|
Home>Products>Journals
and Reviews>I.Re.Co.S.>Latest
issue
International Review
on Computers and Software - Papers
 Arabic
Part Of Speech Disambiguation: A Survey
by Mohammed
Albared, Nazlia Omar, Mohd. Juzaiddin Ab Aziz
Vol. 4. n. 5, pp. 517-532
Abstract -
After several decades of heavy
research activities in part of speech tagging of rich-resources
languages, various supervised and unsupervised machine learning
approaches have been proposed to solve this problem. In contrast, a few
researches have been investigated in Arabic part of speech tagging with
modest effort. Thus, the aim of this paper is to summarize and organize
the information available in the literature to motivate researchers to
look into these techniques, adopt them for other languages and to
develop more advanced ones. The paper also surveys Arabic part of speech
tagging and discuss their limitation. In addition, open areas and future
work directions are discussed.
Copyright © 2009 Praise
Worthy Prize S.r.l. - All rights reserved
Keywords:
Part of Speech Tagging, Supervised and Unsupervised Machine
Learning.
Modeling
Segmental Duration by Statistical Learning for an Arabic Text-to-Speech
System
by Zied Mnasri,
Fatouma Boukadida, Noureddine Ellouze
Vol. 4. n. 5, pp. 533-542
Abstract - Duration
modeling is a key task to generate prosody in text-to-speech systems. It
can be performed by rules or statistically. Statistical learning is
parametric or non–parametric and can be achieved using global or local
models. The research topic of this paper shows the results of the
application of global non-parametric learning algorithms, neural
networks and SVM, and local ones, CART and MARS, to model the Arabic
speech segmental duration. Twofold evaluation was carried out, first by
measuring the mean absolute error, the correlation coefficient and the
standard deviation, and secondly by weighting the contribution of every
factor or class of factors in each model. The results confirm the
statisticians’ preference for local models in non-parametric learning
problems.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Arabic text-to-speech system, Segmental duration, Statistical
learning, Non-parametric regression, Global and local models, Neural
Networks, SVM, CART, MARS.
A
Suitable Model of Evolutionary SOM for Phonemes Recognition
by Mohamed
Salah Salhi, Najet Arous, Noureddine Ellouze
Vol. 4. n. 5, pp. 543-550
Abstract - The
neural networks, particularly Kohonen maps known by SOM, are considered
as nonlinear parametric models for universal recognition which is highly
qualified in the analysis of great volumes data flow. But what is
undeniable that these models offer a limited recognition rates to the
local optima. In another case, the application of genetic algorithm GA,
that’s in fact inspired from natural selection and genetics, has a
premature convergence, but it may sometimes have inexact results. These
problems can be solved through a promising idea which consists in
introducing the approaches, making the hybridization of the two models
SOM and GA to get profit from their common advantages.
This paper, suggests a general representation of various approaches that
are the target of this hybridization. It’s highly important to bear in
mind that we are introducing a method which guaranties population
diversity over generations. By intervening on the possible SOM
parameters, the genetic algorithms brought to this hybridization the
property of total research. The obtained architectures make it possible
to simulate the human intelligence in a human way.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Kohonen Map SOM, Genetic Algorithm GA, hybrid Models GA-SOM.
Using
Finite State Machines in Processing Continuous Queries
by Ali
A. Safaei, Mohammad Ghalambor Dezfuli, Mostafa S. Haghjoo, Fatemeh Abdi
Vol. 4. n. 5, pp. 551-556
Abstract - Data
stream is a continuous, rapid, time-varying sequence of data elements
which should be processed in an on-line manner. But the well-known data
stream processing models don't consider on-line query processing. This
is a major deficit in data stream management systems.
In this paper, on-line processing of continuous queries over data
streams is presented. Architecture of the proposed query processing
model, structure of the finite state automata, and relevant algorithms
for on-line data stream query processing are presented. It is tested
using an on-line auction environment and its performance in term of
system's response time, tuple latency, number of operations and required
memory space is evaluated and compared with the existing system via
simulation.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Query Processing, Data Stream, Finite State Automata, On-line
Processing.
by Ziad A. Alqadi,
Akram A. Moustafa, Majed Alduari, Rushdi abu Zneit
Vol. 4. n. 5, pp. 557-562
Abstract - Noise
elimination from the color and gray-scale images some time causes
loosing some information from the original image thus it is important to
find a method that minimized an information-loss during image
reconstruction. This paper proposed a methodology of noise elimination
for color images using new color model and new morphological filter that
allows providing the reconstruction of an image in order to accept
result without loosing any piece of information. The proposed
methodology was implemented, tested and experimental results show that
the method is optimal for color image reconstruction in terms of quality
(to compare with well known methods such as Gaussian, Sobel, Prewitt,
Median, laplacian, log, average, unsharp methods which are used as a
special filters to eliminate the noise from the gray image) and in terms
of speed.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Morphological operations, Image reconstruction, Image denoising,
Direct and inverse transformations, Filter, Pixels.
