[Link to demo and python package.]
I built a small python package that estimates the readability of Japanese text.
The model used for predicting the readability was developed by Jae-ho Lee and Yoichiro Hasebe and was originally built using passages from various JLPT-aligned textbooks. You can read more about their model here and here. They also have a very useful site for analyzing Japanese text. Unfortunately there just wasn't any python implementation of their model that I could find, which is why I went and made one 🙂
by joshdavham