Please use this identifier to cite or link to this item:
http://hdl.handle.net/20.500.11889/4395
Title: | Tools for Arabic People Names Processing and Retrieval A Statistical Approach | Authors: | Salhi, Ali Yahya, Adnan |
Keywords: | Names, Personal - Arabic;Statistics - Databases;Names Correction;Names translation;Names - Gender etection;Names extraction;Natural language processing | Issue Date: | 9-Oct-2011 | Abstract: | Arabic web content has been rapidly growing, generating a need for tools to overcome the many challenges of processing and retrieving Arabic content: challenges related to Arabic Language Processing, Search and Query Analysis. An important part of dealing with Arabic digital content is processing and analyzing Arabic people names. This paper reports on our work aimed at designing name pre-processing tools that are able to efficiently identify and process Arabic people names in queries and documents. We try to address challenges such as Name Gender Detection, Translation (Arabic to English), Correction, Auto Suggestion and Extraction from text. All through, we employ a statistical approach based on data obtained from High School student names lists in Palestine and Birzeit University student names lists. Based on this information we constructed different types of databases of Arabic names and used them as the infrastructure for the well structured names tools which are capable of being integrated into existing web search engines and document processing systems. We have been experimenting with some of the developed tools in our online application process at Birzeit University, with encouraging preliminary results. | URI: | http://hdl.handle.net/20.500.11889/4395 |
Appears in Collections: | Fulltext Publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
ALTICPaperOct2011yahyaAliVer2.pdf | 643.31 kB | Adobe PDF | View/Open |
Page view(s)
128
Last Week
0
0
Last month
2
2
checked on Mar 25, 2024
Download(s)
171
checked on Mar 25, 2024
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.