In this project, I use 4 stacking classifiers and combine them to get the final prediction. I use majority voting as the meta classifier. I use 4 datasets in this project. The highest accuracy I achieved for validation data in dataset 1 is 90.58%, in dataset 2 is 98.78%, in dataset 3 is 97.50%, and in dataset 4 is 73.64%. Show More
Supervisor: Dr. Soumitra Samanta
In this project, we build a PDF chatbot that accepts multiple PDFs from users and provides Q&A based on the content. We also use a MySQL database to store the chat history. The frontend is built with React, and the backend server is implemented using Flask.
The project aims to provide users with a platform using HTML, CSS, JS to read on various topics such as IPL, Finance, Politics, Earth, and Travel. Users can navigate through different categories and also search for specific news articles using the search bar.
This project is a Movie Recommendation System built using Streamlit, a popular Python library for building interactive web applications. The recommendation system suggests movies to users based on their preferences and past interactions with the platform.
The project utilizes the following technologies: Streamlit, Python, Pandas, NumPy, and OpenAI.
This project contains code for building a model for river and non-river classification of satellite images. We use a Gaussian Naive Bayes (GNB) classifier implemented from scratch. The model achieves an accuracy of 98.50%.
The dataset used in this project is sourced from https://www.isical.ac.in/~murthy/.
This Streamlit application predicts whether a breast cancer tumor is benign or malignant based on input data. The app uses a Logistic Regression machine learning model trained on the Breast Cancer Wisconsin dataset to classify tumors based on various features related to cell nuclei measurements.
The dataset used is the Breast Cancer Wisconsin (Diagnostic) Data , containing various cell nuclei features critical for tumor classification.