Table Of ContentHandbook of Statistics
Series Editor
C.R. Rao
C.R. Rao AIMSCS, University of Hyderabad Campus,
Hyderabad, India
North-HollandisanimprintofElsevier
Radarweg29,POBox211,1000AEAmsterdam, Netherlands
TheBoulevard,LangfordLane,Kidlington,OxfordOX51GB,UnitedKingdom
Copyright©2016Elsevier B.V.Allrightsreserved.
Nopart ofthispublicationmaybereproducedortransmittedinanyformorby
anymeans,electronicormechanical, includingphotocopying,recording,orany
informationstorageandretrieval system, withoutpermissioninwritingfromthe
publisher.Detailsonhowtoseekpermission,furtherinformationaboutthePublisher’s
permissionspoliciesandourarrangementswithorganizationssuchastheCopyright
Clearance CenterandtheCopyrightLicensingAgency, canbefoundatourwebsite:
www.elsevier.com/permissions.
Thisbookandtheindividualcontributionscontainedinit areprotected under
copyrightbythePublisher (otherthanasmaybenotedherein).
Notices
Knowledgeandbestpracticeinthisfieldareconstantlychanging.Asnewresearchand
experiencebroadenourunderstanding, changesinresearchmethods,professional
practices,ormedicaltreatment maybecome necessary.
Practitionersandresearchersmustalwaysrelyontheirownexperienceandknowledge
inevaluating andusinganyinformation,methods,compounds,orexperiments
describedherein.Inusingsuchinformationormethodstheyshouldbemindfuloftheir
ownsafetyandthesafetyofothers,includingpartiesforwhomtheyhavea
professionalresponsibility.
Tothefullestextentofthelaw,neitherthePublishernortheauthors,contributors,or
editors,assumeanyliabilityforanyinjuryand/ordamagetopersonsorpropertyasa
matterofproductsliability,negligence orotherwise,orfromanyuseoroperationof
anymethods,products,instructions, orideascontainedinthematerialherein.
ISBN:978-0-444-63744-4
ISSN:0169-7161
ForinformationonallNorth-Holland publications
visitourwebsiteat https://www.elsevier.com/
Publisher:Zoe Kruze
AcquisitionEditor:KirstenShankland
EditorialProjectManager:HannahColford
ProductionProjectManager:RadhakrishnanLakshmanan
CoverDesigner: VictoriaPearson
TypesetbySPi Global,India
Contributors
NumbersinParenthesesindicatethepagesonwhichtheauthor’scontributionsbegin.
S. Chakraborty (229),Universityof Missouri, Columbia,MO, UnitedStates
E. Fathi(169),EastCarolina University, Greenville, NC,UnitedStates
J.-B.Fiot(255),IBM Research, Dublin,Ireland
R.N.Gottumukkala (81), UniversityofLouisiana atLafayette, Lafayette, LA,
United States
V.N.Gudivada(3,169, 283,301),EastCarolina University, Greenville, NC,
United States
E. Harley (117),The MITRECorporation, McLean, VA,UnitedStates
F.D. Hudson (341), University Corporation for Advanced Internet Development,
White Plains,NY, UnitedStates
M.T.Irfan (169,283),Bowdoin College,Brunswick, ME,UnitedStates
S. Jothilakshmi (301),Annamalai University, Chidambaram,Tamil Nadu,India
M. Limiero(117),The MITRECorporation,McLean, VA, UnitedStates
V.P.A. Lonij (255),IBM Research,Dublin,Ireland
A.S. Maida (39), The Center for Advanced Computer Studies, The University of
Louisiana atLafayette, Lafayette, LA,UnitedStates
E.W. Nichols (341), University Corporation for Advanced Internet Development,
White Plains,NY, UnitedStates
S. Noel (117), TheMITRECorporation, McLean, VA,UnitedStates
V.V.Raghavan (81), UniversityofLouisiana atLafayette, Lafayette, LA,
United States
D.L. Rao (169), EastCarolinaUniversity, Greenville, NC,UnitedStates
M. Share (117),The MITRECorporation, McLean, VA,UnitedStates
S. Suthaharan (207), The University of North Carolina at Greensboro, Greensboro,
NC, UnitedStates
K.H.Tam(117),The MITRECorporation,McLean, VA, UnitedStates
S.R. Venna (81),Universityof LouisianaatLafayette, Lafayette, LA,United States
xiii
Preface
Cognitive computing is a nascent interdisciplinary domain. It is a confluence
ofcognitivescience,neuroscience,datascience,andcloudcomputing.Cogni-
tive science is the study of mind and offers theories, and mathematical and
computationalmodelsofhumancognition.Cognitivescienceitselfisaninter-
disciplinary domain and draws upon philosophy, linguistics, psychology, and
anthropology, among others.
