OpenAIRE ScholeXplorer Service: Scholix JSON Dump

This dataset contains the GZ-compressed dump of the Scholix links (schema Version 4) exposed by the OpenAIRE ScholeXplorer service. It consists of 417+Mi bi-directional links (i.e. 975+Mi directed links) between literature-dataset and dataset-dataset involving 24+ Mi literature objects and 37+ Mi datasets (showing an increase of around 160Mi links wrt the previous release). Links are collected from publishers (CrossRef, EventData), data centers (DataCite and data centers), institutional/thematic repositories (OpenAIRE), life-science databases (EMBL-EBI), and inferred by OpenAIRE via text-mining around 14Mi publication's PDFs. The dataset is structured in 30 compressed files, each of at most ~10 Gb, for a total of ~328GB. Note that the dataset matches a new version of the schema (schema Version 4). Changes are minor, backward compatible, and regard optional fields and extensions of vocabularies. The readme.doc file includes a description of the schema changes and statistics about the dataset.When citing this dataset, please cite the related publication also: "Uwe Schindler, Hylke Koers, Sandro La Bruzzo, Adrian Burton, Michael Diepenbroek, Amir Aryani, & Paolo Manghi. (2020). The data-literature interlinking service: Towards a common infrastructure for sharing data-article links. Program, Emerald Publisher, 51 (0033–0337), 75–100. Doi: 10.1108/prog-06-2016-0048