Jiawei Han - Böcker

Our ability to generate and collect data has been increasing rapidly. Not only are all of our business, scientific, and government transactions now computerized, but the widespread use of digital cameras, publication tools, and bar codes also generate data. On the collection side, scanned text and image platforms, satellite remote sensing systems, and the World Wide Web have flooded us with a tremendous amount of data. This explosive growth has generated an even more urgent need for new techniques and automated tools that can help us transform this data into useful information and knowledge.

Like the first edition, voted the most popular data mining book by KD Nuggets readers, this book explores concepts and techniques for the discovery of patterns hidden in large data sets, focusing on issues relating to their feasibility, usefulness, effectiveness, and scalability. However, since the publication of the first edition, great progress has been made in the development of new data mining methods, systems, and applications. This new edition substantially enhances the first edition, and new chapters have been added to address recent developments on mining complex types of data- including stream data, sequence data, graph structured data, social network data, and multi-relational data.

A comprehensive, practical look at the concepts and techniques you need to know to get the most out of real business data Updates that incorporate input from readers, changes in the field, and more material on statistics and machine learning Dozens of algorithms and implementation examples, all in easily understood pseudo-code and suitable for use in real-world, large-scale data mining projects Complete classroom support for instructors at www.mkp.com/datamining2e companion site

This book brings all of the elements of data mining together in a single volume, saving the reader the time and expense of making multiple purchases. It consolidates both introductory and advanced topics, thereby covering the gamut of data mining and machine learning tactics ? from data integration and pre-processing, to fundamental algorithms, to optimization techniques and web mining methodology. The proposed book expertly combines the finest data mining material from the Morgan Kaufmann portfolio. Individual chapters are derived from a select group of MK books authored by the best and brightest in the field. These chapters are combined into one comprehensive volume in a way that allows it to be used as a reference work for those interested in new and developing aspects of data mining. This book represents a quick and efficient way to unite valuable content from leading data mining experts, thereby creating a definitive, one-stop-shopping opportunity for customers to receive the information they would otherwise need to round up from separate sources. Chapters contributed by various recognized experts in the field let the reader remain up to date and fully informed from multiple viewpoints. Presents multiple methods of analysis and algorithmic problem-solving techniques, enhancing the reader's technical expertise and ability to implement practical solutions. Coverage of both theory and practice brings all of the elements of data mining together in a single volume, saving the reader the time and expense of making multiple purchases.

Data Mining: Concepts and Techniques, Fourth Edition introduces concepts, principles, and methods for mining patterns, knowledge, and models from various kinds of data for diverse applications. Specifically, it delves into the processes for uncovering patterns and knowledge from massive collections of data, known as knowledge discovery from data, or KDD. It focuses on the feasibility, usefulness, effectiveness, and scalability of data mining techniques for large data sets.

After an introduction to the concept of data mining, the authors explain the methods for preprocessing, characterizing, and warehousing data. They then partition the data mining methods into several major tasks, introducing concepts and methods for mining frequent patterns, associations, and correlations for large data sets; data classificcation and model construction; cluster analysis; and outlier detection. Concepts and methods for deep learning are systematically introduced as one chapter. Finally, the book covers the trends, applications, and research frontiers in data mining.

Presents a comprehensive new chapter on deep learning, including improving training of deep learning models, convolutional neural networks, recurrent neural networks, and graph neural networksAddresses advanced topics in one dedicated chapter: data mining trends and research frontiers, including mining rich data types (text, spatiotemporal data, and graph/networks), data mining applications (such as sentiment analysis, truth discovery, and information propagattion), data mining methodologie and systems, and data mining and societyProvides a comprehensive, practical look at the concepts and techniques needed to get the most out of your dataVisit the author-hosted companion site, https://hanj.cs.illinois.edu/bk4/ for downloadable lecture slides and errata

