Text Data Management and Analysis: A Practical Introduction to Information Retrieval and Text Mining

531 Pages·2016·27.22 MB·English

by  ChengXiang Zhai

Z
H
Text Data Management and Analysis A
I • Text Data
A Practical Introduction to Information Retrieval and Text Mining M
A
ChengXiang Zhai and Sean Massung S
S
Management
U
N
Recent years have seen a dramatic growth of natural language text data, including web pages, G
news articles, scientific literature, emails, enterprise documents, and social media such as
blog articles, forum posts, product reviews, and tweets. This has led to an increasing demand and Analysis
for powerful software tools to help people manage and analyze vast amounts of text data ef- T
e
fectively and efficiently. Unlike data generated by a computer system or sensors, text data are
x
usually generated directly by humans, and capture semantically rich content. As such, text t

data are especially valuable for discovering knowledge about human opinions and preferenc- D A Practical Introduction
es, in addition to many other kinds of knowledge that we encode in text. In contrast to struc- a
t
tured data, which conform to well-defined schemas (thus are relatively easy for computers to a to Information Retrieval
handle), text has less explicit structure, requiring computer processing toward understanding M
of the content encoded in text. The current technology of natural language processing has
a and Text Mining
not yet reached a point to enable a computer to precisely understand natural language text, n
but a wide range of statistical and heuristic approaches to management and analysis of text a
g
data have been developed over the past few decades. They are usually very robust and can be
e
applied to analyze and manage text data in any natural language, and about any topic. m

This book provides a systematic introduction to many of these approaches, with an em- e
n
phasis on covering the most useful knowledge and skills required to build a variety of prac-
t
tically useful text information systems. Because humans can understand natural languages
a
far better than computers can, effective involvement of humans in a text information system n
is generally needed and text information systems often serve as intelligent assistants for hu- d

mans. Depending on how a text information system collaborates with humans, we distinguish A
two kinds of text information systems. The first is information retrieval systems which include n
a
search engines and recommender systems; they assist users in finding from a large collection
l
of text data the most relevant text data that are actually needed for solving a specific applica- y
s
tion problem, thus effectively turning big raw text data into much smaller relevant text data
i
s
that can be more easily processed by humans. The second is text mining application systems;
they can assist users in analyzing patterns in text data to extract and discover useful action-
able knowledge directly useful for task completion or decision making, thus providing more ChengXiang Zhai
direct task support for users. A
C
M Sean Massung
|
ABOUT ACM BOOKS M
O
R
ACM Books is a new series of high quality books for
G
M the computer science community, published by ACM A
&Cin collaboration with Morgan & Claypool Publishers. N
&
ACM Books publications are widely distributed in C
both print and digital formats through booksellers L M
and to libraries (and library consortia) and individual ACM members via the ACM ISBN: 978-1-970009-010160-07 AY &C
Digital Library platform. P
O
O
BOOKS.ACM.ORG • WWW.MORGANCLAYPOOL.COM 9 781970 001167 L
Text Data Management
and Analysis

