RelSUE is a system able to search for the most relevant entities to a query entity provided by a user. It can leverage user-provided examples to learn the user's preference. Our tool is open-source under the Apache License. [Download]
We provide all the 8 groups of queries used in our experiments. In each file, odd-numbered lines correspond to query entities, while even-numbered lines correspond to relevant entities. Each entity is associated with a relevance score, in the form of uri:relevance where relevance is an integer. To use those queries for evaluation, please randomly pick some of them as examples; the rest form answers. [Download]
How to cite: Yu Gu, Tianshuo Zhou, Gong Cheng, Ziyang Li, Jeff Z. Pan, Yuzhong Qu. Relevance Search over Schema-Rich Knowledge Graphs. In: WSDM 2019. [Camera-ready version]
Errata: Fig. 3 and Fig. 4 in the proceedings version contain several incorrect numbers. The following figures are corrected.
Contact Yu Gu and Gong Cheng with any questions or comments.