A Comparative Review of Various Machine Learning Approaches for Improving the Performance of Stego Anomaly Detection

In recent years, steganalyzers are intelligently detecting the stego images with high detection rate using high dimensional cover representation. And so the steganographers are working towards this issue to protect the cover element dependency and to protect the detection of hiding secret messages. Any steganalysis algorithm may achieve its success in two ways: 1) extracting the most sensitive features to expose the footprints of message hiding; 2) designing or building an effective classifier engine to favorably detect the stego images through learning all the stego sensitive features. In this chapter, the authors improve the stego anomaly detection using the second approach. This chapter presents a comparative review of application of the machine learning tools for steganalysis problem and recommends the best classifier that produces a superior detection rate.

Impact of Big Data on Security

Big data has great commercial importance to major businesses, but security and privacy challenges are also daunting this storage, processing, and communication. Big data encapsulate organizations' most important and sensitive data with multi-level complex implementation. The challenge for any organization is securing access to the data while allowing end user to extract valuable insights. Unregulated access privileges to the big data leads to loss or theft of valuable and sensitive. Privilege escalation leads to insider threats. Also, the computing architecture of big data is not focusing on session recording; therefore, it is becoming a challenge to identify potential security issues and to take remedial and mitigation mechanisms. Therefore, various big data security issues and their defense mechanisms are discussed in this chapter.

Impact of Big Data on Security

The motivation behind this chapter is to highlight the qualities, security issue, advantages, and disadvantages of big data. In the recent researches, the issue and challenges are due to the exponential growth of social media data and other images and videos. Big data security threats are rising, which is affecting the data heterogeneity adaptability and privacy preservation analytics. Big data analytics helps cyber security, but no new application can be envisioned without delivering new types of information, working on data-driven calculations and expending determined measure of information. This chapter demonstrates how innate attributes of big data are protected.

Detection of Botnet Based Attacks on Network

This chapter describes how with Botnets becoming more and more the leading cyber threat on the web nowadays, they also serve as the key platform for carrying out large-scale distributed attacks. Although a substantial amount of research in the fields of botnet detection and analysis, bot-masters inculcate new techniques to make them more sophisticated, destructive and hard to detect with the help of code encryption and obfuscation. This chapter proposes a new model to detect botnet behavior on the basis of traffic analysis and machine learning techniques. Traffic analysis behavior does not depend upon payload analysis so the proposed technique is immune to code encryption and other evasion techniques generally used by bot-masters. This chapter analyzes the benchmark datasets as well as real-time generated traffic to determine the feasibility of botnet detection using traffic flow analysis. Experimental results clearly indicate that a proposed model is able to classify the network traffic as a botnet or as normal traffic with a high accuracy and low false-positive rates.

Artificial Intelligence Based Intrusion Detection System to Detect Flooding Attack in VANETs

This chapter describes how Vehicular Ad hoc Networks (VANETs) are classes of ad hoc networks that provides communication among various vehicles and roadside units. VANETs being decentralized are susceptible to many security attacks. A flooding attack is one of the major security threats to the VANET environment. This chapter proposes a hybrid Intrusion Detection System which improves accuracy and other performance metrics using Artificial Neural Networks as a classification engine and a genetic algorithm as an optimization engine for feature subset selection. These performance metrics have been calculated in two scenarios, namely misuse and anomaly. Various performance metrics are calculated and compared with other researchers' work. The results obtained indicate a high accuracy and precision and negligible false alarm rate. These performance metrics are used to evaluate the intrusion system and compare with other existing algorithms. The classifier works well for multiple malicious nodes. Apart from machine learning techniques, the effect of the network parameters like throughput and packet delivery ratio is observed.

Flawed Security of Social Network of Things

This chapter deals with the security flaws of social network of things. The network of things (NoT) is a dynamic structure that is basically an interface of real world and virtual world having capabilities of collection and sharing data over a shared network. The social network of things (SNoT) is a versatile way of connecting virtual and real world. Like any other device connected to internet, objects in SNoT are also vulnerable to the various security and privacy attacks. Generally, to secure Social Network of Things in which human intervention is absent, data capturing devices must be avoided. Types of security attacks that are huge threats to NoT as well as SNoT will be discussed in the chapter. The huge collection of information without necessary security measures allows an intruder to misuse the personal data of owner. Different types of attacks with reference to the different layers are also discussed in detail. The best possible potential solutions for the security of devices in SNoT will be considered.

A Robust Biometrics System Using Finger Knuckle Print

Among various biometric indicators, hand-based biometrics has been widely used and deployed for last two decades. Hand-based biometrics are very popular because of their higher acceptance among the population because of their ease of use, high performance, less expensive, etc. This chapter presents a new hand-based biometric known as finger-knuckle-print (FKP) for a person authentication system. FKP are the images obtained from the one's fingers phalangeal joints and are characterized by internal skin pattern. Like other biometrics discrimination ability, FKP also has the capability of high discrimination. The proposed system consists of four modules: image acquisition, extraction of ROI, selection and extraction of features, and their matching. New features based on information theory are proposed for matching. The performance of the proposed system is evaluated using experiment performed on a database of 7920 images from 660 different fingers. The efficacy of the proposed system is evaluated in terms of matching rate and compromising results are obtained.

The Impact of Big Data Analytics and Challenges to Cyber Security

In order to scrutinize or evaluate an extremely high quantity of an ever-present and diversified nature of data, new technologies are developed. With the application of these technologies, called big data technologies, to the constantly developing various internal as well as external sources of data, concealed correlations between data can be identified, and promising strategies can be developed, which is necessary for economic growth and new innovations. This chapter deals with the analysis of the real-time uses of big data to both individual persons and the society too, while concentrating on seven important areas of key usage: big data for business optimization and customer analytics, big data and healthcare, big data and science, big data and finance, big data as enablers of openness and efficiency in government, big data and the emerging energy distribution systems, and big data security.

Successful Computer Forensics Analysis on the Cyber Attack Botnet

The success of computer forensics lies in the complete analysis of the evidence that is available. This is done by not only analyzing the evidence which is available but also searching for new concrete evidence. The evidence is obtained through the logs of the data during the cyberattack. When performing analysis of the cyberattack especially the botnet attacks, there are many challenges. First and the foremost is that it hides the identity of the mastermind, the botmaster. It issues the command to be executed using its subordinate, the command and control (C&C). The traceback of C&C itself is a complex task. Secondly, it victimizes the innocent compromised device zombies. This chapter discusses the analysis done in both proactive and reactive ways to resolve these challenges. The chapter ends by discussing the analysis to find the real mastermind to protect the innocent compromised system and to protect the victim system/organization affected by the botnet cyberattack.

Digital Forensics in Distributed Environment

This chapter is about an ongoing implementation of a digital forensic framework that could be used with standalone systems as well as in distributed environments, including cloud systems. It is oriented towards combining concepts of cyber forensics and security frameworks in operating systems. The framework consists of kernel mechanisms for data and event monitoring. The system monitoring is done in kernel mode by various kernel modules and forensic model mapping is done in user mode using the data collected by those kernel modules. Further, the authors propose a crime model mapping mechanism that makes use of rule sets that are derived from common cyber/digital crime patterns. The decision-making algorithm can be easily extended from a node in a computing cluster, to a cloud. The authors discuss the challenges to digital forensics in distributed environment and cloud extensions and provide some case studies where the proposed framework is applied.