A General Framework for Knowledge Discovery Using High Performance Machine Learning Algorithms

The aim of this paper is to propose a general framework for storing, analyzing, and extracting knowledge from two-dimensional echocardiographic images, color Doppler images, non-medical images, and general data sets. A number of high performance data mining algorithms have been used to carry out this task. Our framework encompasses four layers namely physical storage, object identification, knowledge discovery, user level. Techniques such as active contour model to identify the cardiac chambers, pixel classification to segment the color Doppler echo image, universal model for image retrieval, Bayesian method for classification, parallel algorithms for image segmentation, etc., were employed. Using the feature vector database that have been efficiently constructed, one can perform various data mining tasks like clustering, classification, etc. with efficient algorithms along with image mining given a query image. All these facilities are included in the framework that is supported by state-of-the-art user interface (UI). The algorithms were tested with actual patient data and Coral image database and the results show that their performance is better than the results reported already.

Epileptic Seizure Prediction by Exploiting Signal Transitions Phenomena

A seizure prediction method is proposed by extracting global features using phase correlation between adjacent epochs for detecting relative changes and local features using fluctuation/ deviation within an epoch for determining fine changes of different EEG signals. A classifier and a regularization technique are applied for the reduction of false alarms and improvement of the overall prediction accuracy. The experiments show that the proposed method outperforms the state-of-the-art methods and provides high prediction accuracy (i.e., 97.70%) with low false alarm using EEG signals in different brain locations from a benchmark data set.

Variational EM Inference Algorithm for Gaussian Process Classification Model with Multiclass and Its Application to Human Action Classification

In this paper, we propose the variational EM inference algorithm for the multi-class Gaussian process classification model that can be used in the field of human behavior recognition. This algorithm can drive simultaneously both a posterior distribution of a latent function and estimators of hyper-parameters in a Gaussian process classification model with multiclass. Our algorithm is based on the Laplace approximation (LA) technique and variational EM framework. This is performed in two steps: called expectation and maximization steps. First, in the expectation step, using the Bayesian formula and LA technique, we derive approximately the posterior distribution of the latent function indicating the possibility that each observation belongs to a certain class in the Gaussian process classification model. Second, in the maximization step, using a derived posterior distribution of latent function, we compute the maximum likelihood estimator for hyper-parameters of a covariance matrix necessary to define prior distribution for latent function. These two steps iteratively repeat until a convergence condition satisfies. Moreover, we apply the proposed algorithm with human action classification problem using a public database, namely, the KTH human action data set. Experimental results reveal that the proposed algorithm shows good performance on this data set.

Modeling and Simulation of Acoustic Link Using Mackenize Propagation Speed Equation

Underwater acoustic networks have attracted great attention in the last few years because of its numerous applications. High data rate can be achieved by efficiently modeling the physical layer in the network protocol stack. In Acoustic medium, propagation speed of the acoustic waves is dependent on many parameters such as temperature, salinity, density, and depth. Acoustic propagation speed cannot be modeled using standard empirical formulas such as Urick and Thorp descriptions. In this paper, we have modeled the acoustic channel using real time data of temperature, salinity, and speed of Bay of Bengal (Indian Coastal Region). We have modeled the acoustic channel by using Mackenzie speed equation and real time data obtained from National Institute of Oceanography and Technology. It is found that acoustic propagation speed varies between 1503 m/s to 1544 m/s as temperature and depth differs. The simulation results show that temperature, salinity, depth plays major role in acoustic propagation and data rate increases with appropriate data sets substituted in the simulated model.

Comparison of Polynomial and Radial Basis Kernel Functions based SVR and MLR in Modeling Mass Transfer by Vertical and Inclined Multiple Plunging Jets

Presently various computational techniques are used in modeling and analyzing environmental engineering data. In the present study, an intra-comparison of polynomial and radial basis kernel functions based on Support Vector Regression and, in turn, an inter-comparison with Multi Linear Regression has been attempted in modeling mass transfer capacity of vertical (θ = 90O) and inclined (θ multiple plunging jets (varying from 1 to 16 numbers). The data set used in this study consists of four input parameters with a total of eighty eight cases, forty four each for vertical and inclined multiple plunging jets. For testing, tenfold cross validation was used. Correlation coefficient values of 0.971 and 0.981 along with corresponding root mean square error values of 0.0025 and 0.0020 were achieved by using polynomial and radial basis kernel functions based Support Vector Regression respectively. An intra-comparison suggests improved performance by radial basis function in comparison to polynomial kernel based Support Vector Regression. Further, an inter-comparison with Multi Linear Regression (correlation coefficient = 0.973 and root mean square error = 0.0024) reveals that radial basis kernel functions based Support Vector Regression performs better in modeling and estimating mass transfer by multiple plunging jets.

