Wikimedia Research/Showcase/Archive/2019/03

March 2019

March 20, 2019 Video: YouTube

Learning How to Correct a Knowledge Base from the Edit History

slides

By Thomas Pellissier Tanon (Télécom ParisTech), Camille Bourgaux (DI ENS, CNRS, ENS, PSL Univ. & Inria), Fabian Suchanek (Télécom ParisTech), WWW'19.

The curation of Wikidata (and other knowledge bases) is crucial to keep the data consistent, to fight vandalism and to correct good faith mistakes. However, manual curation of the data is costly. In this work, we propose to take advantage of the edit history of the knowledge base in order to learn how to correct constraint violations automatically. Our method is based on rule mining, and uses the edits that solved violations in the past to infer how to solve similar violations in the present. For example, our system is able to learn that the value of the sex or gender property woman should be replaced by female. We provide a Wikidata game that suggests our corrections to the users in order to improve Wikidata. Both the evaluation of our method on past corrections, and the Wikidata game statistics show significant improvements over baselines.

TableNet

An Approach for Determining Fine-grained Relations for Wikipedia Tables

slides

By Besnik Fetahu

Wikipedia tables represent an important resource, where information is organized w.r.t table schemas consisting of columns. In turn each column, may contain instance values that point to other Wikipedia articles or primitive values (e.g. numbers, strings etc.). In this work, we focus on the problem of interlinking Wikipedia tables for two types of table relations: equivalent and subPartOf. Through such relations, we can further harness semantically related information by accessing related tables or facts therein. Determining the relation type of a table pair is not trivial, as it is dependent on the schemas, the values therein, and the semantic overlap of the cell values in the corresponding tables. We propose TableNet, an approach that constructs a knowledge graph of interlinked tables with subPartOf and equivalent relations. TableNet consists of two main steps: (i) for any source table we provide an efficient algorithm to find all candidate related tables with high coverage, and (ii) a neural based approach, which takes into account the table schemas, and the corresponding table data, we determine with high accuracy the table relation for a table pair. We perform an extensive experimental evaluation on the entire Wikipedia with more than 3.2 million tables. We show that with more than 88% we retain relevant candidate tables pairs for alignment. Consequentially, with an accuracy of 90% we are able to align tables with subPartOf or equivalent relations. Comparisons with existing competitors show that TableNet has superior performance in terms of coverage and alignment accuracy.