Drawn from the US National Science Foundation’s Symposium on Next Generation of Data Mining and Cyber-Enabled Discovery for Innovation (NGDM 07), Next Generation of Data Mining explores emerging technologies and applications in data mining as well as potential challenges faced by the field.Gathering perspectives from top experts across different disciplines, the book debates upcoming challenges and outlines computational methods. The contributors look at how ecology, astronomy, social science, medicine, finance, and more can benefit from the next generation of data mining techniques. They examine the algorithms, middleware, infrastructure, and privacy policies associated with ubiquitous, distributed, and high performance data mining. They also discuss the impact of new technologies, such as the semantic web, on data mining and provide recommendations for privacy-preserving mechanisms.The dramatic increase in the availability of massive, complex data from various sources is creating computing, storage, communication, and human-computer interaction challenges for data mining. Providing a framework to better understand these fundamental issues, this volume surveys promising approaches to data mining problems that span an array of disciplines.

An emerging topic in software engineering and data mining, specification mining tackles software maintenance and reliability issues that cost economies billions of dollars each year. The first unified reference on the subject, Mining Software Specifications: Methodologies and Applications describes recent approaches for mining specifications of software systems. Experts in the field illustrate how to apply state-of-the-art data mining and machine learning techniques to address software engineering concerns.In the first set of chapters, the book introduces a number of studies on mining finite state machines that employ techniques, such as grammar inference, partial order mining, source code model checking, abstract interpretation, and more. The remaining chapters present research on mining temporal rules/patterns, covering techniques that include path-aware static program analyses, lightweight rule/pattern mining, statistical analysis, and other interesting approaches. Throughout the book, the authors discuss how to employ dynamic analysis, static analysis, and combinations of both to mine software specifications.According to the US National Institute of Standards and Technology in 2002, software bugs have cost the US economy 59.5 billion dollars a year. This volume shows how specification mining can help find bugs and improve program understanding, thereby reducing unnecessary financial losses. The book encourages the industry adoption of specification mining techniques and the assimilation of these techniques in standard integrated development environments (IDEs).

The Definitive Volume on Cutting-Edge Exploratory Analysis of Massive Spatial and Spatiotemporal DatabasesSince the publication of the first edition of Geographic Data Mining and Knowledge Discovery, new techniques for geographic data warehousing (GDW), spatial data mining, and geovisualization (GVis) have been developed. In addition, there has been a rise in the use of knowledge discovery techniques due to the increasing collection and storage of data on spatiotemporal processes and mobile objects. Incorporating these novel developments, this second edition reflects the current state of the art in the field.New to the Second Edition Updated material on geographic knowledge discovery (GKD), GDW research, map cubes, spatial dependency, spatial clustering methods, clustering techniques for trajectory data, the INGENS 2.0 software, and GVis techniquesNew chapter on data quality issues in GKDNew chapter that presents a tree-based partition querying methodology for medoid computation in large spatial databasesNew chapter that discusses the use of geographically weighted regression as an exploratory techniqueNew chapter that gives an integrated approach to multivariate analysis and geovisualizationFive new chapters on knowledge discovery from spatiotemporal and mobile objects databasesGeographic data mining and knowledge discovery is a promising young discipline with many challenging research problems. This book shows that this area represents an important direction in the development of a new generation of spatial analysis tools for data-rich environments. Exploring various problems and possible solutions, it will motivate researchers to develop new methods and applications in this emerging field.

Drawn from the US National Science Foundation’s Symposium on Next Generation of Data Mining and Cyber-Enabled Discovery for Innovation (NGDM 07), Next Generation of Data Mining explores emerging technologies and applications in data mining as well as potential challenges faced by the field.Gathering perspectives from top experts across different disciplines, the book debates upcoming challenges and outlines computational methods. The contributors look at how ecology, astronomy, social science, medicine, finance, and more can benefit from the next generation of data mining techniques. They examine the algorithms, middleware, infrastructure, and privacy policies associated with ubiquitous, distributed, and high performance data mining. They also discuss the impact of new technologies, such as the semantic web, on data mining and provide recommendations for privacy-preserving mechanisms.The dramatic increase in the availability of massive, complex data from various sources is creating computing, storage, communication, and human-computer interaction challenges for data mining. Providing a framework to better understand these fundamental issues, this volume surveys promising approaches to data mining problems that span an array of disciplines.

