Abstract: The self-organizing map (SOM) model is a well-known neural network model with wide spread of applications. The main characteristics of SOM are two-fold, namely dimension reduction and topology preservation. Using SOM, a high-dimensional data space will be mapped to some low-dimensional space. Meanwhile, the topological relations among data will be preserved. With such characteristics, the SOM was usually applied on data clustering and visualization tasks. However, the SOM has main disadvantage of the need to know the number and structure of neurons prior to training, which are difficult to be determined. Several schemes have been proposed to tackle such deficiency. Examples are growing/expandable SOM, hierarchical SOM, and growing hierarchical SOM. These schemes could dynamically expand the map, even generate hierarchical maps, during training. Encouraging results were reported. Basically, these schemes adapt the size and structure of the map according to the distribution of training data. That is, they are data-driven or dataoriented SOM schemes. In this work, a topic-oriented SOM scheme which is suitable for document clustering and organization will be developed. The proposed SOM will automatically adapt the number as well as the structure of the map according to identified topics. Unlike other data-oriented SOMs, our approach expands the map and generates the hierarchies both according to the topics and their characteristics of the neurons. The preliminary experiments give promising result and demonstrate the plausibility of the method.
Abstract: A new code for spectral-amplitude coding optical
code-division multiple-access system is proposed called Random
diagonal (RD) code. This code is constructed using code segment and
data segment. One of the important properties of this code is that the
cross correlation at data segment is always zero, which means that
Phase Intensity Induced Noise (PIIN) is reduced. For the performance
analysis, the effects of phase-induced intensity noise, shot noise, and
thermal noise are considered simultaneously. Bit-error rate (BER)
performance is compared with Hadamard and Modified Frequency
Hopping (MFH) codes. It is shown that the system using this new
code matrices not only suppress PIIN, but also allows larger number
of active users compare with other codes. Simulation results shown
that using point to point transmission with three encoded channels,
RD code has better BER performance than other codes, also its found
that at 0 dbm PIIN noise are 10-10 and 10-11 for RD and MFH
respectively.
Abstract: An adaptive spatial Gaussian mixture model is proposed for clustering based color image segmentation. A new clustering objective function which incorporates the spatial information is introduced in the Bayesian framework. The weighting parameter for controlling the importance of spatial information is made adaptive to the image content to augment the smoothness towards piecewisehomogeneous region and diminish the edge-blurring effect and hence the name adaptive spatial finite mixture model. The proposed approach is compared with the spatially variant finite mixture model for pixel labeling. The experimental results with synthetic and Berkeley dataset demonstrate that the proposed method is effective in improving the segmentation and it can be employed in different practical image content understanding applications.
Abstract: This article describes Uruk, the virtual museum of
Iraq that we developed for visual exploration and retrieval of image
collections. The system largely exploits the loosely-structured
hierarchy of XML documents that provides a useful representation
method to store semi-structured or unstructured data, which does not
easily fit into existing database. The system offers users the
capability to mine and manage the XML-based image collections
through a web-based Graphical User Interface (GUI). Typically, at an
interactive session with the system, the user can browse a visual
structural summary of the XML database in order to select interesting
elements. Using this intermediate result, queries combining structure
and textual references can be composed and presented to the system.
After query evaluation, the full set of answers is presented in a visual
and structured way.
Abstract: This paper proposes a method that predicts attractive
evaluation objects. In the learning phase, the method inductively
acquires trend rules from complex sequential data. The data is
composed of two types of data. One is numerical sequential data.
Each evaluation object has respective numerical sequential data. The
other is text sequential data. Each evaluation object is described in
texts. The trend rules represent changes of numerical values related
to evaluation objects. In the prediction phase, the method applies
new text sequential data to the trend rules and evaluates which
evaluation objects are attractive. This paper verifies the effect of the
proposed method by using stock price sequences and news headline
sequences. In these sequences, each stock brand corresponds to an
evaluation object. This paper discusses validity of predicted attractive
evaluation objects, the process time of each phase, and the possibility
of application tasks.
Abstract: The purpose of this paper is to detect human in images.
This paper proposes a method for extracting human body feature descriptors consisting of projected edge component series. The feature descriptor can express appearances and shapes of human with local
and global distribution of edges. Our method evaluated with a linear SVM classifier on Daimler-Chrysler pedestrian dataset, and test with
various sub-region size. The result shows that the accuracy level of
proposed method similar to Histogram of Oriented Gradients(HOG)
feature descriptor and feature extraction process is simple and faster than existing methods.
Abstract: This paper reports the three-phase (gas + liquid +
hydrate) equilibrium pressure versus temperature data for a (O3 + O2 +
CO2 + H2O) system for developing the hydrate-based technology to
preserve ozone, a chemically unstable substance, for various
industrial, medical and consumer uses. These data cover the
temperature range from 272 K to 277 K, corresponding to pressures
from 1.6 MPa to 3.1 MPa, for each of the three different (O3 +
O2)-to-CO2 or O2-to-CO2 molar ratios in the gas phase, which are
approximately 4 : 6, 5 : 5, respectively. The mole fraction of ozone in
the gas phase was ~0.03 , which are the densest ozone fraction to
artificially form O3 containing hydrate ever reported in the literature.
Based on these data, the formation of hydrate containing
high-concentration ozone, as high as 1 mass %, will be expected.
Abstract: The main aim of Supply Chain Management (SCM) is
to produce, distribute, logistics and deliver goods and equipment in
right location, right time, right amount to satisfy costumers, with
minimum time and cost waste. So implementing techniques that
reduce project time and cost, and improve productivity and
performance is very important. Emerging technologies such as the
Radio Frequency Identification (RFID) are now making it possible to
automate supply chains in a real time manner and making them more
efficient than the simple supply chain of the past for tracing and
monitoring goods and products and capturing data on movements of
goods and other events. This paper considers concepts, components
and RFID technology characteristics by concentration of warehouse
and inventories management. Additionally, utilization of RFID in the
role of improving information management in supply chain is
discussed. Finally, the facts of installation and this technology-s
results in direction with warehouse and inventory management and
business development will be presented.
Abstract: Recently the use of data mining to scientific bibliographic data bases has been implemented to analyze the pathways of the knowledge or the core scientific relevances of a laureated novel or a country. This specific case of data mining has been named citation mining, and it is the integration of citation bibliometrics and text mining. In this paper we present an improved WEB implementation of statistical physics algorithms to perform the text mining component of citation mining. In particular we use an entropic like distance between the compression of text as an indicator of the similarity between them. Finally, we have included the recently proposed index h to characterize the scientific production. We have used this web implementation to identify users, applications and impact of the Mexican scientific institutions located in the State of Morelos.
Abstract: User-based Collaborative filtering (CF), one of the
most prevailing and efficient recommendation techniques, provides
personalized recommendations to users based on the opinions of other
users. Although the CF technique has been successfully applied in
various applications, it suffers from serious sparsity problems. The
cloud-model approach addresses the sparsity problems by
constructing the user-s global preference represented by a cloud
eigenvector. The user-based CF approach works well with dense
datasets while the cloud-model CF approach has a greater
performance when the dataset is sparse. In this paper, we present a
hybrid approach that integrates the predictions from both the
user-based CF and the cloud-model CF approaches. The experimental
results show that the proposed hybrid approach can ameliorate the
sparsity problem and provide an improved prediction quality.
Abstract: In mobile environments, unspecified numbers of transactions
arrive in continuous streams. To prove correctness of their
concurrent execution a method of modelling an infinite number of
transactions is needed. Standard database techniques model fixed
finite schedules of transactions. Lately, techniques based on temporal
logic have been proposed as suitable for modelling infinite schedules.
The drawback of these techniques is that proving the basic
serializability correctness condition is impractical, as encoding (the
absence of) conflict cyclicity within large sets of transactions results
in prohibitively large temporal logic formulae. In this paper, we show
that, under certain common assumptions on the graph structure of
data items accessed by the transactions, conflict cyclicity need only
be checked within all possible pairs of transactions. This results in
formulae of considerably reduced size in any temporal-logic-based
approach to proving serializability, and scales to arbitrary numbers
of transactions.
Abstract: This paper presents the design and implementation of
the WebGD, a CORBA-based document classification and retrieval
system on Internet. The WebGD makes use of such techniques as Web,
CORBA, Java, NLP, fuzzy technique, knowledge-based processing
and database technology. Unified classification and retrieval model,
classifying and retrieving with one reasoning engine and flexible
working mode configuration are some of its main features. The
architecture of WebGD, the unified classification and retrieval model,
the components of the WebGD server and the fuzzy inference engine
are discussed in this paper in detail.
Abstract: This paper describes an automatic algorithm to restore
the shape of three-dimensional (3D) left ventricle (LV) models created
from magnetic resonance imaging (MRI) data using a geometry-driven
optimization approach. Our basic premise is to restore the LV shape
such that the LV epicardial surface is smooth after the restoration. A
geometrical measure known as the Minimum Principle Curvature (κ2)
is used to assess the smoothness of the LV. This measure is used to
construct the objective function of a two-step optimization process.
The objective of the optimization is to achieve a smooth epicardial
shape by iterative in-plane translation of the MRI slices.
Quantitatively, this yields a minimum sum in terms of the magnitude
of κ
2, when κ2 is negative. A limited memory quasi-Newton algorithm,
L-BFGS-B, is used to solve the optimization problem. We tested our
algorithm on an in vitro theoretical LV model and 10 in vivo
patient-specific models which contain significant motion artifacts. The
results show that our method is able to automatically restore the shape
of LV models back to smoothness without altering the general shape of
the model. The magnitudes of in-plane translations are also consistent
with existing registration techniques and experimental findings.
Abstract: The aim of this research is to design a collaborative
framework that integrates risk analysis activities into the geospatial
database design (GDD) process. Risk analysis is rarely undertaken
iteratively as part of the present GDD methods in conformance to
requirement engineering (RE) guidelines and risk standards.
Accordingly, when risk analysis is performed during the GDD, some
foreseeable risks may be overlooked and not reach the output
specifications especially when user intentions are not systematically
collected. This may lead to ill-defined requirements and ultimately in
higher risks of geospatial data misuse. The adopted approach consists
of 1) reviewing risk analysis process within the scope of RE and
GDD, 2) analyzing the challenges of risk analysis within the context
of GDD, and 3) presenting the components of a risk-based
collaborative framework that improves the collection of the
intended/forbidden usages of the data and helps geo-IT experts to
discover implicit requirements and risks.
Abstract: In order to develop forest management strategies in
tropical forest in Malaysia, surveying the forest resources and
monitoring the forest area affected by logging activities is essential.
There are tremendous effort has been done in classification of land
cover related to forest resource management in this country as it is a
priority in all aspects of forest mapping using remote sensing and
related technology such as GIS. In fact classification process is a
compulsory step in any remote sensing research. Therefore, the main
objective of this paper is to assess classification accuracy of
classified forest map on Landsat TM data from difference number of
reference data (200 and 388 reference data). This comparison was
made through observation (200 reference data), and interpretation
and observation approaches (388 reference data). Five land cover
classes namely primary forest, logged over forest, water bodies, bare
land and agricultural crop/mixed horticultural can be identified by
the differences in spectral wavelength. Result showed that an overall
accuracy from 200 reference data was 83.5 % (kappa value
0.7502459; kappa variance 0.002871), which was considered
acceptable or good for optical data. However, when 200 reference
data was increased to 388 in the confusion matrix, the accuracy
slightly improved from 83.5% to 89.17%, with Kappa statistic
increased from 0.7502459 to 0.8026135, respectively. The accuracy
in this classification suggested that this strategy for the selection of
training area, interpretation approaches and number of reference data
used were importance to perform better classification result.
Abstract: Construction projects generally take place in
uncontrolled and dynamic environments where construction waste is
a serious environmental problem in many large cities. The total
amount of waste and carbon dioxide emissions from transportation
vehicles are still out of control due to increasing construction
projects, massive urban development projects and the lack of
effective tools for minimizing adverse environmental impacts in
construction. This research is about utilization of the integrated
applications of automated advanced tracking and data storage
technologies in the area of environmental management to monitor
and control adverse environmental impacts such as construction
waste and carbon dioxide emissions. Radio Frequency Identification
(RFID) integrated with the Global Position System (GPS) provides
an opportunity to uniquely identify materials, components, and
equipments and to locate and track them using minimal or no worker
input. The transmission of data to the central database will be carried
out with the help of Global System for Mobile Communications
(GSM).
Abstract: The statistical distributions are modeled in explaining
nature of various types of data sets. Although these distributions are
mostly uni-modal, it is quite common to see multiple modes in the
observed distribution of the underlying variables, which make the
precise modeling unrealistic. The observed data do not exhibit
smoothness not necessarily due to randomness, but could also be due
to non-randomness resulting in zigzag curves, oscillations, humps
etc. The present paper argues that trigonometric functions, which
have not been used in probability functions of distributions so far,
have the potential to take care of this, if incorporated in the
distribution appropriately. A simple distribution (named as, Sinoform
Distribution), involving trigonometric functions, is illustrated in the
paper with a data set. The importance of trigonometric functions is
demonstrated in the paper, which have the characteristics to make
statistical distributions exotic. It is possible to have multiple modes,
oscillations and zigzag curves in the density, which could be suitable
to explain the underlying nature of select data set.
Abstract: In this work, we developed the concept of
supercompression, i.e., compression above the compression standard
used. In this context, both compression rates are multiplied. In fact,
supercompression is based on super-resolution. That is to say,
supercompression is a data compression technique that superpose
spatial image compression on top of bit-per-pixel compression to
achieve very high compression ratios. If the compression ratio is very
high, then we use a convolutive mask inside decoder that restores the
edges, eliminating the blur. Finally, both, the encoder and the
complete decoder are implemented on General-Purpose computation
on Graphics Processing Units (GPGPU) cards. Specifically, the
mentio-ned mask is coded inside texture memory of a GPGPU.
Abstract: Face recognition is a technique to automatically
identify or verify individuals. It receives great attention in
identification, authentication, security and many more applications.
Diverse methods had been proposed for this purpose and also a lot of
comparative studies were performed. However, researchers could not
reach unified conclusion. In this paper, we are reporting an extensive
quantitative accuracy analysis of four most widely used face
recognition algorithms: Principal Component Analysis (PCA),
Independent Component Analysis (ICA), Linear Discriminant
Analysis (LDA) and Support Vector Machine (SVM) using AT&T,
Sheffield and Bangladeshi people face databases under diverse
situations such as illumination, alignment and pose variations.
Abstract: This paper analysed the food security situation among
Nigerian rural farmers. Data collected on 202 rural farmers from
Benue State were analysed using descriptive and inferential statistics.
The study revealed that majority of the respondents (60.83%) had
medium dietary diversity. Furthermore, household daily calorie
requirement for the food secure households was 10,723 and the
household daily calorie consumption was 12,598, with a surplus
index of 0.04. The food security index was 1.16. The Household
daily per capita calorie consumption was 3,221.2. For the food
insecure households, the household daily calorie requirement was
20,213 and the household daily calorie consumption was 17,393. The
shortfall index was 0.14. The food security index was 0.88. The
Household daily per capita calorie consumption was 2,432.8. The
most commonly used coping strategies during food stress included
intercropping (99.2%), reliance on less preferred food (98.1%),
limiting portion size at meal times (85.8%) and crop diversification
(70.8%).