Bøger i SpringerBriefs in Computer Science serien

This book presents the fundamentals of exception handling with examples written in C++ and Python. Starting with its history and evolution, it explores the many facets of exception handling, such as its syntax, semantics, challenges, best practices, and implementation patterns.The book is composed of five chapters: Chapter 1 provides an introduction, covering the history, various definitions, and challenges of exception handling. Chapter 2 then delves into the basics, offering insights into the foundational concepts and techniques. Subsequently, chapter 3 touches upon the best practices for exception handling, including the differences between errors and exceptions, the use of assertions, and how to provide meaningful error messages. Chapter 4 takes a deep dive into advanced exception-handling techniques, exploring e.g. patterns, guard clauses, and hierarchical exception handling. Eventually, chapter 5 focuses on the complexities of exception handling in real-time and embedded systems.This book is mainly written for both students and professionals. Its readers will understand the nuances between syntax and semantic errors, learn how to employ try-catch blocks effectively, grasp the importance of logging exceptions, and delve into advanced exception-handling techniques. This way, they will be enabled to handle exceptions effectively and thus write more robust, reliable, and resilient code.

This book provides a comprehensive and systematic exploration of next-generation Edge Intelligence (EI) Networks. It delves deep into the critical design considerations within this context, emphasizing the necessity for functional and dependable interactions between networking strategies and the diverse application scenarios. This should help assist to encompass a wide range of environments.This book also discusses topics such as resource optimization, incentive mechanisms, channel prediction and cutting-edge technologies, which includes digital twins and advanced machine learning techniques. It underscores the importance of functional integration to facilitate meaningful collaborations between networks and systems, while operating across heterogeneous environments aiming support novel and disruptive human-oriented services and applications. Valuable insights into the stringent requirements for intelligence capabilities, communication latency and real-time response are discussed. This characterizes the new EI era, driving the creation of comprehensive cross-domain architectural ecosystems that infuse human-like intelligence into every aspect of emerging EI systems.This book primarily targets advanced-level students as well as postdoctoral researchers, who are new to this field and are searching for a comprehensive understanding of emerging EI systems. Practitioners seeking guidance in the development and implementation of EI systems in practical contexts will also benefit from this book.

This book offers an overview of the field of continuous biometric authentication systems, which capture and continuously authenticate biometrics from user devices. This book first covers the traditional methods of user authentication and discusses how such techniques have become cumbersome in the world of mobile devices and short usage sessions. The concept of continuous biometric authentication systems is introduced and their construction is discussed. The different biometrics that these systems may utilise (e.g.: touchscreen-gesture interactions) are described and relevant studies surveyed. It also surveys important considerations and challenges.This book brings together a wide variety of key motivations, components and advantages of continuous biometric authentication systems. The overview is kept high level, so as not to limit the scope to any single device, biometric trait, use-case, or scenario. Therefore, the contents of this book are applicable todevices ranging from smartphones to desktop computers, utilising biometrics ranging from face recognition to keystroke dynamics. It also provides metrics from a variety of existing systems such that users can identify the advantages and disadvantages of different approaches.This book targets researchers and lecturers working in authentication, as well as advanced-level students in computer science interested in this field. The book will also be of interest to technical professionals working in cyber security.

This book focuses on online transaction processing indexes designed for scalable, byte-addressable non-volatile memory (NVM) and provides a systematic review and summary of the fundamental principles and techniques as well as an outlook on the future of this research area.In this book, the authors divide the development of NVM indexes into three ¿eras¿¿ pre-Optane, Optane and post-Optane¿based on when the first major scalable NVM device (Optane) became commercially available and when it was announced to be discontinued. The book will analyze the reasons for the slow adoption of NVM and give an outlook for indexing techniques in the post-Optane era.The book assumes only basic undergraduate-level understanding on indexing (e.g., B+-trees, hash tables) and database systems in general. It is otherwise self-contained with the necessary background information, including an introduction to NVM hardware and software/programming issues, a detailed description of different indexes in highly concurrent systems for non-experts and new researchers to get started in this area.