An emerging topic in software engineering and data mining, specification mining tackles software maintenance and reliability issues that cost economies billions of dollars each year. The first unified reference on the subject, Mining Software Specifications: Methodologies and Applications describes recent approaches for mining specifications of software systems. Experts in the field illustrate how to apply state-of-the-art data mining and machine learning techniques to address software engineering concerns.In the first set of chapters, the book introduces a number of studies on mining finite state machines that employ techniques, such as grammar inference, partial order mining, source code model checking, abstract interpretation, and more. The remaining chapters present research on mining temporal rules/patterns, covering techniques that include path-aware static program analyses, lightweight rule/pattern mining, statistical analysis, and other interesting approaches. Throughout the book, the authors discuss how to employ dynamic analysis, static analysis, and combinations of both to mine software specifications.According to the US National Institute of Standards and Technology in 2002, software bugs have cost the US economy 59.5 billion dollars a year. This volume shows how specification mining can help find bugs and improve program understanding, thereby reducing unnecessary financial losses. The book encourages the industry adoption of specification mining techniques and the assimilation of these techniques in standard integrated development environments (IDEs).

Machine Learning and Knowledge Discovery for Engineering Systems Health Management presents state-of-the-art tools and techniques for automatically detecting, diagnosing, and predicting the effects of adverse events in an engineered system. With contributions from many top authorities on the subject, this volume is the first to bring together the two areas of machine learning and systems health management.Divided into three parts, the book explains how the fundamental algorithms and methods of both physics-based and data-driven approaches effectively address systems health management. The first part of the text describes data-driven methods for anomaly detection, diagnosis, and prognosis of massive data streams and associated performance metrics. It also illustrates the analysis of text reports using novel machine learning approaches that help detect and discriminate between failure modes. The second part focuses on physics-based methods for diagnostics and prognostics, exploring how these methods adapt to observed data. It covers physics-based, data-driven, and hybrid approaches to studying damage propagation and prognostics in composite materials and solid rocket motors. The third part discusses the use of machine learning and physics-based approaches in distributed data centers, aircraft engines, and embedded real-time software systems.Reflecting the interdisciplinary nature of the field, this book shows how various machine learning and knowledge discovery techniques are used in the analysis of complex engineering systems. It emphasizes the importance of these techniques in managing the intricate interactions within and between the systems to maintain a high degree of reliability.

