Semi-supervised learning with the clustering and Decision Trees classifier for the task of cognitive workload study

The paper is focused on application of the clustering algorithm and Decision Tress classifier (DTs) as a semi-supervised method for the task of cognitive workload level classification. The analyzed data were collected during examination of Digit Symbol Substitution Test (DSST) with use of eye-tracker device. 26 participants took part in examination as volunteers. There were conducted three parts of DSST test with different levels of difficulty. As a results there were obtained three versions of data: low, middle and high level of cognitive workload. The case study covered clustering of collected data by using k-means algorithm to detect three clusters or more. The obtained clusters were evaluated by three internal indices to measure the quality of clustering. The David-Boudin index detected the best results in case of four clusters. Based on this information it is possible to formulate the hypothesis of the existence of four clusters. The obtained clusters were adopted as classes in supervised learning and have been subjected to classification. The DTs was applied in classification. There were obtained the 0.85 mean accuracy for three-class classification and 0.73 mean accuracy for four-class classification.

Diabetes prediction using Data mining Classification Techniques

Diabetes is one of the second largest disease in the world. In the recent survey it shows that there are overall 246 million people affected with this and in that women ratio is more. By the report of WHO, this figure is going to reach to 380 million by 2025. According to the American Diabetes Association,6% of the population are not aware that there are victims of diabetes and also every 21 sec at least for an individual diabetic test result is positive. With the technology advancement in the field of medical information, data is well maintained in the databases. This paper focuses on to diagnose data to provide the solution by observing the patterns in the data using various datamining classification techniques such as Naïve basis, Logistic regression, Decision tress etc

Web-of-Service Software Reusability Prediction using Heterogenous Ensemble Classifier

WoS computing environment is expected to have numerous parallel computing engines. Presently, software professionals or developers often want to reuse existing software components to exhibit a task with time-efficient and cost effective solutions. However, software component reusability in uncontrolled manner leads to failure, premature shutdown and software smells or aging. This paper develops a novel evolutionary computing assisted ensemble classification system for WoS software reusability prediction. This applies different base learners such asNaïve Bayes (NB), Linear Regression (LR), Decision Tress (DT),Logarithmic Regression (LOGR),and Support Vector Machine (SVM),Multivariate Adaptive Regression Spline (MARS). Once training the base learners, the outputs of each classifier have been processed with majority vote.The computation in conjunction with weighted sum enabled final labelling of each software class. The performance results affirmed that the present work ensemble classifier has better performance with respect to base classifiers.

Cyber Space Security Assessment Case Study

Cyberspace is known as the digital electronic medium for the knowing range of securing in the cyberspace. Therefore the importance of inferring the reference measure in the form of assessment procedure to improve the knowledge and making the decision for the e- government services. A series of the standards build on the application of data mining methods specifically represented as decision tress model, Logistic regression, association rules model, Bayesian network for making reference measurements, to measure the extent of securing the data, and the provided services. The authors discuss various types of cyber-attacks describing how data mining helps in detection and prevention of these attacks. A comparative analysis between a set of selected frameworks is presented. Finally this chapter imparts numbers of applications for the data mining Methodologies in Cyber Security. Results applied on the site of the authority for cleaning and beautifying Cairo governorate in Egypt.

Cyber Space Security Assessment Case Study

Cyberspace is known as the digital electronic medium for the knowing range of securing in the cyberspace. Therefore the importance of inferring the reference measure in the form of assessment procedure to improve the knowledge and making the decision for the e- government services. A series of the standards build on the application of data mining methods specifically represented as decision tress model, Logistic regression, association rules model, Bayesian network for making reference measurements, to measure the extent of securing the data, and the provided services. The authors discuss various types of cyber-attacks describing how data mining helps in detection and prevention of these attacks. A comparative analysis between a set of selected frameworks is presented. Finally this chapter imparts numbers of applications for the data mining Methodologies in Cyber Security. Results applied on the site of the authority for cleaning and beautifying Cairo governorate in Egypt.