This book discusses the interplay between statistics, data science, machine learning and artificial intelligence, with a focus on environmental science, the natural sciences, and technology. It covers the state of the art from both a theoretical and a practical viewpoint and describes how to successfully apply machine learning methods, demonstrating the benefits of statistics for modeling and analyzing high-dimensional and big data. The book's expert contributions include theoretical studies of machine learning methods, expositions of general methodologies for sound statistical analyses of data as well as novel approaches to modeling and analyzing data for specific problems and areas. In terms of applications, the contributions deal with data as arising in industrial quality control, autonomous driving, transportation and traffic, chip manufacturing, photovoltaics, football, transmission of infectious diseases, Covid-19 and public health. The book will appeal to statisticians and data scientists, as well as engineers and computer scientists working in related fields or applications.

Tactile Sensing, Skill Learning and Robotic Dexterous Manipulation focuses on cross-disciplinary lines of research and groundbreaking research ideas in three research lines: tactile sensing, skill learning and dexterous control. The book introduces recent work about human dexterous skill representation and learning, along with discussions of tactile sensing and its applications on unknown objects' property recognition and reconstruction. Sections also introduce the adaptive control schema and its learning by imitation and exploration. Other chapters describe the fundamental part of relevant research, paying attention to the connection among different fields and showing the state-of-the-art in related branches. The book summarizes the different approaches and discusses the pros and cons of each. Chapters not only describe the research but also include basic knowledge that can help readers understand the proposed work, making it an excellent resource for researchers and professionals who work in the robotics industry, haptics and in machine learning.

A remarkable facet of the human brain is its ability to manage multiple tasks with apparent simultaneity. Knowledge learned from one task can then be used to enhance problem-solving in other related tasks. In machine learning, the idea of leveraging relevant information across related tasks as inductive biases to enhance learning performance has attracted significant interest. In contrast, attempts to emulate the human brain's ability to generalize in optimization - particularly in population-based evolutionary algorithms - have received little attention to date. Recently, a novel evolutionary search paradigm, Evolutionary Multi-Task (EMT) optimization, has been proposed in the realm of evolutionary computation. In contrast to traditional evolutionary searches, which solve a single task in a single run, evolutionary multi-tasking algorithm conducts searches concurrently on multiple search spaces corresponding to different tasks or optimization problems, each possessing a unique function landscape. By exploiting the latent synergies among distinct problems, the superior search performance of EMT optimization in terms of solution quality and convergence speed has been demonstrated in a variety of continuous, discrete, and hybrid (mixture of continuous and discrete) tasks. This book discusses the foundations and methodologies of developing evolutionary multi-tasking algorithms for complex optimization, including in domains characterized by factors such as multiple objectives of interest, high-dimensional search spaces and NP-hardness.

Intelligent decision support is based on human knowledge related to a specific part of a real or abstract world. When the knowledge is gained by experience, it is induced from empirical data. The data structure, called an information system, is a record of objects described by a set of attributes. Knowledge is understood here as an ability to classify objects. Objects being in the same class are indiscernible by means of attributes and form elementary building blocks (granules, atoms). In particular, the granularity of knowledge causes that some notions cannot be expressed precisely within available knowledge and can be defined only vaguely. In the rough sets theory created by Z. Pawlak each imprecise concept is replaced by a pair of precise concepts called its lower and upper approximation. These approximations are fundamental tools and reasoning about knowledge. The rough sets philosophy turned out to be a very effective, new tool with many successful real-life applications to its credit. It is worthwhile stressing that no auxiliary assumptions are needed about data, like probability or membership function values, which is its great advantage. The present book reveals a wide spectrum of applications of the rough set concept, giving the reader the flavor of, and insight into, the methodology of the newly developed disciplines. Although the book emphasizes applications, comparison with other related methods and further developments receive due attention.

