Big Data Management – serie

This book introduces readers to a workload-aware methodology for large-scale graph algorithm optimization in graph-computing systems, and proposes several optimization techniques that can enable these systems to handle advanced graph algorithms efficiently.

This book introduces readers to a workload-aware methodology for large-scale graph algorithm optimization in graph-computing systems, and proposes several optimization techniques that can enable these systems to handle advanced graph algorithms efficiently.

In this context, various advanced techniques, including graph embedding and graph neural networks, have recently been proposed to improve the performance of graph data mining.

In this context, various advanced techniques, including graph embedding and graph neural networks, have recently been proposed to improve the performance of graph data mining.

This book presents the state of the art in distributed machine learning algorithms that are based on gradient optimization methods. In the big data era, large-scale datasets pose enormous challenges for the existing machine learning systems. As such, implementing machine learning algorithms in a distributed environment has become a key technology, and recent research has shown gradient-based iterative optimization to be an effective solution. Focusing on methods that can speed up large-scale gradient optimization through both algorithm optimizations and careful system implementations, the book introduces three essential techniques in designing a gradient optimization algorithm to train a distributed machine learning model: parallel strategy, data compression and synchronization protocol.Written in a tutorial style, it covers a range of topics, from fundamental knowledge to a number of carefully designed algorithms and systems of distributed machine learning. It will appealto a broad audience in the field of machine learning, artificial intelligence, big data and database management.

This book presents the state of the art in distributed machine learning algorithms that are based on gradient optimization methods. In the big data era, large-scale datasets pose enormous challenges for the existing machine learning systems. As such, implementing machine learning algorithms in a distributed environment has become a key technology, and recent research has shown gradient-based iterative optimization to be an effective solution. Focusing on methods that can speed up large-scale gradient optimization through both algorithm optimizations and careful system implementations, the book introduces three essential techniques in designing a gradient optimization algorithm to train a distributed machine learning model: parallel strategy, data compression and synchronization protocol.Written in a tutorial style, it covers a range of topics, from fundamental knowledge to a number of carefully designed algorithms and systems of distributed machine learning. It will appealto a broad audience in the field of machine learning, artificial intelligence, big data and database management.

The development of information technology has made it possible to collect large amounts of spatial data on a daily basis. It is of enormous significance when it comes to discovering implicit, non-trivial and potentially valuable information from this spatial data. Spatial co-location patterns reveal the distribution rules of spatial features, which can be valuable for application users. This book provides commercial software developers with proven and effective algorithms for detecting and filtering these implicit patterns, and includes easily implemented pseudocode for all the algorithms. Furthermore, it offers a basis for further research in this promising field.Preference-based co-location pattern mining refers to mining constrained or condensed co-location patterns instead of mining all prevalent co-location patterns. Based on the authors’ recent research, the book highlights techniques for solving a range of problems in this context, including maximal co-location pattern mining, closed co-location pattern mining, top-k co-location pattern mining, non-redundant co-location pattern mining, dominant co-location pattern mining, high utility co-location pattern mining, user-preferred co-location pattern mining, and similarity measures between spatial co-location patterns.Presenting a systematic, mathematical study of preference-based spatial co-location pattern mining, this book can be used both as a textbook for those new to the topic and as a reference resource for experienced professionals.

The development of information technology has made it possible to collect large amounts of spatial data on a daily basis. It is of enormous significance when it comes to discovering implicit, non-trivial and potentially valuable information from this spatial data. Spatial co-location patterns reveal the distribution rules of spatial features, which can be valuable for application users. This book provides commercial software developers with proven and effective algorithms for detecting and filtering these implicit patterns, and includes easily implemented pseudocode for all the algorithms. Furthermore, it offers a basis for further research in this promising field.Preference-based co-location pattern mining refers to mining constrained or condensed co-location patterns instead of mining all prevalent co-location patterns. Based on the authors’ recent research, the book highlights techniques for solving a range of problems in this context, including maximal co-location pattern mining, closed co-location pattern mining, top-k co-location pattern mining, non-redundant co-location pattern mining, dominant co-location pattern mining, high utility co-location pattern mining, user-preferred co-location pattern mining, and similarity measures between spatial co-location patterns.Presenting a systematic, mathematical study of preference-based spatial co-location pattern mining, this book can be used both as a textbook for those new to the topic and as a reference resource for experienced professionals.

Whether the objective is to refine comprehension of online learning engagement dynamics or to deploy AI-infused strategies in educational settings, this text serves as an invaluable resource for revolutionizing education with AI.

