Abstract: In this work, we begin with the presentation of the
Tθ family of usual similarity measures concerning multidimensional
binary data. Subsequently, some properties of these measures are
proposed. Finally the impact of the use of different inter-elements
measures on the results of the Agglomerative Hierarchical Clustering
Methods is studied.
Abstract: Quantification of cardiac function is performed by
calculating blood volume and ejection fraction in routine clinical
practice. However, these works have been performed by manual
contouring, which requires computational costs and varies on the
observer. In this paper, an automatic left ventricle segmentation
algorithm on cardiac magnetic resonance images (MRI) is presented.
Using knowledge on cardiac MRI, a K-mean clustering technique is
applied to segment blood region on a coil-sensitivity corrected image.
Then, a graph searching technique is used to correct segmentation
errors from coil distortion and noises. Finally, blood volume and
ejection fraction are calculated. Using cardiac MRI from 15 subjects,
the presented algorithm is tested and compared with manual
contouring by experts to show outstanding performance.
Abstract: The ultrasound imaging is very popular to diagnosis
the disease because of its non-invasive nature. The ultrasound
imaging slowly produces low quality images due to the presence of
spackle noise and wave interferences. There are several algorithms to
be proposed for the segmentation of ultrasound carotid artery images
but it requires a certain limit of user interaction. The pixel in an
image is highly correlated so the spatial information of surrounding
pixels may be considered in the process of image segmentation which
improves the results further. When data is highly correlated, one pixel
may belong to more than one cluster with different degree of
membership. There is an important step to computerize the evaluation
of arterial disease severity using segmentation of carotid artery lumen
in 2D and 3D ultrasonography and in finding vulnerable
atherosclerotic plaques susceptible to rupture which can cause stroke.
Abstract: The goal of image segmentation is to cluster pixels
into salient image regions. Segmentation could be used for object
recognition, occlusion boundary estimation within motion or stereo
systems, image compression, image editing, or image database lookup.
In this paper, we present a color image segmentation using
support vector machine (SVM) pixel classification. Firstly, the pixel
level color and texture features of the image are extracted and they
are used as input to the SVM classifier. These features are extracted
using the homogeneity model and Gabor Filter. With the extracted
pixel level features, the SVM Classifier is trained by using FCM
(Fuzzy C-Means).The image segmentation takes the advantage of
both the pixel level information of the image and also the ability of
the SVM Classifier. The Experiments show that the proposed method
has a very good segmentation result and a better efficiency, increases
the quality of the image segmentation compared with the other
segmentation methods proposed in the literature.
Abstract: In this paper, an analysis of some model order
reduction techniques is presented. A new hybrid algorithm for model
order reduction of linear time invariant systems is compared with the
conventional techniques namely Balanced Truncation, Hankel Norm
reduction and Dominant Pole Algorithm (DPA). The proposed hybrid
algorithm is known as Clustering Dominant Pole Algorithm (CDPA),
is able to compute the full set of dominant poles and its cluster center
efficiently. The dominant poles of a transfer function are specific
eigenvalues of the state space matrix of the corresponding dynamical
system. The effectiveness of this novel technique is shown through
the simulation results.
Abstract: Neurons in the nervous system communicate with
each other by producing electrical signals called spikes. To
investigate the physiological function of nervous system it is essential
to study the activity of neurons by detecting and sorting spikes in the
recorded signal. In this paper a method is proposed for considering
the spike sorting problem which is based on the nonlinear modeling
of spikes using exponential autoregressive model. The genetic
algorithm is utilized for model parameter estimation. In this regard
some selected model coefficients are used as features for sorting
purposes. For optimal selection of model coefficients, self-organizing
feature map is used. The results show that modeling of spikes with
nonlinear autoregressive model outperforms its linear counterpart.
Also the extracted features based on the coefficients of exponential
autoregressive model are better than wavelet based extracted features
and get more compact and well-separated clusters. In the case of
spikes different in small-scale structures where principal component
analysis fails to get separated clouds in the feature space, the
proposed method can obtain well-separated cluster which removes
the necessity of applying complex classifiers.
Abstract: Geometric and mechanical properties all influence the
resistance of RC structures and may, in certain combination of
property values, increase the risk of a brittle failure of the whole
system.
