dkolovos.7: Using Machine Learning for HTML/CSS Linting


HTML/CSS/JS frameworks such as Bootstrap [1] provide a wealth of built-in components and styles for developing modern web-based UIs. Similar frameworks have been developed internally in many organisations to standardise the look and feel of large websites. Such frameworks come with a lot of implicit styling rules (e.g. in Bootstrap, "column" divs are expected to be found only under "row" divs).

The aim of this project is to use machine learning techniques to learn such rules by examining valid pages and to then check for violations in previously unseen pages.


Suitable applicants should have strong object-oriented skills. Some experience with HTML/CSS/JavaScript would be useful but is not essential.


This project meets the project specifications of the following courses:


  1. Twitter Bootstrap
  2. S. C. Johnson. Lint, a C Program Checker

 Printer-friendly version