Analytical
and Comparative Study of Data Mining Algorithms Using TANAGRA Tool For Farm DSS
by K.S Dhindsa,
Parneet Kaur,Sukhwinder Singh
Vol. 4. n. 5, pp. 563-566
Abstract -
Sustainable agriculture decision-making is a day today practice. That is
why Decision Support Systems are widespread and cover most of the crop
production activities concerning plant growth and development and their
relationship to the environment (agricultural, economic, and social).In
this paper we study, analyze and compare different types of data mining
algorithms using TANAGRA tool. The clustering algorithm used for
analysis in this paper are K-Mean, EM ,SOM, HAC, VARCHA and VARCLUS. The
factors used to compare in this paper are accuracy ,average accuracy and
probability distribution of finding members in different clusters. By
selecting the best technique and implementing in the TANAGRA tool to
obtain the graphical result in the Dendogram form which will help the
researcher to classify the data table very effectively. This paper will
suggest us suitable algorithm for designing of DSS used in agriculture.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Algorithm, Crop and DSS.
On
Global Clustering Algorithm: Layer-Oriented Approach for Multi Hop Wireless
Sensor Networks
by F. Tashtarian,
A. T. Haghighat, M. H. Yaghmaee, S. M. Mazinani, M. T. Honary
Vol. 4. n. 5, pp. 567-576
Abstract -
Clustering methods are potentially the framework for power-conserving
wireless sensor networks. These methods are among the most effective
approaches aiming for prolonging the lifetime. Power-limited nature of
the employed devices in wireless networks made the conventional
clustering approaches to mainly focus on selecting cluster heads -which
are the central points within a cluster- among the nodes with more
remaining energy. Periodic rotation of this task -being a cluster head-
between nodes leads to distribute the energy consumption. In multihop
wireless sensor networks, a Hot-Spot problem arises when the closest
cluster heads to the base station tend to be excessively utilized by
standing the relaying traffic loads of other nodes and consequently they
drain their energy rapidly. The main purpose of this paper is to provide
a layering-based clustering algorithm that aim at directly optimizing
the network's longevity by considering the relationship between the
width of each layer and its overall lifetime. Analytical methods are
employed to study the impacts of both intra-cluster and inter-cluster
traffic loads on the average layer’s energy consumption that is
formulated as an optimization problem.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Wireless Sensor Network, Layering, Clustering, Lifetime, Multihop.
Essential
Characteristics of Software Model that Provide the Software Quality
Assurance
by Yas
A. Alsultanny, Ahmed M. Wohaishi
Vol. 4. n. 5, pp. 577-585
Abstract - Software
quality model is a process that evaluates, asses, and improves the
quality of software. Software quality is the degree to which software
meets requirements for reliability, maintainability and
transportability. In this paper a proposed model to provide software
quality assurance was introduced to satisfy quality requirements, where
one of the most important problems arise is the software complexity.
Which is an inherent complexity for each project, and further complexity
is added during the design and code phase of the software development
process. The proposed model shows how to provide safe, reliable and
quality engineering software product is to develop, by taking in
consideration all the factors affecting software quality to reach the
ISO 9126.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Software quality, Software complexity, ISO.
Learner
Modelling Authoring System for Interactive Learning Environments
by Douadi
Bourouaieh, Tahar Bensebaa
Vol. 4. n. 5, pp. 586-593
Abstract - This
paper presents Learner Modelling Tool Services (LMoTS), a set of tools
and process for authoring learner models. LMoTS presents novel aspects
in learner modelling so that it allows authoring learner models in
incremental, continuous and contextual way. There are three main
purposes of this paper. First, we will briefly present learner modelling
issues. Second, we describe a novel system and tools designed to aid in
learner modelling by authoring activities. Our aim in this system,
embodied in a set of tools and engine called LMoTS, is to ease the
learner modelling task, often referred as difficult and useful in
interactive learning environments. Third, we present an exploratory test
of the system’s efficacy. Specifically, we consider the different
services and features that such a system has to offer, as well as
mechanisms for its implementation. Design activities are allowed during
the design and exploitation phase of the learning environment.
Essentially, we highlight the relegation of the hypothetical aspect of
the learner model to second order consideration, proposing the concept
of responsibility on the designed models, designed models quality and a
new role namely learner model designer.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Learner Modelling, Authoring Systems.