With the recent ?ourishing research activities on Web search and mining, social networkanalysis,informationnetworkanalysis,informationretrieval,linkana- sis,andstructuraldatamining,researchonlinkmininghasbeenrapidlygrowing, forminganew?eldofdatamining. Traditionaldataminingfocuseson"?at"or"isolated"datainwhicheachdata objectisrepresentedasanindependentattributevector. However,manyreal-world data sets are inter-connected, much richer in structure, involving objects of h- erogeneoustypesandcomplexlinks. Hence,thestudyoflinkminingwillhavea highimpactonvariousimportantapplicationssuchasWebandtextmining,social networkanalysis,collaborative?ltering,andbioinformatics. Asanemergingresearch?eld,therearecurrentlynobooksfocusingonthetheory andtechniquesaswellastherelatedapplicationsforlinkmining,especiallyfrom aninterdisciplinarypointofview. Ontheotherhand,duetothehighpopularity oflinkagedata,extensiveapplicationsrangingfromgovernmentalorganizationsto commercial businesses to people's daily life call for exploring the techniques of mininglinkagedata.Therefore,researchersandpractitionersneedacomprehensive booktosystematicallystudy,furtherdevelop,andapplythelinkminingtechniques totheseapplications. Thisbookcontainscontributedchaptersfromavarietyofprominentresearchers inthe?eld. Whilethechaptersarewrittenbydifferentresearchers,thetopicsand contentareorganizedinsuchawayastopresentthemostimportantmodels,al- rithms,andapplicationsonlinkmininginastructuredandconciseway. Giventhe lackofstructurallyorganizedinformationonthetopicoflinkmining,thebookwill provideinsightswhicharenoteasilyaccessibleotherwise. Wehopethatthebook willprovideausefulreferencetonotonlyresearchers,professors,andadvanced levelstudentsincomputersciencebutalsopractitionersinindustry. Wewouldliketoconveyourappreciationtoallauthorsfortheirvaluablec- tributions. WewouldalsoliketoacknowledgethatthisworkissupportedbyNSF throughgrantsIIS-0905215,IIS-0914934,andDBI-0960443. Chicago,Illinois PhilipS. Yu Urbana-Champaign,Illinois JiaweiHan Pittsburgh,Pennsylvania ChristosFaloutsos v Contents Part I Link-Based Clustering 1 Machine Learning Approaches to Link-Based Clustering...3 Zhongfei(Mark)Zhang,BoLong,ZhenGuo,TianbingXu, andPhilipS.Yu 2 Scalable Link-Based Similarity Computation and Clustering...45 XiaoxinYin,JiaweiHan,andPhilipS. Yu 3 Community Evolution and Change Point Detection in Time-Evolving Graphs...73 JimengSun,SpirosPapadimitriou,PhilipS. Yu,andChristosFaloutsos Part II Graph Mining and Community Analysis 4 A Survey of Link Mining Tasks for Analyzing Noisy and Incomplete Networks...107 GalileoMarkNamata,HossamSharara,andLiseGetoor 5 Markov Logic: A Language and Algorithms for Link Mining...135 PedroDomingos,DanielLowd,StanleyKok,AniruddhNath,Hoifung Poon,MatthewRichardson,andParagSingla 6 Understanding Group Structures and Properties in Social Media...163 LeiTangandHuanLiu 7 Time Sensitive Ranking with Application to Publication Search...187 XinLi,BingLiu,andPhilipS. Yu 8 Proximity Tracking on Dynamic Bipartite Graphs: Problem De?nitions and Fast Solutions...211 Hanghang Tong, Spiros Papadimitriou, Philip S. Yu, andChristosFaloutsos vii viii Contents 9 Discriminative Frequent Pattern-Based Graph Classi?cation...237 HongCheng,XifengYan,andJiaweiHan Part III Link Analysis for Data Cleaning and Information Integration 10 Information Integration for Graph Databases...265 Ee-PengLim,AixinSun,AnwitamanDatta,andKuiyuChang 11 Veracity Analysis and Object Distinction...283 XiaoxinYin,JiaweiHan,andPhilipS. Yu Part IV Social Network Analysis 12 Dynamic Community Identi?cation...

