Full Automatic Arabic Text Tagging System

Part-of-Speech tagging is the process of assigning grammatical part-of-speech tags to words based on their context. Many automated tagging systems have been developed for English and many other western languages, and for some Asian languages, and have achieved accuracy rates ranging from 95% to 98%. A tagged corpus has more useful information than untagged corpus; so, tagged corpus can be used to extract grammatical and linguistic information from the corpus. Then, it can be used for many applications such as creating dictionaries and grammars of a language using real language data. Tagged corpora are also useful for detailed quantitative analysis of text. In this project, we have described a system for recognizing names, verbs, particles, and proper names in Arabic language text through a combination of high precision morphological analysis and a subsequent component which recognizes the named entities. Although highly deterministic and not taking account of context, the morphological analysis component removes a great deal of morpho-lexical ambiguity; yet, has the side-effect of demonstrating that the true difficulties in Arabic morphological ambiguity might be limited to specific contexts. We have shown that morphological information is crucially important to effective Arabic name recognition. We have tested our system by using some vowelized and non-vowelized text documents, and achieved an accuracy rate of about 93%. We have stated the factors of errors and how this accuracy rate can be enhanced.