We propose methods for extracting triples from Wikipedia’s HTML tables using a reference knowledge graph. Our methods use a distant-supervision approach to find existing triples in the knowledge graph for pairs of entities on the same row of a table, postulating the corresponding relation for pairs of entities from other rows in the corresponding columns, thus extracting novel candidate triples. Binary classifiers are applied on these candidates to detect correct triples and thus increase the precision of the output triples. We extend this approach with a preliminary step where we first group and merge similar tables, thereafter applying extraction on the larger merged tables. More specifically, we propose anobserved schemafor individual tables, which is used to group and merge tables. We compare the precision and number of triples extracted with and without table merging, where we show that with merging, we can extract a larger number of triples at a similar precision. Ultimately, from the tables of English Wikipedia, we extract 5.9 million novel and unique triples for Wikidata at an estimated precision of 0.718.

Henry Rosales-Méndez, Aidan Hogan, Barbara Poblete. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

The validation of overlapping clusterings has gained considerable attention in the last years, as it represents a fundamental step towards achieving reliable clustering algorithms, still being largely considered as an open question. A number of measures have been proposed for validating overlapping clusterings, but it remains unclear which one is the most convenient. Most existing external measures aim to validate partitional clusterings, even when most real-life problem involve overlapping clustering scenarios. In this paper, we propose a new external measure specifically designed for validating overlapping clusterings, which fulfills the main set of desirable conditions presented in the literature. We also show that our proposal correctly handles situations where previous measures display undesirable performances.