الفهرس | يوجد فقط 14 صفحة متاحة للعرض العام |
المستخلص Named Entity Recognition (NER) task has become essential to improve the performance of many Natural Language Processing (NLP) tasks. Its aim is to come up with a solution to increase the accuracy of extracted named entities{u2019} identification. This thesis presents the first step to extract useful information for a researcher who is interested in the Egyptian People{u2019}s Assembly by creating a new corpus of the Egyptian People{u2019}s Assembly and presenting a novel solution for Arabic Named Entity Recognition (ANER). The solution uses a Conditional Random Field (CRF) sequence-labeling model by training it on mixing feature, morphological, gazetteers, and using character n-gram of leading and trailing letters in words. The results in this thesis show that the F-measure of mixing features running on the datasets of the Egyptian People{u2019}s Assembly is the better F- measure than other features run on the datasets as we are going to show in this thesis |