This book demystifies the core principles and architectural design of spatiotemporal big data systems, offering incisive analysis of the unique attributes of spatiotemporal data and its broad applicability in real-world scenarios, aiming to impart a profound understanding of the intrinsic value of spatiotemporal big data and its far-reaching impact across various fields.In the core technology section, particular emphasis is placed on the pivotal role of GIS (Geographic Information Systems) technology in the processing of spatiotemporal big data, comprehensively covering the entire process from data collection, pre-processing, in-depth analysis to final visualization. Through carefully selected real-world cases and detailed technical explanations, readers will gain proficiency in leveraging GIS technology to uncover the latent value of spatiotemporal big data.The book also delves into the essential techniques and algorithms required for building efficient spatiotemporal big data systems, such as efficient data storage and management, intelligent data mining and analysis, alongside specific application cases under the smart city framework, including advanced practices in urban planning optimization, traffic management innovation, and environmental monitoring upgrades. Rich in content and blending theoretical depth with practical guidance, it serves as a valuable resource and reference guide for both academic experts and practitioners in the field, as well as a quality read for those intrigued by spatiotemporal big data systems.The translation was done with the help of artificial intelligence. A subsequent human revision was done primarily in terms of content.

Blockchain, a decentralized ledger technology based on cryptographic algorithms, ensures the creation of immutable and tamper-proof ledgers in decentralized systems. The transparent nature of blockchain allows public access to transaction records, providing unprecedented opportunities for blockchain data analytics and mining. The primary value of blockchain transaction data analytics lies in two aspects: 1) by delving into the details of blockchain transaction data, we can extensively explore various types of user behavior patterns and the evolutionary process of blockchain transaction networks; and 2) analyzing blockchain transaction data aids in identifying illicit activities, offering effective regulatory solutions for the establishment of a healthier blockchain ecosystem. This book focuses on data analytics based on network-based approaches, providing a comprehensive analysis of blockchain data analytics problems, key technologies, and future directions. Different from most existing book, this book takes a unique approach to blockchain data analysis research, focusing on data analytics based on network-based approaches. Leveraging network analysis methods, the book concentrates on three main aspects of blockchain transaction data analytics and mining: (1) transaction network modelling and pattern mining, including macro and micro-level account attributes, money laundering network patterns, and network evolution patterns; (2) account business classification, such as account label prediction based on graph neural networks; and (3) anomaly behavior identification, covering phishing detection, risk scoring, and transaction tracking. Designed as a valuable resource for students, researchers, engineers, and policymakers in various fields related to blockchain data analytics, this book holds significant importance for understanding blockchain transaction behavior and addressing the detection of illicit activities in the blockchain space.

Blockchain, a decentralized ledger technology based on cryptographic algorithms, ensures the creation of immutable and tamper-proof ledgers in decentralized systems. The transparent nature of blockchain allows public access to transaction records, providing unprecedented opportunities for blockchain data analytics and mining. The primary value of blockchain transaction data analytics lies in two aspects: 1) by delving into the details of blockchain transaction data, we can extensively explore various types of user behavior patterns and the evolutionary process of blockchain transaction networks; and 2) analyzing blockchain transaction data aids in identifying illicit activities, offering effective regulatory solutions for the establishment of a healthier blockchain ecosystem. This book focuses on data analytics based on network-based approaches, providing a comprehensive analysis of blockchain data analytics problems, key technologies, and future directions. Different from most existing book, this book takes a unique approach to blockchain data analysis research, focusing on data analytics based on network-based approaches. Leveraging network analysis methods, the book concentrates on three main aspects of blockchain transaction data analytics and mining: (1) transaction network modelling and pattern mining, including macro and micro-level account attributes, money laundering network patterns, and network evolution patterns; (2) account business classification, such as account label prediction based on graph neural networks; and (3) anomaly behavior identification, covering phishing detection, risk scoring, and transaction tracking. Designed as a valuable resource for students, researchers, engineers, and policymakers in various fields related to blockchain data analytics, this book holds significant importance for understanding blockchain transaction behavior and addressing the detection of illicit activities in the blockchain space.

This book describes theoretical elements, practical approaches, and specialized tools that systematically organize, characterize, and analyze big data gathered from educational affairs and settings. Moreover, the book shows several inference criteria to leverage and produce descriptive, explanatory, and predictive closures to study and understand education phenomena at in classroom and online environments.This is why diverse researchers and scholars contribute with valuable chapters to ground with well-–sounded theoretical and methodological constructs in the novel field of Educational Data Science (EDS), which examines academic big data repositories, as well as to introduces systematic reviews, reveals valuable insights, and promotes its application to extend its practice. EDS as a transdisciplinary field relies on statistics, probability, machine learning, data mining, and analytics, in addition to biological, psychological, and neurological knowledge aboutlearning science. With this in mind, the book is devoted to those that are in charge of educational management, educators, pedagogues, academics, computer technologists, researchers, and postgraduate students, who pursue to acquire a conceptual, formal, and practical landscape of how to deploy EDS to build proactive, real- time, and reactive applications that personalize education, enhance teaching, and improve learning!Chapter “Sync Ratio and Cluster Heat Map for Visualizing Student Engagement” is available open access under a Creative Commons Attribution 4.0 International License via link.springer.com.

