n-duncan/GOLD-Digger

GOLD-Digger

Abstract

• We will utilize data from the UCSB course catalog to determine which factors (i.e. time, location, subject, days of the week) affect student enrollment.

Contributors:

• Wei Tung Chen
• Nicholas Duncan

Agenda:

1. scrap data from UCSB course catalog x
2. parse scraped data form UCSB course catalog x
3. clean the data 1-2 weeks x
   • remove empty rows
   • convert the strings into meaningful data
   • parse filenames into quarters and years
4. build models 2 - 4 weeks
   • we will use the 'pandas' python module to analyze the collected data
   • TBD
5. analyze built models 1 - week
   • create a graphical representation of the data
   • TBD
6. PUBLISH IT! 1 - 2 weeks

Scraper and Parser Script Usage

• execution (normal): python driver.py
• execution (with single quarter flag): singleQuarterFlag=True python driver.py
• set up automatic scheduler every hour:
  1. open crontab editor: crontab -e
  2. copy and paste line: 0 * * * * export DISPLAY=:0 && cd /path/to/directory/GOLD-Digger && cronFlag=True singleQuarterFlag=True python driver.py

Heroku Server Instructions

• Execution (local):
  1. Install libraries: npm install
  2. Start node server: npm start
  3. Open browser at http://localhost:5000/.
• Execution (heroku):
  • heroku open or
  • Open browser at https://ucsb-gold-digger.herokuapp.com/.
• Git Instruction:
  • Push changes to GitHub: git push origin master
  • Push changes to Heroku: git push heroku master