Title |
Person Named Entity Generation and Recognition for Arabic Language |
Authors |
Ibrahim Alkharashi |
Abstract |
Arabic person named entities have unique characteristics that govern the generation and analyzing processes. This work presents an infrastructure that has been developed to assist and simplify the processes related to Arabic person named entities such as generation, recognition, translation, transliteration and correction. The infrastructure consists of a single two-dimensional person name map-table that maps the root of a given Arabic person name to its valid pattern. Each valid Arabic person name or Arabic name fraction is projected onto that table. Each projected entry is assigned some properties that identify some affixation and gender characteristics. The paper will focus on two main parts regarding the Arabic person name map-table. The first part will be devoted to discuss the person named entity production behavior in the Arabic language with emphasis on some statistical findings on the capabilities of roots in producing person names, and patterns usability as a model for person names. The second part will discuss the usability of the Arabic person name map-table as a training set to assist the process of person named entity recognition, correction and transliteration. |
Topics |
Definition and requirements for a Basic LAnguage Resource Kit (BLARK) for Arabic, Extraction and acquisition of knowledge (e.g. terms, lexical information, language modelling) from LRs, Methods, tools and procedures for acquisition, creation, management, access, distribution and use of Arabic LRs |
Full paper |
Person Named Entity Generation and Recognition for Arabic Language |
Bibtex |
@InProceedings{ALKHARASHI09.44,
author = {Ibrahim Alkharashi},
title = {Person Named Entity Generation and Recognition for Arabic Language},
booktitle = {Proceedings of the Second International Conference on Arabic Language Resources and Tools},
year = {2009},
month = {April},
date = {22-23},
address = {Cairo, Egypt},
editor = {Khalid Choukri and Bente Maegaard},
publisher = {The MEDAR Consortium},
isbn = {2-9517408-5-9},
language = {english}
} |