This paper presents a statistical and probabilistic investigation on
the resistance of RC beams designed according to Eurocodes 2 and 8,
and subjected to multiple failure modes, under both the natural
variation of material properties and the uncertainty associated with
cross-section and transverse reinforcement geometry. A full
probabilistic model based on JCSS Probabilistic Model Code is
derived. Different beams are studied through material nonlinear
analysis via Monte Carlo simulations. The resistance model is
consistent with Eurocode 2. Both a multivariate statistical evaluation
and the data clustering analysis of outcomes are then performed.
Results show that the ultimate load behaviour of RC beams
subjected to flexural and shear failure modes seems to be mainly
influenced by the combination of the mechanical properties of both
longitudinal reinforcement and stirrups, and the tensile strength of
concrete, of which the latter appears to affect the overall response of
the system in a nonlinear way. The model uncertainty of the
resistance model used in the analysis plays undoubtedly an important
role in interpreting results.
Abstract: Clustering involves the partitioning of n objects into k
clusters. Many clustering algorithms use hard-partitioning techniques
where each object is assigned to one cluster. In this paper we propose
an overlapping algorithm MCOKE which allows objects to belong to
one or more clusters. The algorithm is different from fuzzy clustering
techniques because objects that overlap are assigned a membership
value of 1 (one) as opposed to a fuzzy membership degree. The
algorithm is also different from other overlapping algorithms that
require a similarity threshold be defined a priori which can be
difficult to determine by novice users.
Abstract: Vancron 40, a nitrided powder metallurgical tool
Steel, is used in cold work applications where the predominant failure
mechanisms are adhesive wear or galling. Typical applications of
Vancron 40 are among others fine blanking, cold extrusion, deep
drawing and cold work rolls for cluster mills. Vancron 40 positive
results for cold work rolls for cluster mills and as a tool for some
severe metal forming process makes it competitive compared to other
type of work rolls that require higher precision, among others in cold
rolling of thin stainless steel, which required high surface finish
quality. In this project, three roll materials for cold rolling of stainless
steel strip was examined, Vancron 40, Narva 12B (a high-carbon,
high-chromium tool steel alloyed with tungsten) and Supra 3 (a
Chromium-molybdenum tungsten-vanadium alloyed high speed
steel). The purpose of this project was to study the depth profiles of
the ironed stainless steel strips, emergence of galling and to study the
lubrication performance used by steel industries. Laboratory
experiments were conducted to examine scratch of the strip, galling
and surface roughness of the roll materials under severe tribological
conditions. The critical sliding length for onset of galling was
estimated for stainless steel with four different lubricants. Laboratory
experiments result of performance evaluation of resistance capability
of rolls toward adhesive wear under severe conditions for low and
high reductions. Vancron 40 in combination with cold rolling
lubricant gave good surface quality, prevents galling of
metal surfaces and good bearing capacity.
Abstract: Job Scheduling plays an important role for efficient
utilization of grid resources available across different domains and
geographical zones. Scheduling of jobs is challenging and NPcomplete.
Evolutionary / Swarm Intelligence algorithms have been
extensively used to address the NP problem in grid scheduling.
Artificial Bee Colony (ABC) has been proposed for optimization
problems based on foraging behaviour of bees. This work proposes a
modified ABC algorithm, Cluster Heterogeneous Earliest First Min-
Min Artificial Bee Colony (CHMM-ABC), to optimally schedule
jobs for the available resources. The proposed model utilizes a novel
Heterogeneous Earliest Finish Time (HEFT) Heuristic Algorithm
along with Min-Min algorithm to identify the initial food source.
Simulation results show the performance improvement of the
proposed algorithm over other swarm intelligence techniques.
Abstract: Leukaemia is a blood cancer disease that contributes
to the increment of mortality rate in Malaysia each year. There are
two main categories for leukaemia, which are acute and chronic
leukaemia. The production and development of acute leukaemia cells
occurs rapidly and uncontrollable. Therefore, if the identification of
acute leukaemia cells could be done fast and effectively, proper
treatment and medicine could be delivered. Due to the requirement of
prompt and accurate diagnosis of leukaemia, the current study has
proposed unsupervised pixel segmentation based on clustering
algorithm in order to obtain a fully segmented abnormal white blood
cell (blast) in acute leukaemia image. In order to obtain the
segmented blast, the current study proposed three clustering
algorithms which are k-means, fuzzy c-means and moving k-means
algorithms have been applied on the saturation component image.
