Measuring text readability with machine comprehension: a pilot study

摘要

This article studies the relationship between text readability levels and automatic machine understanding systems. Our hypothesis is that the simpler a text is, the better it should be understood by a machine. We thus expect a strong correlation between readability levels on the one hand, and performance of automatic reading systems on the other hand. We test this hypothesis with several understanding systems based on language models of varying strengths, measuring this correlation on two corpora of journalistic texts. Our results suggest that this correlation is quite small and that existing comprehension systems are far to reproduce the gradual improvement of their performance on texts of decreasing complexity.