Abstract: The problem of spam has been seriously troubling the Internet community during the last few years and currently reached an alarming scale. Observations made at CERN (European Organization for Nuclear Research located in Geneva, Switzerland) show that spam mails can constitute up to 75% of daily SMTP traffic. A naïve Bayesian classifier based on a Bag Of Words representation of an email is widely used to stop this unwanted flood as it combines good performance with simplicity of the training and classification processes. However, facing the constantly changing patterns of spam, it is necessary to assure online adaptability of the classifier. This work proposes combining such a classifier with another NBC (naïve Bayesian classifier) based on pairs of adjacent words. Only the latter will be retrained with examples of spam reported by users. Tests are performed on considerable sets of mails both from public spam archives and CERN mailboxes. They suggest that this architecture can increase spam recall without affecting the classifier precision as it happens when only the NBC based on single words is retrained.
Abstract: Salient points are frequently used to represent local
properties of the image in content-based image retrieval. In this paper,
we present a reduction algorithm that extracts the local most salient
points such that they not only give a satisfying representation of an
image, but also make the image retrieval process efficiently. This
algorithm recursively reduces the continuous point set by their
corresponding saliency values under a top-down approach. The
resulting salient points are evaluated with an image retrieval system
using Hausdoff distance. In this experiment, it shows that our method
is robust and the extracted salient points provide better retrieval
performance comparing with other point detectors.
Abstract: This paper proposes an Interactive Chinese Character
Learning System (ICCLS) based on pictorial evolution as an
edutainment concept in computer-based learning of language. The
advantage of the language origination itself is taken as a learning
platform due to the complexity in Chinese language as compared to
other types of languages. Users especially children enjoy more by
utilize this learning system because they are able to memories the
Chinese Character easily and understand more of the origin of the
Chinese character under pleasurable learning environment, compares
to traditional approach which children need to rote learning Chinese
Character under un-pleasurable environment. Skeletonization is used
as the representation of Chinese character and object with an animated
pictograph evolution to facilitate the learning of the language. Shortest
skeleton path matching technique is employed for fast and accurate
matching in our implementation. User is required to either write a
word or draw a simple 2D object in the input panel and the matched
word and object will be displayed as well as the pictograph evolution
to instill learning. The target of computer-based learning system is for
pre-school children between 4 to 6 years old to learn Chinese
characters in a flexible and entertaining manner besides utilizing
visual and mind mapping strategy as learning methodology.
Abstract: This paper looks into areas not covered by prominent
Agent-Oriented Software Engineering (AOSE) methodologies.
Extensive paper review led to the identification of two issues, first
most of these methodologies almost neglect semantic web and
ontology. Second, as expected, each one has its strength and
weakness and may focus on some phases of the development
lifecycle but not all of the phases. The work presented here builds
extensions to a highly regarded AOSE methodology (MaSE) in order
to cover the areas that this methodology does not concentrate on. The
extensions include introducing an ontology stage for semantic
representation and integrating early requirement specification from a
methodology which mainly focuses on that. The integration involved
developing transformation rules (with the necessary handling of nonmatching
notions) between the two sets of representations and
building the software which automates the transformation. The
application of this integration on a case study is also presented in the
paper. The main flow of MaSE stages was changed to smoothly
accommodate the new additions.
Abstract: Hybrid algorithm is the hot issue in Computational
Intelligence (CI) study. From in-depth discussion on Simulation
Mechanism Based (SMB) classification method and composite patterns,
this paper presents the Mamdani model based Adaptive Neural
Fuzzy Inference System (M-ANFIS) and weight updating formula in
consideration with qualitative representation of inference consequent
parts in fuzzy neural networks. M-ANFIS model adopts Mamdani
fuzzy inference system which has advantages in consequent part.
Experiment results of applying M-ANFIS to evaluate traffic Level
of service show that M-ANFIS, as a new hybrid algorithm in computational
intelligence, has great advantages in non-linear modeling,
membership functions in consequent parts, scale of training data and
amount of adjusted parameters.
