@inproceedings{barsotti2017predicting,
title={Predicting invariant nodes in large scale semantic knowledge graphs},
author={Barsotti, Damian and Dominguez, Martin Ariel and Duboue, Pablo Ariel},
booktitle={Annual International Symposium on Information Management and Big Data},
pages={48--60},
year={2017},
organization={Springer, Cham},
url={http://ceur-ws.org/Vol-2029/paper4.pdf}
}
This 2017 paper tracks DBpedia evolution, using Spark. It is joint work with the large scale social and semantic networks research group at FaMAF.
The paper was presented at the SIMBig conference in Peru (remotely).
While the paper underwent two rounds of reviews and it had three authors that actively contributed to the code-base, it has a major target-leak in the sparql queries that invalidate its results. We haven't been able to compute updated results. While the results themselves are not representative, the technique presented is sound.
A preprint is available on-line.