistical analysis project applying hypothesis testing, data visualization, and statistical inference to industry problems. Analyzes sports tournaments and startup success rates to provide strategic investment insights.

Statistics Hypothesis Testing Data Visualization Statistical Inference Business Intelligence

Portfolio Website

Modern, responsive personal portfolio website built with Jekyll and hosted on GitHub Pages. Features a clean design with blog integration.

Jekyll SCSS GitHub Pages Responsive Design

---

More Projects Coming Soon!

I’m constantly working on new projects and experiments. Check back regularly or follow me on GitHub to see what I’m building next.

Want to Collaborate?

I’m always open to interesting project ideas and collaborations. If you have a project in mind or want to work together, feel free to get in touch!