This book focuses on the model and algorithm aspects of securing Unmanned Aerial Vehicle Networks (UAV). To equip readers with the essential knowledge required for conducting research in this field, it covers the foundational concepts of this topic as well. This book also offers a detailed insight into UAV networks from the physical layer security point of view.The authors discuss the appropriate channel models for characterizing various propagation environments that UAV networks are exposed. The state-of-the-art technologies, such as UAV trajectory design, cooperative jamming and reconfigurable intelligent surfaces are covered. The corresponding algorithms for significantly improving the security of UAV networks, along with practical case studies on topics such as energy-efficient and secure UAV networks, elevation angle-distance trade-off for securing UAV networks and securing UAV networks with the aid of reconfigurable intelligent surfaces are presented as well. Last, this book outlines the future challenges and research directions to facilitate further studies on secure UAV networks. This book is suitable reading for graduate students and researchers who are interested in the research areas of UAV networking and communications, IoT security, and physical layer security in wireless networks. Professionals working within these related fields will also benefit from this book.

This book provides an overview of both experimental and commercial real-time database systems (RTDBs) and a systematic approach to understanding, designing, and implementing them.To this end, the book is composed of four chapters: Chapter 1 ¿An Overview of Real-Time Database Systems¿ delves into the realm of RTDBs and discusses the specific requirements, transaction models, and scheduling algorithms that set RTDBs apart from conventional DBMs. Chapter 2 on ¿Experimental Real-Time Databases¿ presents various experimental RTDBs developed in academia with their architectures, features, and implementations, while chapter 3 on ¿Commercial Real-Time Databases¿ does so for systems developed and offered by commercial vendors as products or services. Eventually, chapter 4 on ¿Applications of Real-Time Database Systems¿ showcases various applications of RTDBs across different domains.This book will help researchers, graduate students and advanced professionals to get anoverview of the area and to understand the main challenges and systems available.

One of the main benefits of this book is that it presents a comprehensive and innovative eHealth framework that leverages deep learning and IoT wearable devices for the evaluation of Parkinson's disease patients. This framework offers a new way to assess and monitor patients' motor deficits in a personalized and automated way, improving the efficiency and accuracy of diagnosis and treatment.Compared to other books on eHealth and Parkinson's disease, this book offers a unique perspective and solution to the challenges facing patients and healthcare providers. It combines state-of-the-art technology, such as wearable devices and deep learning algorithms, with clinical expertise to develop a personalized and efficient evaluation framework for Parkinson's disease patients.This book provides a roadmap for the integration of cutting-edge technology into clinical practice, paving the way for more effective and patient-centered healthcare. To understand this book, readers should have a basic knowledge of eHealth, IoT, deep learning, and Parkinson's disease. However, the book provides clear explanations and examples to make the content accessible to a wider audience, including researchers, practitioners, and students interested in the intersection of technology and healthcare.

This book provides readers with the basic understanding regarding the threats to the voice processing systems, the state-of-the-art defense methods as well as the current research results on securing voice processing systems.It also introduces three mechanisms to secure the voice processing systems against malicious voice attacks under different scenarios, by utilizing time-domain signal waves, frequency-domain spectrum features, and acoustic physical attributes.First, the authors uncover the modulated replay attack, which uses an inverse filter to compensate for the spectrum distortion caused by the replay attacks to bypass the existing spectrum-based defenses. The authors also provide an effective defense method that utilizes both the time-domain artifacts and frequency-domain distortion to detect the modulated replay attacks. Second, the book introduces a secure automatic speech recognition system for driverless car to defeat adversarial voice command attacks launched from car loudspeakers, smartphones, and passengers. Third, it provides an acoustic compensation system design to reduce the effects from the spectrum reduction attacks, by the audio spectrum compensation and acoustic propagation principle. Finally, the authors conclude with their research effort on defeating the malicious voice attacks and provide insights into more secure voice processing systems.This book is intended for security researchers, computer scientists, and electrical engineers who are interested in the research areas of biometrics, speech signal processing, IoT security, and audio security. Advanced-level students who are studying these topics will benefit from this book as well.