Spatio-Temporal Data Mining with Association Rules for Lake Van

People, throughout the history, have made estimates and inferences about the future by using their past experiences. Developing information technologies and the improvements in the database management systems make it possible to extract useful information from knowledge in hand for the strategic decisions. Therefore, different methods have been developed. Data mining by association rules learning is one of such methods. Apriori algorithm, one of the well-known association rules learning algorithms, is not commonly used in spatio-temporal data sets. However, it is possible to embed time and space features into the data sets and make Apriori algorithm a suitable data mining technique for learning spatiotemporal association rules. Lake Van, the largest lake of Turkey, is a closed basin. This feature causes the volume of the lake to increase or decrease as a result of change in water amount it holds. In this study, evaporation, humidity, lake altitude, amount of rainfall and temperature parameters recorded in Lake Van region throughout the years are used by the Apriori algorithm and a spatio-temporal data mining application is developed to identify overflows and newlyformed soil regions (underflows) occurring in the coastal parts of Lake Van. Identifying possible reasons of overflows and underflows may be used to alert the experts to take precautions and make the necessary investments.

Online Optic Disk Segmentation Using Fractals

Optic disk segmentation plays a key role in the mass screening of individuals with diabetic retinopathy and glaucoma ailments. An efficient hardware-based algorithm for optic disk localization and segmentation would aid for developing an automated retinal image analysis system for real time applications. Herein, TMS320C6416DSK DSP board pixel intensity based fractal analysis algorithm for an automatic localization and segmentation of the optic disk is reported. The experiment has been performed on color and fluorescent angiography retinal fundus images. Initially, the images were pre-processed to reduce the noise and enhance the quality. The retinal vascular tree of the image was then extracted using canny edge detection technique. Finally, a pixel intensity based fractal analysis is performed to segment the optic disk by tracing the origin of the vascular tree. The proposed method is examined on three publicly available data sets of the retinal image and also with the data set obtained from an eye clinic. The average accuracy achieved is 96.2%. To the best of the knowledge, this is the first work reporting the use of TMS320C6416DSK DSP board and pixel intensity based fractal analysis algorithm for an automatic localization and segmentation of the optic disk. This will pave the way for developing devices for detection of retinal diseases in the future.

Is HR in a State of Transition? An International Comparative Study on the Development of HR Competencies

Research Objectives: The roles and activities of Human Resource Management (HRM) have changed a lot in the past years. Driven by a changing environment and therefore new business requirements, the scope of human resource (HR) activities has widened. The extent to which these activities should focus on strategic issues to support the long term success of a company has been discussed in science for many years. As many economies of Central and Eastern Europe (CEE) experienced a phase of transition after the socialist era and are now recovering from the 2008 global crisis it is needed to examine the current state of HR positioning. Furthermore a trend in HR work developing from rather administrative units to being strategic partners of management can be noticed. This leads to the question of better understanding the underlying competencies which are necessary to support organisations. This topic was addressed by the international study “HR Competencies in international comparison”. The quantitative survey was conducted by the Institute for Human Resources & Organisation of FHWien University of Applied Science of WKW (A) in cooperation with partner universities in the countries Bosnia- Herzegovina, Croatia, Serbia and Slovenia. Methodology: Using the questionnaire developed by Dave Ulrich we tested whether the HR Competency model can be used for Austria, Bosnia and Herzegovina, Croatia, Serbia and Slovenia. After performing confirmatory and exploratory factor analysis for the whole data set containing all five countries we could clearly distinguish between four competencies. In a further step our analysis focused on median and average comparisons between the HR competency dimensions. Conclusion: Our literature review, in alignment with other studies, shows a relatively rapid pace of development of HR Roles and HR Competencies in BCSS in the past decades. Comparing data from BCSS and Austria we still can notice that regards strategic orientation there is a lack in BCSS countries, thus competencies are not as developed as in Austria. This leads us to the tentative conclusion that HR has undergone a rapid change but is still in a State of Transition from being a rather administrative unit to performing the role of a strategic partner.