Then, median filter and seeded region growing area extraction
algorithms have been applied, to smooth the region of segmented
blast and to remove the large unwanted regions from the image,
respectively. Comparisons among the three clustering algorithms are
made in order to measure the performance of each clustering
algorithm on segmenting the blast area. Based on the good sensitivity
value that has been obtained, the results indicate that moving kmeans
clustering algorithm has successfully produced the fully
segmented blast region in acute leukaemia image. Hence, indicating
that the resultant images could be helpful to haematologists for
further analysis of acute leukaemia.
Abstract: A mixed method by combining modified pole
clustering technique and modified cauer continued fraction is
proposed for reducing the order of the large-scale dynamic systems.
The denominator polynomial of the reduced order model is obtained
by using modified pole clustering technique while the coefficients of
the numerator are obtained by modified cauer continued fraction.
This method generated 'k' number of reduced order models for kth
order reduction. The superiority of the proposed method has been
elaborated through numerical example taken from the literature and
compared with few existing order reduction methods.
Abstract: The star network is one of the promising
interconnection networks for future high speed parallel computers, it
is expected to be one of the future-generation networks. The star
network is both edge and vertex symmetry, it was shown to have
many gorgeous topological proprieties also it is owns hierarchical
structure framework. Although much of the research work has been
done on this promising network in literature, it still suffers from
having enough algorithms for load balancing problem. In this paper
we try to work on this issue by investigating and proposing an
efficient algorithm for load balancing problem for the star network.
The proposed algorithm is called Star Clustered Dimension Exchange
Method SCDEM to be implemented on the star network. The
proposed algorithm is based on the Clustered Dimension Exchange
Method (CDEM). The SCDEM algorithm is shown to be efficient in
redistributing the load balancing as evenly as possible among all
nodes of different factor networks.
Abstract: Many of the ever-growing elderly population require
exercise, such as running, for health management. One important
element of a runner’s training is the choice of shoes for exercise; shoes
are important because they provide the interface between the feet and
road. When we purchase shoes, we may instinctively choose a pair
after trying on many different pairs of shoes. Selecting the shoes
instinctively may work, but it does not guarantee a suitable fit for
running activities. Therefore, if we could select suitable shoes for each
runner from the viewpoint of brain activities, it would be helpful for
validating shoe selection. In this paper, we describe how brain
activities show different characteristics during particular task,
corresponding to different properties of shoes. Using five subjects, we
performed a verification experiment, applying weight, softness, and
flexibility as shoe properties. In order to affect the shoe property’s
differences to the brain, subjects run for 10 min. Before and after
running, subjects conducted a paced auditory serial addition task
(PASAT) as the particular task; and the subjects’ brain activities
during the PASAT are evaluated based on oxyhemoglobin and
deoxyhemoglobin relative concentration changes, measured by
near-infrared spectroscopy (NIRS). When the brain works actively,
oxihemoglobin and deoxyhemoglobin concentration drastically
changes; therefore, we calculate the maximum values of concentration
changes. In order to normalize relative concentration changes after
running, the maximum value are divided by before running maximum
value as evaluation parameters. The classification of the groups of
shoes is expressed on a self-organizing map (SOM). As a result,
deoxyhemoglobin can make clusters for two of the three types of
shoes.
Abstract: In this paper, Fuzzy C-Means clustering with
Expectation Maximization-Gaussian Mixture Model based hybrid
modeling algorithm is proposed for Continuous Tamil Speech
Recognition. The speech sentences from various speakers are used
for training and testing phase and objective measures are between the
proposed and existing Continuous Speech Recognition algorithms.
From the simulated results, it is observed that the proposed algorithm
improves the recognition accuracy and F-measure up to 3% as
compared to that of the existing algorithms for the speech signal from
various speakers. In addition, it reduces the Word Error Rate, Error
Rate and Error up to 4% as compared to that of the existing
algorithms. In all aspects, the proposed hybrid modeling for Tamil
speech recognition provides the significant improvements for speechto-
text conversion in various applications.