Neuroscienceisthestudyofthenervoussystemincludingitsdevelopment,
structure,andfunction.Morespecifically,neuroscientistsstudythestructureof
thebrain,andhowbehaviorandcognitivefunctionsareregulatedbythebrain.
Brain imaging techniques such as positron emission tomography, functional
magneticresonanceimaging,diffusiontensorimaging,andarterialspinlabel-
ingenableprobingbrainfunctionsbothqualitativelyandquantitatively.
Data science is also an interdisciplinary domain. It deals with processes
and systems to extract information and knowledge from structured and
unstructured data using machine learning algorithms and statistical methods.
The end goal is to discover patterns, generate actionable insights, and answer
predictive questions.
Cloudcomputingprovidesturnkeysolutionssuchasplatform-as-a-service,
infrastructure-as-a-service,andsoftware-as-a-service.Ituseshigh-performance
CPUs,GPUs,neuromorphicprocessors,virtuallyunlimitedmemoryandstor-
age, and high-speed networks to provide computing resources on demand.
A fixed pool of these resources is dynamically provisioned among various
applicationsandcontinuallyadjustedsothattheapplicationscanguaranteeper-
formanceamidfluctuatingworkloads.Cloudcomputingachieveseconomiesof
scale and helps cognitive computing applications to perform at scale without
upfrontcomputinginvestments.Applications arebilledfor onlythe resources
theyactuallyuse.
Broadly,therearetwolinesofresearchinthecognitivecomputingdiscipline.
Thefirstoneiscenteredoncognitivescienceasthefoundationandencompasses
neuroscience, philosophy, psychology, anthropology, and linguistics research.
Thesecondoneismorerecentandisbasedoncomputerscienceasthefoundation.
It encompasses data science, statistics, and subdisciplines of computer science
suchashigh-performancecomputing,cloudcomputing,naturallanguageproces-
sing,computervision,machinelearning,informationretrieval,anddatamanage-
ment. These two lines of research are not only complementary but mutually
helpingtoacceleratediscoveriesandinnovation.
xv
xvi Preface
It is this synergistic confluence that makes cognitive computing powerful
andhasthepotentialforgroundbreakingdiscoveriesandadvances.Especially
theadvancesinthecomputingdisciplinearepoisedtobringabouttransforma-
tionalchangestothewayresearchisconductedinthediscipline.Forexample,
IBM’s TrueNorth cognitive computing system is a case in point. Its design is
inspired by the function and efficiency of the human brain. The TrueNorth
architectureprovidesaspikingneuronmodelasabuildingblock.Itsprogram-
ming paradigm is based on an abstraction called corelet, which represents a
network of neurosynaptic cores. The corelet encapsulates all details except
the external inputs and outputs. An object-oriented language is available for
programming corelets. A library of reusable corelets as well as an integrated
development environment help accelerate the development of cognitive com-
puting applications. Using this environment, IBM has already implemented
several algorithms including hidden Markov models, convolution networks,
andrestrictedBoltzmannmachines.Thesealgorithmshave beenincorporated
into applications such as speaker recognition, sequence prediction, and colli-
sion avoidance.
As of this writing, Nvidia released Tesla P100 GPU, which specifically
targets machine learning algorithmsthatemploydeep learning.P100features
150 billion transistors on a single chip. Also, Google released Natural
Language API, a cloud service that provides application developers access
topretrainedalgorithmsfor sentiment analysis, entityrecognition,andsyntax
analysis. Likewise, Speech API, Translate API, and Vision API are public
cloud services for speech to text conversion, translation between natural lan-
guages,andimage analysis, respectively. TheSpeech APIenables converting
audio to text for over 80 languages and the Translate API provides machine
translation between these languages. Applications can use the Vision API to
perform image analysis tasks including object detection and classification.
These computing advances will propel further research in cognitive and
neurosciences.
The goal of this handbook is to bring together a coherent body of knowl-
edge and recent research in cognitive computing. It promotes a unified view
ofthedomainandlaysthefoundationforcognitivecomputingasanacademic
disciplineandaresearchenterprise.Tothebestoftheeditors’knowledge,this
handbook is the first in formally defining cognitive computing and providing
anacademic expositionofthefield.Thehandbookaimstoserveasacatalyst
for advancing research in cognitive computing.
AUDIENCE
The handbook aims to meet the needs of both students and industry practi-
tioners. Especially it is suited for students in advanced undergraduate and
beginninggraduatecoursesoncognitivecomputing,neuroscience,andcogni-
tivescience.Itisalsoagoodsourceforgraduatestudentswhoplantopursue
Preface xvii
research in cognitive computing. The handbook is also a good reference for
industry practitioners who desire to learn about cognitive computing.
ORGANIZATION
Thehandbookiscomprisedof11chapters,whichareorganizedinto3sections.