Visual Question Answering (VQA) usually combines visual inputs like image and video with a natural language question concerning the input and generates a natural language answer as the output. This is by nature a multi-disciplinary research problem, involving computer vision (CV), natural language processing (NLP), knowledge representation and reasoning (KR), etc. Further, VQA is an ambitious undertaking, as it must overcome the challenges of general image understanding and the question-answering task, as well as the difficulties entailed by using large-scale databases with mixed-quality inputs. However, with the advent of deep learning (DL) and driven by the existence of advanced techniques in both CV and NLP and the availability of relevant large-scale datasets, we have recently seen enormous strides in VQA, with more systems and promising results emerging. This book provides a comprehensive overview of VQA, covering fundamental theories, models, datasets, and promising future directions. Given its scope, it can be used as a textbook on computer vision and natural language processing, especially for researchers and students in the area of visual question answering. It also highlights the key models used in VQA.

Optimum-Path Forest: Theory, Algorithms, and Applications was first published in 2008 in its supervised and unsupervised versions with applications in medicine and image classification. Since then, it has expanded to a variety of other applications such as remote sensing, electrical and petroleum engineering, and biology. In recent years, multi-label and semi-supervised versions were also developed to handle video classification problems. The book presents the principles, algorithms and applications of Optimum-Path Forest, giving the theory and state-of-the-art as well as insights into future directions.

Machine Learning for Biometrics: Concepts, Algorithms and Applications highlights the fundamental concepts of machine learning, processing and analyzing data from biometrics and provides a review of intelligent and cognitive learning tools which can be adopted in this direction. Each chapter of the volume is supported by real-life case studies, illustrative examples and video demonstrations. The book elucidates various biometric concepts, algorithms and applications with machine intelligence solutions, providing guidance on best practices for new technologies such as e-health solutions, Data science, Cloud computing, and Internet of Things, etc. In each section, different machine learning concepts and algorithms are used, such as different object detection techniques, image enhancement techniques, both global and local feature extraction techniques, and classifiers those are commonly used data science techniques. These biometrics techniques can be used as tools in Cloud computing, Mobile computing, IOT based applications, and e-health care systems for secure login, device access control, personal recognition and surveillance.

The evolution of deep learning models, combined with with advances in the Internet of Things and sensor technology, has gained more importance for weather forecasting, plant disease detection, underground water detection, soil quality, crop condition monitoring, and many other issues in the field of agriculture. agriculture. Deep Learning for Sustainable Agriculture discusses topics such as the impactful role of deep learning during the analysis of sustainable agriculture data and how deep learning can help farmers make better decisions. It also considers the latest deep learning techniques for effective agriculture data management, as well as the standards established by international organizations in related fields. The book provides advanced students and professionals in agricultural science and engineering, geography, and geospatial technology science with an in-depth explanation of the relationship between agricultural inference and the decision-support amenities offered by an advanced mathematical evolutionary algorithm.

Deep Learning in Bioinformatics: Techniques and Applications in Practice introduces the topic in an easy-to-understand way, exploring how it can be utilized for addressing important problems in bioinformatics, including drug discovery, de novo molecular design, sequence analysis, protein structure prediction, gene expression regulation, protein classification, biomedical image processing and diagnosis, biomolecule interaction prediction, and in systems biology. The book also presents theoretical and practical successes of deep learning in bioinformatics, pointing out problems and suggesting future research directions. Dr. Izadkhah provides valuable insights and will help researchers use deep learning techniques in their biological and bioinformatics studies.

Advanced Data Mining Tools and Methods for Social Computing explores advances in the latest data mining tools, methods, algorithms and the architectures being developed specifically for social computing and social network analysis. The book reviews major emerging trends in technology that are supporting current advancements in social networks, including data mining techniques and tools. It also aims to highlight the advancement of conventional approaches in the field of social networking. Chapter coverage includes reviews of novel techniques and state-of-the-art advances in the area of data mining, machine learning, soft computing techniques, and their applications in the field of social network analysis.

