My |

Developed an AI powered clothing search platform enabling users to find apparel using natural language queries based on color, fabric, price range, and style. Implemented NLP based query understanding, intelligent filtering, and data extraction from multiple brands. Integrated the Gemini 2.5 Flash API to enhance query interpretation and deliver accurate, personalized search results through a modern web interface.

Developed an image classification system to automatically classify waste products into recyclable and organic categories using transfer learning. Fine tuned a pre-trained VGG16 model to improve classification accuracy, enabling efficient and scalable waste sorting through computer vision.

Developed a menu driven image processing system that allows users to apply various image transformations, filters, and segmentation techniques interactively. The system demonstrates practical applications of digital image processing concepts in a structured and user friendly manner.

Built a convolutional neural network to recognize handwritten digits using the MNIST dataset. Preprocessed the data, trained and evaluated the model, and achieved high accuracy in digit classification, demonstrating practical applications of deep learning for image recognition.

A machine learning system that classifies images of drones and fighter jets using SIFT feature extraction, data augmentation, and SVM classification. The project demonstrates practical applications of feature based image analysis and supervised learning for object recognition tasks.

Fine-tuned a pre-trained ResNet50 model for object classification on the CIFAR-10 dataset using transfer learning. Applied data augmentation, two stage training, and performance evaluation with accuracy metrics and visualizations, achieving high classification accuracy on multi class image data.

Developed a machine learning based predictive system to analyze heart disease patterns across age, gender, and geographic datasets. Performed exploratory data analysis to identify risk trends, age distributions, and demographic differences, enabling data driven health insights and early risk assessment.

Built a machine learning based model to predict airline ticket prices using factors such as flight duration, departure date, and arrival/departure timings. Analyzed historical flight data to uncover pricing patterns and trends, enabling accurate fare estimation and data driven travel insights.

Developed an image classification system to identify plant diseases from leaf images using transfer learning with VGG16. Applied data preprocessing and augmentation, evaluated model performance with visual metrics, and demonstrated the use of deep learning for real world agricultural applications.

Built a Linear Regression model to predict house prices based on features like median income and average rooms. Performed exploratory data analysis, visualized correlations, and evaluated model performance using R² and MSE, providing insights into key factors affecting housing prices.

I analyzed and visualized customer data to identify distinct segments based on age, income, and spending behavior. Used clustering results and visual insights to highlight customer patterns, supporting targeted marketing strategies and data driven business decisions.

I Developed supervised learning models to classify forest cover types using cartographic variables from U.S. Forest Service data. Trained and evaluated Random Forest and XGBoost classifiers, comparing their performance through accuracy scores and confusion matrices to identify the most effective model.