Abstract: Automated operations based on voice commands will become more and more important in many applications, including robotics, maintenance operations, etc. However, voice command recognition rates drop quite a lot under non-stationary and chaotic noise environments. In this paper, we tried to significantly improve the speech recognition rates under non-stationary noise environments. First, 298 Navy acronyms have been selected for automatic speech recognition. Data sets were collected under 4 types of noisy environments: factory, buccaneer jet, babble noise in a canteen, and destroyer. Within each noisy environment, 4 levels (5 dB, 15 dB, 25 dB, and clean) of Signal-to-Noise Ratio (SNR) were introduced to corrupt the speech. Second, a new algorithm to estimate speech or no speech regions has been developed, implemented, and evaluated. Third, extensive simulations were carried out. It was found that the combination of the new algorithm, the proper selection of language model and a customized training of the speech recognizer based on clean speech yielded very high recognition rates, which are between 80% and 90% for the four different noisy conditions. Fourth, extensive comparative studies have also been carried out.
Abstract: Missing data is a persistent problem in almost all
areas of empirical research. The missing data must be treated very
carefully, as data plays a fundamental role in every analysis.
Improper treatment can distort the analysis or generate biased results.
In this paper, we compare and contrast various imputation techniques
on missing data sets and make an empirical evaluation of these
methods so as to construct quality software models. Our empirical
study is based on NASA-s two public dataset. KC4 and KC1. The
actual data sets of 125 cases and 2107 cases respectively, without
any missing values were considered. The data set is used to create
Missing at Random (MAR) data Listwise Deletion(LD), Mean
Substitution(MS), Interpolation, Regression with an error term and
Expectation-Maximization (EM) approaches were used to compare
the effects of the various techniques.
Abstract: In biological and biomedical research motif finding tools are important in locating regulatory elements in DNA sequences. There are many such motif finding tools available, which often yield position weight matrices and significance indicators. These indicators, p-values and E-values, describe the likelihood that a motif alignment is generated by the background process, and the expected number of occurrences of the motif in the data set, respectively. The various tools often estimate these indicators differently, making them not directly comparable. One approach for comparing motifs from different tools, is computing the E-value as the product of the p-value and the number of possible alignments in the data set. In this paper we explore the combinatorics of the motif alignment models OOPS, ZOOPS, and ANR, and propose a generic algorithm for computing the number of possible combinations accurately. We also show that using the wrong alignment model can give E-values that significantly diverge from their true values.
Abstract: In this paper we address the issue of classifying the fluorescent intensity of a sample in Indirect Immuno-Fluorescence (IIF). Since IIF is a subjective, semi-quantitative test in its very nature, we discuss a strategy to reliably label the image data set by using the diagnoses performed by different physicians. Then, we discuss image pre-processing, feature extraction and selection. Finally, we propose two ANN-based classifiers that can separate intrinsically dubious samples and whose error tolerance can be flexibly set. Measured performance shows error rates less than 1%, which candidates the method to be used in daily medical practice either to perform pre-selection of cases to be examined, or to act as a second reader.
Abstract: In this study, the Multi-Layer Perceptron (MLP)with Back-Propagation learning algorithm are used to classify to effective diagnosis Parkinsons disease(PD).It-s a challenging problem for medical community.Typically characterized by tremor, PD occurs due to the loss of dopamine in the brains thalamic region that results in involuntary or oscillatory movement in the body. A feature selection algorithm along with biomedical test values to diagnose Parkinson disease.Clinical diagnosis is done mostly by doctor-s expertise and experience.But still cases are reported of wrong diagnosis and treatment. Patients are asked to take number of tests for diagnosis.In many cases,not all the tests contribute towards effective diagnosis of a disease.Our work is to classify the presence of Parkinson disease with reduced number of attributes.Original,22 attributes are involved in classify.We use Information Gain to determine the attributes which reduced the number of attributes which is need to be taken from patients.The Artificial neural networks is used to classify the diagnosis of patients.Twenty-Two attributes are reduced to sixteen attributes.The accuracy is in training data set is 82.051% and in the validation data set is 83.333%.
