Abstract: In this paper we propose a new classification method for automatic sleep scoring using an artificial neural network based decision tree. It attempts to treat sleep scoring progress as a series of two-class problems and solves them with a decision tree made up of a group of neural network classifiers, each of which uses a special feature set and is aimed at only one specific sleep stage in order to maximize the classification effect. A single electroencephalogram (EEG) signal is used for our analysis rather than depending on multiple biological signals, which makes greatly simplifies the data acquisition process. Experimental results demonstrate that the average epoch by epoch agreement between the visual and the proposed method in separating 30s wakefulness+S1, REM, S2 and SWS epochs was 88.83%. This study shows that the proposed method performed well in all the four stages, and can effectively limit error propagation at the same time. It could, therefore, be an efficient method for automatic sleep scoring. Additionally, since it requires only a small volume of data it could be suited to pervasive applications.
Abstract: The residue number system (RNS) is popular in high performance computation applications because of its carry-free nature. The challenges of RNS systems design lie in the moduli set selection and in the reverse conversion from residue representation to weighted representation. In this paper, we proposed a fully parallel reverse conversion algorithm for the moduli set {rn - 2, rn - 1, rn}, based on simple mathematical relationships. Also an efficient hardware realization of this algorithm is presented. Our proposed converter is very faster and results to hardware savings, compared to the other reverse converters.
Abstract: A challenged control problem is when the
performance is pushed to the limit. The state-derivative feedback
control strategy directly uses acceleration information for feedback
and state estimation. The derivative part is concerned with the rateof-
change of the error with time. If the measured variable approaches
the set point rapidly, then the actuator is backed off early to allow it
to coast to the required level. Derivative action makes a control
system behave much more intelligently. A sensor measures the
variable to be controlled and the measured in formation is fed back to
the controller to influence the controlled variable. A high gain
problem can be also formulated for proportional plus derivative
feedback transformation. Using MATLAB Simulink dynamic
simulation tool this paper examines a system with a proportional plus
derivative feedback and presents an automatic implementation of
finding an acceptable controlled system. Using feedback
transformations the system is transformed into another system.
Abstract: Over the early years of the 21st century, cities
throughout the Middle East, particularly in the Gulf region have
expanded more rapidly than ever before. Given the presence of a
large volume of high-rise buildings allover the region, the local
authority aims to set a new standard for sustainable development;
with an integrated approach to maintain a balance between economy,
quality, environmental protection and safety of life. In the very near
future, as mandatory requirements, sustainability will be the criteria
that should be included in all building projects. It is well known in
the building sustainability topics that structural design engineers do
not have a key role in this matter. In addition, the LEED (Leadership
in Energy and Environmental Design) has looked almost exclusively
on the environmental components and materials specifications. The
objective of this paper is to focus and establish groundwork for
sustainability techniques and applications related to the RC high-rise
buildings design, from the structural point of view. A set of
recommendations related to local conditions, structural modeling and
analysis is given, and some helpful suggestions for structural design
team work are addressed. This paper attempts to help structural
engineers in identifying the building sustainability design, in order to
meet local needs and achieve alternative solutions at an early stage of
project design.
Abstract: Most known methods for measuring the structural similarity of document structures are based on, e.g., tag measures, path metrics and tree measures in terms of their DOM-Trees. Other methods measures the similarity in the framework of the well known vector space model. In contrast to these we present a new approach to measuring the structural similarity of web-based documents represented by so called generalized trees which are more general than DOM-Trees which represent only directed rooted trees.We will design a new similarity measure for graphs representing web-based hypertext structures. Our similarity measure is mainly based on a novel representation of a graph as strings of linear integers, whose components represent structural properties of the graph. The similarity of two graphs is then defined as the optimal alignment of the underlying property strings. In this paper we apply the well known technique of sequence alignments to solve a novel and challenging problem: Measuring the structural similarity of generalized trees. More precisely, we first transform our graphs considered as high dimensional objects in linear structures. Then we derive similarity values from the alignments of the property strings in order to measure the structural similarity of generalized trees. Hence, we transform a graph similarity problem to a string similarity problem. We demonstrate that our similarity measure captures important structural information by applying it to two different test sets consisting of graphs representing web-based documents.
