project-examples.md

General Assembly Data Science Course: Example Student Projects

This repository is a collection of example projects from students in General Assembly's part-time Data Science course. Its purpose is to give incoming students a sense of the variety and scope of past projects, and to spur their thinking for their own projects.

• NPR One: App Store Reviews Text Analysis
  • Can we use app store review data to identify the app features that are critical to its success?
• NPR One: Predicting User Behavior with N-Nearest Neighbor Stories
  • Can we predict which stories to play for an app user by measuring the similarity between stories?
• MetroMetric
  • How accurate are NextBus predictions of bus arrival time, and can predictions be improved using other data?
  • Other links: code, data
• Forecasting the All-NBA Teams
  • Can we predict the 2015 All-NBA Teams based on player statistics?
  • Other links: paper, code, data
• Billboard Top 40 Song Analysis
  • Can we predict the popularity, decade, and genre of hit songs from the past 54 years based on song characteristics?
• Social media and brand marketing in the hotel industry
  • How are different hotel brands engaging users on Twitter, and what is the customer response?
  • Other links: paper, code
• Predicting Kickstarter
  • Can we predict whether a Kickstarter project will be funded based on the project characteristics?
  • Other links: code
• Selling Ideas
  • Can we predict the amount of venture funding a startup will receive based on its patent applications?
  • Other links: paper, code, data
• Kaggle Allstate Purchase Prediction Challenge
  • Can we predict which car insurance options a customer will buy based on the quotes they reviewed?
  • Other links: paper, code
• Kaggle Driver Telematics Analysis
  • Can drivers be identified using their driving characteristics?
  • Other links: paper, code
• Kaggle Seizure Prediction Challenge
  • Can impending seizures be predicted using brain activity data?
  • Other links: paper, code, data
• Predicting Loan Defaults in Peer-to-Peer Lending Markets
  • Can we predict which Lending Club loans will default based on the loan and borrower characteristics?
  • Other links: paper
• Machine Technical Analysis
  • Can visual features be used for predictive modeling of security prices?
  • Other links: paper, code, data
• Predicting Treasury Yields
  • Can we predict US Treasury yields using macroeconomic data?
  • Other links: paper, code, data
• Predicting User Churn
  • Can we predict which LearnZillion users will turn into engaged users based on their actions?
• Twitter Music Recommendation Engine
  • Can we recommend music for a Twitter user based on their similarity to Twitter feeds of artists?
• Travel Recommendations Using Social Media Data
  • Can we recommend countries to visit based on a user's travel history as captured on social media?
  • Other links: paper
• Predicting Flight Delays
  • Can we predict flight delays using environmental factors and flight characteristics?
  • Other links: paper, code, data
• Bike Sharing in Mexico City
  • Can we predict demand for the Mexico City bikesharing system based on weather data and user characteristics?
  • Other links: code
• Data Athletics
  • Can we predict an athlete's future performance based on their race history and training data?
  • Other links: code, data
• Classifying PDFs as Likely Malicious or Likely Benign
  • Can we predict whether a PDF contains malware based on its characteristics?