With the recent ?ourishing research activities on Web search and mining, social networkanalysis,informationnetworkanalysis,informationretrieval,linkana- sis,andstructuraldatamining,researchonlinkmininghasbeenrapidlygrowing, forminganew?eldofdatamining. Traditionaldataminingfocuseson"?at"or"isolated"datainwhicheachdata objectisrepresentedasanindependentattributevector. However,manyreal-world data sets are inter-connected, much richer in structure, involving objects of h- erogeneoustypesandcomplexlinks. Hence,thestudyoflinkminingwillhavea highimpactonvariousimportantapplicationssuchasWebandtextmining,social networkanalysis,collaborative?ltering,andbioinformatics. Asanemergingresearch?eld,therearecurrentlynobooksfocusingonthetheory andtechniquesaswellastherelatedapplicationsforlinkmining,especiallyfrom aninterdisciplinarypointofview. Ontheotherhand,duetothehighpopularity oflinkagedata,extensiveapplicationsrangingfromgovernmentalorganizationsto commercial businesses to people's daily life call for exploring the techniques of mininglinkagedata.Therefore,researchersandpractitionersneedacomprehensive booktosystematicallystudy,furtherdevelop,andapplythelinkminingtechniques totheseapplications. Thisbookcontainscontributedchaptersfromavarietyofprominentresearchers inthe?eld. Whilethechaptersarewrittenbydifferentresearchers,thetopicsand contentareorganizedinsuchawayastopresentthemostimportantmodels,al- rithms,andapplicationsonlinkmininginastructuredandconciseway. Giventhe lackofstructurallyorganizedinformationonthetopicoflinkmining,thebookwill provideinsightswhicharenoteasilyaccessibleotherwise. Wehopethatthebook willprovideausefulreferencetonotonlyresearchers,professors,andadvanced levelstudentsincomputersciencebutalsopractitionersinindustry. Wewouldliketoconveyourappreciationtoallauthorsfortheirvaluablec- tributions. WewouldalsoliketoacknowledgethatthisworkissupportedbyNSF throughgrantsIIS-0905215,IIS-0914934,andDBI-0960443. Chicago,Illinois PhilipS. Yu Urbana-Champaign,Illinois JiaweiHan Pittsburgh,Pennsylvania ChristosFaloutsos v Contents Part I Link-Based Clustering 1 Machine Learning Approaches to Link-Based Clustering...3 Zhongfei(Mark)Zhang,BoLong,ZhenGuo,TianbingXu, andPhilipS.Yu 2 Scalable Link-Based Similarity Computation and Clustering...45 XiaoxinYin,JiaweiHan,andPhilipS. Yu 3 Community Evolution and Change Point Detection in Time-Evolving Graphs...73 JimengSun,SpirosPapadimitriou,PhilipS. Yu,andChristosFaloutsos Part II Graph Mining and Community Analysis 4 A Survey of Link Mining Tasks for Analyzing Noisy and Incomplete Networks...107 GalileoMarkNamata,HossamSharara,andLiseGetoor 5 Markov Logic: A Language and Algorithms for Link Mining...135 PedroDomingos,DanielLowd,StanleyKok,AniruddhNath,Hoifung Poon,MatthewRichardson,andParagSingla 6 Understanding Group Structures and Properties in Social Media...163 LeiTangandHuanLiu 7 Time Sensitive Ranking with Application to Publication Search...187 XinLi,BingLiu,andPhilipS. Yu 8 Proximity Tracking on Dynamic Bipartite Graphs: Problem De?nitions and Fast Solutions...211 Hanghang Tong, Spiros Papadimitriou, Philip S. Yu, andChristosFaloutsos vii viii Contents 9 Discriminative Frequent Pattern-Based Graph Classi?cation...237 HongCheng,XifengYan,andJiaweiHan Part III Link Analysis for Data Cleaning and Information Integration 10 Information Integration for Graph Databases...265 Ee-PengLim,AixinSun,AnwitamanDatta,andKuiyuChang 11 Veracity Analysis and Object Distinction...283 XiaoxinYin,JiaweiHan,andPhilipS. Yu Part IV Social Network Analysis 12 Dynamic Community Identi?cation...

Departing from many existing network models that view interconnected data as homogeneous graphs or networks, our semi-structured heterogeneous information network model leverages the rich semantics of typed nodes and links in a network and uncovers surprisingly rich knowledge from the network.

