Research on the Evaluation of Textual Comprehensibility Based on French Corpus

The study on the degree of the textual comprehensibility based on French corpus comes under the umbrella of corpus-involved text research. A systematic random sampling is employed in the present paper to compare the six different groups extracted from one and the same sampled French text. The formula, viz. A+BX≤C, is provided here to equidistantly extract linguistic fragments from the famous French novel L'Assommoir. "A" means the starting page, "B" the distance of extraction, "C" the total number of the pages contained in a text, "X" the set of pages available. If the condition is fully met by B∈(10, 5, 4, 3, 2, 1), the corresponding sampling ratios are 9.55%, 20.29%, 25.13%, 33.13%, 50.36% and 100% respectively, which also bring different values of textual comprehensibility, notably 0.0897, 0.0841, 0.0854, 0.0847, 0.0848, 0.0854. The non-relevance between textual comprehensibility and the sampling ratio verifies that the system of textual comprehensibility evaluation based on English and Japanese corpora can successfully extend to the French corpus, which helps the literature critics to analyze French texts according to a pre-set procedure via computers.