Executive Summary
Automating Gap Analysis
The primary focus of this project was to apply machine learning and Natural Language Processing (NLP) approaches to assist post-secondary institutions with process of analyzing the differences and similarities in learning outcomes between two separate programs.
Ideally, the project team envisioned that Transfer Pathway development projects between partnering postsecondary institutions would benefit from a system where:
1) All project partners could upload their course information into a shared database;
2) An artificial intelligence provides an estimation of the semantic similarity between multiple courses; and
3) Recommendations and rankings of courses could be generated to guide discussion and content expert reviews of course equivalencies.
The results of this project include:
- The development of an unsupervised NLP algorithm that calculates the semantic similarity between learning outcomes;
- An online user interface that allows users to:
- Upload course outlines and learning outcomes;
- Complete a semantic analysis of course level learning outcomes on a course by course or program by program basis; and
- Receive output in the form of reports and visualizations that summarize the differences between two separate sets of courses and learning outcomes.
- Three separate test cases that apply historical results of transfer pathway gap analysis and content expert rankings; and
- The development of a field specific ‘corpus’ to adjust the NLP algorithm to highly specified fields in post-secondary education.