The book presents the confluence of wearable and wireless inertial sensor systems, such as a smartphone, for deep brain stimulation for treating movement disorders, such as essential tremor, and machine learning. The machine learning distinguishes between distinct deep brain stimulation settings, such as 'On' and 'Off' status. This achievement demonstrates preliminary insight with respect to the concept of Network Centric Therapy, which essentially represents the Internet of Things for healthcare and the biomedical industry, inclusive of wearable and wireless inertial sensor systems, machine learning, and access to Cloud computing resources.Imperative to the realization of these objectives is the organization of the software development process. Requirements and pseudo code are derived, and software automation using Python for post-processing the inertial sensor signal data to a feature set for machine learning is progressively developed. A perspective of machine learning in terms of a conceptual basis and operational overview is provided. Subsequently, an assortment of machine learning algorithms is evaluated based on quantification of a reach and grasp task for essential tremor using a smartphone as a wearable and wireless accelerometer system.Furthermore, these skills regarding the software development process and machine learning applications with wearable and wireless inertial sensor systems enable new and novel biomedical research only bounded by the reader's creativity.

This book is about conformal prediction, an approach to prediction that originated in machine learning in the late 1990s. The main feature of conformal prediction is the principled treatment of the reliability of predictions. The prediction algorithms described - conformal predictors - are provably valid in the sense that they evaluate the reliability of their own predictions in a way that is neither over-pessimistic nor over-optimistic (the latter being especially dangerous). The approach is still flexible enough to incorporate most of the existing powerful methods of machine learning. The book covers both key conformal predictors and the mathematical analysis of their properties. Algorithmic Learning in a Random World contains, in addition to proofs of validity, results about the efficiency of conformal predictors. The only assumption required for validity is that of "randomness" (the prediction algorithm is presented with independent and identically distributed examples); in later chapters, even the assumption of randomness is significantly relaxed. Interesting results about efficiency are established both under randomness and under stronger assumptions. Since publication of the First Edition in 2005 conformal prediction has found numerous applications in medicine and industry, and is becoming a popular machine-learning technique. This Second Edition contains three new chapters. One is about conformal predictive distributions, which are more informative than the set predictions produced by standard conformal predictors. Another is about the efficiency of ways of testing the assumption of randomness based on conformal prediction. The third new chapter harnesses conformal testing procedures for protecting machine-learning algorithms against changes in the distribution of the data. In addition, the existing chapters have been revised, updated, and expanded.

This book aims to solve some key problems in the decision and optimization procedure for power market organizers and participants in data-driven approaches. It begins with an overview of the power market data and analyzes on their characteristics and importance for market clearing. Then, the first part of the book discusses the essential problem of bus load forecasting from the perspective of market organizers. The related works include load uncertainty modeling, bus load bad data correction, and monthly load forecasting. The following part of the book answers how much information can be obtained from public data in locational marginal price (LMP)-based markets. It introduces topics such as congestion identification, componential price forecasting, quantifying the impact of forecasting error, and financial transmission right investment. The final part of the book answers how to model the complex market bidding behaviors. Specific works include pattern extraction, aggregated supply curve forecasting, market simulation, and reward function identification in bidding. These methods are especially useful for market organizers to understand the bidding behaviors of market participants and make essential policies. It will benefit and inspire researchers, graduate students, and engineers in the related fields.