This book describes theoretical elements, practical approaches, and specialized tools that systematically organize, characterize, and analyze big data gathered from educational affairs and settings. Moreover, the book shows several inference criteria to leverage and produce descriptive, explanatory, and predictive closures to study and understand education phenomena at in classroom and online environments.This is why diverse researchers and scholars contribute with valuable chapters to ground with well-–sounded theoretical and methodological constructs in the novel field of Educational Data Science (EDS), which examines academic big data repositories, as well as to introduces systematic reviews, reveals valuable insights, and promotes its application to extend its practice. EDS as a transdisciplinary field relies on statistics, probability, machine learning, data mining, and analytics, in addition to biological, psychological, and neurological knowledge aboutlearning science. With this in mind, the book is devoted to those that are in charge of educational management, educators, pedagogues, academics, computer technologists, researchers, and postgraduate students, who pursue to acquire a conceptual, formal, and practical landscape of how to deploy EDS to build proactive, real- time, and reactive applications that personalize education, enhance teaching, and improve learning!Chapter “Sync Ratio and Cluster Heat Map for Visualizing Student Engagement” is available open access under a Creative Commons Attribution 4.0 International License via link.springer.com.

This open access book systematically investigates the topic of entity alignment, which aims to detect equivalent entities that are located in different knowledge graphs. Entity alignment represents an essential step in enhancing the quality of knowledge graphs, and hence is of significance to downstream applications, e.g., question answering and recommender systems. Recent years have witnessed a rapid increase in the number of entity alignment frameworks, while the relationships among them remain unclear. This book aims to fill that gap by elaborating the concept and categorization of entity alignment, reviewing recent advances in entity alignment approaches, and introducing novel scenarios and corresponding solutions.Specifically, the book includes comprehensive evaluations and detailed analyses of state-of-the-art entity alignment approaches and strives to provide a clear picture of the strengths and weaknesses of the currently available solutions, so as to inspire follow-upresearch. In addition, it identifies novel entity alignment scenarios and explores the issues of large-scale data, long-tail knowledge, scarce supervision signals, lack of labelled data, and multimodal knowledge, offering potential directions for future research.The book offers a valuable reference guide for junior researchers, covering the latest advances in entity alignment, and a valuable asset for senior researchers, sharing novel entity alignment scenarios and their solutions. Accordingly, it will appeal to a broad audience in the fields of knowledge bases, database management, artificial intelligence and big data.

This open access book systematically investigates the topic of entity alignment, which aims to detect equivalent entities that are located in different knowledge graphs. Entity alignment represents an essential step in enhancing the quality of knowledge graphs, and hence is of significance to downstream applications, e.g., question answering and recommender systems. Recent years have witnessed a rapid increase in the number of entity alignment frameworks, while the relationships among them remain unclear. This book aims to fill that gap by elaborating the concept and categorization of entity alignment, reviewing recent advances in entity alignment approaches, and introducing novel scenarios and corresponding solutions.Specifically, the book includes comprehensive evaluations and detailed analyses of state-of-the-art entity alignment approaches and strives to provide a clear picture of the strengths and weaknesses of the currently available solutions, so as to inspire follow-upresearch. In addition, it identifies novel entity alignment scenarios and explores the issues of large-scale data, long-tail knowledge, scarce supervision signals, lack of labelled data, and multimodal knowledge, offering potential directions for future research.The book offers a valuable reference guide for junior researchers, covering the latest advances in entity alignment, and a valuable asset for senior researchers, sharing novel entity alignment scenarios and their solutions. Accordingly, it will appeal to a broad audience in the fields of knowledge bases, database management, artificial intelligence and big data.

With the growing advances in technology and transformation to digital services, the world is becoming more connected and more complex. Huge heterogeneous data are generated at rapid speed from various types of sensors. Augmented with artificial intelligence and machine learning and internet of things, latent relations, and new insights can be captured helping in optimizing plans and resource utilization, improving infrastructure, and enhancing quality of services.A “spatial data management system” is a way to take care of data that has something to do with space. This could include data such as maps, satellite images, and GPS data. A temporal data management system is a system designed to manage data that has a temporal component. This could include data such as weather data, financial data, and social media data. Some advanced techniques used in spatial and temporal data management systems include geospatial indexing for efficient querying and retrieval of location-based data, time-series analysis for understanding and predicting temporal patterns in datasets like weather or financial trends, machine learning algorithms for uncovering hidden patterns and correlations in large and complex datasets, and integration with Internet of Things (IoT) technologies for real-time data collection and analysis. These techniques, augmented with artificial intelligence, enable the extraction of latent relations and insights, thereby optimizing plans, improving infrastructure, and enhancing the quality of services. This book provides essential technical knowledge, best practices, and case studies on the state-of-the-art techniques of artificial intelligence and machine learning for spatiotemporal data analysis and modeling. The book is composed of several chapters written by experts in their fields and focusing on several applications including recommendation systems, big data analytics, supply chains and e-commerce, energy consumption and demand forecasting,and traffic and environmental monitoring. It can be used as academic reference at graduate level or by professionals in science and engineering related fields such as data science and engineering, big data analytics and mining, artificial intelligence, machine learning and deep learning, cloud computing, and internet of things.