Comments, Questions, Discussion:
WSEAS TRANSACTIONS on COMPUTERS. Issue 1, Volume 7, January 2008
Title of the Paper: Constraint Satisfaction Problem
Using Modified Branch and Bound Algorithm
Authors: Azlinah Mohamed, Marina
Yusoff, Itaza Afiani Mohtar, Sofianita Mutalib, Shuzlina Abdul Rahman
Abstract: A constraint satisfaction problem (CSP) involves assigning possible
values to a set of variables without defying any constraints. There are various
techniques available to solve or give partial solution to CSP. This paper
presents a modification of branch and bound algorithm, which is used to solve a
constraint satisfaction problem in map colouring problem. There are two
constraints involved which are only three colours are allowed to be used and
adjacent regions in the map must not be of the same colour. The modified branch
and bound algorithm uses back jumping when it encounters a dead-end in the
search. Static variable ordering was also applied to aid the searching process.
The modified branch and bound algorithm shows a better result in terms of the
number of nodes instantiated and the reduced number of backtracking at dead
ends. The result illustrated that the modified branch and bound algorithm with
the use of variable ordering technique is better if compared to backjumping.
Thus, it is concluded that the modified branch and bound algorithm would improve
constraint satisfaction problem.
Keywords: Backjumping, Branch and Bound Algorithm, Constraint Satisfaction
Problem, and Static Variable Ordering
Title of the Paper:
An Application of Type-2 Fuzzy Notions in Website
Structures Selection: Utilizing Extended TOPSIS Method
Authors: Hamed
Qahri Saremi, Gholam Ali Montazer
Abstract: Giving more effectiveness to e-commerce sites increases customer
satisfaction as visitors can navigate the website easier and find their targets
in less time and cost. With the development of web content, the structure of a
website becomes more complex and critical to both web designers and users in a
way that has made the prioritizing of various options of website structure a
pivotal decision making problem incorporating large uncertainty in judgment.
Finding a response to such a need, TOPSIS, a conventional MADM technique, had
been a mere remedy for years, solving the problems with a more or less adequate
accuracy. In an aspiring step toward an improvement in this method, Fuzzy TOPSIS
method, a combination of ordinary TOPSIS method and Fuzzy theory, could heal
some of shortcomings of uncertainties and ordinary TOPSIS in decision making.
However there are still lots of occasions in which decision making is faced with
lots of shadowiness making the Fuzzy TOPSIS method not sufficiently receptive.
As a sensible response to this drawback, in this paper, we utilize a brand-new
extension to TOPSIS and Fuzzy TOPSIS methods, based on type-2 fuzzy notions with
ability to cope with type-2 fuzzy environment and data incorporating much more
fuzziness in decision making. We apply this method to a vague case of the real
world and discuss its results against the other previously-developed TOPSIS
methods.
Keywords: Decision Making; Website
Structure; TOPSIS; Type-2 Fuzzy Sets; Interval Valued Fuzzy Sets; MADM; Fuzzy
TOPSIS; IVF-TOPSIS
Title of the Paper:
A Novel Robust watermarking
technique Using IntDCT Based AC Prediction
Authors:
Kuo-Ming Hung
Abstract: Because of the blocking artifacts resulted from 8x8 Discrete Cosine
Transform (DCT) , most watermarking technologies have been proposed using DCT
whose image quality was not very good. In 1990, Gonzales et al. described a
technique which predicts a few low frequency AC coefficients. The AC predictor
uses the dequantized DC values of a 3x3 neighborhood of 8x8 blocks to predict
the AC values in the center block. Wang proposed a data hiding scheme using the
AC prediction technology in 2005. But it is unable to be suitable for the image
of all types to predict AC coefficients accurately. We propose a new
watermarking system by using the technologies of 4x4 integer DCT transform and
adaptive AC estimation. We use 4x4 INTDCT transform to reduce blocking artifacts
caused from 8x8 DCT transform and improve Imperceptibility and watermark
capacity greatly. Moreover, we utilize AC prediction value as error-checking
code to enhance robustness of watermark.
Keywords: Watermark; DCT; IntDCT; AC
prediction; H.264
Title of the Paper:
Implementation Feasibility of Convex Recursive
Deletion Regions Using Multi-Layer Perceptrons
Authors: Che-Chern Lin
Abstract: A constructive algorithm to implement convex recursive deletion
regions via two-layer perceptrons has been presented in a recent study. In the
algorithm, the absolute values of the weights become larger and larger when the
number of nested layers of a convex recursive deletion region increases. In
addition, the absolute values of the weights are determined according to the
complexity of the structure of the convex recursive deletion region. More
complicated convex recursive deletion regions result in larger values of
weights. Besides, the constructive procedure is needed to get the parameters
(weights and thresholds) for the neural networks. In this paper, we propose a
simple three-layer network structure to implement the convex recursive deletion
regions in which all weights of the second and third layers are all 1�s and the
thresholds for the nodes in the second layer are pre-determined according to the
structures of the convex recursive deletion regions. This paper also provides
the activation function for the output node. In brief, all of parameters
(weights and activation functions) in the proposed structure are pre-determined
and no constructive algorithm is needed for solving the convex recursive
deletion region problems. We prove the feasibility of the proposed structure and
give an illustrative example to demonstrate how the proposed structure
implements the convex recursive deletion regions. Finally, we provide the
conceptual diagram of the hardware implementation of the proposed network
structure.
Keywords: Multi-layer perceptrons, nested
decision region, convex recursive deletion region, hardware implementation.
Title of the Paper:
Multi-layer perceptrons, nested decision region, convex recursive deletion
region, hardware implementation.
Authors: Yanjing Sun, Xiangping Gu,
Jiansheng Qian
Abstract: Virtual backbone has been used extensively in various aspects for
wireless ad hoc or sensor networks recently. We propose an approximation
solution to construct a virtual backbone based on a more generalized and
realistic model of polynomial bounded growth. A localized distributed algorithm
of MCDS_GBG for computing a Minimum Connected Dominating Sets as backbone in the
growth-bounded graph is presented. This approach consists of three stages:
firstly construct an MIS by network decomposition scheme; secondly compute a
minimum dominating set in 2-separated collection with r transmission range and
finally use Marking process and ruling k to reduce the virtual backbone with 3r
transmission range. The computed Connected Dominating Set guarantees a constant
stretch factor on the length of a shortest path and induces a subgraph of
constant degree while the nodes only require direct neighborhood information.
The efficiency of our approach is confirmed through both theoretical analysis
and comparison study.
Keywords: Virtual backbone; Growth-bounded
graph; Connected dominating sets; Maximal independent sets; Wireless ad hoc
sensor network; Network decomposition
1 comment:
The paper "Improving Customer Satisfaction by the Expert System Using Artificial Neural Networks" is very important
Congratulation to the authors
Post a Comment