This book provides a coherent and complete overview of various Question Answering (QA) systems. It covers three main categories based on the source of the data that can be unstructured text (TextQA), structured knowledge graphs (KBQA), and the combination of both. Developing a QA system usually requires using a combination of various important techniques, including natural language processing, information retrieval and extraction, knowledge graph processing, and machine learning. After a general introduction and an overview of the book in Chapter 1, the history of QA systems and the architecture of different QA approaches are explained in Chapter 2. It starts with early close domain QA systems and reviews different generations of QA up to state-of-the-art hybrid models. Next, Chapter 3 is devoted to explaining the datasets and the metrics used for evaluating TextQA and KBQA. Chapter 4 introduces the neural and deep learning models used in QA systems. This chapter includes the required knowledge of deep learning and neural text representation models for comprehending the QA models over text and QA models over knowledge base explained in Chapters 5 and 6, respectively. In some of the KBQA models the textual data is also used as another source besides the knowledge base; these hybrid models are studied in Chapter 7. In Chapter 8, a detailed explanation of some well-known real applications of the QA systems is provided. Eventually, open issues and future work on QA are discussed in Chapter 9. This book delivers a comprehensive overview on QA over text, QA over knowledge base, and hybrid QA systems which can be used by researchers starting in this field. It will help its readers to follow the state-of-the-art research in the area by providing essential and basic knowledge.

Learn how to apply the principles of machine learning to time series modeling with this indispensable resource Machine Learning for Time Series Forecasting with Python is an incisive and straightforward examination of one of the most crucial elements of decision-making in finance, marketing, education, and healthcare: time series modeling. Despite the centrality of time series forecasting, few business analysts are familiar with the power or utility of applying machine learning to time series modeling. Author Francesca Lazzeri, a distinguished machine learning scientist and economist, corrects that deficiency by providing readers with comprehensive and approachable explanation and treatment of the application of machine learning to time series forecasting. Written for readers who have little to no experience in time series forecasting or machine learning, the book comprehensively covers all the topics necessary to: Understand time series forecasting concepts, such as stationarity, horizon, trend, and seasonality Prepare time series data for modeling Evaluate time series forecasting models' performance and accuracy Understand when to use neural networks instead of traditional time series models in time series forecasting Machine Learning for Time Series Forecasting with Python is full real-world examples, resources and concrete strategies to help readers explore and transform data and develop usable, practical time series forecasts. Perfect for entry-level data scientists, business analysts, developers, and researchers, this book is an invaluable and indispensable guide to the fundamental and advanced concepts of machine learning applied to time series modeling.

This book describes a novel machine-learning based approach to answer some traditional archaeological problems, relating to archaeological site detection and site locational preferences. Institutional data collected from six Swiss regions (Zurich, Aargau, Grisons, Vaud, Geneva and Fribourg) have been analyzed with an original conceptual framework based on the Random Forest algorithm. It is shown how the algorithm can assist in the modelling process in connection with heterogeneous, incomplete archaeological datasets and related cultural heritage information. Moreover, an in-depth review of past and more recent works of quantitative methods for archaeological predictive modelling is provided. The book guides the readers to set up their own protocol for: i) dealing with uncertain data, ii) predicting archaeological site location, iii) establishing environmental features importance, iv) and suggest a model validation procedure. It addresses both academics and professionals in archaeology and cultural heritage management, and offers a source of inspiration for future research directions in the field of digital humanities and computational archaeology.

Currently, machine learning is playing a pivotal role in the progress of genomics. The applications of machine learning are helping all to understand the emerging trends and the future scope of genomics. This book provides comprehensive coverage of machine learning applications such as DNN, CNN, and RNN, for predicting the sequence of DNA and RNA binding proteins, expression of the gene, and splicing control. In addition, the book addresses the effect of multiomics data analysis of cancers using tensor decomposition, machine learning techniques for protein engineering, CNN applications on genomics, challenges of long noncoding RNAs in human disease diagnosis, and how machine learning can be used as a tool to shape the future of medicine. More importantly, it gives a comparative analysis and validates the outcomes of machine learning methods on genomic data to the functional laboratory tests or by formal clinical assessment. The topics of this book will cater interest to academicians, practitioners working in the field of functional genomics, and machine learning. Also, this book shall guide comprehensively the graduate, postgraduates, and Ph.D. scholars working in these fields.

