The design of a Lexical Difficulty Filter for language learning on the Internet

Abstract

In this paper, we describe the design and implementation of a tool called the Lexical Dificulty Filter (LDF) intended to help learners or teachers in selecting sentences appropriate to the level of students for English vocabulary learning. The LDF serves as a bridge between learner and information resource bank. It selects suitable sentences from the output of standard English corpus concordancers. The core technology in the design of LDF is a f u u y expert system. The results obtained are very encouraging in this pioneering work. We achieve 94.33% accuracy rate in imitating the judgments of a human expert in determining the degree of dificulty of a sentence for a given learner level.

Sentence Difficulty Evaluation for a Learner's Dictionary

“…Kuo et al [3] developed a system to classify English sentences by difficulty with a fuzzy expert system: It was trained on a set of sentences classified in three difficulty classes by English teachers, and used features such as sentence length and proportion of words above a frequency threshold to determine difficulty. Unlike readability indexes, this concept can be applied to Chinese…”