Abstract: In this paper, an approach to reduce the computation steps required by fast neural networksfor the searching process is presented. The principle ofdivide and conquer strategy is applied through imagedecomposition. Each image is divided into small in sizesub-images and then each one is tested separately usinga fast neural network. The operation of fast neuralnetworks based on applying cross correlation in thefrequency domain between the input image and theweights of the hidden neurons. Compared toconventional and fast neural networks, experimentalresults show that a speed up ratio is achieved whenapplying this technique to locate human facesautomatically in cluttered scenes. Furthermore, fasterface detection is obtained by using parallel processingtechniques to test the resulting sub-images at the sametime using the same number of fast neural networks. Incontrast to using only fast neural networks, the speed upratio is increased with the size of the input image whenusing fast neural networks and image decomposition.
Abstract: This paper applies the homotopy analysis method method to a nonlinear differential-difference equation arising in nanotechnology. Continuum hypothesis on nanoscales is invalid, and a differential-difference model is considered as an alternative approach to describing discontinued problems. Comparison of the approximate solution with the exact one reveals that the method is very effective.
Abstract: This paper presents a possibilistic (fuzzy) model in optimal siting and sizing of Distributed Generation (DG) for loss reduction and improve voltage profile in power distribution system. Multi-objective problem is developed in two phases. In the first one, the set of non-dominated planning solutions is obtained (with respect to the objective functions of fuzzy economic cost, and exposure) using genetic algorithm. In the second phase, one solution of the set of non-dominated solutions is selected as optimal solution, using a suitable max-min approach. This method can be determined operation-mode (PV or PQ) of DG. Because of considering load uncertainty in this paper, it can be obtained realistic results. The whole process of this method has been implemented in the MATLAB7 environment with technical and economic consideration for loss reduction and voltage profile improvement. Through numerical example the validity of the proposed method is verified.
Abstract: With the tremendous growth of World Wide Web
(WWW) data, there is an emerging need for effective information
retrieval at the document level. Several query languages such as
XML-QL, XPath, XQL, Quilt and XQuery are proposed in recent
years to provide faster way of querying XML data, but they still lack of
generality and efficiency. Our approach towards evolving a framework
for querying semistructured documents is based on formal query
algebra. Two elements are introduced in the proposed framework:
first, a generic and flexible data model for logical representation of
semistructured data and second, a set of operators for the manipulation
of objects defined in the data model. In additional to accommodating
several peculiarities of semistructured data, our model offers novel
features such as bidirectional paths for navigational querying and
partitions for data transformation that are not available in other
proposals.
Abstract: A lot of matching algorithms with different characteristics have been introduced in recent years. For real time systems these algorithms are usually based on minutiae features. In this paper we introduce a novel approach for feature extraction in which the extracted features are independent of shift and rotation of the fingerprint and at the meantime the matching operation is performed much more easily and with higher speed and accuracy. In this new approach first for any fingerprint a reference point and a reference orientation is determined and then based on this information features are converted into polar coordinates. Due to high speed and accuracy of this approach and small volume of extracted features and easily execution of matching operation this approach is the most appropriate for real time applications.
Abstract: The paper deals with results of a project “Interoperability Workplaces to Support Teaching of Security Management in a Computer Network". This project is focused on the perspectives and possibilities of "new approaches" to education, training and crisis communication of rescue teams in the Czech Republic. It means that common technologies considering new perspectives are used to educate selected members of crisis management. The main part concentrates on possibilities of application of new technology and computer-aided tools to education and training of Integrated Rescue System teams.This project uses the COST principle for the creation of specialized centers and for all communication between these workplaces.
Abstract: This paper evaluates the performance of a novel
algorithm for tracking of a mobile node, interms of execution time
and root mean square error (RMSE). Particle Filter algorithm is used
to track the mobile node, however a new technique in particle filter
algorithm is also proposed to reduce the execution time. The
stationary points were calculated through trilateration and finally by
averaging the number of points collected for a specific time, whereas
tracking is done through trilateration as well as particle filter
algorithm. Wi-Fi signal is used to get initial guess of the position of
mobile node in x-y coordinates system. Commercially available
software “Wireless Mon" was used to read the WiFi signal strength
from the WiFi card. Visual Cµ version 6 was used to interact with
this software to read only the required data from the log-file
generated by “Wireless Mon" software. Results are evaluated through
mathematical modeling and MATLAB simulation.
