Search In this Thesis
   Search In this Thesis  
العنوان
Arabic named entity recognition and classification using semi-supervised learning /
الناشر
Noha Ahmed Saad Eldeen ,
المؤلف
Noha Ahmed Saad Eldeen
تاريخ النشر
2015
عدد الصفحات
97 Leaves :
الفهرس
يوجد فقط 14 صفحة متاحة للعرض العام

from 113

from 113

المستخلص

Arabic Named Entity Recognition (ANER) task aims to detect and classify atomic information such as Arabic proper names or numerical expressionsfrom unstructured texts. There has been growing interest in this field of research since the time Named Entity Recognition (NER) task was born in 1996. NER is considered an essential subtask for many NLP tasks because it greatly improves their performance. Most of researches were done towards English language and other languages that are similar to English making use of available annotated data. Recently Arabic started to take attention from researchers in this field. However Arabic is a complex language and not sufficient amount of Arabic Resources are available. Research on Arabic has to face all of Arabic Challenges and overcome them. In other words, recent machine learning approaches have a challenge with annotated data availability which is a serious problem in building and maintaining large scale of ANER systems