Scholarly triage: Advances in manuscript submission using text analytics techniques

The drastic increase in the volume of submissions for inclusion in scholarly journals offers new challenges for scholarly publishers. These include sorting through thousands of new manuscripts to prioritize those most likely to be published first and detecting dubious research. Although recent advances in peer review and editorial management platforms have advanced processes to help alleviate the problem, new solutions to prioritize high-value papers – and to flag suspect ones – are emerging. Using text analytics methods grounded in natural language processing (NLP) and other techniques to augment the submission review process can include leveraging taxonomy-based indexing terms to match manuscripts to appropriate reviewers and editors, preventing fraud by detecting machine-generated entries, screening for irreproducible research practices, and predicting the likelihood of acceptance by examining non-content factors.