Interpretation of Computer Science Educational Data for the State of Wisconsin and a Static Analysis Driven Approach to Classifying Student Code

From REU@MU
Revision as of 16:55, 18 June 2020 by Alexanderr (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Project Description

Computer Science education has never been so critical to the next generation of academics and industry leaders as it is today. 67% of all new STEM jobs created are in computing, but only 11% of STEM bachelor degrees are in computer science. This massive disparity is exacerbated by a fundamental lack of CS education in middle schools and high schools across the country. This project aims to uncover new information about how CS education has changed in the state of Wisconsin, and visualize the data in such a way as to encourage policy makers to take a serious look at the approach schools take with respect to computer science. Furthermore, this project also seeks to aid CS education at the university level by integrating static analysis techniques into Marquette's TABot, used in the grading of student-written C code in classes such as Operating Systems.


Project Goals

  • Gather, clean, and interpret CS education data concerning teacher licensure, enrollment statistics, program quality, and geographic location
  • Develop models trained on the data to predict what CS education might look like in the future
  • Integrate existing static analysis tools with TABot and analyze the pros and cons of particular methods

Milestones:

Week Description
Week 1
  • Initial meetings with mentor
  • Data science crash course
Week 2
  • Read related literature on CS education
  • Practice machine learning techniques
  • Begin scraping/cleaning data
Week 3
  • Automate process of data scraping
  • Read related literature on static code analysis
Week 4
  • Visualize/interpret collected data
  • Experiment with different static code analysis techniques on sample code