ACM Books
EditorinChief
M.TamerO¨zsu,UniversityofWaterloo
ACMBooksisanewseriesofhigh-qualitybooksforthecomputersciencecommunity,
published by ACM in collaboration with Morgan & Claypool Publishers. ACM Books
publicationsarewidelydistributedinbothprintanddigitalformatsthroughbooksellers
andtolibraries(andlibraryconsortia)andindividualACMmembersviatheACMDigital
Libraryplatform.
Text Data Management and Analysis: A Practical Introduction to Information
RetrievalandTextMining
ChengXiangZhai,UniversityofIllinoisatUrbana–Champaign
SeanMassung,UniversityofIllinoisatUrbana–Champaign
2016
AnArchitectureforFastandGeneralDataProcessingonLargeClusters
MateiZaharia,MassachusettsInstituteofTechnology
2016
ReactiveInternetProgramming:StateChartXMLinAction
FranckBarbier,UniversityofPau,France
2016
VerifiedFunctionalProgramminginAgda
AaronStump,TheUniversityofIowa
2016
TheVRBook:Human-CenteredDesignforVirtualReality
JasonJerald,NextGenInteractions
2016
Ada’sLegacy:CulturesofComputingfromtheVictoriantotheDigitalAge
RobinHammerman,StevensInstituteofTechnology
AndrewL.Russell,StevensInstituteofTechnology
2016
EdmundBerkeleyandtheSocialResponsibilityofComputerProfessionals
BernadetteLongo,NewJerseyInstituteofTechnology
2015
CandidateMultilinearMaps
SanjamGarg,UniversityofCalifornia,Berkeley
2015
SmarterthanTheirMachines:OralHistoriesofPioneersinInteractiveComputing
John Cullinane, Northeastern University; Mossavar-Rahmani Center for Business
andGovernment,JohnF.KennedySchoolofGovernment,HarvardUniversity
2015
AFrameworkforScientificDiscoverythroughVideoGames
SethCooper,UniversityofWashington
2014
Trust Extension as a Mechanism for Secure Code Execution on Commodity
Computers
BryanJeffreyParno,MicrosoftResearch
2014
EmbracingInterferenceinWirelessSystems
ShyamnathGollakota,UniversityofWashington
2014
Text Data Management
and Analysis
A Practical Introduction to Information
Retrieval and Text Mining
ChengXiang Zhai
UniversityofIllinoisatUrbana–Champaign
Sean Massung
UniversityofIllinoisatUrbana–Champaign
ACMBooks#12
Copyright©2016bytheAssociationforComputingMachinery
andMorgan&ClaypoolPublishers
Allrightsreserved.Nopartofthispublicationmaybereproduced, storedinaretrieval
system,ortransmittedinanyformorbyanymeans—electronic,mechanical,photocopy,
recording,oranyotherexceptforbriefquotationsinprintedreviews—withouttheprior
permissionofthepublisher.
Designations used by companies to distinguish their products are often claimed as
trademarksorregisteredtrademarks.InallinstancesinwhichMorgan&Claypoolisaware
ofaclaim,theproductnamesappearininitialcapitalorallcapitalletters.Readers,however,
should contact the appropriate companies for more complete information regarding
trademarksandregistration.
TextDataManagementandAnalysis
ChengXiangZhaiandSeanMassung
books.acm.org
www.morganclaypoolpublishers.com
ISBN:978-1-97000-119-8 hardcover
ISBN:978-1-97000-116-7 paperback
ISBN:978-1-97000-117-4 ebook
ISBN:978-1-97000-118-1 ePub
SeriesISSN: 2374-6769print 2374-6777electronic
DOIs:
10.1145/2915031 Book 10.1145/2915031.2915044 Chapter12
10.1145/2915031.2915032 Preface 10.1145/2915031.2915045 Chapter13
10.1145/2915031.2915033 Chapter1 10.1145/2915031.2915046 Chapter14
10.1145/2915031.2915034 Chapter2 10.1145/2915031.2915047 Chapter15
10.1145/2915031.2915035 Chapter3 10.1145/2915031.2915048 Chapter16
10.1145/2915031.2915036 Chapter4 10.1145/2915031.2915049 Chapter17
10.1145/2915031.2915037 Chapter5 10.1145/2915031.2915050 Chapter18
10.1145/2915031.2915038 Chapter6 10.1145/2915031.2915051 Chapter19
10.1145/2915031.2915039 Chapter7 10.1145/2915031.2915052 Chapter20
10.1145/2915031.2915040 Chapter8 10.1145/2915031.2915053 Appendices
10.1145/2915031.2915041 Chapter9 10.1145/2915031.2915054 References
10.1145/2915031.2915042 Chapter10 10.1145/2915031.2915055 Index
10.1145/2915031.2915043 Chapter11
ApublicationintheACMBooksseries,#12
EditorinChief:M.TamerO¨zsu,UniversityofWaterloo
AreaEditor:EdwardA.Fox,VirginiaTech
FirstEdition
10 9 8 7 6 5 4 3 2 1
ToMeiandAlex
ToKai

Contents
Preface xv
Acknowledgments xviii
PARTI OVERVIEW AND BACKGROUND 1
Chapter1 Introduction 3
1.1 FunctionsofTextInformationSystems 7
1.2 ConceptualFrameworkforTextInformationSystems 10
1.3 OrganizationoftheBook 13
1.4 HowtoUsethisBook 15
BibliographicNotesandFurtherReading 18
Chapter2 Background 21
2.1 BasicsofProbabilityandStatistics 21
2.2 InformationTheory 31
2.3 MachineLearning 34
BibliographicNotesandFurtherReading 36
Exercises 37
Chapter3 TextDataUnderstanding 39
3.1 HistoryandStateoftheArtinNLP 42
3.2 NLPandTextInformationSystems 43
3.3 TextRepresentation 46
3.4 StatisticalLanguageModels 50
BibliographicNotesandFurtherReading 54
Exercises 55