SectionA:FundamentalsandPrinciplesconsistsoftwochapters,providesan
introductiontocognitivecomputing,andsetsthebackdropforreadingrestof
the handbook. Section B: Complex Analytics and Machine Learning is com-
prisedoffivechapters.Complexanalyticsandmachinelearningareasaredis-
cussed in this section. Lastly, Section C: Applications discusses applications
ofcognitivecomputingandfourchaptersaredevotedtothesetopics.
Section A: Fundamentals and Principles
Chapter 1: Cognitive Computing: Concepts, Architectures,
Systems, and Applications
This chapter provides an interdisciplinary introduction to cognitive comput-
ing. The aim of the chapter is to provide a unified view of the discipline. It
begins with an overview of cognitive science, data science, and cognitive
computing. Principal technology enablers of cognitive computing, an over-
viewofthreemajorcategoriesofcognitivearchitectures,cognitivecomputing
systems, and their applications are discussed. Current trends and future
research directions in cognitive computing are indicated. The chapter con-
cludes by listing various cognitive computing resources.
Chapter 2: Cognitive Computing and Neural Networks: Reverse
Engineering the Brain
IBM, Nvidia, and Qualcomm have developed microprocessors which mimic
neurons and synapses of the human brain. These microprocessors are called
neuromorphic chips, and IBM’s TrueNorth and the HumanBrain Project’s
SpiNNaker are examples. This chapter presents principles and theory needed
as a backdrop to understanding these advances from a cognitive science
perspective. Neural networks found within the mammalian neocortex, and
associated formal and computational models that appear to form the basis of
human cognition are described.
Section B: Complex Analytics and Machine Learning
Chapter 3: Visual Analytic Decision-Making Environments for
Large-Scale Time-Evolving Graphs
Data scientists are faced with the challenge of analyzing large-scale graphs
that change dynamically. Existing tools and metaphors for data collection,
processing, storage, and analysis are not suitable for handling large-scale
xviii Preface
evolutionary graphs. This chapter describes visual analytics as a cognitive
computing approach to improving decision making with large-scale dynamic
graphs. It provides a conceptual introduction to time-varying graphs,
describes functional components of systems for visual analytics including
performance considerations, and presents a visual graph analytics sandbox
architecture and sample applications implemented within it.
Chapter 4: CyGraph: Graph-Based Analytics and Visualization
for Cybersecurity
Theadversarialnatureandcomplexinterdependenciesofnetworkedmachines
demandacognitivesystemsapproachtocybersecurity.Thischapterdescribes
CyGraph, a graph-based cognitive system for protecting mission-critical
computing assets and applications. CyGraph brings together isolated data
and events into a comprehensive property-graph model, providing an overall
picture for decision support and situational awareness. CyGraph features
CyQL (CyGraph Query Language), a domain-specific query language for
expressing graph patterns of interest, with interactive visualization of query
results. CyGraph integrates with third-party tools for visualizing graph state
changes.CyGraphcanalsosynthesizegraphmodelswithparticularstatistical
properties.
Chapter 5: Cognitive Analytics: Going Beyond Big Data Analytics
and Machine Learning
Traditional data analytics evolved from the database domain and exclusively
focused on structured data stored in relational databases. It was propelled to
the next stage in its evolution with the advent of data warehouses and data
mining. Cognitive analytics is the third stage in this evolutionary path and
goesbeyondstructureddata.Itintegratessemistructuredandunstructureddata
into the analytic process. This chapter provides an introduction to cognitive
analytics. It describestypes oflearning and classes ofmachine learning algo-
rithms in the context of cognitive analytics. It proposes a reference architec-
ture for cognitive analytics and indicates ways to implementing it. It also
describes a few cognitive analytics applications.
Chapter 6: A Cognitive Random Forest: An Intra- and
Intercognitive Computing for Big Data Classification
Under Cune Condition
This chapter address the classification problem in the big data context. The
data is often noisy, inconsistent, and incomplete. To solve the classification
problem, a cognitive model (called STE-M) is proposed in this chapter.
Also, a cognitive computing architecture, called Cognitive Random Forest,
is proposed to implement STE-M. The architecture amalgamates the STE-M
model and a set of random forest classifiers to enhance continuous learning.
The architecture is implemented and validated.
Preface xix
Chapter 7: Bayesian Additive Regression Tree for Seemingly
Unrelated Regression with Automatic Tree Selection
This chapter introduces a flexible Bayesian additive regression tree (seem-
ingly unrelated regression) model, called BART-SUR, which is suitable for
situations where the response variable is a vector and the components of the
vector are highly correlated. BART-SUR can jointly model the correlation
structure among the related response variables and provide a highly flexible
and nonlinear regression structure for each of the individual regression func-
tions. The number of trees in BART-SUR is selected adaptively by treating
it as a model parameter and assigning a prior distribution on it. The adaptive
treeselectionmakesBART-SURextremelyfast.Theauthordemonstratesthe
superiorityofBART-SURoverseveraloutoftheshelvepopularmethodslike
the random forest, neural network, wavelet regression, and support vector
machine through two simulation studies and three real data applications.
