Towards Arabic Sentence Simplification via Classification and Generative Approaches

TL;DR

This paper explores Arabic sentence simplification using classification and generative methods, leveraging Arabic-BERT, fastText, and mT5 models, with promising results demonstrated through BERTScore evaluation and manual analysis.

Contribution

It introduces a novel Arabic sentence simplification system combining lexical and generative approaches, with a new aligned corpus from an Arabic novel.

Findings

The lexical approach with Arabic-BERT and fastText achieved high BERTScore (F-1 0.97).

The generative mT5 model achieved moderate BERTScore (F-1 0.70).

Manual error analysis provided insights into model performance.

Abstract

This paper presents an attempt to build a Modern Standard Arabic (MSA) sentence-level simplification system. We experimented with sentence simplification using two approaches: (i) a classification approach leading to lexical simplification pipelines which use Arabic-BERT, a pre-trained contextualised model, as well as a model of fastText word embeddings; and (ii) a generative approach, a Seq2Seq technique by applying a multilingual Text-to-Text Transfer Transformer mT5. We developed our training corpus by aligning the original and simplified sentences from the internationally acclaimed Arabic novel "Saaq al-Bambuu". We evaluate effectiveness of these methods by comparing the generated simple sentences to the target simple sentences using the BERTScore evaluation metric. The simple sentences produced by the mT5 model achieve P 0.72, R 0.68 and F-1 0.70 via BERTScore, while, combining Arabic-BERT and fastText achieves P 0.97, R 0.97 and F-1 0.97. In addition, we report a manual error analysis for these experiments. \url{https://github.com/Nouran-Khallaf/Lexical_Simplification}