Abstract: Malaysia has successfully applied economic planning
to guide the development of the country from an economy of
agriculture and mining to a largely industrialised one. Now, with its
sights set on attaining the economic level of a fully developed nation
by 2020, the planning system must be made even more efficient and
focused.
It must ensure that every investment made in the country, contribute
towards creating the desirable objective of a strong, modern,
internationally competitive, technologically advanced, post-industrial
economy. Cities in Malaysia must also be fully aware of the enormous
competition it faces in a region with rapidly expanding and
modernising economies, all contending for the same pool of potential
international investments.
Efficiency of urban governance is also fundamental issue in
development characterized by sustainability, subsidiarity, equity,
transparency and accountability, civic engagement and citizenship, and
security. As described above, city competitiveness is harnessed
through 'city marketing and city management'.
High technology and high skilled industries, together with finance,
transportation, tourism, business, information and professional
services shopping and other commercial activities, are the principal
components of the nation-s economy, which must be developed to a
level well beyond where it is now. In this respect, Kuala Lumpur being
the premier city must play the leading role.
Abstract: Debates on residential satisfaction topic have been
vigorously discussed in family house setting. Nonetheless, less or
lack of attention was given to survey on student residential
satisfaction in the campus house setting. This study, however, tried to
fill in the gap by focusing more on the relationship between students-
socio-economic backgrounds and student residential satisfaction with
their on-campus student housing facilities. Two-stage cluster
sampling method was employed to classify the respondents. Then,
self-administered questionnaires were distributed face-to-face to the
students. In general, it was confirmed that the students- socioeconomic
backgrounds have significantly influence the students-
satisfaction with their on-campus student housing facilities. The main
influential factors were revealed as the economic status, sense of
sharing, and the ethnicity of roommates. Likewise, this study could
also provide some useful feedback for the universities administration
in order to improve their student housing facilities.
Abstract: An adaptive Chinese hand-talking system is presented
in this paper. By analyzing the 3 data collecting strategies for new
users, the adaptation framework including supervised and unsupervised
adaptation methods is proposed. For supervised adaptation,
affinity propagation (AP) is used to extract exemplar subsets, and enhanced
maximum a posteriori / vector field smoothing (eMAP/VFS)
is proposed to pool the adaptation data among different models. For
unsupervised adaptation, polynomial segment models (PSMs) are
used to help hidden Markov models (HMMs) to accurately label
the unlabeled data, then the "labeled" data together with signerindependent
models are inputted to MAP algorithm to generate
signer-adapted models. Experimental results show that the proposed
framework can execute both supervised adaptation with small amount
of labeled data and unsupervised adaptation with large amount
of unlabeled data to tailor the original models, and both achieve
improvements on the performance of recognition rate.
Abstract: In this paper, we investigated the characteristic of a
clinical dataseton the feature selection and classification
measurements which deal with missing values problem.And also
posed the appropriated techniques to achieve the aim of the activity;
in this research aims to find features that have high effect to mortality
and mortality time frame. We quantify the complexity of a clinical
dataset. According to the complexity of the dataset, we proposed the
data mining processto cope their complexity; missing values, high
dimensionality, and the prediction problem by using the methods of
missing value replacement, feature selection, and classification.The
experimental results will extend to develop the prediction model for
cardiology.
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,
we present three feature selection methods: Information Gain,
Support Vector Machine feature selection called (SVM_FS) and
Genetic Algorithm with SVM (called GA_SVM). We show that the
best results were obtained with GA_SVM method for a relatively
small dimension of the feature vector.
