Back

WikiProject Clinical Trials for Wikidata

Rasberry, L.; Tibbs, S.; Hoos, W.; Westermann, A.; Keefer, J.; Baskauf, S. J.; Anderson, C. B.; Walker, P. D.; Kwok, C.; Mietchen, D.

2022-04-04 health informatics
10.1101/2022.04.01.22273328 medRxiv
Show abstract

WikiProject Clinical Trials is a Wikidata community project to integrate clinical trials metadata with the Wikipedia ecosystem. Using Wikidata methods for data modeling, import, querying, curating, and profiling, the project brought ClinicalTrials.gov records into Wikidata and enriched them. The motivation for the project was gaining the benefits of hosting in Wikidata, which include distribution to new audiences and staging the content for the Wikimedia editor community to develop it further. Project pages present options for engaging with the content in the Wikidata environment. Example applications include generation of web-based profiles of clinical trials by medical condition, research intervention, research site, principal investigator, and funder. The projects curation workflows including entity disambiguation and language translation could be expanded when there is a need to make subsets of clinical trial information more accessible to a given community. This projects methods could be adapted for other clinical trial registries, or as a model for using Wikidata to enrich other metadata collections.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.