“The curse of dimensionality ” is pertinent to many learning algorithms, and it denotes the drastic raise of computational complexity and classification error in high dimensions. In this paper, different feature extraction (FE) techniques are analyzed as means of dimensionality reduction, and constructive induction with respect to the performance of Naïve Bayes classifier. When a data set contains a large number of instances, some sampling approach is applied to address the computational complexity of FE and classification processes. The main goal of this paper is to show the impact of sample reduction on the process of FE for supervised learning. In our study we analyzed the conventional PCA and two eigenvector-based approaches that take into account class information. The first class-conditional approach is parametric and optimizes the ratio of between-class variance to the within-class variance of the transformed data. The second approach is a nonparametric modification of the first one based on the local calculation of the between-class covariance matrix. The experiments are conducted on ten UCI data sets, using four different strategies to select samples: (1) random sampling, (2) stratified random sampling, (3) kd-tree based selective sampling, and (4) stratified sampling with kd-tree based selection. Our experiments show that if the sample size for FE model construction is small then it is important to take into account both class information and data distribution. Further, for supervised learning the nonparametric FE approach needs much less instances to produce a new representation space that result in the same or higher classification accuracy than the other FE approaches.

Cyber-Physical Systems (CPS) are sensing, communication and processing platforms, deeply embedded in physical processes and provide real-time monitoring and actuation services. Such systems are becoming increasing common in enabling many of the pervasive computing technologies that are becoming available today such as, smart-homes, smart-vehicles, pervasive health monitoring systems. Given the automation that CPSs introduce in managing physical processes, and the detail of information available to them for carrying out their tasks, securing them is of prime importance. In this dissertation, a novel security paradigm for CPSs is proposed, called Cyber-Physical Security (CYPSec). CYPSec solutions are unique in that they take they take into account the environmentally-coupled nature of CPSs in enabling security solutions. This dissertation explores CYPSec solutions for two diverse but related problems. The first is a usable and secure key agreement protocol called Physiological Signal based Key Agreement (PSKA), which combines signal processing and cryptographic primitives to enable automated key agreement between sensors in a Body Area Network (BAN) without any form of external user involvement. It uses specific physiological stimuli-based features (Photoplethsymogram and Electrocardiogram) from the human body for its task. The second is an access control model called Criticality Aware Access Control (CAAC), which facilitates a more adaptive and proactive provisioning of authorizations-

Abstract—Research in the fields of software quality, maintainability requires the analysis of large quantity of data, which originate from software projects. It is a challenging task in pre-processing the data and synthesizing the composite results. It is very often an error prone task. Data mining techniques are generally considered as the best for preprocessing data. But there may be case that data is not up to the mark for that, in this context an improvised core tool will be used to facilitate software engineering research on large and diverse data set.There may be the chance of not getting the proper result for boosting up the software industry, but upto certain extent it will be surely helpful in research field or to the researchers for further innovative development.

Abstract not found