Abstract: In this paper we use exponential particle swarm
optimization (EPSO) to cluster data. Then we compare between
(EPSO) clustering algorithm which depends on exponential variation
for the inertia weight and particle swarm optimization (PSO)
clustering algorithm which depends on linear inertia weight. This
comparison is evaluated on five data sets. The experimental results
show that EPSO clustering algorithm increases the possibility to find
the optimal positions as it decrease the number of failure. Also show
that (EPSO) clustering algorithm has a smaller quantization error
than (PSO) clustering algorithm, i.e. (EPSO) clustering algorithm
more accurate than (PSO) clustering algorithm.
Abstract: A concern that researchers usually face in different
applications of Artificial Neural Network (ANN) is determination of
the size of effective domain in time series. In this paper, trial and
error method was used on groundwater depth time series to determine
the size of effective domain in the series in an observation well in
Union County, New Jersey, U.S. different domains of 20, 40, 60, 80,
100, and 120 preceding day were examined and the 80 days was
considered as effective length of the domain. Data sets in different
domains were fed to a Feed Forward Back Propagation ANN with
one hidden layer and the groundwater depths were forecasted. Root
Mean Square Error (RMSE) and the correlation factor (R2) of
estimated and observed groundwater depths for all domains were
determined. In general, groundwater depth forecast improved, as
evidenced by lower RMSEs and higher R2s, when the domain length
increased from 20 to 120. However, 80 days was selected as the
effective domain because the improvement was less than 1% beyond
that. Forecasted ground water depths utilizing measured daily data
(set #1) and data averaged over the effective domain (set #2) were
compared. It was postulated that more accurate nature of measured
daily data was the reason for a better forecast with lower RMSE
(0.1027 m compared to 0.255 m) in set #1. However, the size of input
data in this set was 80 times the size of input data in set #2; a factor
that may increase the computational effort unpredictably. It was
concluded that 80 daily data may be successfully utilized to lower the
size of input data sets considerably, while maintaining the effective
information in the data set.
Abstract: Ensemble learning algorithms such as AdaBoost and
Bagging have been in active research and shown improvements in
classification results for several benchmarking data sets with mainly
decision trees as their base classifiers. In this paper we experiment to
apply these Meta learning techniques with classifiers such as random
forests, neural networks and support vector machines. The data sets
are from MAGIC, a Cherenkov telescope experiment. The task is to
classify gamma signals from overwhelmingly hadron and muon
signals representing a rare class classification problem. We compare
the individual classifiers with their ensemble counterparts and
discuss the results. WEKA a wonderful tool for machine learning has
been used for making the experiments.
Abstract: We have developed a computer program consisting of
6 subtests assessing the children hand dexterity applicable in the
rehabilitation medicine. We have carried out a normative study on a
representative sample of 285 children aged from 7 to 15 (mean age
11.3) and we have proposed clinical standards for three age groups
(7-9, 9-11, 12-15 years). We have shown statistical significance of
differences among the corresponding mean values of the task time
completion. We have also found a strong correlation between the task
time completion and the age of the subjects, as well as we have
performed the test-retest reliability checks in the sample of 84
children, giving the high values of the Pearson coefficients for the
dominant and non-dominant hand in the range 0.740.97 and
0.620.93, respectively.
A new MATLAB-based programming tool aiming at analysis of
cardiologic RR intervals and blood pressure descriptors, is worked
out, too. For each set of data, ten different parameters are extracted: 2
in time domain, 4 in frequency domain and 4 in Poincaré plot
analysis. In addition twelve different parameters of baroreflex
sensitivity are calculated. All these data sets can be visualized in time
domain together with their power spectra and Poincaré plots. If
available, the respiratory oscillation curves can be also plotted for
comparison. Another application processes biological data obtained
from BLAST analysis.
