November 2009, No. 5
Print E-mail

Arabic Anaphora Resolution: Corpora
Annotation with Coreferential Links

Souha Hammami, Lamia Belguith, and Abdelmajid Ben Hamadou
 LARIS-MIRACL Laboratory, University of Sfax, Tunisia


Abstract:Annotated resources are much needed for evaluation and training of anaphora resolution systems. The coreferential chain annotation is a difficult task which can not be realised without an appropriate tool. In this paper, we present our work on Arabic corpora annotation with anaphoric links (i.e., the annotation of the identity relation between the anaphors and their antecedents). In particular, we propose an anaphoric annotating tool for Arabic. Anaphoric annotating tool for Arabic has the advantage of automatic detection of Arabic pronouns and allows the human annotator to select several anaphoric pronouns related to the same antecedent. Our aim is to build a real corpus which will be used for anaphora resolution (i.e., either for system training or evaluation).

Keywords: Anaphora resolution, Arabic language, corpus annotation tool, pronominal anaphora, lexical anaphora.

Received December 18, 2008; accepted June 24, 2009

Read more...
 
Print E-mail

A Combination Approach of Gaussian
Mixture Models and Support Vector
Machines for Speaker Identifica

Rafik Djemili1, Hocine Bourouba2, and Amara Korba3
1Electronics Department, University of 20 Août, Algeria
2Electronics Department, University Mentouri of Constantine, Algeria
3LASA, University Badji Mokhtar of Annaba, Algeria

Abstract: Gaussian mixture models are commonly used in speaker identification and verification systems. However, owing to their non discriminant nature, Gaussian mixture models still give greater identification errors in the evaluation process. Partitioning speakers database in clusters based on some proximity criteria where only a single cluster Gaussian mixture models is run in every test, have been suggested in literature generally to speed up the identification process for very large databases. In this paper, we propose a hierarchical clustering scheme using the discriminant power of support vector machines. Speakers are divided into small subsets and evaluation is then processed by GMMs. Experimental results show that the proposed method reduced significantly the error in overall speaker identification tests.   

Keywords:

Speaker identification, GMM, SVM.

Received December 18, 2008; accepted June 29, 2009

Read more...
 
Print E-mail

Using Maximality-Based Labelled Transition System as a Model for Petri Nets

Djamel Eddine Saidouni, Nabil Belala, and Messaouda Bouneb
 Computer Science Department, University of Mentouri, Algeria

Abstract: This work deals with the specification and the verification of concurrent systems. Our goal is to exploit an implementable model, namely the maximality -based labelled transition system, which permits expressing true-concurrency in a natural way without splitting actions on their start and end events. To do this, we give an operational semantics to build maximality-based labelled transition systems for place/transition Petri nets. 

Keywords:

Maximality-based labelled transition systems, maximality bisimulation, Petri nets.

Received December 18, 2008; accepted June 16, 2009

Read more...
 
Print E-mail

Cursor Movement Control Development
by Using ANFIS Algorithm

Suhail Odeh, Joseph Hodali, Maha Sleibi, and Ilyaa Salsa
Faculty of Science, Computer Information Systems Department, Bethlehem University, Palestine


Abstract:Our non-invasive brain computer interface uses EEG signals and beta frequency bands over sensorimotor cortex to control cursor movement horizontally (i.e., one-dimension). The main goal of this study is to help people with sever motor disabilities (i.e., Spinal cord injuries) and provide them a new way of communication and control options by which they can move the cursor in one dimension. In this study, offline analysis of the data collected was used to make the user able of controlling the movement of the cursor horizontally (i.e., one dimension). The data was collected during a session in which the user selected among two targets by thinking and moving either the right hand little finger or the left hand little finger. The Adaptive-Network based fuzzy inference system algorithm was examined for the classification method with some parameters. In the offline analysis, the method used showed a significant performance in the classification accuracy level and it gave an accuracy level of more than 80%.This result suggests that using the adaptive-network based fuzzy inference system algorithm will improve online operation of the current BCI system. 

Keywords: Brain-Computer interface, ANFIS algorithm, fuzzy logic, electroencephalogram.

Received December 18, 2008; accepted June 18, 2009

Read more...
 