This SpringerBrief is a computational study of significant concerns and their role in forming long-term relationships between intelligent entities. Significant concerns include attitudes, preferences, affinities, and values that are held to be highly valued and meaningful: The means through which a person may find deeply held identity, purpose, and transformation. Significant concerns always engage the emotions and senses in a way that simply holding an opinion may or may not. For example, experiencing a significant concern may provoke deep feelings of awe and wonder in a way that deciding what to have for lunch probably does not, even if the lunch decision involves a rich array of preferences and values. Significant concerns also include what Emmons has called ultimate concerns.The author builds upon this base by considering the hypothetical case of intelligence in androids. An android is defined as a human-like robot that humans would accept as equal to humans in how theyperform and behave in society. An android as defined in this book is not considered to be imitating a human, nor is its purpose to deceive humans into believing that it is a human. Instead, the appropriately programmed android self-identifies as a non-human with its own integrity as a person. Therefore, a computational understanding of personhood and how persons ¿ whether human or android ¿ participate in relationships is essential to this perspective on artificial intelligence.Computational Modelling of Robot Personhood and Relationality describes in technical detail an implementation of a computational model called Affinity that takes the form of a simulation of a population of entities that form, maintain, and break relationships with each other depending upon a rich range of values, motivations, attitudes, and beliefs. Future experimentation and improvements of this model may be used not only to gain a wider understanding of human persons but may also form a preliminary cognitive model of the reasoning process of an android.

Mobile crowdsensing (MCS) is emerging as a novel sensing paradigm in the Internet of Things (IoTs) due to the proliferation of smart devices (e.g., smartphones, wearable devices) in people¿s daily lives. These ubiquitous devices provide an opportunity to harness the wisdom of crowds by recruiting mobile users to collectively perform sensing tasks, which largely collect data about a wide range of human activities and the surrounding environment. However, users suffer from resource consumption such as battery, processing power, and storage, which discourages users¿ participation. To ensure the participation rate, it is necessary to employ an incentive mechanism to compensate users¿ costs such that users are willing to take part in crowdsensing.This book sheds light on the design of incentive mechanisms for MCS in the context of game theory. Particularly, this book presents several game-theoretic models for MCS in different scenarios. In Chapter 1, the authors present an overview of MCS and state the significance of incentive mechanism for MCS. Then, in Chapter 2, 3, 4, and 5, the authors propose a long-term incentive mechanism, a fair incentive mechanism, a collaborative incentive mechanism, and a coopetition-aware incentive mechanism for MCS, respectively. Finally, Chapter 6 summarizes this book and point out the future directions.This book is of particular interest to the readers and researchers in the field of IoT research, especially in the interdisciplinary field of network economics and IoT.

As a privacy-preserving and illumination-robust manner, WiFi signal-based user authentication has become a new direction for ubiquitous user authentication to protect user privacy and security. It gradually turns into an important option for addressing the security concern of IoT environment.However, due to the limited sensing capability of WiFi signals and wide application scenarios, WiFi signal-based user authentication suffers from practical issues of diversified behaviors and complex scenarios.Therefore, it is necessary to address the issues and build integrated systems for user authentication using WiFi signals. In this book, the development and progress of WiFi signal-based user authentication systems in extensive scenarios are presented, which provides a new direction and solution for ubiquitous security and privacy protection. This book gives strong motivation of leveraging WiFi signals to sense human activities for user authentication, and presents the keyissues of WiFi-based user authentication in diversified behaviors and complex scenarios. This book provides the approaches for digging WiFi signals to sense human activities and extract features, realizing user authentication under fine-grained finger gestures, undefined body gestures, and multi-user scenarios. State-of-the-art researches and future directions involved with WiFi signal-based user authentication are presented and discussed as well. This book will benefit researchers and practitioners in the related field.

This book mainly shows readers how to calibrate and control robots. In this regard, it proposes three control schemes: an error-summation enhanced Newton algorithm for model predictive control; RNN for solving perturbed time-varying underdetermined linear systems; and a new joint-drift-free scheme aided with projected ZNN, which can effectively improve robot control accuracy. Moreover, the book develops four advanced algorithms for robot calibration ¿ Levenberg-Marquarelt with diversified regularizations; improved covariance matrix adaptive evolution strategy; quadratic interpolated beetle antennae search algorithm; and a novel variable step-size Levenberg-Marquardt algorithm ¿ which can effectively enhance robot positioning accuracy.In addition, it is exceedingly difficult for experts in other fields to conduct robot arm calibration studies without calibration data. Thus, this book provides a publicly available dataset to assist researchers from other fields in conductingcalibration experiments and validating their ideas. The book also discusses six regularization schemes based on its robot error models, i.e., L1, L2, dropout, elastic, log, and swish. Robots¿ positioning accuracy is significantly improved after calibration. Using the control and calibration methods developed here, readers will be ready to conduct their own research and experiments.