Studies of Rule Induction by STRIM from the Decision Table with Contaminated Attribute Values from Missing Data and Noise — In the Case of Critical Dataset Size —

STRIM (Statistical Test Rule Induction Method) has been proposed as a method to effectively induct if-then rules from the decision table which is considered as a sample set obtained from the population of interest. Its usefulness has been confirmed by simulation experiments specifying rules in advance, and by comparison with conventional methods. However, scope for future development remains before STRIM can be applied to the analysis of real-world data sets. The first requirement is to determine the size of the dataset needed for inducting true rules, since finding statistically significant rules is the core of the method. The second is to examine the capacity of rule induction from datasets with contaminated attribute values created by missing data and noise, since real-world datasets usually contain such contaminated data. This paper examines the first problem theoretically, in connection with the rule length. The second problem is then examined in a simulation experiment, utilizing the critical size of dataset derived from the first step. The experimental results show that STRIM is highly robust in the analysis of datasets with contaminated attribute values, and hence is applicable to real-world data

The Classification Performance in Parametric and Nonparametric Discriminant Analysis for a Class- Unbalanced Data of Diabetes Risk Groups

The problems arising from unbalanced data sets generally appear in real world applications. Due to unequal class distribution, many researchers have found that the performance of existing classifiers tends to be biased towards the majority class. The k-nearest neighbors’ nonparametric discriminant analysis is a method that was proposed for classifying unbalanced classes with good performance. In this study, the methods of discriminant analysis are of interest in investigating misclassification error rates for classimbalanced data of three diabetes risk groups. The purpose of this study was to compare the classification performance between parametric discriminant analysis and nonparametric discriminant analysis in a three-class classification of class-imbalanced data of diabetes risk groups. Data from a project maintaining healthy conditions for 599 employees of a government hospital in Bangkok were obtained for the classification problem. The employees were divided into three diabetes risk groups: non-risk (90%), risk (5%), and diabetic (5%). The original data including the variables of diabetes risk group, age, gender, blood glucose, and BMI were analyzed and bootstrapped for 50 and 100 samples, 599 observations per sample, for additional estimation of the misclassification error rate. Each data set was explored for the departure of multivariate normality and the equality of covariance matrices of the three risk groups. Both the original data and the bootstrap samples showed nonnormality and unequal covariance matrices. The parametric linear discriminant function, quadratic discriminant function, and the nonparametric k-nearest neighbors’ discriminant function were performed over 50 and 100 bootstrap samples and applied to the original data. Searching the optimal classification rule, the choices of prior probabilities were set up for both equal proportions (0.33: 0.33: 0.33) and unequal proportions of (0.90:0.05:0.05), (0.80: 0.10: 0.10) and (0.70, 0.15, 0.15). The results from 50 and 100 bootstrap samples indicated that the k-nearest neighbors approach when k=3 or k=4 and the defined prior probabilities of non-risk: risk: diabetic as 0.90: 0.05:0.05 or 0.80:0.10:0.10 gave the smallest error rate of misclassification. The k-nearest neighbors approach would be suggested for classifying a three-class-imbalanced data of diabetes risk groups.

New Hybrid Method to Model Extreme Rainfalls

Modeling and forecasting dynamics of rainfall occurrences constitute one of the major topics, which have been largely treated by statisticians, hydrologists, climatologists and many other groups of scientists. In the same issue, we propose, in the present paper, a new hybrid method, which combines Extreme Values and fractal theories. We illustrate the use of our methodology for transformed Emberger Index series, constructed basing on data recorded in Oujda (Morocco). The index is treated at first by Peaks Over Threshold (POT) approach, to identify excess observations over an optimal threshold u. In the second step, we consider the resulting excess as a fractal object included in one dimensional space of time. We identify fractal dimension by the box counting. We discuss the prospect descriptions of rainfall data sets under Generalized Pareto Distribution, assured by Extreme Values Theory (EVT). We show that, despite of the appropriateness of return periods given by POT approach, the introduction of fractal dimension provides accurate interpretation results, which can ameliorate apprehension of rainfall occurrences.