Print E-mail

Using Language Independent and Language Specific Features to Enhance Arabic Named
 Entity Recognition

Yassine Benajiba2, Mona Diab2,  and Paolo Rosso1
1Natural Language Engineering Laboratory, ELiRF, Universidad Politécnica Valencia, Spain
2Center of Computational Learning Systems, Columbia University, USA


Abstract:The Named entity recognition task has been garnering significant attention as it has been shown to help improve the performance of many natural language processing applications. More recently, we are starting to see a surge in developing named entity recognition systems for languages other than English. With the relative abundance of resources for the Arabic language and a certain degree of maturation in the state of the art for processing Arabic, it is natural to see interest in developing NER systems for the language.  In this paper, we investigate the impact of using different sets of features that are both language independent and language specific in a discriminative machine learning framework, namely, Support Vector Machines. We explore lexical, contextual and morphological features and nine data-sets of different genres and annotations. We systematically measure the impact of the different features in isolation and combined. We achieve the highest performance using a combination of all features, F1=82.71. Essentially combining language independent features with language specific ones yields the best performance on all the genres of text we investigate. However, on a class level, we observe that the different classes of named entities benefit differently from the morphological features employed.

Keywords: Arabic natural language processing, classification, information extraction, named entity recognition.

Received December 18, 2008; accepted June 21, 2009

Read more...
 
Print E-mail

A Modified Spiral Model Using PSP, TSP, and
Six Sigma Process Model for
Embedded Systems

Adnan Shaout and Tejas Chhaya
The University of Michigan-Dearborn, Michigan, USA

Abstract: The first goal of this paper is to understand, practice, and find out the suitability of the PSP and TSP software processes in real world environment for an embedded system project in automotive industry while working on ‘Moderately complex and medium size’ project and compare it’s statistical results, advantages, and short falls with other software process models. The second goal of this paper is to present a modified software process model using the personal software process SM, team software process SM and six sigma. The new process model was used for an embedded systems project in automotive industry with ‘moderately complex and medium size’. The result of using this new process model has show 70% improvement in defects/KLOC.

Keywords:

Software processes, embedded systems controls, spiral model, automotive, quality.

Received December 18, 2008; accepted June 22, 2009

 
Print E-mail

Performance Optimization in Structured
Wireless Sensor Networks

Amine Moussa and Hoda Maalouf
Computer Science Department, Notre Dame University, Lebanon


Abstract: This paper presents a performance-optimization technique in a structured wireless sensor network that has a circular form and consists of several rings with many clusters. Optimization is done by varying the number of rings and by applying aggregation on the data transferred across the network up to the destination node. The main reason for this study is that the energy-limited sensor nodes may need to operate in time-constrained environments, in which both energy and time must be conserved. In this paper, we analyze the impact of ring density and data aggregation on energy consumption and transfer time. We also find an optimum structure consisting of a certain number of rings, whereby the sensor network operates with low consumption of energy and little transfer time.

Keywords:

Wireless sensor network, performance optimization, circular network model, data aggregation, energy consumption

Received December 18, 2008; accepted June 21, 2009

Read more...
 
Print E-mail

Multimedia Courses Generator for Deaf Children

                                                          Oussama El Ghoul and Mohamed Jemni
                     Research Laboratory of IT, Ecole Supérieure des Sciences et Techniques de Tunis, Tunisie

 

Abstract:In this paper, we present an environment developed in the research laboratory of technologies of information and communication [13] of the university of Tunis to aid deaf people in improving their social integration and communication capabilities. In fact, it is proved that sign language is not innate at deaf children and therefore it needs methodic and specific training. In this context, our environment is a specialized Learning Content Management System that generates multimedia courses to teach and learn sign language. According to our survey we did not find an Learning Content Management System witch can help teachers to generate courses to person with hearing disability. Moreover our tool is original in case that the sign animation is generated automatically from a textual description. The generated courses can be used either by deaf pupils to learn (or e-learn) sign language or also by hearing people to be able to communicate with deaf people. This educational environment uses mainly a web-based interpreter of sign language developed in our research laboratory and called websign [2, 6]. It is a tool that permits to interpret automatically written texts in visual-gestured-spatial language using avatar technology. 

Keywords: Hearing impaired, sign language, LCMS, e-Learning.

Received December 16, 2008; accepted June 21, 2009

Read more...
 
Print E-mail

2009 Index
The International Arab Journal for Information Technology Vol. 6

Full Text

 
Print E-mail

Reviewers List*

Full Text

 
<
 
Copyright 2006-2009 Zarqa Private University. All rights reserved.
Print ISSN: 1683-3198.
 
 
Warning: fsockopen(): php_network_getaddresses: getaddrinfo failed: Name or service not known in /hsphere/local/home/ccis2k/ccis2k.org/iajit/templates/rt_chromatophore/index.php on line 251 Warning: fsockopen(): unable to connect to oucha.net:80 (php_network_getaddresses: getaddrinfo failed: Name or service not known) in /hsphere/local/home/ccis2k/ccis2k.org/iajit/templates/rt_chromatophore/index.php on line 251 skterr