This book is intended as an introduction to a versatile model for temporal data. It exhibits an original lattice structure on the space of chronicles and proposes new counting approach for multiple occurrences of chronicle occurrences. This book also proposes a new approach for frequent temporal pattern mining using pattern structures. This book was initiated by the work of Ch. Dousson in the 1990¿s. At that time, the prominent format was Temporal Constraint Networks for which the article by Richter, Meiri and Pearl is seminal.Chronicles do not conflict with temporal constraint networks, they are closely related. Not only do they share a similar graphical representation, they also have in common a notion of constraints in the timed succession of events. However, chronicles are definitely oriented towards fairly specific tasks in handling temporal data, by making explicit certain aspects of temporal data such as repetitions of an event. The notion of chronicle has been applied both for situation recognition and temporal sequence abstraction. The first challenge benefits from the simple but expressive formalism to specify temporal behavior to match in a temporal sequence. The second challenge aims to abstract a collection of sequences by chronicles with the objective to extract characteristic behaviors.This book targets researchers and students in computer science (from logic to data science). Engineers who would like to develop algorithms based on temporal models will also find this book useful.

Document layout analysis (DLA) is a crucial step towards the development of an effective document image processing system. In the early days of document image processing, DLA was not considered as a complete and complex research problem, rather just a pre-processing step having some minor challenges. The main reason for that is the type of layout being considered for processing was simple. Researchers started paying attention to this complex problem as they come across a large variety of documents. This book presents a clear view of the past, present, and future of DLA, and it also discusses two recent methods developed to address the said problem.

This book provides a broad overview of the state of the art of the research in generative methods for the analysis of social media data. It especially includes two important aspects that currently gain importance in mining and modelling social media: dynamics and networks.The book is divided into five chapters and provides an extensive bibliography consisting of more than 250 papers. After a quick introduction and survey of the book in the first chapter, chapter 2 is devoted to the discussion of data models and ontologies for social network analysis. Next, chapter 3 deals with text generation and generative text models and the dangers they pose to social media and society at large. Chapter 4 then focuses on topic modelling and sentiment analysis in the context of social networks. Finally, Chapter 5 presents graph theory tools and approaches to mine and model social networks. Throughout the book, open problems, highlighting potential future directions, are clearly identified.The book aims at researchers and graduate students in social media analysis, information retrieval, and machine learning applications.

This book is meant for readers who want to understand GANs without the need for a strong mathematical background. Moreover, it covers the practical applications of GANs, making it an excellent resource for beginners. A Primer on Generative Adversarial Networks is suitable for researchers, developers, students, and anyone who wishes to learn about GANs. It is assumed that the reader has a basic understanding of machine learning and neural networks. The book comes with ready-to-run scripts that readers can use for further research. Python is used as the primary programming language, so readers should be familiar with its basics.The book starts by providing an overview of GAN architecture, explaining the concept of generative models. It then introduces the most straightforward GAN architecture, which explains how GANs work and covers the concepts of generator and discriminator. The book then goes into the more advanced real-world applications of GANs, such as human face generation, deep fake, CycleGANs, and more.By the end of the book, readers will have an essential understanding of GANs and be able to write their own GAN code. They can apply this knowledge to their projects, regardless of whether they are beginners or experienced machine learning practitioners.

Natural interaction is one of the hottest research issues in human-computer interaction. At present, there is an urgent need for intelligent devices (service robots, virtual humans, etc.) to be able to understand intentions in an interactive dialogue. Focusing on human-computer understanding based on deep learning methods, the book systematically introduces readers to intention recognition, unknown intention detection, and new intention discovery in human-computer dialogue. This book is the first to present interactive dialogue intention analysis in the context of natural interaction. In addition to helping readers master the key technologies and concepts of human-machine dialogue intention analysis and catch up on the latest advances, it includes valuable references for further research.