Abstract: In this work we evaluate the possibility of predicting
the emotional state of a person based on the EEG. We investigate
the problem of classifying valence from EEG signals during
the presentation of affective pictures, utilizing the "frontal EEG
asymmetry" phenomenon. To distinguish positive and negative
emotions, we applied the Common Spatial Patterns algorithm.
In contrast to our expectations, the affective pictures did not
reliably elicit changes in frontal asymmetry. The classifying task
thereby becomes very hard as reflected by the poor classifier
performance. We suspect that the masking of the source of the
brain activity related to emotions, coming mostly from deeper
structures in the brain, and the insufficient emotional engagement
are among main reasons why it is difficult to predict the emotional
state of a person.
Abstract: The internet has become an attractive avenue for
global e-business, e-learning, knowledge sharing, etc. Due to
continuous increase in the volume of web content, it is not practically
possible for a user to extract information by browsing and integrating
data from a huge amount of web sources retrieved by the existing
search engines. The semantic web technology enables advancement
in information extraction by providing a suite of tools to integrate
data from different sources. To take full advantage of semantic web,
it is necessary to annotate existing web pages into semantic web
pages. This research develops a tool, named OWIE (Ontology-based
Web Information Extraction), for semantic web annotation using
domain specific ontologies. The tool automatically extracts
information from html pages with the help of pre-defined ontologies
and gives them semantic representation. Two case studies have been
conducted to analyze the accuracy of OWIE.
Abstract: The contribution deals with analysis of identity style
at adolescents (N=463) at the age from 16 to 19 (the average age is
17,7 years). We used the Identity Style Inventory by Berzonsky,
distinguishing three basic, measured identity styles: informational,
normative, diffuse-avoidant identity style and also commitment. The
informational identity style influencing on personal adaptability,
coping strategies, quality of life and the normative identity style, it
means the style in which an individual takes on models of authorities
at self-defining were found to have the highest representation in the
studied group of adolescents by higher scores at girls in comparison
with boys. The normative identity style positively correlates with the
informational identity style. The diffuse-avoidant identity style was
found to be positively associated with maladaptive decisional
strategies, neuroticism and depressive reactions. There is the style,
in which the individual shifts aside defining his personality. In our
research sample the lowest score represents it and negatively
correlates with commitment, it means with coping strategies, thrust in
oneself and the surrounding world. The age of adolescents did not
significantly differentiate representation of identity style. We were
finding the model, in which informational and normative identity
style had positive relationship and the informational and diffuseavoidant
style had negative relationship, which were determinated
with commitment. In the same time the commitment is influenced
with other outside factors.
Abstract: Facial features are frequently used to represent local
properties of a human face image in computer vision applications. In
this paper, we present a fast algorithm that can extract the facial
features online such that they can give a satisfying representation of a
face image. It includes one step for a coarse detection of each facial
feature by AdaBoost and another one to increase the accuracy of the
found points by Active Shape Models (ASM) in the regions of interest.
The resulted facial features are evaluated by matching with artificial
face models in the applications of physiognomy. The distance measure
between the features and those in the fate models from the database is
carried out by means of the Hausdorff distance. In the experiment, the
proposed method shows the efficient performance in facial feature
extractions and online system of physiognomy.
Abstract: CIM is the standard formalism for modeling management
information developed by the Distributed Management Task
Force (DMTF) in the context of its WBEM proposal, designed to
provide a conceptual view of the managed environment. In this
paper, we propose the inclusion of formal knowledge representation
techniques, based on Description Logics (DLs) and the Web Ontology
Language (OWL), in CIM-based conceptual modeling, and then we
examine the benefits of such a decision. The proposal is specified as a
CIM metamodel level mapping to a highly expressive subset of DLs
capable of capturing all the semantics of the models. The paper shows
how the proposed mapping can be used for automatic reasoning
about the management information models, as a design aid, by means
of new-generation CASE tools, thanks to the use of state-of-the-art
automatic reasoning systems that support the proposed logic and use
algorithms that are sound and complete with respect to the semantics.
Such a CASE tool framework has been developed by the authors and
its architecture is also introduced. The proposed formalization is not
only useful at design time, but also at run time through the use of
rational autonomous agents, in response to a need recently recognized
by the DMTF.
