الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 226 |  |  |
| | | from | 226 | | |
Abstract
A Named Entity is a word, or sequence of words that can be classified as a name of a person, organization, location, date, time, percentage, or quantity. Named entity recognition (NER) systems aim to automatically identify and classify the proper nouns in text. The Named Entity Recognition (NER) task has been gaining huge attention in Natural Language Processing (NLP) as it proved to be valuable in several NLP applications such as Information Retrieval, Question Answering tasks, and text clustering.
The aim of this study is to design and to construct a system that can efficiently and effectively be employed to Arabic Named Entity Recognition systems whose results could be easily used and embedded in any further NLP tool. Unlike most of the previous NER systems, the system concentrates more in-depth on classification of NEs categories and constrains. It also employs a rule based approach that helped solving some of the challenges posed by Arabic language and made a successful implementation of ANER system. Furthermore; the system improves the performance when dealing with NE types that appear without any trigger words by using semantic tags. In contrast with the majority of the systems in this field, this system did not use any predefined person name gazetteers. It uses only gazetteers of parts of trigger words not the whole trigger word this way was used to eliminate the number of trigger words used in search for NEs that reduces the time of tagging process.
The evaluation results of the system designed for the purpose of this study showed that the system can be considered as an accurate automatic tagging tool. It is designed in a simplified way that enables increasing the number of rules. In addition, the design of the database and the implementation of the system allows for adding more trigger words that could enhance the tagging process. Therefore it is more flexible and can account for different types.