This tool aims to quantify the readability of Wikipedia articles in different languages.
Existing approaches based on readability formulas use simple text-based features such average sentence length to yield a score that captures the difficulty of a text. Often this score is given as an approximate representation of the US grade level needed to comprehend the text (such as the Flesch-Kincaid grade level). The problem is that these formulas are often tailored specifically towards English and are thus not easily applicable for texts of other languages.
We developed a multi-lingual model for scoring readability of articles supporting more than 100 languages. You can read more about the model here: the model card, the research paper, or the project page on MetaWiki.
You can test out the tool below with your own article -- e.g., Toni Morrison (en) -- or leave the title field blank for a random article. The model's output will consist of: i) a readability score ranging from negative (easy to read) to positive (difficult to read), ii) a predicted grade level (i.e. roughly capturing the number of years of education generally required to understand this text) similar to the Flesch-Kincaid grade level for English.