Micro- and Macro-Level Features of NLP-Based Writing Tools in Higher Education

Abstract

This paper reviews tools that use natural language processing (NLP) to support writing in higher education. These NLP-based writing tools usually include various features such as grammar and spelling suggestions, but also general feedback such as the overall readability or consistency of the text. In contrast to previous studies, these feedback features are analyzed and described in detail to give a more accurate overview of the current capabilities of these tools. In a systematic analysis, we present 14 NLP-based writing tools that provide different types of micro-level feedback (e.g., grammar, spelling), macro-level feedback (e.g., argumentative structure, rhetorical moves), as well as additional services (e.g., plagiarism checks, access to writing templates). In contrast to previous findings, the state-of-the-art analysis revealed that many writing tools provide feedback at the micro-level and the macro-level. However, macro-level feedback and further additional services are often only included in paid subscription plans. Overall, the presented features of NLP-based writing tools could be helpful in designing own use cases as they may serve as a checklist and guideline to develop tools that go beyond what already exists.