28 May 2019
American Journal Experts (AJE) has developed and released a new tool that evaluates the quality of English in a manuscript and will suggest whether the paper needs language editing. The tool was built using deep learning in collaboration with Google and is designed to save publishers time by allowing their editorial teams to determine a paper's need for language editing prior to peer review.
The language assessment tool was built using deep learning AI to assess language quality. The tool was trained on a dataset of over 100,000 papers. From this large dataset of variable quality papers, the model has learned how to score the language quality. By working with publishers, AJE aims to optimise the language assessment tool to provide the most value to each publisher/journal/society/association by collecting feedback on desired output and function of the tool.
The language assessment tool is available for publishers, journals, and scholarly associations to trial now. AJE expects to make it available for use with institutions and individual authors later this year. For more information or to explore trial opportunities, contact Lindsay Miller, Strategic Partnerships Manager, lindsay.miller@aje.com.