STATISTICA Software: A State of the Art Review

Data mining idea is mounting rapidly in admiration and also in their popularity. The foremost aspire of data mining method is to extract data from a huge data set into several forms that could be comprehended for additional use. The data mining is a technology that contains with rich potential resources which could be supportive for industries and businesses that pay attention to collect the necessary information of the data to discover their customer’s performances. For extracting data there are several methods are available such as Classification, Clustering, Association, Discovering, and Visualization… etc., which has its individual and diverse algorithms towards the effort to fit an appropriate model to the data. STATISTICA mostly deals with excessive groups of data that imposes vast rigorous computational constraints. These results trials challenge cause the emergence of powerful STATISTICA Data Mining technologies. In this survey an overview of the STATISTICA software is illustrated along with their significant features.

An ANN-Based Predictive Model for Diagnosis and Forecasting of Hypertension

The effects of hypertension are often lethal thus its early detection and prevention is very important for everybody. In this paper, a neural network (NN) model was developed and trained based on a dataset of hypertension causative parameters in order to forecast the likelihood of occurrence of hypertension in patients. Our research goal was to analyze the potential of the presented NN to predict, for a period of time, the risk of hypertension or the risk of developing this disease for patients that are or not currently hypertensive. The results of the analysis for a given patient can support doctors in taking pro-active measures for averting the occurrence of hypertension such as recommendations regarding the patient behavior in order to lower his hypertension risk. Moreover, the paper envisages a set of three example scenarios in order to determine the age when the patient becomes hypertensive, i.e. determine the threshold for hypertensive age, to analyze what happens if the threshold hypertensive age is set to a certain age and the weight of the patient if being varied, and, to set the ideal weight for the patient and analyze what happens with the threshold of hypertensive age.

Influential Parameters in Estimating Soil Properties from Cone Penetrating Test: An Artificial Neural Network Study

The Cone Penetration Test (CPT) is a common in-situ test which generally investigates a much greater volume of soil more quickly than possible from sampling and laboratory tests. Therefore, it has the potential to realize both cost savings and assessment of soil properties rapidly and continuously. The principle objective of this paper is to demonstrate the feasibility and efficiency of using artificial neural networks (ANNs) to predict the soil angle of internal friction (Φ) and the soil modulus of elasticity (E) from CPT results considering the uncertainties and non-linearities of the soil. In addition, ANNs are used to study the influence of different parameters and recommend which parameters should be included as input parameters to improve the prediction. Neural networks discover relationships in the input data sets through the iterative presentation of the data and intrinsic mapping characteristics of neural topologies. General Regression Neural Network (GRNN) is one of the powerful neural network architectures which is utilized in this study. A large amount of field and experimental data including CPT results, plate load tests, direct shear box, grain size distribution and calculated data of overburden pressure was obtained from a large project in the United Arab Emirates. This data was used for the training and the validation of the neural network. A comparison was made between the obtained results from the ANN's approach, and some common traditional correlations that predict Φ and E from CPT results with respect to the actual results of the collected data. The results show that the ANN is a very powerful tool. Very good agreement was obtained between estimated results from ANN and actual measured results with comparison to other correlations available in the literature. The study recommends some easily available parameters that should be included in the estimation of the soil properties to improve the prediction models. It is shown that the use of friction ration in the estimation of Φ and the use of fines content in the estimation of E considerable improve the prediction models.

Evaluation of Model Evaluation Criterion for Software Development Effort Estimation

Estimation of model parameters is necessary to predict the behavior of a system. Model parameters are estimated using optimization criteria. Most algorithms use historical data to estimate model parameters. The known target values (actual) and the output produced by the model are compared. The differences between the two form the basis to estimate the parameters. In order to compare different models developed using the same data different criteria are used. The data obtained for short scale projects are used here. We consider software effort estimation problem using radial basis function network. The accuracy comparison is made using various existing criteria for one and two predictors. Then, we propose a new criterion based on linear least squares for evaluation and compared the results of one and two predictors. We have considered another data set and evaluated prediction accuracy using the new criterion. The new criterion is easy to comprehend compared to single statistic. Although software effort estimation is considered, this method is applicable for any modeling and prediction.