Cognitive Big Data Intelligence with a Metaheuristic Approach presents an exact and compact organization of content relating to the latest metaheuristics methodologies based on new challenging big data application domains and cognitive computing. The combined model of cognitive big data intelligence with metaheuristics methods can be used to analyze emerging patterns, spot business opportunities, and take care of critical process-centric issues in real-time. Various real-time case studies and implemented works are discussed in this book for better understanding and additional clarity. This book presents an essential platform for the use of cognitive technology in the field of Data Science. It covers metaheuristic methodologies that can be successful in a wide variety of problem settings in big data frameworks.

Cyber-Physical Systems: AI and COVID-19 highlights original research which addresses current data challenges in terms of the development of mathematical models, cyber-physical systems-based tools and techniques, and the design and development of algorithmic solutions, etc. It reviews the technical concepts of gathering, processing and analyzing data from cyber-physical systems (CPS) and reviews tools and techniques that can be used. This book will act as a resource to guide COVID researchers as they move forward with clinical and epidemiological studies on this outbreak, including the technical concepts of gathering, processing and analyzing data from cyber-physical systems (CPS). The major problem in the identification of COVID-19 is detection and diagnosis due to non-availability of medicine. In this situation, only one method, Reverse Transcription Polymerase Chain Reaction (RT-PCR) has been widely adopted and used for diagnosis. With the evolution of COVID-19, the global research community has implemented many machine learning and deep learning-based approaches with incremental datasets. However, finding more accurate identification and prediction methods are crucial at this juncture.

Representation learning in heterogeneous graphs (HG) is intended to provide a meaningful vector representation for each node so as to facilitate downstream applications such as link prediction, personalized recommendation, node classification, etc. This task, however, is challenging not only because of the need to incorporate heterogeneous structural (graph) information consisting of multiple types of node and edge, but also the need to consider heterogeneous attributes or types of content (e.g. text or image) associated with each node. Although considerable advances have been made in homogeneous (and heterogeneous) graph embedding, attributed graph embedding and graph neural networks, few are capable of simultaneously and effectively taking into account heterogeneous structural (graph) information as well as the heterogeneous content information of each node. In this book, we provide a comprehensive survey of current developments in HG representation learning. More importantly, we present the state-of-the-art in this field, including theoretical models and real applications that have been showcased at the top conferences and journals, such as TKDE, KDD, WWW, IJCAI and AAAI. The book has two major objectives: (1) to provide researchers with an understanding of the fundamental issues and a good point of departure for working in this rapidly expanding field, and (2) to present the latest research on applying heterogeneous graphs to model real systems and learning structural features of interaction systems. To the best of our knowledge, it is the first book to summarize the latest developments and present cutting-edge research on heterogeneous graph representation learning. To gain the most from it, readers should have a basic grasp of computer science, data mining and machine learning.

This book is based on the author's Ph.D. dissertation 56]. The the sis research was conducted while the author was a graduate student in the Department of Computer Science at Rutgers University. The book was pre pared at the University of Massachusetts at Amherst where the author is currently an Assistant Professor in the Department of Computer and Infor mation Science. Programs that learn concepts from examples are guided not only by the examples (and counterexamples) that they observe, but also by bias that determines which concept is to be considered as following best from the ob servations. Selection of a concept represents an inductive leap because the concept then indicates the classification of instances that have not yet been observed by the learning program. Learning programs that make undesir able inductive leaps do so due to undesirable bias. The research problem addressed here is to show how a learning program can learn a desirable inductive bias."

Cartesian Genetic Programming (CGP) is a highly effective and increasingly popular form of genetic programming. It represents programs in the form of directed graphs, and a particular characteristic is that it has a highly redundant genotype-phenotype mapping, in that genes can be noncoding. It has spawned a number of new forms, each improving on the efficiency, among them modular, or embedded, CGP, and self-modifying CGP. It has been applied to many problems in both computer science and applied sciences. This book contains chapters written by the leading figures in the development and application of CGP, and it will be essential reading for researchers in genetic programming and for engineers and scientists solving applications using these techniques. It will also be useful for advanced undergraduates and postgraduates seeking to understand and utilize a highly efficient form of genetic programming.