A
Model for Exploring Individual’s Self-Disclosure Online
by Sheng-Fei Hsu,
Dong-Her Shih
Vol. 4. n. 5, pp. 594-599
Abstract -
Individual’s behavior on self-disclosure online was affected by not only
the personal psychological factors, but also individual’s computer
knowledge and techniques. However, the latter always ignored in previous
researches. For investigating how the technological and psychological
factors simultaneously affected one’s privacy concern and restrain the
behavior on disclosure, this study applied computer self-efficacy to
modifying previous privacy researches. In the meantime, research
proposed a hypothesized model to describe the relationships between
related constructs. The result of this study showed that, except
individual computer self-efficacy had negative impact on self-disclosure
online, all the other hypotheses indicated positive casual effect.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Privacy concern, Perceived privacy, Trust, Computer self-efficacy,
Self-disclosure online.
Evaluating
SVM and BPNN Classifiers for Remote Sensing Data
by O. Marrakchi,
A. Agina, A. Elghali
Vol. 4. n. 5, pp. 600-605
Abstract - In this
paper, Support Vector Machines (SVM) and Back Propagation Neuronal
Networks (BPNN) performances were assessed in order to identify
heterogeneous and highly correlated textures of Remote Sensing (RS)
images: Terra–Aster and airborne- SAR data. The use of two
classification methods based on non–linear separation principle, as SVM
and BPNN methods, made the identification of this kind of textures
possible. Textural analysis was carried out through local measures
extracted by observations within sampling windows. In the learning stage
of the SVM and BPNN nets, the convergence of the algorithms depends
essentially on the number of samples presented to the networks and on
the complexity of the textural information related to the heterogeneity
of the texture. Classification results show that SVM classifiers are
more performing in terms of learning time compared to the BPNN
classifiers but identification rates values are comparable. Indeed, for
most RS images textures and for both classification methods,
classification rates are near and equal to 100%.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
BPNN, SVM, Classification, Aster-Terra, SAR, Correlated textures.
Partial
Runtime Reconfiguration of FPGA, Applications and a Fault Emulation Case
Study
by Uroš Legat,
Anton Biasizzo, Franc Novak
Vol. 4. n. 5, pp. 606-611
Abstract - The paper
surveys the basic principles of partial runtime reconfiguration of FPGA
and comments on their possible applications in practice. The dynamic
partial reconfiguration of FPGA is a process of reconfiguring a part of
FPGA logic, while the rest of the logic is unaffected by the
configuration. Three different successful examples of reconfigurable
applications are presented. The paper also presents a fault emulation
case study. By using partial runtime reconfiguration the fault injection
time is significantly reduced. Fault is injected during run-time using
an embedded microprocessor and only the resource affected by the fault
is reconfigured. The approach is demonstrated using a hardware
implementation of Advanced Encryption Standard (AES) algorithm. This
work has been done as a preliminary feasibility study for concurrent
test solutions in mission critical applications.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Runtime reconfiguration, Partial reconfiguration, FPGA, Fault
injection.
Model-Driven
Transformations: From Analysis to MVC 2 Web Model
by S. Mbarki, M.
Erramdani
Vol. 4. n. 5, pp. 612-620
Abstract - Web
applications have the complexity of designing, developing, maintaining,
and managing. These systems have increased significantly as well. To
cope with this complexity, several frameworks (supporting the MVC 2
pattern) have been elaborated. Facing this diversity and incessant
improvement of Web technology, we are in need of developing a tool which
is able to produce the code from analysis-level UML models. In this
paper, we apply MDA approach for generating code from requirements to
MVC 2 Web implementation. That is why we have developed two metamodels
handling UML class diagrams and MVC 2 Web applications, then we have to
set up two transformations. The first one is a PIM to PIM transformation
allowing the refinement of the analysis-level class diagram. The second
one is a PIM to PSM transformation for MVC 2 Web applications.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Model Driven Engineering, Meta-models, rules transformation, Web
Applications development, Software Engineering.
About
Applets Application in Natural Science in 1st Level of Basic School
by Milan Bernát,
Renáta Bernátová, Danica Fazekašová, Roman Cimbala
Vol. 4. n. 5, pp. 621-626
Abstract - The paper
reports the aspects related to Java applets creation and their
application in teaching natural subjects. The author of the paper also
presents the Java applets he himself created and applied in the natural
teaching process. At the same time he emphasises the irreplaceability of
didactic and professional mastership of a teacher in the teaching
process using Java applet programmes.
Copyright © 2009 Praise Worthy Prize S.r.l. -
All rights reserved
Keywords:
Java Applets, Natural Science, Basic School.
| 