Abstract: K-Modes is an extension of K-Means clustering algorithm, developed to cluster the categorical data, where the mean is replaced by the mode. The similarity measure proposed by Huang is the simple matching or mismatching measure. Weight of attribute values contribute much in clustering; thus in this paper we propose a new weighted dissimilarity measure for K-Modes, based on the ratio of frequency of attribute values in the cluster and in the data set. The new weighted measure is experimented with the data sets obtained from the UCI data repository. The results are compared with K-Modes and K-representative, which show that the new measure generates clusters with high purity.
Abstract: As it is known, buoyancy and drag forces rule bubble's rise velocity in a liquid column. These forces are strongly dependent on fluid properties, gravity as well as equivalent's diameter. This study reports a set of bubble rising velocity experiments in a liquid column using water or glycerol. Several records of terminal velocity were obtained. The results show that bubble's rise terminal velocity is strongly dependent on dynamic viscosity effect. The data set allowed to have some terminal velocities data interval of 8.0 ? 32.9 cm/s with Reynolds number interval 1.3 -7490. The bubble's movement was recorded with a video camera. The main goal is to present an original set data and results that will be discussed based on two-phase flow's theory. It will also discussed, the prediction of terminal velocity of a single bubble in liquid, as well as the range of its applicability. In conclusion, this study presents general expressions for the determination of the terminal velocity of isolated gas bubbles of a Reynolds number range, when the fluid proprieties are known.
Abstract: Medical Decision Support Systems (MDSSs) are sophisticated, intelligent systems that can provide inference due to lack of information and uncertainty. In such systems, to model the uncertainty various soft computing methods such as Bayesian networks, rough sets, artificial neural networks, fuzzy logic, inductive logic programming and genetic algorithms and hybrid methods that formed from the combination of the few mentioned methods are used. In this study, symptom-disease relationships are presented by a framework which is modeled with a formal concept analysis and theory, as diseases, objects and attributes of symptoms. After a concept lattice is formed, Bayes theorem can be used to determine the relationships between attributes and objects. A discernibility relation that forms the base of the rough sets can be applied to attribute data sets in order to reduce attributes and decrease the complexity of computation.
Abstract: This study examined a habitat-suitability assessment method namely the Ecological Niche Factor Analysis (ENFA). A virtual species was created and then dispatched in a geographic information system model of a real landscape in three historic scenarios: (1) spreading, (2) equilibrium, and (3) overabundance. In each scenario, the virtual species was sampled and these simulated data sets were used as inputs for the ENFA to reconstruct the habitat suitability model. The 'equilibrium' scenario gives the highest quantity and quality among three scenarios. ENFA was sensitive to the distribution scenarios but not sensitive to sample sizes. The use of a virtual species proved to be a very efficient method, allowing one to fully control the quality of the input data as well as to accurately evaluate the predictive power of the analyses.
Abstract: In this paper, the application of multiple Elman neural networks to time series data regression problems is studied. An ensemble of Elman networks is formed by boosting to enhance the performance of the individual networks. A modified version of the AdaBoost algorithm is employed to integrate the predictions from multiple networks. Two benchmark time series data sets, i.e., the Sunspot and Box-Jenkins gas furnace problems, are used to assess the effectiveness of the proposed system. The simulation results reveal that an ensemble of boosted Elman networks can achieve a higher degree of generalization as well as performance than that of the individual networks. The results are compared with those from other learning systems, and implications of the performance are discussed.
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: Clustering is a very well known technique in data mining. One of the most widely used clustering techniques is the k-means algorithm. Solutions obtained from this technique are dependent on the initialization of cluster centers. In this article we propose a new algorithm to initialize the clusters. The proposed algorithm is based on finding a set of medians extracted from a dimension with maximum variance. The algorithm has been applied to different data sets and good results are obtained.
