Course Blog‎ > ‎

Final Projects

posted May 4, 2015, 12:55 PM by Jen Mankoff   [ updated May 4, 2015, 1:43 PM ]
The final projects from this class made me so proud, you each clearly worked hard and applied many of the ideas we spent the semester learning. Here I want to provide a brief summary of them.

World Bank Development Indicators: Provides a comprehensive visualization of a country's development over the last two decades over multiple spheres, allowing for easy cross-metric and cross-country comparisons. A highlight of this project is the sheer number of types of indicators than can be explored and the careful analysis of Singapore in particular. 

Measuring Immigration: The project visualizes the impact of immigrants in United States: Who are they? What do they do? What is their impact? The highlight of this project is the linked data table and parallel coordinates graph which supports filtering and allows rapid exploration. The data set is scraped from wikipedia and fairly unique as well. 

Pantheon is a visualization and exploration tool designed around the dataset by MIT Media labs about historical figures from Wikipedia: A lens into history through cultural figures from Wikipedia. A highlight of this project is the easy interface for exploring the data, sorting and filtering by attributes such as country or time period. 
ISIS twitter network analysis: This project used data from the Twitter REST API to train a machine learning algorithm to identify ISIS propagandists from data sampled. Highlights include the visualization showing in-degree and out-degree data for accounts (seen at left), and the live interface being used to collect crowd labels for further enhancement of the data set. 

Which State Has the Best Pizza? A comparison and analysis of restaurant ratings and reviews for each state's most populous city. This website shows data for 7 cuisines across the country. A highlight is the surprising results around what is most popular where (though many questions still need to be answered about the impact of the approach to data collection as well as the variable number of reviews and restaurants per state for different cuisines). 

Ask Us Anything for Reddit AMA Visualizes shared stories and experiences on the online knowledge sharing platform Reddit, specifically focusing on 'Ask Me Anything' threads. The thread visualized at left for example shows a question to a US Navy sailor stationed at Guantanimo Bay. Visualizations explore sentiment, upvotes, and thread structure. 

What to do! Discover and plan social events in your city! This application is something we could all use -- it lets people search for events by data and location across multiple sources and helps plan an evening out or a day trip by pointing to other necessities (such as restaurants). In addition, you can star the stuff you are most interested in and generate a calendar from them. 

Analyzing factors that influence diabetes patients readmission: Some of the highlights of this project include the descriptive analysis of the features and the feature engineering done to improve the machine learning. 

Healt$care lets policy makers explore what factors lead to better quality healthcare?  At a time when increasing amounts of data on healthcare are becoming available to the public, we demonstrate ways to use this data to ensure that we provide higher quality care across the country. Highlights of this project include the multiple visualizations and careful story telling connected to them.