Project Objectives

1. Analyze and create visualizations on an Olympic dataset from 1896 to 2016 to uncover which countries stay at the top overtime, by season, and by sport; what it takes for an Olympian to be at the top for each sport, sport popularity, and the relationship between Olympic Medal counts and country’s GDP.


2. Additionally from the Olympic dataset, machine learning was incorporated to predict the amount of medals 25 countries would obtain in the 2020 Tokyo Olympics with Linear Regression, Logistic Regression, and Auto Regressive Integrated Moving Average(ARIMA) model.


GitHub - Visualizations GitHub - Machine Learning Presentaion - Visualizations Presentaion - Machine Learning

Team

Aastha Arora

Dianne Jardinez

Duong
Luu

Ritika Bhansali

Swarna Guntaka


Key Takeaways

What went well

  • Creating the following charts: Racing barchart, static bar and line chart, and Leaflet map with two different layers(street map and choropleth) with Plot.js, D3.js, Chart.js, and Leaflet.js

Challenges we overcame

  • Utilizing and connecting Python Flask-powered RESTful API

  • Web browser cache accurately displaying visualizations

  • New Machine Learning model: ARIMA