A Word-Complexity Lexicon and A Neural Readability Ranking Model for Lexical Simplification
This work addresses lexical simplification for natural language processing applications, offering a more human-aligned approach compared to existing heuristic methods.
The authors tackled the problem of lexical simplification by creating a human-rated word-complexity lexicon of 15,000 English words and proposing a neural readability ranking model that outperforms state-of-the-art systems on various tasks and datasets.
Current lexical simplification approaches rely heavily on heuristics and corpus level features that do not always align with human judgment. We create a human-rated word-complexity lexicon of 15,000 English words and propose a novel neural readability ranking model with a Gaussian-based feature vectorization layer that utilizes these human ratings to measure the complexity of any given word or phrase. Our model performs better than the state-of-the-art systems for different lexical simplification tasks and evaluation datasets. Additionally, we also produce SimplePPDB++, a lexical resource of over 10 million simplifying paraphrase rules, by applying our model to the Paraphrase Database (PPDB).