This book presents the latest research in cognitive security, a rapidly emerging field that addresses the vulnerabilities in human behavior and cognition that can lead to Cyber-Physical Systems (CPS) compromise. This book demonstrates that as adversaries increasingly use manipulative and deceptive information to disrupt human cognitive processes, including sensation, attention, memory, and mental operations, humans are misled into fallacious reasoning and manipulated decisions that can lead to system-level meltdown. Cognitive security aims to protect humans from the exploitation of cognitive vulnerabilities, help them make informed decisions that are free from manipulation and undue influence, and mitigate the aggravating risk in the ensuing steps of the attacker¿s kill chain. This book offers solutions that work across different fields, such as psychology, neuroscience, data science, social science, and game theory, to deal with cognitive threats. It guides the reader through the core ideas with figures, real-life examples, and case studies. Moreover, it formally defines all research questions, presents the results using mathematical theorems and proofs, and obtains insights through numerical validation. This book provides a self-contained and brief overview of essential system-scientific tools for modeling, analyzing, and mitigating cognitive vulnerabilities. The concepts of human cognitive capacities and cognitive vulnerabilities are formally discussed, followed by two case studies in the scenarios of reactive and proactive attention vulnerabilities. This book provides insights and applications on this transdisciplinary topic, with the goal of motivating future research in this emerging area and pushing the frontier of human-technology convergence. This book is a valuable reference for researchers and advanced-level students studying or working in cognitive security and related fields. It is also useful for decision-makers, managers, and professionals working within these related fields.

To be able to merge the psyche with the neural system has been a long-sought goal. There is much scientific literature on results from research on this topic, but the goal of this ¿booklet¿ is to present the subject in a nutshell and to attract a wider audience to this highly complex topic. Scientists often need years to grasp the scope and implications of merging the psyche with the neural system. Does that really have to be the case? What does the simulated model look like? What are the underlying philosophies? Can it be understood without mathematical formalism?Uniting the psyche and neurology in one model, on the one hand, allows psychological and social theories to be tested on a scientific basis using simulation experiments. On the other hand, a model developed according to the functional structures of the human brain, let us call it the Psi-Organ, which comprises neurology and psyche as one unit, can serve as a basis for AI systems. These can be systems with cognitive capabilities that save human lives, save energy, ensure safety at airports, provide support in caring for the elderly and much more. In other words, systems that can simplify our lives in the most relevant ways and on a broad basis. This model, the Psi-Organ, goes far beyond today's primarily behavior-based AI methods.The manuscript can serve as an excellent introduction to the problem of understanding and modelling the human mind, and to the problem of achieving artificial ¿intelligence¿ in general, increasing awareness and understanding for the associated challenges. In that regard, it is a valuable supplementary text for advanced students or researchers in the field, notably not only in AI, but also (and perhaps primarily) in the medical fields.

This book introduces readers to the fundamentals of the cross-technology coexistence problem in heterogeneous wireless networks. It also highlights a range of mechanisms designed to combat this problem and improve network performance, including protocol design, theoretical analysis, and experimental evaluation.In turn, the book proposes three mechanisms that can be combined to combat the cross-technology coexistence problem and improve network performance. First, the authors present a fast signal identification method. It provides the basis for the subsequent protocol design and allows heterogeneous devices to adopt proper transmission strategies. Second, the authors present two cross-technology interference management mechanisms in both the time domain and the frequency domain, which can mitigate interference and increase transmission opportunities for heterogeneous devices, thus improving network performance. Third, they present a cross-technology communication mechanism basedon symbol-level energy modulation, which allows heterogeneous devices to transmit information directly without a gateway, improving transmission efficiency and paving the way for new applications in IoT scenarios. Lastly, they outline several potential research directions to further improve the efficiency of cross-technology coexistence. This book is intended for researchers, computer scientists, and engineers who are interested in the research areas of wireless networking, wireless communication, mobile computing, and Internet of Things. Advanced-level students studying these topics will benefit from the book as well.

This open access book provides information how to choose and collect the appropriate metrics for a software project in an organization. There are several kinds of metrics, based on the analysis of source code and developed for different programming paradigms such as structured programming and object-oriented programming (OOP). This way, the book follows three main objectives: (i) to identify existing and easily-collectible measures, if possible in the early phases of software development, for predicting and modeling both the traditional attributes of software systems and attributes specifically related to their efficient use of resources, and to create new metrics for such purposes; (ii) to describe ways to collect these measures during the entire lifecycle of a system, using minimally-invasive monitoring of design-time processes, and consolidate them into conceptual frameworks able to support model building by using a variety of approaches, including statistics, data mining and computational intelligence; and (iii) to present models and tools to support design time evolution of systems based on design-time measures and to empirically validate them.The book provides researchers and advanced professionals with methods for understanding the full implications of alternative choices and their relative attractiveness in terms of enhancing system resilience. It also explores the simultaneous use of multiple models that reflect different system interpretations or stakeholder perspectives.

