Search

Tree-based learning techniques are considered to be one of the best and most used supervised learning methods. Tree-based methods empower predictive models with high accuracy, stability, and ease of interpretation. Unlike linear models, they map non-linear relationships pretty well. These methods are adaptable at solving any kind of problem at hand (classification or regression). Methods like decision trees, random forest, gradient boosting are being widely used in all kinds of machine learning and data science problems. Hence, for every data analyst, it is important to learn these algorithms and use them for modeling. This chapter guide the learner to learn tree-based modeling techniques from scratch.

Our world is increasingly driven by sophisticated networks of advanced computing technology, and the basic operation of everyday society is becoming increasingly vulnerable to those networks shortcomings. The implementation and upkeep of a strong network defense is a substantial challenge, beset not only by economic disincentives, but also by an inherent logistical bias that grants advantage to attackers.Network Security Attacks and Countermeasures discusses the security and optimization of computer networks for use in a variety of disciplines and fields. Touching on such matters as mobile and VPN security, IP spoofing, and intrusion detection, this edited collection emboldens the efforts of researchers, academics, and network administrators working in both the public and private sectors.This edited compilation includes chapters covering topics such as attacks and countermeasures, mobile wireless networking, intrusion detection systems, next-generation firewalls, and more.

Big Data is a relatively recent phenomenon, but has already shown its potential to drastically alter the relationships between businesses, individuals, and governments. The issues surrounding privacy of the online users (Mayer-Shoenberger, Cukier 2013) and the overall ethical challenges involved (Schroeder, 2014) make big data a topical issue, especially in the aftermath of the Snowden revelations. Many organisations now control vast amounts of raw data, and those industry players with the resources to mine that data to create new information have a significant advantage in the big data market. The use of predictive analytics in processing information tracked across different platforms to identify trends in the behaviour of individuals further adds value to big data (Fotopoulou, 2014) and makes it an important asset for any commercial entity. This rapid commodification of personal data has given rise to a new approach with regard its legal protection in the era of big data: a shift from the traditional privacy protection regime to a wider protection under property law is considered by scholars as an appropriate legal response to the phenomenon of monetisation of personal data, once seen through the lens of big data (Victor, 2013).\ud \ud The aim of this chapter is to identify the legal grounds for the ownership of big data: who legally owns the petabytes and exabytes of information created daily? Does this belong to the users, the data analysts or to the data brokers and various infomediaries? The chapter presents a succinct overview of the legal ownership of big data by examining the key players in control of the information at each stage of the processing of big data. It then moves on to describe the current legislative framework with regard to data protection and concludes in additional techno-legal solutions offered to complement the law of big data in this respect, with a particular focus on the European context .