The challenge
Elsevier’s renowned scientific and medical publications include The Lancet, Cell and the ScienceDirect collection of electronic journals. Elsevier strives to uphold the quality and validity of individual articles and published journals.
Elsevier receives millions of articles that must be assessed for publication each year. However, the sheer volume of submissions meant editorial teams struggled to quantify and validate papers in a timely manner.
The solution
Crosstide helped Elsevier’s Data Science Team productionise machine learning (ML) and automate the initial processing of submissions in a GDPR-compliant manner.
Using a data set of just 200 words, the Data Science team now uses Natural Language Processing (NLP) to create a summary abstract of each paper, applying ML to quantify submissions against a variety of metrics, including scope match, trending issues, and author credibility.
The result
Crosstide enabled Elsevier’s Data Science Team to significantly reduce the burden of effort on editorial teams, reducing processing times by half. By embedding DevOps into the Data Science lifecycle, Crosstide enabled the use of new analytical strategies that streamline editors' workloads.
As a result, editors have the time and capacity to conduct rigorous peer reviews and never miss or reject a high-quality document that competitors would otherwise publish.
By embracing new AI-driven technologies that support reviewers and editors in the editorial process, Elsevier has improved how it conducts completeness and plagiarism checks while respecting confidentiality, proprietary rights, and data privacy.
We significantly streamlined workflows, which helped to avoid rejecting high-quality documents that competitors could otherwise publish."
%20Michael%20Seipp.jpg)