We are pleased to announce that Writefull has expanded its set of services for publishers. In addition to the Writefull Revise language check, Writefull now offers APIs for automatic manuscript categorization and metadata extraction. Early adopters are Hindawi and ACS (American Chemical Society). Through automating otherwise manual and time-consuming tasks at scale, the Writefull APIs allow for significant savings throughout the publishing workflow.
The Manuscript Categorization API leverages Writefull’s proprietary language models to evaluate the number and type of language edits a manuscript needs, and then assigns a score or category to the manuscript.
The API classifies manuscripts at scale without editors having to open documents and scan the text, significantly reducing the time spent on manuscript evaluation.
The service can be used at different points in the publishing workflow. ACS, an early adopter that has worked closely with the Writefull team to test and shape this service, uses the API for two use cases:
- Post-acceptance, to evaluate how much language editing manuscripts require, allowing to assign them to designated resources or workflow paths. Read more about this use case here.
- Post-copyediting, to evaluate the language in manuscripts as part of their quality control program.
ACS and Writefull have thoroughly tested the API outputs and found an alignment of over 95% between the classifications of Writefull and ACS’s human-assigned values.
For even more significant time savings, the Manuscript Categorization API can be combined with Writefull’s Language API, which offers automated language revision. The Language API can auto-apply (a subset of) Writefull’s language suggestions before copyediting, and the Manuscript Categorization API can show much copyediting is still needed after this auto-revision.
Metadata extraction and structural checks
In addition to the language services APIs, Writefull recently launched a Metadata API. This API leverages state-of-the-art NLP to automatically extract elements from (batches of) manuscripts. Think of sections, author names, affiliations, and captions. It lists these components in a structured JSON response file.
Publishers can use the API response for different purposes, such as to cross-check and validate manuscripts, to auto-populate systems, or to run structural checks. It offers a scalable, time-saving solution, as publishers no longer need to manually extract elements or review manuscripts for consistency and completeness.
Hindawi uses Writefull’s Metadata service as an author-facing structural checks tool. By using this service, authors can ensure that their manuscript meets the requirements to be handled by an editor. Hindawi, in turn, saves a considerable amount of time checking the submitted manuscripts. Read more about Hindawi’s use case here.
ACS uses Writefull’s Metadata API to assess the quality and accuracy of their copyedited manuscripts, as part of their quality control program. ACS’s manuscript reviewing process, carried out with an elaborate scorecard, is now partly automated thanks to Writefull’s Metadata API. Learn more about ACS’s use case here.
Would you like to learn more, or discuss how Writefull may fit your publishing workflow? Please contact us at firstname.lastname@example.org.