Agentic data science with the Data Agent Kit
cloud.google.com·6h ago
AI Research
arXiv CS.AI·Aug 7·5 min read
CourseGraph: Finding overlaps and differences in Computer Science courses across universities
TL;DR
CourseGraph is a new methodology that automates the evaluation of external Computer Science courses to identify overlaps with a student's home curriculum. It effectively supports curriculum alignment across universities, enhancing student mobility programs.
✦ Why It Matters
Implement CourseGraph to streamline course evaluation processes and enhance student mobility programs today.
Key Takeaways
- Utilize CourseGraph to automate course evaluation for student mobility.
- Extract course data using web scraping techniques for analysis.
- Implement BERT for semantic representation of course information.
- Leverage Random Forest classifiers to assess course overlap.
How It Works
CourseGraph extracts course information from university websites, including titles and descriptions. It uses a BERT-based language model to semantically represent this data, enabling the calculation of similarity between courses.
A Random Forest classifier then evaluates whether a candidate course overlaps with existing courses in a student's home curriculum.