Extracting Arabic Composite Names Using Genitive Principles of Arabic Grammar
Hussein Khalil, Taha Osman, Mohammed Miltan · ACM Transactions on Asian and Low-Resource Language Information Processing · 2020
Named Entity Recognition (NER) is a basic prerequisite of using Natural Language Processing (NLP) for information retrieval. Arabic NER is especially challenging as the language is morphologically rich and has short vowels with no capitalisation convention. This article presents a novel rule-based approach that uses linguistic grammar-based techniques to extract Arabic composite names from Arabic text. Our approach uniquely exploits the genitive Arabic grammar rules; in particular, the rules regarding the identification of definite nouns (معرفة) and indefinite nouns (نكرة) to support the process of extracting composite names. Based on domain knowledge and Arabic Genitive Rules (AGR), the developed approach formalises a set of syntactical rules and linguistic patterns that initially use genitive patterns to classify definiteness within phrases and then extracts proper composite names from the unstructured text. The developed novel approach does not place any constraints on the length of the Arabic composite name and our initial experimentation demonstrated high recall and precision results when the NER algorithm was applied to a financial domain corpus.