This paper is a follow-up to our Florida AI Research Society (FLAIRS) proceedings from 2018.
We used a novel approach to generate human ratings of comparative text difficulty to serve as our benchmarks. We then used a variety of natural language processing tools to extract multidimensional features of language in the texts. These features were submitted to a number of machine learning classification algorithms that classify data in a flat manner in a hierarchical fashion.
Balyan, R., McCarthy, K. S., & McNamara, D. S. (2020). Applying Natural Language Processing and Hierarchical Machine Learning Approaches to Text Difficulty Classification. International Journal of Artificial Intelligence in Education, 1-34. https://doi.org/10.1007/s40593-020-00201-7