Text Classification in Arabic Natural Language Processing: A Review

Authors:

Abstract
Natural language processing holds immense importance across diverse fields, primarily centered around the English language. In contrast, efforts directed at processing Arabic have been comparatively limited. Therefore, our goal is to augment the existing efforts focused on advancing Arabic text processing. Generally speaking, text processing can vary according to specific demands. It encompasses tasks like text recognition, text analysis, or text classification. Among these, text classification has become a key focus in this domain and forms the core subject of this paper. Methods employed span AI to data mining algorithms, selected according to the text and the classification objective. The process consists of four primary stages: dataset selection, preprocessing and representation, text classification, and experimentation/evaluation. The objective of this paper is to compile the latest research that contributes to solving some of the challenges of Arabic text processing in a precise manner and highlight their contributions, findings, methodologies, and advancements in the Arabic Natural Language Processing ANLP field. In addition, we discuss and analyze these studies to find their limitations and make some suggestions.

This publication has 25 references indexed in Scilit: