Abstract: The paper presents an on-line recognition machine
(RM) for continuous/isolated, dynamic and static gestures that arise
in Flight Deck Officer (FDO) training. RM is based on generic pattern
recognition framework. Gestures are represented as templates using
summary statistics. The proposed recognition algorithm exploits temporal
and spatial characteristics of gestures via dynamic programming
and Markovian process. The algorithm predicts corresponding index
of incremental input data in the templates in an on-line mode.
Accumulated consistency in the sequence of prediction provides a
similarity measurement (Score) between input data and the templates.
The algorithm provides an intuitive mechanism for automatic detection
of start/end frames of continuous gestures. In the present paper,
we consider isolated gestures. The performance of RM is evaluated
using four datasets - artificial (W TTest), hand motion (Yang) and
FDO (tracker, vision-based ). RM achieves comparable results which
are in agreement with other on-line and off-line algorithms such as
hidden Markov model (HMM) and dynamic time warping (DTW).
The proposed algorithm has the additional advantage of providing
timely feedback for training purposes.
Abstract: Video streaming over lossy IP networks is very
important issues, due to the heterogeneous structure of networks.
Infrastructure of the Internet exhibits variable bandwidths, delays,
congestions and time-varying packet losses. Because of variable
attributes of the Internet, video streaming applications should not
only have a good end-to-end transport performance but also have a
robust rate control, furthermore multipath rate allocation mechanism.
So for providing the video streaming service quality, some other
components such as Bandwidth Estimation and Adaptive Rate
Controller should be taken into consideration. This paper gives an
overview of video streaming concept and bandwidth estimation tools
and then introduces special architectures for bandwidth adaptive
video streaming. A bandwidth estimation algorithm – pathChirp,
Optimized Rate Controllers and Multipath Rate Allocation Algorithm
are considered as all-in-one solution for video streaming problem.
This solution is directed and optimized by a decision center which is
designed for obtaining the maximum quality at the receiving side.
Abstract: In this paper, application of artificial neural networks
in typical disease diagnosis has been investigated. The real procedure
of medical diagnosis which usually is employed by physicians was
analyzed and converted to a machine implementable format. Then
after selecting some symptoms of eight different diseases, a data set
contains the information of a few hundreds cases was configured and
applied to a MLP neural network. The results of the experiments and
also the advantages of using a fuzzy approach were discussed as
well. Outcomes suggest the role of effective symptoms selection and
the advantages of data fuzzificaton on a neural networks-based
automatic medical diagnosis system.
Abstract: Data mining has been used very frequently to extract
hidden information from large databases. This paper suggests the use
of decision trees for continuously extracting the clinical reasoning in
the form of medical expert-s actions that is inherent in large number
of EMRs (Electronic Medical records). In this way the extracted data
could be used to teach students of oral medicine a number of orderly
processes for dealing with patients who represent with different
problems within the practice context over time.
Abstract: This work proposes an approach to address automatic
text summarization. This approach is a trainable summarizer, which
takes into account several features, including sentence position,
positive keyword, negative keyword, sentence centrality, sentence
resemblance to the title, sentence inclusion of name entity, sentence
inclusion of numerical data, sentence relative length, Bushy path of
the sentence and aggregated similarity for each sentence to generate
summaries. First we investigate the effect of each sentence feature on
the summarization task. Then we use all features score function to
train genetic algorithm (GA) and mathematical regression (MR)
models to obtain a suitable combination of feature weights. The
proposed approach performance is measured at several compression
rates on a data corpus composed of 100 English religious articles.
The results of the proposed approach are promising.
Abstract: The model of neural networks on the small-world
topology, with metric (local and random connectivity) is investigated.
The synaptic weights are random, driving the network towards a
chaotic state for the neural activity. An ordered macroscopic neuron
state is induced by a bias in the network connections. When the
connections are mainly local, the network emulates a block-like
structure. It is found that the topology and the bias compete to
influence the network to evolve into a global or a block activity
ordering, according to the initial conditions.
Abstract: This paper introduces the effective speckle reduction of
synthetic aperture radar (SAR) images using inner product spaces in
undecimated wavelet domain. There are two major areas in projection
onto span algorithm where improvement can be made. First is the use
of undecimated wavelet transformation instead of discrete wavelet
transformation. And second area is the use of smoothing filter namely
directional smoothing filter which is an additional step. Proposed
method does not need any noise estimation and thresholding
technique. More over proposed method gives good results on both
single polarimetric and fully polarimetric SAR images.
Abstract: SAD (Sum of Absolute Difference) algorithm is
heavily used in motion estimation which is computationally highly
demanding process in motion picture encoding. To enhance the
performance of motion picture encoding on a VLIW processor, an
efficient implementation of SAD algorithm on the VLIW processor is
essential. SAD algorithm is programmed as a nested loop with a
conditional branch. In VLIW processors, loop is usually optimized by
software pipelining, but researches on optimal scheduling of software
pipelining for nested loops, especially nested loops with conditional
branches are rare. In this paper, we propose an optimal scheduling and
implementation of SAD algorithm with conditional branch on a VLIW
DSP processor. The proposed optimal scheduling first transforms the
nested loop with conditional branch into a single loop with conditional
branch with consideration of full utilization of ILP capability of the
VLIW processor and realization of earlier escape from the loop. Next,
the proposed optimal scheduling applies a modulo scheduling
technique developed for single loop. Based on this optimal scheduling
strategy, optimal implementation of SAD algorithm on TMS320C67x,
a VLIW DSP is presented. Through experiments on TMS320C6713
DSK, it is shown that H.263 encoder with the proposed SAD
implementation performs better than other H.263 encoder with other
SAD implementations, and that the code size of the optimal SAD
implementation is small enough to be appropriate for embedded
environments.
Abstract: This paper focuses on the data-driven generation
of fuzzy IF...THEN rules. The resulted fuzzy rule base can be
applied to build a classifier, a model used for prediction, or
it can be applied to form a decision support system. Among
the wide range of possible approaches, the decision tree and
the association rule based algorithms are overviewed, and two
new approaches are presented based on the a priori fuzzy
clustering based partitioning of the continuous input variables.
An application study is also presented, where the developed
methods are tested on the well known Wisconsin Breast Cancer
classification problem.
Abstract: This article combines two techniques: data
envelopment analysis (DEA) and Factor analysis (FA) to data
reduction in decision making units (DMU). Data envelopment
analysis (DEA), a popular linear programming technique is useful to
rate comparatively operational efficiency of decision making units
(DMU) based on their deterministic (not necessarily stochastic)
input–output data and factor analysis techniques, have been proposed
as data reduction and classification technique, which can be applied
in data envelopment analysis (DEA) technique for reduction input –
output data. Numerical results reveal that the new approach shows a
good consistency in ranking with DEA.
Abstract: Color image segmentation can be considered as a
cluster procedure in feature space. k-means and its adaptive
version, i.e. competitive learning approach are powerful tools
for data clustering. But k-means and competitive learning suffer
from several drawbacks such as dead-unit problem and need to
pre-specify number of cluster. In this paper, we will explore to
use competitive and cooperative learning approach to perform
color image segmentation. In competitive and cooperative
learning approach, seed points not only compete each other, but
also the winner will dynamically select several nearest
competitors to form a cooperative team to adapt to the input
together, finally it can automatically select the correct number
of cluster and avoid the dead-units problem. Experimental
results show that CCL can obtain better segmentation result.
Abstract: This paper explores the scalability issues associated
with solving the Named Entity Recognition (NER) problem using
Support Vector Machines (SVM) and high-dimensional features. The
performance results of a set of experiments conducted using binary
and multi-class SVM with increasing training data sizes are
examined. The NER domain chosen for these experiments is the
biomedical publications domain, especially selected due to its
importance and inherent challenges. A simple machine learning
approach is used that eliminates prior language knowledge such as
part-of-speech or noun phrase tagging thereby allowing for its
applicability across languages. No domain-specific knowledge is
included. The accuracy measures achieved are comparable to those
obtained using more complex approaches, which constitutes a
motivation to investigate ways to improve the scalability of multiclass
SVM in order to make the solution more practical and useable.
Improving training time of multi-class SVM would make support
vector machines a more viable and practical machine learning
solution for real-world problems with large datasets. An initial
prototype results in great improvement of the training time at the
expense of memory requirements.
Abstract: Conceptualization strengthens intelligent systems in generalization skill, effective knowledge representation, real-time inference, and managing uncertain and indefinite situations in addition to facilitating knowledge communication for learning agents situated in real world. Concept learning introduces a way of abstraction by which the continuous state is formed as entities called concepts which are connected to the action space and thus, they illustrate somehow the complex action space. Of computational concept learning approaches, action-based conceptualization is favored because of its simplicity and mirror neuron foundations in neuroscience. In this paper, a new biologically inspired concept learning approach based on the probabilistic framework is proposed. This approach exploits and extends the mirror neuron-s role in conceptualization for a reinforcement learning agent in nondeterministic environments. In the proposed method, instead of building a huge numerical knowledge, the concepts are learnt gradually from rewards through interaction with the environment. Moreover the probabilistic formation of the concepts is employed to deal with uncertain and dynamic nature of real problems in addition to the ability of generalization. These characteristics as a whole distinguish the proposed learning algorithm from both a pure classification algorithm and typical reinforcement learning. Simulation results show advantages of the proposed framework in terms of convergence speed as well as generalization and asymptotic behavior because of utilizing both success and failures attempts through received rewards. Experimental results, on the other hand, show the applicability and effectiveness of the proposed method in continuous and noisy environments for a real robotic task such as maze as well as the benefits of implementing an incremental learning scenario in artificial agents.
Abstract: As a popular rank-reduced vector space approach,
Latent Semantic Indexing (LSI) has been used in information
retrieval and other applications. In this paper, an LSI-based content
vector model for text classification is presented, which constructs
multiple augmented category LSI spaces and classifies text by their
content. The model integrates the class discriminative information
from the training data and is equipped with several pertinent feature
selection and text classification algorithms. The proposed classifier
has been applied to email classification and its experiments on a
benchmark spam testing corpus (PU1) have shown that the approach
represents a competitive alternative to other email classifiers based
on the well-known SVM and naïve Bayes algorithms.
Abstract: This paper describes the project and development of a
very low-cost and small electronic prototype, especially designed for
monitoring and controlling existing home automation alarm systems
(intruder, smoke, gas, flood, etc.), via TCP/IP, with a typical web
browser. Its use will allow home owners to be immediately alerted
and aware when an alarm event occurs, and being also able to
interact with their home automation alarm system, disarming, arming
and watching event alerts, with a personal wireless Wi-Fi PDA or
smartphone logged on to a dedicated predefined web page, and using
also a PC or Laptop.
Abstract: One of the major challenges in the Information
Retrieval field is handling the massive amount of information
available to Internet users. Existing ranking techniques and strategies
that govern the retrieval process fall short of expected accuracy.
Often relevant documents are buried deep in the list of documents
returned by the search engine. In order to improve retrieval accuracy
we examine the issue of language effect on the retrieval process.
Then, we propose a solution for a more biased, user-centric relevance
for retrieved data. The results demonstrate that using indices based
on variations of the same language enhances the accuracy of search
engines for individual users.
Abstract: We present an Electronic Nose (ENose), which is
aimed at identifying the presence of one out of two gases, possibly
detecting the presence of a mixture of the two. Estimation of the
concentrations of the components is also performed for a volatile
organic compound (VOC) constituted by methanol and acetone, for
the ranges 40-400 and 22-220 ppm (parts-per-million), respectively.
Our system contains 8 sensors, 5 of them being gas sensors (of the
class TGS from FIGARO USA, INC., whose sensing element is a tin
dioxide (SnO2) semiconductor), the remaining being a temperature
sensor (LM35 from National Semiconductor Corporation), a
humidity sensor (HIH–3610 from Honeywell), and a pressure sensor
(XFAM from Fujikura Ltd.).
Our integrated hardware–software system uses some machine
learning principles and least square regression principle to identify at
first a new gas sample, or a mixture, and then to estimate the
concentrations. In particular we adopt a training model using the
Support Vector Machine (SVM) approach with linear kernel to teach
the system how discriminate among different gases. Then we apply
another training model using the least square regression, to predict
the concentrations.
The experimental results demonstrate that the proposed
multiclassification and regression scheme is effective in the
identification of the tested VOCs of methanol and acetone with
96.61% correctness. The concentration prediction is obtained with
0.979 and 0.964 correlation coefficient for the predicted versus real
concentrations of methanol and acetone, respectively.
Abstract: The objective of this paper is the introduction to a
unified optimization framework for research and education. The
OPTILIB framework implements different general purpose algorithms
for combinatorial optimization and minimum search on standard continuous
test functions. The preferences of this library are the straightforward
integration of new optimization algorithms and problems
as well as the visualization of the optimization process of different
methods exploring the search space exclusively or for the real time
visualization of different methods in parallel. Further the usage of
several implemented methods is presented on the basis of two use
cases, where the focus is especially on the algorithm visualization.
First it is demonstrated how different methods can be compared
conveniently using OPTILIB on the example of different iterative
improvement schemes for the TRAVELING SALESMAN PROBLEM.
A second study emphasizes how the framework can be used to find
global minima in the continuous domain.
Abstract: Over the past few years, a number of efforts have
been exerted to build parallel processing systems that utilize the idle
power of LAN-s and PC-s available in many homes and corporations.
The main advantage of these approaches is that they provide cheap
parallel processing environments for those who cannot afford the
expenses of supercomputers and parallel processing hardware.
However, most of the solutions provided are not very flexible in the
use of available resources and very difficult to install and setup.
In this paper, a multi-level web-based parallel processing system
(MWPS) is designed (appendix). MWPS is based on the idea of
volunteer computing, very flexible, easy to setup and easy to use.
MWPS allows three types of subscribers: simple volunteers (single
computers), super volunteers (full networks) and end users. All of
these entities are coordinated transparently through a secure web site.
Volunteer nodes provide the required processing power needed by
the system end users. There is no limit on the number of volunteer
nodes, and accordingly the system can grow indefinitely. Both
volunteer and system users must register and subscribe. Once, they
subscribe, each entity is provided with the appropriate MWPS
components. These components are very easy to install.
Super volunteer nodes are provided with special components that
make it possible to delegate some of the load to their inner nodes.
These inner nodes may also delegate some of the load to some other
lower level inner nodes .... and so on. It is the responsibility of the
parent super nodes to coordinate the delegation process and deliver
the results back to the user.
MWPS uses a simple behavior-based scheduler that takes into
consideration the current load and previous behavior of processing
nodes. Nodes that fulfill their contracts within the expected time get a
high degree of trust. Nodes that fail to satisfy their contract get a
lower degree of trust.
MWPS is based on the .NET framework and provides the minimal
level of security expected in distributed processing environments.
Users and processing nodes are fully authenticated. Communications
and messages between nodes are very secure. The system has been
implemented using C#.
MWPS may be used by any group of people or companies to
establish a parallel processing or grid environment.
Abstract: In order to accelerate the similarity search in highdimensional database, we propose a new hierarchical indexing method. It is composed of offline and online phases. Our contribution concerns both phases. In the offline phase, after gathering the whole of the data in clusters and constructing a hierarchical index, the main originality of our contribution consists to develop a method to construct bounding forms of clusters to avoid overlapping. For the online phase, our idea improves considerably performances of similarity search. However, for this second phase, we have also developed an adapted search algorithm. Our method baptized NOHIS (Non-Overlapping Hierarchical Index Structure) use the Principal Direction Divisive Partitioning (PDDP) as algorithm of clustering. The principle of the PDDP is to divide data recursively into two sub-clusters; division is done by using the hyper-plane orthogonal to the principal direction derived from the covariance matrix and passing through the centroid of the cluster to divide. Data of each two sub-clusters obtained are including by a minimum bounding rectangle (MBR). The two MBRs are directed according to the principal direction. Consequently, the nonoverlapping between the two forms is assured. Experiments use databases containing image descriptors. Results show that the proposed method outperforms sequential scan and SRtree in processing k-nearest neighbors.