Nishant Baruah

MY PROJECTS

Project 1

InfoAssist Bot - RNN-LSTM model

InfoAssist-RNN-model is a chatbot application designed to assist users by answering frequently asked questions (FAQs) and providing support for various queries related to a software product. This project leverages Natural Language Processing (NLP) and deep learning techniques to deliver accurate and contextually relevant responses.

View on GitHub
Project 2

InfoAssist Bot - Pre trained transformer model

This project is an NLP-Transformer based chatbot built using Flask. The chatbot is capable of answering questions based on a predefined dataset. This project uses the all-MiniLM-L6-v2 model from SentenceTransformer. It is a pre-trained model optimized for semantic similarity tasks. The model is efficient and provides a good balance between performance and speed.

View on GitHub
Project 3

MoodLens

MoodLens is an sentiment detection project that classifies images into different emotion categories like happy and sad using a convolutional neural network (CNN). The project includes training data, CNN model, and a web application for testing the model. This project involves manual data scraping from the internet to train the CNN model

View on GitHub
Project 4

Customer Churn Prediction

This project aims to predict customer churn in a telecom company using various machine learning algorithms. The goal is to identify which customers are likely to leave the company and to understand the factors contributing to churn. The project includes a detailed Jupyter notebook that explores different models and evaluates their performance.

View on GitHub
Project 5

Real-Time Data Analysis using PySpark vs Pandas

This project compares the performance of two powerful data processing libraries: PySpark and Pandas. It focuses on real-time data analysis to highlight the strengths and weaknesses of each library in handling large datasets and performing complex data transformations. The ipynb files and results screenshots are in the project github repo.

View on GitHub
Project 6

Optical Mark Recognition Prototype

Optical Marks Recognition is a technology that allows automated grading and analysis of scanned or photographed multiple-choice answer sheets. This project provides a complete end-to-end solution for processing and analyzing OMR sheets. This project uses Opencv and image processing techniques

View on GitHub
Project 7

AI game using hand distance detector

This project is a part of my fifth-semester AI coursework. It includes the implementation of an AI-based game, showcasing various artificial intelligence techniques and algorithms.

View on GitHub
Project 8

Guidance System - Prediction model

I built this project as a research model during my internship at Mahindra. This project aims to predict whether a customer will purchase an SUV car based on their age and salary. The prediction is made using a logistic regression model. The project includes a web application built with Streamlit to allow users to input age and salary values and receive a prediction.

View on GitHub
Project 9

National Health Profile - Data visualization in Tableau

This project focuses on visualizing and analyzing business information related to the National Health Profile. It leverages data visualization tools in tableau to provide insights and enhance the understanding of national health statistics throught interactive dashboards

View on GitHub