Abstract: We have applied new accelerated algorithm for linear
discriminate analysis (LDA) in face recognition with support vector
machine. The new algorithm has the advantage of optimal selection
of the step size. The gradient descent method and new algorithm has
been implemented in software and evaluated on the Yale face
database B. The eigenfaces of these approaches have been used to
training a KNN. Recognition rate with new algorithm is compared
with gradient.
Abstract: This paper proposes rough set models with three
different level knowledge granules in incomplete information system
under tolerance relation by similarity between objects according to
their attribute values. Through introducing dominance relation on the
discourse to decompose similarity classes into three subclasses: little
better subclass, little worse subclass and vague subclass, it dismantles
lower and upper approximations into three components. By using
these components, retrieving information to find naturally hierarchical
expansions to queries and constructing answers to elaborative queries
can be effective. It illustrates the approach in applying rough set
models in the design of information retrieval system to access different
granular expanded documents. The proposed method enhances rough
set model application in the flexibility of expansions and elaborative
queries in information retrieval.
Abstract: Process improvements have drawn much attention in
practical software engineering. The capability maturity levels from
CMMI have become an important index to assess a software company-s
software engineering capability. However, in countries like
Taiwan, customers often have no choices but to deal with vendors that
are not CMMI prepared or qualified. We call these vendors maturitylevel-
1 (ML1) vendors. In this paper, we describe our experience
from consulting an e-school project. We propose an approach to help
our client tackle the ML1 vendors. Through our system analysis, we
produce a design. This design is suggested to be used as part of
contract and a blueprint to guide the implementation.
Abstract: An approach is offered for more precise definition of base lines- borders in handwritten cursive text and general problems of handwritten text segmentation have also been analyzed. An offered method tries to solve problems arose in handwritten recognition with specific slant or in other words, where the letters of the words are not on the same vertical line. As an informative features, some recognition systems use ascending and descending parts of the letters, found after the word-s baseline detection. In such recognition systems, problems in baseline detection, impacts the quality of the recognition and decreases the rate of the recognition. Despite other methods, here borders are found by small pieces containing segmentation elements and defined as a set of linear functions. In this method, separate borders for top and bottom border lines are found. At the end of the paper, as a result, azerbaijani cursive handwritten texts written in Latin alphabet by different authors has been analyzed.
Abstract: In this paper, we propose an approach for the classification of fingerprint databases. It is based on the fact that a fingerprint image is composed of regular texture regions that can be successfully represented by co-occurrence matrices. So, we first extract the features based on certain characteristics of the cooccurrence matrix and then we use these features to train a neural network for classifying fingerprints into four common classes. The obtained results compared with the existing approaches demonstrate the superior performance of our proposed approach.
Abstract: The network of delivering commodities has been an important design problem in our daily lives and many transportation applications. The delivery performance is evaluated based on the system reliability of delivering commodities from a source node to a sink node in the network. The system reliability is thus maximized to find the optimal routing. However, the design problem is not simple because (1) each path segment has randomly distributed attributes; (2) there are multiple commodities that consume various path capacities; (3) the optimal routing must successfully complete the delivery process within the allowable time constraints. In this paper, we want to focus on the design optimization of the Multi-State Flow Network (MSFN) for multiple commodities. We propose an efficient approach to evaluate the system reliability in the MSFN with respect to randomly distributed path attributes and find the optimal routing subject to the allowable time constraints. The delivery rates, also known as delivery currents, of the path segments are evaluated and the minimal-current arcs are eliminated to reduce the complexity of the MSFN. Accordingly, the correct optimal routing is found and the worst-case reliability is evaluated. It has been shown that the reliability of the optimal routing is at least higher than worst-case measure. Two benchmark examples are utilized to demonstrate the proposed method. The comparisons between the original and the reduced networks show that the proposed method is very efficient.
Abstract: Many studies have applied the Theory of Planned
Behavior (TPB) in predicting health behaviors among unique
populations. However, a new paradigm is emerging where focus is
now directed to modification and expansion of the TPB model rather
than utilization of the traditional theory. This review proposes new
models modified from the Theory of Planned Behavior and suggest
an appropriate study design that can be used to test the models within
physical activity and dietary practice domains among Type 2
diabetics in Kenya. The review was conducted by means of literature
search in the field of nutrition behavior, health psychology and
mixed methods using predetermined key words. The results identify
pre-intention and post intention gaps within the TPB model that need
to be filled. Additional psychosocial factors are proposed to be
included in the TPB model to generate new models and the efficacy
of these models tested using mixed methods design.
