No abstract

In the ongoing situation, the volume of information expands step by step. By the year 2020 the volume of Big Data would reach up to 40zb according to International Data Corporation (IDC). Big Data has turned out to be prevalent for handling, putting away and overseeing huge volumes of information. The grouping of datasets has turned into a testing issue in the field of Big Data examination; however, there are entanglements for applying conventional bunching calculations to huge information because of expanding the volume of information step by step. In this manuscript a new hybrid clustering algorithm, namely KCu to combine the features of both K-Means and CURE clustering algorithms is proposed. The proposed algorithm first applies k-means on data set and then applies CURE on resultant clusters from k-means. We experimented KCu and we show that, when compared to k-means and Cure. Which gives accurate results because of CURE? CURE can handle outliers and it gives non spherical shapes it is the disadvantage of other clustering algorithm.

Huge amount of personal data is collected by online applications and its protection based on privacy has brought a lot of major challenging issues. Hence, the [Formula: see text]-anonymization with privacy-preserving data publishing has emerged as an active research field. The published data contains personalized information, which may be used for analysis converting it to useful information. In this paper, Quasi identifier (QI) data publishing with data preservation through the [Formula: see text]-anonymization process is proposed. Moreover, the risks such as the temporal attack in the previous release of re-identifying QI information are evaluated using the [Formula: see text]-anonymity model. The development of independent and ensemble classifiers for finding efficient QI’s to avoid temporal attacks is the major objective of this paper. Therefore, the classifiers like Naïve Bayes, Support Vector Machine, and Multilayer Perceptron are used as base classifiers. An ensemble model based on these base classifiers is also used. The experimental results demonstrate that, the proposed classification approach is an effective K-anonymity tool for the enhancement of sequential release.

No abstract

In recent decades, remote sensing scene type classification becomes a challenging task in remote sensing applications. In this paper, a new model is proposed for multi-class scene type classification in remote sensing images. Firstly, the aerial images are collected from the Aerial Image Dataset (AID), University of California Merced (UC Merced) and REmote Sensing Image Scene Classification 45 (RESISC45) datasets. Next, AlexNet, GoogLeNet, ResNet 18, and Visual Geometric Group (VGG) 19 models are used for extracting feature vectors from the collected aerial images. After feature extraction, the Multi-Trial vector based Differential Evolution (MTDE) algorithm is proposed to choose active feature vectors for better classification and to reduce system complexity and time consumption. The selected active features are fed to the Multi Support Vector Machine (MSVM) for final scene type classification. The simulation results showed that the proposed MTDE-MSVM model obtained high classification accuracy of 99.41%, 99.59% and 99.74% on RESISC45, AID and UC Merced datasets.