Sneha Uppu
Student
My Projects
Similar Image Finder
In today’s world full of images, finding visually or semantically similar images quickly is a big challenge. So, we built Similar Image Finder, an AI-powered image search tool that uses OpenAI’s CLIP model to understand the content of images and K-Nearest Neighbors (KNN) to find the closest matches. Trained on the Unsplash dataset, our system takes any image as input and returns the top 6 most similar ones based on visual and semantic similarity. It’s fast, scalable, and works without any task-specific training, making it perfect for real-world applications like e-commerce or photo management.
Recipe Recommendation System
In today’s busy world, one of the most common questions people face is: “What can I cook with the ingredients I already have at home?” To solve this, we built a Recipe Recommendation System that makes meal planning simple and fun. The system uses the K-Nearest Neighbors (KNN) algorithm to match the ingredients you enter with recipes in the dataset, returning the best possible meal options you can prepare. Beyond just showing direct matches, it also suggests similar recipes to the one you select, helping users discover new dishes they may not have considered. By turning a handful of ingredients into a variety of recipe ideas, this project not only reduces food waste but also enhances convenience for everyday cooking.
PrimeVideo Analysis Dashboard
With the explosion of streaming platforms, understanding content distribution has become essential. I designed an interactive Tableau dashboard that analyzes Prime Video’s global library of over 6,000 titles by genre, release year, rating, and country. The dashboard uncovers insights such as the growth of movies vs. TV shows, regional trends in content availability, and shifts in audience preferences across genres. This project highlights my skills in data cleaning, visualization, and storytelling with data, enabling complex patterns to be communicated in a clear and engaging way.