Abstract: Locality Sensitive Hashing (LSH) is one of the most
promising techniques for solving nearest neighbour search problem in
high dimensional space. Euclidean LSH is the most popular variation
of LSH that has been successfully applied in many multimedia
applications. However, the Euclidean LSH presents limitations that
affect structure and query performances. The main limitation of the
Euclidean LSH is the large memory consumption. In order to achieve
a good accuracy, a large number of hash tables is required. In this
paper, we propose a new hashing algorithm to overcome the storage
space problem and improve query time, while keeping a good
accuracy as similar to that achieved by the original Euclidean LSH.
The Experimental results on a real large-scale dataset show that the
proposed approach achieves good performances and consumes less
memory than the Euclidean LSH.
Abstract: Combining classifiers is a useful method for solving
complex problems in machine learning. The ECOC (Error Correcting
Output Codes) method has been widely used for designing combining
classifiers with an emphasis on the diversity of classifiers. In this
paper, in contrast to the standard ECOC approach in which individual
classifiers are chosen homogeneously, classifiers are selected
according to the complexity of the corresponding binary problem. We
use SATIMAGE database (containing 6 classes) for our experiments.
The recognition error rate in our proposed method is %10.37 which
indicates a considerable improvement in comparison with the
conventional ECOC and stack generalization methods.
Abstract: Home is important for Chinese people. Because the
information regarding the house attributes and surrounding
environments is incomplete in most real estate agency, most house
buyers are difficult to consider the overall factors effectively and only
can search candidates by sorting-based approach. This study aims to
develop a decision support system for housing purchasing, in which
surrounding facilities of each house are quantified. Then, all
considered house factors and customer preferences are incorporated
into Simple Multi-Attribute Ranking Technique (SMART) to support
the housing evaluation. To evaluate the validity of proposed approach,
an empirical study was conducted from a real estate agency. Based on
the customer requirement and preferences, the proposed approach can
identify better candidate house with consider the overall house
attributes and surrounding facilities.
Abstract: Robotic system is an important area in artificial intelligence that aims at developing the performance techniques of the robot and making it more efficient and more effective in choosing its correct behavior. In this paper the distributed learning classifier system is used for designing a simulated control system for robot to perform complex behaviors. A set of enhanced approaches that support default hierarchies formation is suggested and compared with each other in order to make the simulated robot more effective in mapping the input to the correct output behavior.
Abstract: After the terrorist attack on September 11, 2001 in
U.S., the container security issue got high attention, especially by U.S.
government, which deployed a lot of measures to promote or improve
security systems. U.S. government not only enhances its national
security system, but allies with other countries against the potential
terrorist attacks in the future. For example CSI (Container Security
Initiative), it encourages foreign ports outside U.S. to become CSI
ports as a part of U.S. anti-terrorism network. Although promotion of
the security could partly reach the goal of anti-terrorism, that will
influence the efficiency of container supply chain, which is the main
concern when implementing the inspection measurements. This paper
proposes a quick estimation methodology for an inspection service
rate by a berth allocation heuristic such that the inspection activities
will not affect the original container supply chain. Theoretical and
simulation results show this approach is effective.
Abstract: Word sense disambiguation is one of the most important open problems in natural language processing applications such as information retrieval and machine translation. Many approach strategies can be employed to resolve word ambiguity with a reasonable degree of accuracy. These strategies are: knowledgebased, corpus-based, and hybrid-based. This paper pays attention to the corpus-based strategy that employs an unsupervised learning method for disambiguation. We report our investigation of Latent Semantic Indexing (LSI), an information retrieval technique and unsupervised learning, to the task of Thai noun and verbal word sense disambiguation. The Latent Semantic Indexing has been shown to be efficient and effective for Information Retrieval. For the purposes of this research, we report experiments on two Thai polysemous words, namely /hua4/ and /kep1/ that are used as a representative of Thai nouns and verbs respectively. The results of these experiments demonstrate the effectiveness and indicate the potential of applying vector-based distributional information measures to semantic disambiguation.