Abstract: One of the most ancient humankind concerns is knowledge formalization i.e. what a concept is. Concept Analysis, a branch of analytical philosophy, relies on the purpose of decompose the elements, relations and meanings of a concept. This paper aims at presenting a method to make a concept analysis obtaining a knowledge representation suitable to be processed by a computer system using either object-oriented or ontology technologies. Security notion is, usually, known as a set of different concepts related to “some kind of protection". Our method concludes that a more general framework for the concept, despite it is dynamic, is possible and any particular definition (instantiation) depends on the elements used by its construction instead of the concept itself.
Abstract: Under-representation of women in leadership positions" is still a general phenomenon in Germany despite the high number of implemented measures. The under-representation of female executives in the aviation sector is even worse. In this context our research hypothesis is that the representation and acceptance of women in management positions is determined by corporate culture.
Abstract: In our current political climate of assessment and
accountability initiatives we are failing to prepare our children for a
participatory role in the creative economy. The field of education is
increasingly falling prey to didactic methodologies which train a
nation of competent test takers, foregoing the opportunity to educate
students to find problems and develop multiple solutions. No where is
this more evident than in the area of art education. Due to a myriad of
issues including budgetary shortfalls, time constraints and a general
misconception that anyone who enjoys the arts is capable of teaching
the arts, our students are not developing the skills they require to
become fully literate in critical thinking and creative processing.
Although art integrated curriculum is increasingly being viewed as a
reform strategy for motivating students by offering alternative
presentation of concepts and representation of knowledge acquisition,
misinformed administrators are often excluding the art teacher from
the integration equation. The paper to follow addresses the problem
of the need for divergent thinking and conceptualization in our
schools. Furthermore, this paper explores the role of education, and
specifically, art education in the development of a creatively literate
citizenry.
Abstract: This research proposes an algorithm for the simulation
of time-periodic unsteady problems via the solution unsteady Euler
and Navier-Stokes equations. This algorithm which is called Time
Spectral method uses a Fourier representation in time and hence
solve for the periodic state directly without resolving transients
(which consume most of the resources in a time-accurate scheme).
Mathematical tools used here are discrete Fourier transformations. It
has shown tremendous potential for reducing the computational cost
compared to conventional time-accurate methods, by enforcing
periodicity and using Fourier representation in time, leading to
spectral accuracy. The accuracy and efficiency of this technique is
verified by Euler and Navier-Stokes calculations for pitching airfoils.
Because of flow turbulence nature, Baldwin-Lomax turbulence
model has been used at viscous flow analysis. The results presented
by the Time Spectral method are compared with experimental data. It
has shown tremendous potential for reducing the computational cost
compared to the conventional time-accurate methods, by enforcing
periodicity and using Fourier representation in time, leading to
spectral accuracy, because results verify the small number of time
intervals per pitching cycle required to capture the flow physics.
Abstract: In this paper, we propose a robust scheme to work face alignment and recognition under various influences. For face representation, illumination influence and variable expressions are the important factors, especially the accuracy of facial localization and face recognition. In order to solve those of factors, we propose a robust approach to overcome these problems. This approach consists of two phases. One phase is preprocessed for face images by means of the proposed illumination normalization method. The location of facial features can fit more efficient and fast based on the proposed image blending. On the other hand, based on template matching, we further improve the active shape models (called as IASM) to locate the face shape more precise which can gain the recognized rate in the next phase. The other phase is to process feature extraction by using principal component analysis and face recognition by using support vector machine classifiers. The results show that this proposed method can obtain good facial localization and face recognition with varied illumination and local distortion.
Abstract: Functional Magnetic Resonance Imaging(fMRI) is a
noninvasive imaging technique that measures the hemodynamic
response related to neural activity in the human brain. Event-related
functional magnetic resonance imaging (efMRI) is a form of
functional Magnetic Resonance Imaging (fMRI) in which a series of
fMRI images are time-locked to a stimulus presentation and averaged
together over many trials. Again an event related potential (ERP) is a
measured brain response that is directly the result of a thought or
perception. Here the neuronal response of human visual cortex in
normal healthy patients have been studied. The patients were asked
to perform a visual three choice reaction task; from the relative
response of each patient corresponding neuronal activity in visual
cortex was imaged. The average number of neurons in the adult
human primary visual cortex, in each hemisphere has been estimated
at around 140 million. Statistical analysis of this experiment was
done with SPM5(Statistical Parametric Mapping version 5) software.
