Ravi Rai
Hello, welcome to my personal site! I’m data scientist with a background in math and physics of over 7 years, but I’ve been involved in machine learning for almost as long. Below you can see some apps I’ve contributed to, or you can check out my blog for my latest thoughts in the field!
Apps
SentiRec Analytics
A web app of review analytics for headphones. Providing the user an overview of what earbuds are like, aiming to streamline the decision process when purchasing a new pair. See highlighted features below:
- Overall Ratings: Easily compare headphone models with comprehensive overall ratings, helping you make informed purchasing decisions at a glance.
- Sentiment Analysis: Using NLP techniques, scores for specific key aspects of earbuds are shown, like sound quality, comfort, and noise cancellation.
- YouTube Review Summaries: With the help of LLMs, get the gist of what top youtubers are saying with generated concise summaries.
Flavor Quasar
A simple app that generates the calories given just the name of a recipe. The goal is to simplify one’s calorie counting endeavors, making dieting easier. See below the main features (currently still being developed):
- Calorie Estimation: Simply enter a recipe name to get an estimation for the calories in the meal.
- Ingredients List: Given the recipe name, a potential list of ingredients can be generated using LLMs.
Projects
Predicting Optimal Chess Strategies
Sentiment Analysis on Amazon Instruments
Dashboard on Junior Tech Skills
Python project on a large dataset of chess games, where I sought to predict the winner. Some notable techniques here include text pre-processing and XGBoost.
Python sentiment analysis project on Amazon instrument reviews. Some notable techniques here include SMOTE, neural networks, and an LSTM.
Tableau project on analyzing a dataset of junior tech job descriptions. Some notable techniques here include calculated fields, filters, pandas, and matplotlib.