Abstract: As enterprise computing becomes more and more
complex, the costs and technical challenges of IT system maintenance
and support are increasing rapidly. One popular approach to managing
IT system maintenance is to prepare and use a FAQ (Frequently Asked
Questions) system to manage and reuse systems knowledge. Such a
FAQ system can help reduce the resolution time for each service
incident ticket. However, there is a major problem where over time the
knowledge in such FAQs tends to become outdated. Much of the
knowledge captured in the FAQ requires periodic updates in response
to new insights or new trends in the problems addressed in order to
maintain its usefulness for problem resolution. These updates require a
systematic approach to define the exact portion of the FAQ and its
content. Therefore, we are working on a novel method to
hierarchically structure the FAQ and automate the updates of its
structure and content. We use structured information and the
unstructured text information with the timelines of the information in
the service incident tickets. We cluster the tickets by structured
category information, by keywords, and by keyword modifiers for the
unstructured text information. We also calculate an urgency score
based on trends, resolution times, and priorities. We carefully studied
the tickets of one of our projects over a 2.5-year time period. After the
first 6 months we started to create FAQs and confirmed they improved
the resolution times. We continued observing over the next 2 years to
assess the ongoing effectiveness of our method for the automatic FAQ
updates. We improved the ratio of tickets covered by the FAQ from
32.3% to 68.9% during this time. Also, the average time reduction of
ticket resolution was between 31.6% and 43.9%. Subjective analysis
showed more than 75% reported that the FAQ system was useful in
reducing ticket resolution times.
Abstract: The hydrogenated amorphous carbon films (α-C:H)
were deposited on p-type Si (100) substrates at different thicknesses by
radio frequency plasma enhanced chemical vapor deposition
technique (rf-PECVD). Raman spectra display asymmetric
diamond-like carbon (DLC) peaks, representative of the α-C:H films.
The decrease of intensity ID/IG ratios revealed the sp3 content arise at
different thicknesses of the α-C:H films. In terms of mechanical
properties, the high hardness and elastic modulus values showed the
elastic and plastic deformation behaviors related to sp3 content in
amorphous carbon films. Electrochemical properties showed that the
α-C:H films exhibited excellent corrosion resistance in air-saturated
3.5 wt.% NaCl solution for pH 2 at room temperature. Thickness
increasing affected the small sp2 clusters in matrix, restricting the
velocity transfer and exchange of electrons. The deposited α-C:H films
exhibited excellent mechanical properties and corrosion resistance.
Abstract: The world wide web network is a network with a
complex topology, the main properties of which are the distribution
of degrees in power law, A low clustering coefficient and a weak
average distance. Modeling the web as a graph allows locating the
information in little time and consequently offering a help in the
construction of the research engine. Here, we present a model based
on the already existing probabilistic graphs with all the aforesaid
characteristics. This work will consist in studying the web in order to
know its structuring thus it will enable us to modelize it more easily
and propose a possible algorithm for its exploration.
Abstract: The North-eastern part of India, which receives
heavier rainfall than other parts of the subcontinent, is of great
concern now-a-days with regard to climate change. High intensity
rainfall for short duration and longer dry spell, occurring due to
impact of climate change, affects river morphology too. In the present
study, an attempt is made to delineate the North-eastern region of
India into some homogeneous clusters based on the Fuzzy Clustering
concept and to compare the resulting clusters obtained by using
conventional methods and nonconventional methods of clustering.
The concept of clustering is adapted in view of the fact that, impact
of climate change can be studied in a homogeneous region without
much variation, which can be helpful in studies related to water
resources planning and management. 10 IMD (Indian Meteorological
Department) stations, situated in various regions of the North-east,
have been selected for making the clusters. The results of the Fuzzy
C-Means (FCM) analysis show different clustering patterns for
different conditions. From the analysis and comparison it can be
concluded that nonconventional method of using GCM data is
somehow giving better results than the others. However, further
analysis can be done by taking daily data instead of monthly means to
reduce the effect of standardization.
Abstract: Online forum is part of a Learning Management
System (LMS) environment in which students share their opinions.
This study attempts to investigate the perceptions of students towards
online forum and their patterns of listening behavior during the forum
interaction. The students’ perceptions were measured using a
questionnaire, in which seven dimensions were used involving online
experience, benefits of forum participation, cost of participation,
perceived ease of use, usefulness, attitude, and intention. Meanwhile,
their patterns of listening behaviors were obtained using the log file
extracted from the LMS. A total of 25 postgraduate students
undertaking a course were involved in this study, and their activities
in the forum session were recorded by the LMS and used as a log file.
The results from the questionnaire analysis indicated that the students
perceived that the forum is easy to use, useful, and bring benefits to
them. Also, they showed positive attitude towards online forum, and
they have the intention to use it in future. Based on the log data, the
participants were also divided into six clusters of listening behavior,
in which they are different in terms of temporality, breadth, depth and
speaking level. The findings were compared to previous clusters
grouping and future recommendations are also discussed.