BIOF 309 Project Ideas

BIOF 309 Curriculum - back to main page

project ideas for data ninjas

Quantitative _____________.
Use python to crunch a data set and/or perform some calculation
identify a bottleneck, think up a task that requires a lot of data shovelling in your work
Will require writing up a project proposal
There will be a Google document where you'll put the project title and description up to avoid project overlap
You can help each other, but everyone must do a separate project, meaning you write your own code
- Write your code in a nicely packaged module so if you wanted to collaborate in the future, you can exchange modules

Pro-Football-Reference.com NFL stats
- A visualization of NFL stats here
- use machine learning to do fantasy football predictions.
- New York Times 4th down bot
- Come up with a replacement statistic for the much-maligned QB rating, then do some historical revisionism, who was an overrated/underrated QB from NFL history.
Regressing
Flowing Data

"Trendspotting in asset markets" Nobel Prize in Economics 2013 Fama, Hanson, and Shiller (Teach us all how to get rich quick!)
Fraud detection using forensic accounting/Benford's Law

Blokus game simulation, monte carlo, perhaps develop algorithm, perhaps implement the game - "Blokus kinda like tetris except the object is not to be as compact as possible, in fact your pieces can't touch each other at all. No piece is like any other."

Spot the station - Develop an ISS orbiting simulation, or a position of some star as it would appear to us on earch, or graph the procession of the perihelion of mercury
Cannon firing program that gives you angles to aim your cannon, given a projectile initial speed and height, and location of enemy in x,y, & z. (Good opportunity for GUI!)

New Organizing Institute US political data - FEC filings, Governance Project (db of 50,000 elected officials, 2012 nationwide General Election ballot data, polling data from the Voter Information Project (VIP)
Google Trends - See where, when and what people searched for in Google.

Attempt a Kaggle competition
Applying to job/college? Have drunken pictures of you on the internet? Create python automated web crawling program to look for potential offenders, show specificity vs selectivity.

They Might Be Giants Song Lyrics Usage - Each dot is a word, the x-axis is the number of times the word appeared in a They Might Be Giants song, the y-axis is the proportion of the frequency they used it in a song versus frequency of how often it's used in the English language as a whole, based on the Google Books ngrams dataset. This guy created a text scraper, which is a computer program that systematically visits all the links in a web page and captures the information contained thereupon