Section C: Applications
Chapter 8: Cognitive Systems for the Food–Water–Energy Nexus
Meetingthefood,water,andenergyneedsofagrowingworldpopulationisa
grandchallenge.Theseresourcesareoftennotproducedinplaceswherethey
are consumed, which entails transportation and storage costs. One can avoid
storing a resource, if good forecast models for supply and demand exist.
Developing such models requires handling large-scale datasets efficiently,
building forecasting models using machine learning methods, and leveraging
optimization techniques to help incorporate forecasting results into a
decision-making process. Toward these goals, this chapter discusses methods
tomakethemostofthesensordata,includingdatafromsatellitesandground-
based cameras. Forecasting methods ranging from minutes-ahead to days-
ahead are described. Finally, ways to use the outputs of these analytics tools
to help decision-making processes are discussed in the context of energy.
Chapter 9: Cognitive Computing Applications in Education
and Learning
Education and learning applications stand out among many uses of cognitive
computing due to their practical appeal and research challenges. This chapter
discusses the role of cognitive computing in teaching and learning environ-
ments. More specifically, the chapter examines the important roles played
by the Educational Data Mining (EDM) and Learning Analytics (LA)
researchersinimprovingstudentlearning.Itdescribesanarchitectureforper-
sonalized eLearning and summarizes relevant research.
Chapter 10: Large-Scale Data Enabled Evolution of Spoken
Language Research and Applications
Humanlanguagesareusedintwoforms:writtenandspoken.Textandspeech
are the mediums for written and spoken languages, respectively. Human
xx Preface
languages are the most natural means of communication between cognitive
computing systems and their users. The emergence of big data and data sci-
ence is accelerating research and applications in the analysis and understand-
ingofhuman/naturallanguages.Thischapterprovidesanintroductorytutorial
on the core tasks in speech processing, reviews recent large-scale data-driven
approachestosolving problemsinspoken languages, describescurrent trends
in speech research, and indicates future research directions.
Chapter 11: The Internet of Things and Cognitive Computing
Internet of Things (IoT) technologies are now more widely deployed. The
confluence of IoT and cognitive computing provides unprecedented opportu-
nities to develop deeper insights from the data generated by IoT devices.
These actionable insights have the potential for transformationalchanges that
affect people, cities, and industry. This chapter explores the state of the art
and future opportunities to bring IoT and cognitive computing together to
solve a range of problems including smart cities and connected health care.
Venkat N. Gudivada
Vijay V. Raghavan
Venu Govindaraju
C.R. Rao
Chapter 1
Cognitive Computing:
Concepts, Architectures,
Systems, and Applications
V.N. Gudivada1
EastCarolinaUniversity,Greenville,NC,UnitedStates
1Correspondingauthor:e-mail:gudivadav15@ecu.edu
ABSTRACT
Cognitive computing is an emerging field ushered in by the synergistic confluence of
cognitivescience,datascience,andanarrayofcomputingtechnologies.Cognitivesci-
ence theories provide frameworks to describe various models of human cognition
includinghowinformationisrepresentedandprocessedbythebrain.Datasciencepro-
videsprocessesandsystemstoextractknowledgefrombothstructuredandunstructured
data. Cognitive computing employs the computing discipline’s theories, methods, and
tools to model human cognition. The recent advances in data science and computing
disciplines—neuromorphicprocessors,bigdata,predictivemodeling,machinelearning,
natural language understanding, and cloud computing—are accelerating advances in
cognitivescienceandcognitivecomputing.
Theoverarchinggoalofthischapteristoprovideaninterdisciplinaryintroduction
tocognitivecomputing.Thefocusisonbreadthtoprovideaunifiedviewofthedisci-
pline.Thechapterbeginswithanoverviewofcognitivescience,datascience,andcog-
nitive computing. The principal technology enablers of cognitive computing are
presentednext.Anoverviewofthreemajorcategoriesofcognitivearchitecturesispre-
sented, which is followed by a description of cognitive computing systems and their
applications. Trends and future research directions in cognitive computing are dis-
cussed.Thechapterconcludesbylistingvariouscognitivecomputingresources.
Keywords: Cognitive computing, Cognitive architectures, Cognitive models, Cogni-
tivesystems, Cognitiveapplications,Cognitivecomputingsystems, Datascience
1 INTRODUCTION
An autonomous system is a self-contained and self-regulated entity. The sys-
tem continually reconstitutes itself in real time in response to changes in its
HandbookofStatistics,Vol.35.http://dx.doi.org/10.1016/bs.host.2016.07.004
©2016ElsevierB.V.Allrightsreserved. 3