The question
Curriculum reviews involve comparing learning outcomes across many spreadsheets. Keyword matching misses changes in wording that preserve the same meaning.
The approach
Built a Python desktop workflow using the all-MiniLM-L6-v2 Sentence Transformer model, adjustable similarity thresholds, and background processing so large comparisons do not freeze the interface.
The result
A capstone application that lets reviewers select Excel files, inspect semantic matches, and export comparison reports. Similarity scores support human review rather than deciding course equivalency.
Project information from the original portfolio and linked public sources. Read the project documentation ↗