This volume presents the current status of software development in the field of computational and theoretical chemistry and gives an overview of the emerging trends. The challenges of maintaining the legacy codes and their adaptation to the rapidly growing hardware capabilities and the new programming environments are surveyed in a series of topical reviews written by the core developers and maintainers of the popular quantum chemistry and molecular dynamics programs. Special emphasis is given to new computational methodologies and practical aspects of their implementation and application in the computational chemistry codes. Modularity of the computational chemistry software is an emerging concept that enables to bypass the development and maintenance bottleneck of the legacy software and to customize the software using the best available computational procedures implemented in the form of self-contained modules. Perspectives on modular design of the computer programs for modeling molecular electronic structure, non-adiabatic dynamics, kinetics, as well as for data visualization are presented by the researchers actively working in the field of software development and application. This volume is of interest to quantum and computational chemists as well as experimental chemists actively using and developing computational software for their research. Chapters "MLatom 2: An Integrative Platform for Atomistic Machine Learning" and "Evolution of the Automatic Rhodopsin Modeling (ARM) Protocol" are available open access under a CC BY 4.0 License via link.springer.com.

The recent advancement of industrial computerization has significantly helped in resolving the challenges associated with the conventional industrial systems. The industry 4.0 quality standards demand smart and intelligent solutions to revolutionize the industrial applications. Despite, a wide range of available industrial solutions, still the precision, accuracy and speed are a matter of interest to scientists to device novel solutions. In addition, the IoT technologies, though, have gained an inspiration in industrial applications, yet, the time and resource complexity of industrial sensors require intelligent management and monitoring of sensors' data. The integration of machine intelligence and IoT technologies can greatly help in devising cutting edge solutions to very recent issues of industrial applications. Machine intelligence is the most appropriate set of techniques for constructing prediction models due to its capability in handling large-scale and complex datasets. Machine intelligent solution can effectively help the industrial stakeholders in automatic detection of faults, early prediction of errors and risks, tracking industrial shipments through intelligent sensors, health industry risk mitigation employing IoT and machine intelligence, disaster management, and many more. The objectives include: Assessment of limitations of industrial systems, challenges and solutions; Empowerment of industrial systems with machine intelligence to mitigate the risks; Smart safety measures towards industrial systems; and Presentation of recent intelligent systems for a wide range of industrials applications.

This book provides its readers with an introduction to interesting prediction and science dynamics problems in the field of Science of Science. Prediction focuses on the forecasting of future performance (or impact) of an entity, either a research article or a scientist, and also the prediction of future links in collaboration networks or identifying missing links in citation networks. The single chapters are written in a way that help the reader gain a detailed technical understanding of the corresponding subjects, the strength and weaknesses of the state-of-the-art approaches for each described problem, and the currently open challenges. While chapter 1 provides a useful contribution in the theoretical foundations of the fields of scientometrics and science of science, chapters 2-4 turn the focal point to the study of factors that affect research impact and its dynamics. Chapters 5-7 then focus on article-level measures that quantify the current and future impact of scientific articles. Next, chapters 8-10 investigate subjects relevant to predicting the future impact of individual researchers. Finally, chapters 11-13 focus on science evolution and dynamics, leveraging heterogeneous and interconnected data, where the analysis of research topic trends and their evolution has always played a key role in impact prediction approaches and quantitative analyses in the field of bibliometrics. Each chapter can be read independently, since it includes a detailed description of the problem being investigated along with a thorough discussion and study of the respective state-of-the-art. Due to the cross-disciplinary character of the Science of Science field, the book may be useful to interested readers from a variety of disciplines like information science, information retrieval, network science, informetrics, scientometrics, and machine learning, to name a few. The profiles of the readers may also be diverse ranging from researchers and professors in the respective fields to students and developers being curious about the covered subjects.