Imputation Technique for Feature Selection in Microarray Data Set

Analyzing DNA microarray data sets is a great challenge, which faces the bioinformaticians due to the complication of using statistical and machine learning techniques. The challenge will be doubled if the microarray data sets contain missing data, which happens regularly because these techniques cannot deal with missing data. One of the most important data analysis process on the microarray data set is feature selection. This process finds the most important genes that affect certain disease. In this paper, we introduce a technique for imputing the missing data in microarray data sets while performing feature selection.

Retrospective Reconstruction of Time Series Data for Integrated Waste Management

The development, operation and maintenance of Integrated Waste Management Systems (IWMS) affects essentially the sustainable concern of every region. The features of such systems have great influence on all of the components of sustainability. In order to reach the optimal way of processes, a comprehensive mapping of the variables affecting the future efficiency of the system is needed such as analysis of the interconnections among the components and modeling of their interactions. The planning of a IWMS is based fundamentally on technical and economical opportunities and the legal framework. Modeling the sustainability and operation effectiveness of a certain IWMS is not in the scope of the present research. The complexity of the systems and the large number of the variables require the utilization of a complex approach to model the outcomes and future risks. This complex method should be able to evaluate the logical framework of the factors composing the system and the interconnections between them. The authors of this paper studied the usability of the Fuzzy Cognitive Map (FCM) approach modeling the future operation of IWMS’s. The approach requires two input data set. One is the connection matrix containing all the factors affecting the system in focus with all the interconnections. The other input data set is the time series, a retrospective reconstruction of the weights and roles of the factors. This paper introduces a novel method to develop time series by content analysis.

A Tree Based Association Rule Approach for XML Data with Semantic Integration

The use of eXtensible Markup Language (XML) in web, business and scientific databases lead to the development of methods, techniques and systems to manage and analyze XML data. Semi-structured documents suffer due to its heterogeneity and dimensionality. XML structure and content mining represent convergence for research in semi-structured data and text mining. As the information available on the internet grows drastically, extracting knowledge from XML documents becomes a harder task. Certainly, documents are often so large that the data set returned as answer to a query may also be very big to convey the required information. To improve the query answering, a Semantic Tree Based Association Rule (STAR) mining method is proposed. This method provides intentional information by considering the structure, content and the semantics of the content. The method is applied on Reuter’s dataset and the results show that the proposed method outperforms well.

Using Data Mining in Automotive Safety

Safety is one of the most important considerations when buying a new car. While active safety aims at avoiding accidents, passive safety systems such as airbags and seat belts protect the occupant in case of an accident. In addition to legal regulations, organizations like Euro NCAP provide consumers with an independent assessment of the safety performance of cars and drive the development of safety systems in automobile industry. Those ratings are mainly based on injury assessment reference values derived from physical parameters measured in dummies during a car crash test. The components and sub-systems of a safety system are designed to achieve the required restraint performance. Sled tests and other types of tests are then carried out by car makers and their suppliers to confirm the protection level of the safety system. A Knowledge Discovery in Databases (KDD) process is proposed in order to minimize the number of tests. The KDD process is based on the data emerging from sled tests according to Euro NCAP specifications. About 30 parameters of the passive safety systems from different data sources (crash data, dummy protocol) are first analysed together with experts opinions. A procedure is proposed to manage missing data and validated on real data sets. Finally, a procedure is developed to estimate a set of rough initial parameters of the passive system before testing aiming at reducing the number of tests.

SQL Generator Based On MVC Pattern

Structured Query Language (SQL) is the standard de facto language to access and manipulate data in a relational database. Although SQL is a language that is simple and powerful, most novice users will have trouble with SQL syntax. Thus, we are presenting SQL generator tool which is capable of translating actions and displaying SQL commands and data sets simultaneously. The tool was developed based on Model-View-Controller (MVC) pattern. The MVC pattern is a widely used software design pattern that enforces the separation between the input, processing, and output of an application. Developers take full advantage of it to reduce the complexity in architectural design and to increase flexibility and reuse of code. In addition, we use White-Box testing for the code verification in the Model module.