No abstract

No abstract

The world has been drowned by floods of data due to technological development. Consequently, the Big Data term has gotten the expression to portray the gigantic sum. Different sorts of quick data are doubling every second. We have to profit fro m this enormous surge of data to convert it to knowledge. Knowledge Discovery (KDD) can enhance detecting the value of Big Data based on some techniques and technologies like Hadoop, MapReduce, and NoSQL. The use of Big Data value is critical in different fields. This survey discusses the expansion of data that led the world to Big Data exp ression. Big Data has distinctive characteristics as volume, variety, velocity, value, veracity, variab ility, viscosity, virality, amb iguity, and complexity. We will describe the connection between Big Data and KDD techniques to reach data value. Big Data applications that are applied by big organizat ions will be discussed. Characteristics of big data will be introduced, which represent a significant challenge for Big Data management. Finally, some of the impo rtant future directions in Big Data field will be presented.

Currently smartphone' users run many crucial applications (such as banking and emails) which contains a very confidential information. To secure this information, the built in sensors equipped with smartphone devices can be utilized. In this paper, based on these sensors, an implicit authentication system for smartphone's users is proposed. A mobile App is developed to collect the data source of users' biometrics and then features (pressure, position, size, and time) are extracted. classifiers were then applied to decide whether a user is the true owner of device or an impostor. The experimental results showed that our implicit authentication system achieved accuracy of 96.5 % which is better than a related work.

Abstract-Intrusion detection systems (IDS) are gaining attention as network technologies are vastly growing. Most of the research in this field focuses on improving the performance of these systems through various feature selection techniques along with using ensembles of classifiers. An orthogonal problem is to estimate the proper sample sizes to train those classifiers. While this problem has been considered in other disciplines, mainly medical and biological, to study the relation between the sample size and the classifiers accuracy, it has not received a similar attention in the context of intrusion detection as far as we know.In this paper we focus on systems based on Naï ve Bayes classifiers and investigate the effect of the training sample size on the classification performance for the imbalanced NSL-KDD intrusion dataset. In order to estimate the appropriate sample size required to achieve a required classification performance, we constructed the learning curve of the classifier for individual classes in the dataset. For this construction we performed nonlinear least squares curve fitting using two different power law models. Results showed that while the shifted power law outperforms the power law model in terms of fitting performance, it exhibited a poor prediction performance. The power law, on the other hand, showed a significantly better prediction performance for larger sample sizes.