Abstract: A novel behavioral detection framework is proposed
to detect zero day buffer overflow vulnerabilities (based on network
behavioral signatures) using zero-day exploits, instead of the
signature-based or anomaly-based detection solutions currently
available for IDPS techniques. At first we present the detection
model that uses shadow honeypot. Our system is used for the online
processing of network attacks and generating a behavior detection
profile. The detection profile represents the dataset of 112 types of
metrics describing the exact behavior of malware in the network. In
this paper we present the examples of generating behavioral
signatures for two attacks – a buffer overflow exploit on FTP server
and well known Conficker worm. We demonstrated the visualization
of important aspects by showing the differences between valid
behavior and the attacks. Based on these metrics we can detect
attacks with a very high probability of success, the process of
detection is however very expensive.
Abstract: This work presents a methodology for the selection
and design of propeller oriented to the experimental verification of
theoretical results. The problem of propeller selection and design
usually present itself in the following manner: a certain air volume
and static pressure are required for a certain system. Once the
necessity of fan design on a theoretical basis has been recognized, it
is possible to determinate the dimensions for a fan unit so that it will
perform in accordance with a certain set of specifications. The same
procedures in this work then can be applied in other propeller
selection.
Abstract: The robustness of color-based signatures in the presence of a selection of representative distortions is investigated. Considered are five signatures that have been developed and evaluated within a new modular framework. Two signatures presented in this work are directly derived from histograms gathered from video frames. The other three signatures are based on temporal information by computing difference histograms between adjacent frames. In order to obtain objective and reproducible results, the evaluations are conducted based on several randomly assembled test sets. These test sets are extracted from a video repository that contains a wide range of broadcast content including documentaries, sports, news, movies, etc. Overall, the experimental results show the adequacy of color-histogram-based signatures for video fingerprinting applications and indicate which type of signature should be preferred in the presence of certain distortions.
Abstract: The use of neural networks for recognition application is generally constrained by their inherent parameters inflexibility after the training phase. This means no adaptation is accommodated for input variations that have any influence on the network parameters. Attempts were made in this work to design a neural network that includes an additional mechanism that adjusts the threshold values according to the input pattern variations. The new approach is based on splitting the whole network into two subnets; main traditional net and a supportive net. The first deals with the required output of trained patterns with predefined settings, while the second tolerates output generation dynamically with tuning capability for any newly applied input. This tuning comes in the form of an adjustment to the threshold values. Two levels of supportive net were studied; one implements an extended additional layer with adjustable neuronal threshold setting mechanism, while the second implements an auxiliary net with traditional architecture performs dynamic adjustment to the threshold value of the main net that is constructed in dual-layer architecture. Experiment results and analysis of the proposed designs have given quite satisfactory conducts. The supportive layer approach achieved over 90% recognition rate, while the multiple network technique shows more effective and acceptable level of recognition. However, this is achieved at the price of network complexity and computation time. Recognition generalization may be also improved by accommodating capabilities involving all the innate structures in conjugation with Intelligence abilities with the needs of further advanced learning phases.
Abstract: The goal of a network-based intrusion detection
system is to classify activities of network traffics into two major
categories: normal and attack (intrusive) activities. Nowadays, data
mining and machine learning plays an important role in many
sciences; including intrusion detection system (IDS) using both
supervised and unsupervised techniques. However, one of the
essential steps of data mining is feature selection that helps in
improving the efficiency, performance and prediction rate of
proposed approach. This paper applies unsupervised K-means
clustering algorithm with information gain (IG) for feature selection
and reduction to build a network intrusion detection system. For our
experimental analysis, we have used the new NSL-KDD dataset,
which is a modified dataset for KDDCup 1999 intrusion detection
benchmark dataset. With a split of 60.0% for the training set and the
remainder for the testing set, a 2 class classifications have been
implemented (Normal, Attack). Weka framework which is a java
based open source software consists of a collection of machine
learning algorithms for data mining tasks has been used in the testing
process. The experimental results show that the proposed approach is
very accurate with low false positive rate and high true positive rate
and it takes less learning time in comparison with using the full
features of the dataset with the same algorithm.
