Reproduce “Knowledge Translation: Extended Technical Report” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Bahar Ghadiri Bashardoost, Renée J. Miller, Kelly Lyons, Fatemeh Nargesian

Kensho is a Knowledge Mapping Generation Tool (KMGT) that automates the creation of SPARQL mapping rules to translate information between heterogeneous knowledge bases. It achieves this by discovering semantic associations between aligned elements and interpreting correspondences, demonstrating the ability to handle complex scenarios and scale with increasing KB complexity.

Key insights

  • Discovering semantic associations, which capture related concepts, attributes, and relationships through property paths, is crucial for comprehensive knowledge translation.
  • Applying validity constraints during correspondence interpretation significantly reduces the number of generated mapping rules while preserving semantically meaningful ones.
  • Combining ranking heuristics such as path priority, consistency, and coverage helps users identify the most relevant mapping rules among many possibilities.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.