Topic Modeling and Classification (Capstone project @ NJIT)
Topic Modeling using Latent Dirichlet Allocation (LDA) and Topic Classification using Support Vector Machine (SVM) with aim of recommending users to other users with matching profiles.
Spam Detector
Spam Classifier using Naive Bayes from scratch (with Laplace Smoothing) without use of scikit-learn. Live
Face Recognition
Face recognition using python. Identifies an individual in an image based on a previously known image of that individual.
Stress Detection using PPG devices
Detect the presence of stress from heart rate measure by wrist-worn PPG devices (such as Fitbit, Apple watches, etc). This project was part of my Stanford Professional Certification Course in Machine Learning (xCS229ii)
Image Compression with K-Means
Apply the K-means algorithm to lossy image compression, by reducing the number of colors used in an image.
Chest Xray Binary Classifier using Transfer Learning
A binary classifier implemented using resnet50 pre-trained model and transfer learning, that can detect whether a person has pneumonia or not based on their chest X-Ray image.
Feature Maps
Demonstrating the use of feature maps for classifying non-linear (sinusoidal) data.
GPT-2 text generator
A GPT-2 based text-generator that pretends to reside in the year 2032.
Housing Prices Prediction in time of Covid-19
In this project we created a model to predict the per-square-foot housing prices from the NYC rolling sales data and the pluto dataset, taking into consideration the number of covid cases per day.
MySQL project (Public Library)
Project to write MySQL queries and a basic UI for public library, per the problem statement. (Part of coursework @NJIT).
