A full-stack website I architected and built that contains it's own local user authorization and authentication microservice that I wrote myself. Users can log in and chat with each other as well as chat with ChatGPT. More features and enhancements to come in the future. The front-end is written in Next.JS, React, and Typescript. The back-end consists of several microservices written in either Node/Express that are hosted on Heroku. The databases being used are MongoDB, PostgreSQL and Firebase for messages/authentication as well as authentication. Unit and integration tests are being done with Mocha and Chai. AWS is being used for pipelines, servers and domain names (CodePipeline, EC2, ELB, Route 53, S3, Codebuild, VPC, etc.)
This is a supervised learning project that uses 3 algorithms on two classification problems. The 3 supervised algorithms are Neural Networks, Support Vector Machines, k-Nearest Neighbors. Different hyperparameters are explored for each algorithm. To see which ones exactly refer to the pdf explaining the experiment in its entirety.