← Back to Research
Academic project

Academic Paper Ranking and Recommendation System

Question

Rank academic papers by citation-graph influence and combine that ranking with content-based recommendation.

Role

Contributed the ranking component to a group system as part of an MA 544 (Numerical Linear Algebra for Big Data) group project.

Method

Sparse PageRank via power iteration over a 1.7-million-node academic citation graph, combined with TF-IDF and word-vector similarity for content-based recommendation.

Division of work

Human

Implemented the sparse PageRank component using CSR matrices and power iteration.

AI

None disclosed; this is a course project.

Verification and limitations

Course deliverable; independent review beyond the course has not occurred.

Related