Although deep learning models have achieved great progress in vision, speech, language, planning, control, and many other areas, there still exists a large performance gap between deep learning models and the human cognitive system. Many researchers argue that one of the major reasons accounting for the performance gap is that deep learning models and the human cognitive system process visual information in very different ways.To mimic the performance gap, since 2014, there has been a trend to model various cognitive mechanisms from cognitive neuroscience, e.g., attention, memory, reasoning, and decision, based on deep learning models. This book unifies these new kinds of deep learning models and calls them deep cognitive networks, which model various human cognitive mechanisms based on deep learning models. As a result, various cognitive functions are implemented, e.g., selective extraction, knowledge reuse, and problem solving, for more effective information processing.This book first summarizes existing evidence of human cognitive mechanism modeling from cognitive psychology and proposes a general framework of deep cognitive networks that jointly considers multiple cognitive mechanisms. Then, it analyzes related works and focuses primarily but not exclusively, on the taxonomy of four key cognitive mechanisms (i.e., attention, memory, reasoning, and decision) surrounding deep cognitive networks. Finally, this book studies two representative cases of applying deep cognitive networks to the task of image-text matching and discusses important future directions.

This book investigates intelligent network resource management for IoV, with the objective of maximizing the communication and computing performance of vehicle users. Focusing on two representative use cases in IoV, i.e., safety message broadcast and autonomous driving, the authors propose link-layer protocol design and application-layer computing task scheduling to achieve the objective given the unique characteristics and requirements of IoV. In particular, this book illustrates the challenges of resource management for IoV due to network dynamics, such as time-varying traffic intensity and vehicle mobility, and presents intelligent resource management solutions to adapt to the network dynamics. The Internet of Vehicles (IoV) enables vehicle-to-everything connectivity and supports a variety of applications for vehicles on the road.Intelligent resource management is critical for satisfying demanding communication and computing requirements on IoV, while the highly dynamic network environments pose challenges to the design of resource management schemes. This book provides insights into the significance of adaptive resource management in improving the performance of IoV. The customized communication protocol and computing scheduling scheme are designed accordingly by taking the network dynamics information as an integral design factor. Moreover, the decentralized designs of the proposed solutions guarantee low signaling overhead and high scalability.A comprehensive literature review summarizing recent resource management schemes in IoV, followed by the customized design of communication and computing solutions for the two IoV use cases is included which can serve as a useful reference for professionals from both academia and industry in the area of IoV and resource management. Researchers working within this field and computer science and electrical engineering students will find this book useful as well.

Objective Information Theory (OIT) is proposed to represent and compute the information in a large-scale complex information system with big data in this monograph. To formally analyze, design, develop, and evaluate the information, OIT interprets the information from essential nature, measures the information from mathematical properties, and models the information from concept, logic, and physic. As the exemplified applications, Air Traffic Control System (ATCS) and Smart Court SoSs (System of Systems) are introduced for practical OITs. This Open Access book can be used as a technical reference book in the field of information science and also a reference textbook for senior students and graduate ones in related majors.

Wineinformatics is a new data science application with a focus on understanding wine through artificial intelligence. Thousands of new wine reviews are produced monthly, which benefits the understanding of wine through wine experts for winemakers and consumers. This book systematically investigates how to process human language format reviews and mine useful knowledge from a large volume of processed data.This book presents a human language processing tool named Computational Wine Wheel to process professional wine reviews and three novel Wineinformatics studies to analyze wine quality, price and reviewers. Through the lens of data science, the author demonstrates how the wine receives 90+ scores out of 100 points from Wine Spectator, how to predict a wine's specific grade and price through wine reviews and how to rank a group of wine reviewers. The book also shows the advanced application of the Computational Wine Wheel to capture more information hidden in wine reviews and the possibility of extending the wheel to coffee, tea beer, sake and liquors.This book targets computer scientists, data scientists and wine industrial researchers, who are interested in Wineinformatics. Senior data science undergraduate and graduate students may also benefit from this book.