Abstract: Sparse representation which can represent high dimensional
data effectively has been successfully used in computer vision
and pattern recognition problems. However, it doesn-t consider the
label information of data samples. To overcome this limitation,
we develop a novel dimensionality reduction algorithm namely
dscriminatively regularized sparse subspace learning(DR-SSL) in this
paper. The proposed DR-SSL algorithm can not only make use of
the sparse representation to model the data, but also can effective
employ the label information to guide the procedure of dimensionality
reduction. In addition,the presented algorithm can effectively deal
with the out-of-sample problem.The experiments on gene-expression
data sets show that the proposed algorithm is an effective tool for
dimensionality reduction and gene-expression data classification.
Abstract: Transcription factors are a group of proteins that
helps for interpreting the genetic information in DNA.
Protein-protein interactions play a major role in the execution
of key biological functions of a cell. These interactions are
represented in the form of a graph with nodes and edges.
Studies have showed that some nodes have high degree of
connectivity and such nodes, known as hub nodes, are the
inevitable parts of the network. In the present paper a method
is proposed to identify hub transcription factor proteins using
sequence information. On a complete data set of transcription
factor proteins available from the APID database, the
proposed method showed an accuracy of 77%, sensitivity of
79% and specificity of 76%.
Abstract: In this paper, the implementation of a rule-based
intuitive reasoner is presented. The implementation included two
parts: the rule induction module and the intuitive reasoner. A large
weather database was acquired as the data source. Twelve weather
variables from those data were chosen as the “target variables"
whose values were predicted by the intuitive reasoner. A “complex"
situation was simulated by making only subsets of the data available
to the rule induction module. As a result, the rules induced were
based on incomplete information with variable levels of certainty.
The certainty level was modeled by a metric called "Strength of
Belief", which was assigned to each rule or datum as ancillary
information about the confidence in its accuracy. Two techniques
were employed to induce rules from the data subsets: decision tree
and multi-polynomial regression, respectively for the discrete and the
continuous type of target variables. The intuitive reasoner was tested
for its ability to use the induced rules to predict the classes of the
discrete target variables and the values of the continuous target
variables. The intuitive reasoner implemented two types of
reasoning: fast and broad where, by analogy to human thought, the
former corresponds to fast decision making and the latter to deeper
contemplation. . For reference, a weather data analysis approach
which had been applied on similar tasks was adopted to analyze the
complete database and create predictive models for the same 12
target variables. The values predicted by the intuitive reasoner and
the reference approach were compared with actual data. The intuitive
reasoner reached near-100% accuracy for two continuous target
variables. For the discrete target variables, the intuitive reasoner
predicted at least 70% as accurately as the reference reasoner. Since
the intuitive reasoner operated on rules derived from only about 10%
of the total data, it demonstrated the potential advantages in dealing
with sparse data sets as compared with conventional methods.
Abstract: This paper reports work done to improve the modeling of complex processes when only small experimental data sets are available. Neural networks are used to capture the nonlinear underlying phenomena contained in the data set and to partly eliminate the burden of having to specify completely the structure of the model. Two different types of neural networks were used for the application of pulping problem. A three layer feed forward neural networks, using the Preconditioned Conjugate Gradient (PCG) methods were used in this investigation. Preconditioning is a method to improve convergence by lowering the condition number and increasing the eigenvalues clustering. The idea is to solve the modified odified problem M-1 Ax= M-1b where M is a positive-definite preconditioner that is closely related to A. We mainly focused on Preconditioned Conjugate Gradient- based training methods which originated from optimization theory, namely Preconditioned Conjugate Gradient with Fletcher-Reeves Update (PCGF), Preconditioned Conjugate Gradient with Polak-Ribiere Update (PCGP) and Preconditioned Conjugate Gradient with Powell-Beale Restarts (PCGB). The behavior of the PCG methods in the simulations proved to be robust against phenomenon such as oscillations due to large step size.