Outlier (or anomaly) detection is a very broad field which has been studied in the context of a large number of research areas like statistics, data mining, sensor networks, environmental science, distributed systems, spatio-temporal mining, etc. Initial research in outlier detection focused on time series-based outliers (in statistics). Since then, outlier detection has been studied on a large variety of data types including high-dimensional data, uncertain data, stream data, network data, time series data, spatial data, and spatio-temporal data. While there have been many tutorials and surveys for general outlier detection, we focus on outlier detection for temporal data in this book. A large number of applications generate temporal datasets. For example, in our everyday life, various kinds of records like credit, personnel, financial, judicial, medical, etc., are all temporal. This stresses the need for an organized and detailed study of outliers with respect to such temporal data.In the past decade, there has been a lot of research on various forms of temporal data including consecutive data snapshots, series of data snapshots and data streams. Besides the initial work on time series, researchers have focused on rich forms of data including multiple data streams, spatio-temporal data, network data, community distribution data, etc. Compared to general outlier detection, techniques for temporal outlier detection are very different. In this book, we will present an organized picture of both recent and past research in temporal outlier detection. We start with the basics and then ramp up the reader to the main ideas in state-of-the-art outlier detection techniques. We motivate the importance of temporal outlier detection and brief the challenges beyond usual outlier detection. Then, we list down a taxonomy of proposed techniques for temporal outlier detection. Such techniques broadly include statistical techniques (like AR models, Markov models, histograms, neuralnetworks), distance- and density-based approaches, grouping-based approaches (clustering, community detection), network-based approaches, and spatio-temporal outlier detection approaches. We summarize by presenting a wide collection of applications where temporal outlier detection techniques have been applied to discover interesting outliers. Table of Contents: Preface / Acknowledgments / Figure Credits / Introduction and Challenges / Outlier Detection for Time Series and Data Sequences / Outlier Detection for Data Streams / Outlier Detection for Distributed Data Streams / Outlier Detection for Spatio-Temporal Data / Outlier Detection for Temporal Network Data / Applications of Outlier Detection for Temporal Data / Conclusions and Research Directions / Bibliography / Authors' Biographies

The "big data" era is characterized by an explosion of information in the form of digital data collections, ranging from scientific knowledge, to social media, news, and everyone's daily life.

A lot of digital ink has been spilled on "big data" over the past few years. Most of this surge owes its origin to the various types of unstructured data in the wild, among which the proliferation of text-heavy data is particularly overwhelming, attributed to the daily use of web documents, business reviews, news, social posts, etc., by so many people worldwide.A core challenge presents itself: How can one efficiently and effectively turn massive, unstructured text into structured representation so as to further lay the foundation for many other downstream text mining applications?In this book, we investigated one promising paradigm for representing unstructured text, that is, through automatically identifying high-quality phrases from innumerable documents. In contrast to a list of frequent n-grams without proper filtering, users are often more interested in results based on variable-length phrases with certain semantics such as scientific concepts, organizations, slogans,and so on. We propose new principles and powerful methodologies to achieve this goal, from the scenario where a user can provide meaningful guidance to a fully automated setting through distant learning. This book also introduces applications enabled by the mined phrases and points out some promising research directions.

Departing from many existing structure extraction methods that have heavy reliance on human annotated data for model training, our effort-light approach leverages human-curated facts stored in external knowledge bases as distant supervision and exploits rich data redundancy in large text corpora for context understanding.

However, acquiring such multidimensional knowledge from massive text data remains a challenging task.This book presents data mining techniques that turn unstructured text data into multidimensional knowledge.

It is ubiquitous in our daily life, ranging from product taxonomies used by online retailers, topic taxonomies deployed by news outlets and social media, as well as scientific taxonomies deployed by digital libraries across various domains.

It is ubiquitous in our daily life, ranging from product taxonomies used by online retailers, topic taxonomies deployed by news outlets and social media, as well as scientific taxonomies deployed by digital libraries across various domains.

Each chapter is self-contained, and synthesizes one aspect of frequent pattern mining. Key topics include: Pattern Growth Methods, Frequent Pattern Mining in Data Streams, Mining Graph Patterns, Big Data Frequent Pattern Mining, Algorithms for Data Clustering and more.