Social sensing is one of the newest concepts of the Internet, fueled by the growth of sensors. The increasing need for communication and connectivity while one is on the move makes up a very important characteristic of social sensing. However, this need is opposed to the need to provide security services such as integrity, confidentiality, privacy, and reliability.

This book offers solutions in this direction, combining social sensing analytics with machine learning, information fusion, data mining, and statistics.

In a clear and concise style, with well-founded and exemplified pieces of information, its 12 chapters allow the reader to identify and analyze a range of solutions regarding social sensing problems. The book is recommended both to cyber-physical systems researchers and sensor network researchers, but also to people involved in business analytics. After finishing this book, readers will better understand the fundamental concepts related to the subject and also receive new interpretations of, and solutions to, the problems encountered.

Chapter 1 presents an overview and a concise state of the art, which familiarize the reader with basic knowledge regarding social sensing. The authors also examine the underlying technical enablers and motivations of social sensing applications. A review of social sensing trends and its applications is made available to readers in chapter 2. Social data scavenging is also presented in a broader context, which is stimulated by the increasing popularity of online social networks and the growth of opportunities for the propagation of information.

In chapter 3, the authors go through the mathematical foundations and the principal technologies that will help the reader fully understand the next chapters. Methods of calculating probabilities, which represent a measure of quality of information (QoI), are presented in chapter 4. The next three chapters focus mainly on the maximum likelihood estimation (MLE) model, algorithms, its capabilities and limitations, and a generalization of it. The authors consolidate the theory by bringing up real-world examples of applications and simulations, which are very well explained, analyzed, and richly illustrated.

Recent work about the underlying information dissemination topology focusing on solving reliability-related social sensing problems is addressed in chapter 8. A real-world case study using the Twitter platform is also reviewed in this chapter. The book continues with an exploration of the physical dependencies between observed variables, presenting the cyber-physical approach. The end of chapter 9 presents another example that involves this approach in a crowd-sensing application.

Chapter 10 presents work in recursive fact-finding and real-time streaming data challenges in the area. It also explains the expectation maximization (EM) algorithm, which is shown to achieve a high performance tradeoff between estimation accuracy and its execution time. Areas such as estimation theory, data quality, trust analysis, outlier and attack detection, recommender systems, and survey opinion polling are fully introduced to the reader in chapter 11. A final contribution, represented within a summary of the theories, techniques, and methods presented, is made in the last chapter. This is a valuable opportunity for the audience to reread and review the principal problems and solutions of this book.

To sum up, the book will be a real challenge for readers, both in the main topic addressed, social sensing, and the beneficial way in which the authors discuss it and contribute to a better understanding of it.