Data Science Projects

Recidivism Forecasting Challenge - Complete

https://nij.ojp.gov/funding/recidivism-forecasting-challenge

• Prizewinner of Machine Learning Contest hosted by National Institute of Justice ($19,500)
• Predicted recidivism using person and place-based variables with the goal of improving outcomes for those serving a community supervision sentence.
• Utilised Xgboost, Adaboost, LightGBM, CatBoost, Autoencoder, and Logistic Regression algorithms using Python libraries
• You can find the code on my github page

School Closure and Digital Learning Marker - Complete

In this work, we analyzed the impact of the COVID-19 on the digital learning market. Mainly, we tested how the school closure order has changed the useage of digital learning platforms. To examine this hypothesis, we utilized the data provided by the LearnPlatform organization. The code and Data visualization is publicly available on my Kaggle page

Recommendation System - In Progress

with 211 LA Data Team

• Developed Recommender System using cutting-edge Deep Learning and Machine Learning models in Pytorch and Sklearn
• Deployed the model using Flask, Docker and Google Cloud Platforms
• Develop A/B test plans and manage implementation