Abstract: This paper proposes a novel improvement of forecasting approach based on using time-invariant fuzzy time series. In contrast to traditional forecasting methods, fuzzy time series can be also applied to problems, in which historical data are linguistic values. It is shown that proposed time-invariant method improves the performance of forecasting process. Further, the effect of using different number of fuzzy sets is tested as well. As with the most of cited papers, historical enrollment of the University of Alabama is used in this study to illustrate the forecasting process. Subsequently, the performance of the proposed method is compared with existing fuzzy time series time-invariant models based on forecasting accuracy. It reveals a certain performance superiority of the proposed method over methods described in the literature.
Abstract: Artificial neural networks (ANN) have the ability to model input-output relationships from processing raw data. This characteristic makes them invaluable in industry domains where such knowledge is scarce at best. In the recent decades, in order to overcome the black-box characteristic of ANNs, researchers have attempted to extract the knowledge embedded within ANNs in the form of rules that can be used in inference systems. This paper presents a new technique that is able to extract a small set of rules from a two-layer ANN. The extracted rules yield high classification accuracy when implemented within a fuzzy inference system. The technique targets industry domains that possess less complex problems for which no expert knowledge exists and for which a simpler solution is preferred to a complex one. The proposed technique is more efficient, simple, and applicable than most of the previously proposed techniques.
Abstract: In this paper, several improvements are proposed to
previous work of automated classification of alcoholics and nonalcoholics.
In the previous paper, multiplayer-perceptron neural
network classifying energy of gamma band Visual Evoked Potential
(VEP) signals gave the best classification performance using 800
VEP signals from 10 alcoholics and 10 non-alcoholics. Here, the
dataset is extended to include 3560 VEP signals from 102 subjects:
62 alcoholics and 40 non-alcoholics. Three modifications are
introduced to improve the classification performance: i) increasing
the gamma band spectral range by increasing the pass-band width of
the used filter ii) the use of Multiple Signal Classification algorithm
to obtain the power of the dominant frequency in gamma band VEP
signals as features and iii) the use of the simple but effective knearest
neighbour classifier. To validate that these two modifications
do give improved performance, a 10-fold cross validation
classification (CVC) scheme is used. Repeat experiments of the
previously used methodology for the extended dataset are performed
here and improvement from 94.49% to 98.71% in maximum
averaged CVC accuracy is obtained using the modifications. This
latest results show that VEP based classification of alcoholics is
worth exploring further for system development.
Abstract: Multi-Radio Multi-Channel (MRMC) systems are key to power control problems in wireless mesh networks (WMNs). In this paper, we present asynchronous multiple-state based power control for MRMC WMNs. First, WMN is represented as a set of disjoint Unified Channel Graphs (UCGs). Second, each network interface card (NIC) or radio assigned to a unique UCG adjusts transmission power using predicted multiple interaction state variables (IV) across UCGs. Depending on the size of queue loads and intra- and inter-channel states, each NIC optimizes transmission power locally and asynchronously. A new power selection MRMC unification protocol (PMMUP) is proposed that coordinates interactions among radios. The efficacy of the proposed method is investigated through simulations.
Abstract: The equilibrium process of plasma nitrogen species by
chemical kinetic reactions along various pressures is successfully
investigated. The equilibrium process is required in industrial
application to obtain the stable condition when heating up the
material for having homogenous reaction. Nitrogen species densities
is modeled by a continuity equation and extended Arrhenius form.
These equations are used to integrate the change of density over the
time. The integration is to acquire density and the reaction rate of
each reaction where temperature and time dependence are imposed.
A comparison is made with global model within pressure range of 1-
100mTorr and the temperature of electron is set to be higher than
other nitrogen species. The results shows that the chemical kinetic
model only agrees for high pressure because of no power imposed;
while the global model considers the external power along the
pressure range then the electron and nitrogen species give highly
quantity densities by factor of 3 to 5.