The Real Estate Price Prediction website is built by creating a model using sklearn and linear regression. The website is built using HTML, CSS, and Javascript that allows users to enter the area in square feet, bedrooms, etc, and calls a Python Flask server to retrieve the predicted price and deployed it to production AWS. It incorporates the concepts of data cleaning, outlier detection, and removal, dimensionality reduction, GridSearchCV for hyperparameter tuning, and used the k-fold cross-validation algorithm.
This website helps to classify sports celebrities based on their images. This model classifies 5 celebrities only. The UI is created using HTML, CSS, Javascript, and JQuery. It makes use of the OpenCV library for face and eyes detection and feature engineering using wavelet transforms. If the image is not clear, the model doesn't classify the images. However, if a match is found, the name of the celebrity is mentioned along with its matching percentage of comparison with other celebrities.
Retail Store Analysis
I created Microsoft Power BI dashboards on a retail sample store database provided through a MySQL file. It was an individual contribution. I was supposed to find out the weaker section or areas of a region where the retail store can improve its product and improve its sales. I was also required to find out which region causes a lot of unnecessary transportation expenditure. I learned how to focus on important data columns and how some data can be combined so it can be better for visualization purposes and can provide a better understanding to others.
I created Created a BMI Calculator app using Android using Intents and Explicit, Broadcast Receiver, Geospatial API, etc which calculates the BMI based on your height and weight and is connected to an external website for the latest trends related to diet.
Online Food Ordering website using HTML, CSS, Bootstrap, PHP, JavaScript which lets the registered customers select food from a variety of cuisines present and delivers to you.
I created an online website named "E-Voting using Blockchain" in the domain of Cyber Security using HTML, CSS, ReactJS, and NodeJS programming languages and then deployed on the Heroku server and used PostgreSQL for the database. Its main emphasis was to assign registered citizens a token to cast their vote and then these votes are added to the blockchain. It made use of SHA-256 for hashing, ECDSA algorithm for digital signature and shows vote count in a graph.
