Course Search Engine

The Jaccard Similarity Coefficient is a statistical formula used to measure the structural similarity between two sets. In this search engine; Set A represents the unique keywords of the first course, and Set B represents the second course's unique keywords found in the catalog (excluding conjunctions).

The algorithm calculates a deterministic overlap ratio by dividing the number of common words by the total number of unique words in both courses. This method does not perform semantic analysis; it is directly based on the terminological intersection of the curriculum.

This model is an advanced Jaccard algorithm that takes into account the hierarchical structure of the curriculum. Unlike standard calculation, it weights the term according to where it appears. Matching words are weighted 3x if they are in the Root (Course Code), 2x in the Trunk (Course Title), and 1x in the Leaf (Content). Thus, going beyond mere word counting, the most appropriate results for academic reference are highlighted.