Incomplete big data are frequently encountered in many industrial applications, such as recommender systems, the Internet of Things, intelligent transportation, cloud computing, and so on. It is of great significance to analyze them for mining rich and valuable knowledge and patterns. Latent feature analysis (LFA) is one of the most popular representation learning methods tailored for incomplete big data due to its high accuracy, computational efficiency, and ease of scalability. The crux of analyzing incomplete big data lies in addressing the uncertainty problem caused by their incomplete characteristics. However, existing LFA methods do not fully consider such uncertainty.In this book, the author introduces several robust latent feature learning methods to address such uncertainty for effectively and efficiently analyzing incomplete big data, including robust latent feature learning based on smooth L1-norm, improving robustness of latent feature learning using L1-norm, improving robustness of latent feature learning using double-space, data-characteristic-aware latent feature learning, posterior-neighborhood-regularized latent feature learning, and generalized deep latent feature learning. Readers can obtain an overview of the challenges of analyzing incomplete big data and how to employ latent feature learning to build a robust model to analyze incomplete big data. In addition, this book provides several algorithms and real application cases, which can help students, researchers, and professionals easily build their models to analyze incomplete big data.

Emerging machine learning techniques bring new opportunities to flexible network control and management. This book focuses on using state-of-the-art machine learning-based approaches to improve the performance of Software-Defined Networking (SDN). It will apply several innovative machine learning methods (e.g., Deep Reinforcement Learning, Multi-Agent Reinforcement Learning, and Graph Neural Network) to traffic engineering and controller load balancing in software-defined wide area networks, as well as flow scheduling, coflow scheduling, and flow migration for network function virtualization in software-defined data center networks. It helps readers reflect on several practical problems of deploying SDN and learn how to solve the problems by taking advantage of existing machine learning techniques. The book elaborates on the formulation of each problem, explains design details for each scheme, and provides solutions by running mathematical optimization processes, conducting simulated experiments, and analyzing the experimental results.

Latent factor analysis models are an effective type of machine learning model for addressing high-dimensional and sparse matrices, which are encountered in many big-data-related industrial applications. The performance of a latent factor analysis model relies heavily on appropriate hyper-parameters. However, most hyper-parameters are data-dependent, and using grid-search to tune these hyper-parameters is truly laborious and expensive in computational terms. Hence, how to achieve efficient hyper-parameter adaptation for latent factor analysis models has become a significant question.This is the first book to focus on how particle swarm optimization can be incorporated into latent factor analysis for efficient hyper-parameter adaptation, an approach that offers high scalability in real-world industrial applications.The book will help students, researchers and engineers fully understand the basic methodologies of hyper-parameter adaptation via particle swarm optimization in latent factor analysis models. Further, it will enable them to conduct extensive research and experiments on the real-world applications of the content discussed.

This book addresses several knowledge discovery problems on multi-sourced data where the theories, techniques, and methods in data cleaning, data mining, and natural language processing are synthetically used. This book mainly focuses on three data models: the multi-sourced isomorphic data, the multi-sourced heterogeneous data, and the text data. On the basis of three data models, this book studies the knowledge discovery problems including truth discovery and fact discovery on multi-sourced data from four important properties: relevance, inconsistency, sparseness, and heterogeneity, which is useful for specialists as well as graduate students. Data, even describing the same object or event, can come from a variety of sources such as crowd workers and social media users. However, noisy pieces of data or information are unavoidable. Facing the daunting scale of data, it is unrealistic to expect humans to "e;label"e; or tell which data source is more reliable. Hence, it is crucial to identify trustworthy information from multiple noisy information sources, referring to the task of knowledge discovery. At present, the knowledge discovery research for multi-sourced data mainly faces two challenges. On the structural level, it is essential to consider the different characteristics of data composition and application scenarios and define the knowledge discovery problem on different occasions. On the algorithm level, the knowledge discovery task needs to consider different levels of information conflicts and design efficient algorithms to mine more valuable information using multiple clues. Existing knowledge discovery methods have defects on both the structural level and the algorithm level, making the knowledge discovery problem far from totally solved.