aubarrios142@gmail.com
Ensuring machine learning models are fair should be a main priority by those implementing them, but it is often overlooked. This project aims to solve that problem by showing how a machine learning model can be both accurate and fair. This project equalizes loan selection rates across different races whilst also maintaining accuracy in its loan predictions using HMDA California 2017 data.
GitHub Repository | Blog☑ pandas ☑ fairlearn package ☑ ML model ☑ Ethical ML model building ☑ Data wrangling
This project was aimed at implementing a data visualization dashboard from scratch. Rather than use software that builds dashboards for you, I implemented one using JavaScript and the Highcharts library. This dashboard aims to provide statistics on teams for the current Premier League season. Using the Python pandas package the original csv files were converted to json files for each visualization. With HTML and CSS it was possible to outline the dashboard to fit all visualizations.
GitHub Repository | Dashboard☑ HTML ☑ CSS ☑ JavaScript ☑ Data Visualization ☑ Highcharts ☑ Pandas
In this project a Twitch streamer recommender system is built using item based collaborative filtering. The data for this recommender system is scraped off Twitter using the tweepy Twitter API. Data collected is a series of connections between users that follow twitch streamers. Each account that the user is following is collected to establish connections between streamers followed by the same person. These connections are then evaluated and used to recommend new streamers that a user might like. The recommendor system is then deployed onto a website using Flask. Feel free to check out the website below. If you want to see the inner workings of the recommendor system check out the GitHub repository.
GitHub Repository | Recommender☑ Twitter Scraping ☑ Flask Deployment ☑ Recommendor Systems ☑ Machine Learning
The aim of this project is to show a rudimentary approach to a machine learning task from start to finish. This project should provide the basics on which one can build upon to meet their machine learning needs. This project shows how to visualize a datasets features, build upon those features and ultimately build a machine learning model that converges and does not overfit a dataset. For more information on the kaggle competion visit the link.
GitHub Repository☑ pandas ☑ TensorFlow ☑ Sklearn ☑ Feature Engineering ☑ Data wrangling
Eager to kickstart my career. Feel free to email me any time, regarding anything you found interesting. Thanks for taking the time to look through my personal website.
Collection of blogs, articles, videos or anything I find interesting at the moment
Look for me on LinkedIn