|
[9000883.] رقم البحث : 9000883 - |
|
Bilingual Language Model for English Arabic Technical Translation / |
| تخصص البحث : Machine Translation |
| |
هندسة اللغة: / عدد (2) - مجلد (2) - سبتمبر 2015 |
| |
تاريخ تقديم البحث 23/10/2015 |
| |
تاريخ قبول البحث 23/10/2015 |
| |
عدد صفحات البحث 10 |
| |
Marwa N. Refaie ( basmallah@hotmail.com - ) |
| |
Ibrahim F. Imam ( ifi05@yahoo.com - ) |
| |
Ibrahim F. Eissa ( i.farag@fci-cu.edu.eg - ) |
| |
Statistical machine translation, Domain adaptation, Bilingual Model, Fill-up phrase table |
| |
Abstract:The massive fast of new scientific publications increase the need to a reliable effective automatic machine translation (AMT) system, which translates from English, as the common language of publications, to other different languages. Statistical machine translation (SMT) model crafted to deal with certain domain of text often fails when subjected to another domain. The paper addresses the characterization of language domains and their behavior in SMT, experiments the management of SMT model to translate scientific text collected from artificial intelligence publications. The effectiveness of Bilingual language model is tested against the typical N-gram language model, in addition to utilizing the fill-up and back-off techniques to handle different phrase tables from different domains. As not every human capable to translate artificial intelligence book, should have strong knowledge in the field, We suggest that in order AMT can handle different domains it must be trained by in-domain parallel data, adjusting weights for the words on different domains to learn the model how to differentiate between different meaning of same word in different domains. |
| |
 Download Paper |
|
[9000884.] رقم البحث : 9000884 - |
|
MASHAEIR: A Corpus-Based Multi-Dialect Fine-Grained Emotion Thesaurus for Arabic Social Media Emotion Recognition / |
| تخصص البحث : Social Networks Contents Analysis |
| |
هندسة اللغة: / عدد (2) - مجلد (2) - سبتمبر 2015 |
| |
تاريخ تقديم البحث 23/10/2015 |
| |
تاريخ قبول البحث 23/10/2015 |
| |
عدد صفحات البحث 12 |
| |
KhaledElghamry ( elghamryk@gmail.com - ) |
| |
The user-generated content on social media sites, e.g. Twitter and Facebook, provides a rich source of people’s emotions towards products, issues, people and major events. Accordingly, the focus of more research has moved from negative-positive sentiment classification tasks to tasks of recognizing more fine-grained emotions. However, research on and resources for fine-grained emotion identification in Arabic texts are still lacking. To fill in this gap, this paper introduces MASHAEIR (an Arabic word that means ‘emotions’), a corpus-based multi-dialect fine-grained emotion thesaurus for Arabic. MASHAEIR was bootstrapped using ’big data’ from Arabic Twitter from January 2007 to July 2015. The thesaurus is enriched with (i) different types of single- as well as multi-word terms expressing emotions, (ii) Arabic dialectal variations in the expression of emotions and (iii) scores that reflect the intensity of the emotions conveyed through these units. The paper also presents a simple evaluation of the thesaurus coverage on a sample Twitter corpus. MASHAEIR is intended to present an outline of a large-scale and easy-to-update emotion thesaurus for Arabic that could also be enriched in the future with more information such as gender and age preferences in expressing emotions |
| |
Download Paper |
|
[9000885.] رقم البحث : 9000885 - |
|
Statistics in the Greek Linguistic Studies: A Computational Approach / |
| تخصص البحث : Natural Language Processing |
| |
هندسة اللغة: / عدد (2) - مجلد (2) - سبتمبر 2015 |
| |
تاريخ تقديم البحث 23/10/2015 |
| |
تاريخ قبول البحث 23/10/2015 |
| |
عدد صفحات البحث 9 |
| |
Abdelmonem Ahmed Zaki ( abdelmoneam.ahmed@art.asu.edu.eg - ) |
| |
Statistical Linguistics, Linguistic Phenomenon, Greek Linguistics, Computational Approach. Morphological Classification, Semantic Classification. |
| |
This paper aims to explain why we would use statistical methods for Ancient Greek linguistics? The statistics are typically using a large machine-readable corpus, in order to discover general principles of linguistic behavior, genre difference, etc. The paper also sets out to prove some hypotheses, or identify some linguistic phenomena, such as morphological, syntactic, and semantic phenomena.
The proposed paper will consist of the following points:
- What kinds of linguistic data can they handle?
- What are the advantages and disadvantages of statistical linguistics?
- What is the nature of the assumptions they require of the analyst?
- What is the strategy for studying of linguistic phenomena? |
| |
Download Paper |
|
[9000886.] رقم البحث : 9000886 - |
|
A Proposed E-Learning System for Arabic Sign Language / |
| تخصص البحث : Language Generation |
| |
هندسة اللغة: / عدد (2) - مجلد (2) - سبتمبر 2015 |
| |
تاريخ تقديم البحث 24/10/2015 |
| |
تاريخ قبول البحث 24/10/2015 |
| |
عدد صفحات البحث 10 |
| |
A. S. Elons ( ahmed.new80@hotmail.com - ) |
| |
, M. F. Tolba ( fahmytolba@gmail.com - ) |
| |
Language Knowledge, Arabic Sign Language (ArSL), E-Learning, Hearing Impaired (HI), Sign Language Recognition and Sign Language Synthesis. |
| |
There is no doubt that the Middle East region suffers from many educational problems especially within the field of people with disabilities. Hearing Impaired (HI) people (especially children) have the right to a quality education, with the same content and to the same academic standards as hearing people. HI education is the education of students with various hearing levels in a way that addresses the students´ individual differences and needs. Unfortunately, the courses in different education levels are not available to HI students in a form they can understand and there are also too few qualified teachers that teach in HI Language Schools. The proposed solution is to provide an eLearning solution that leverages the current IT infrastructures and presents enhanced pedagogical aspects of the e-Course for Arabic Sign Language (ArSL). The proposed system modules will be hosted on a public cloud for security and resource management purposes and will be the base for constructing an e-Learning center for hearing impaired individual education. 3 modules have been built; sign to text, text to sign and student scoring. The system is initially built for 50 signs and two students were exposed to the system. Translation accuracy reached 88%. The student can use either a digital camera or leap motion for signing. |
| |
Download Paper |
تسجيل دخول 