The result shows a robust design of imaging the neuronal activity of
human visual cortex.
Abstract: We show that Chebyshev Polynomials are a practical representation of computable functions on the computable reals. The paper presents error estimates for common operations and demonstrates that Chebyshev Polynomial methods would be more efficient than Taylor Series methods for evaluation of transcendental functions.
Abstract: The enthusiasm for gluten avoidance in a growing
market is met by improvements in sensitive detection methods for
analysing gluten content. Paradoxically, manufacturers employ no
such systems in the production process but continue to market their
product as gluten free, a significant risk posed to an undetermined
coeliac population. This paper resonates with an immunological
response that causes gastrointestinal scarring and villous atrophy with
the conventional description of personal injury. This thesis divulges
into evaluating potential inadequacies of gluten labelling laws which
not only present a diagnostic challenge for general practitioners in the
UK but it also exposes a less than adequate form of available legal
protection to those who suffer adverse reactions as a result of gluten
digestion. Central to this discussion is whether a claim brought in
misrepresentation, negligence and/or under the Consumer Protection
Act 1987 could be sustained. An interesting comparison is then made
with the legal regimes of neighboring jurisdictions furthering the
theme of a legally un-catered for gluten kingdom.
Abstract: Text categorization is the problem of classifying text
documents into a set of predefined classes. After a preprocessing
step, the documents are typically represented as large sparse vectors.
When training classifiers on large collections of documents, both the
time and memory restrictions can be quite prohibitive. This justifies
the application of feature selection methods to reduce the
dimensionality of the document-representation vector. In this paper,
three feature selection methods are evaluated: Random Selection,
Information Gain (IG) and Support Vector Machine feature selection
(called SVM_FS). We show that the best results were obtained with
SVM_FS method for a relatively small dimension of the feature
vector. Also we present a novel method to better correlate SVM
kernel-s parameters (Polynomial or Gaussian kernel).
Abstract: This paper presents the methodology from machine
learning approaches for short-term rain forecasting system. Decision
Tree, Artificial Neural Network (ANN), and Support Vector Machine
(SVM) were applied to develop classification and prediction models
for rainfall forecasts. The goals of this presentation are to
demonstrate (1) how feature selection can be used to identify the
relationships between rainfall occurrences and other weather
conditions and (2) what models can be developed and deployed for
predicting the accurate rainfall estimates to support the decisions to
launch the cloud seeding operations in the northeastern part of
Thailand. Datasets collected during 2004-2006 from the
Chalermprakiat Royal Rain Making Research Center at Hua Hin,
Prachuap Khiri khan, the Chalermprakiat Royal Rain Making
Research Center at Pimai, Nakhon Ratchasima and Thai
Meteorological Department (TMD). A total of 179 records with 57
features was merged and matched by unique date. There are three
main parts in this work. Firstly, a decision tree induction algorithm
(C4.5) was used to classify the rain status into either rain or no-rain.
The overall accuracy of classification tree achieves 94.41% with the
five-fold cross validation. The C4.5 algorithm was also used to
classify the rain amount into three classes as no-rain (0-0.1 mm.),
few-rain (0.1- 10 mm.), and moderate-rain (>10 mm.) and the overall
accuracy of classification tree achieves 62.57%. Secondly, an ANN
was applied to predict the rainfall amount and the root mean square
error (RMSE) were used to measure the training and testing errors of
the ANN. It is found that the ANN yields a lower RMSE at 0.171 for
daily rainfall estimates, when compared to next-day and next-2-day
estimation. Thirdly, the ANN and SVM techniques were also used to
classify the rain amount into three classes as no-rain, few-rain, and
moderate-rain as above. The results achieved in 68.15% and 69.10%
of overall accuracy of same-day prediction for the ANN and SVM
models, respectively. The obtained results illustrated the comparison
of the predictive power of different methods for rainfall estimation.