Fundamentals of Data Analytics chapter 2

this chapter is about the fundamental of data analytics in internet of things

1. Fundamentals
of Data
Analytics
CSC1212 IOT DATA
ANALYTICS

2. Learning objectives
At the end of this chapter, student should be able to:
1.Define the IoT analytics
2.Determine the categorization of data
3.Explain the challenges of IoT Analytics

3. IMAGINE:
WHAT IF THINGS START TO THINK???

7. An introduction to IoT Data
Analytics
▪inIoTworld,thecreationofmassiveamountsofdata
fromsensorsiscommonandoneofthebiggest
challenges–notonlyfromatransportperspectivebut
alsofromadatamanagementstandpoint.
▪Example:Modernjetenginesarefittedwiththousands
ofsensorsthatgenerateawhopping10GBofdataper
second
▪analysingthisamountofdatainthemostefficient
mannerpossiblefallsundertheumbrellaofdata
analytics

8. An introduction to IoT Data
Analytics
▪Notalldataisthesame;itcanbecategorizedandthusanalyzedin
differentways.
▪Dependingonhowdataiscategorized,variousdataanalyticstoolsand
processingmethodscanbeapplied.
▪TWOimportantcategorizationfromanIoTperspectivearewhetherthedata
isstructuredorunstructuredANDwhetheritisinmotionoratrest.

9. Structured vs Unstructured Data
❑structuredandunstructureddataareimportantclassificationsasthey
typicallyrequiredifferenttoolsetfromadataanalyticsperspective.
❑structureddatameansthedatafollowsamodelorschemathatdefines
howthedataisrepresentedororganized,meaningitfitswellwitha
traditionalrelationaldatabasemanagementsystem(RDBMS)
❑thestructureddatainasimpletabularform–forexample,spreadsheet
wheredataoccupiesaspecificcellandcanbeexplicitlydefinedand
referenced

10. Structured vs Unstructured Data
❑structureddataiseasilyformatted,stored,queriedandprocessed.
❑becauseofthehighlyorganizationalformatofstructureddata,awidearray
ofdataanalyticstoolsarereadilyavailableforprocessingthistypeofdata
❑MicrosoftExcel
❑Tableau

11. Structured vs Unstructured Data
❑Unstructureddatalacksalogicalschemaforunderstandinganddecoding
thedatathroughprogramming.
❑Example,text,speech,imagesandvideos.
❑Asageneralrule,anydatadoesnotfitneatlyintoapredefineddatamodel
isclassifiedasunstructureddata.

12. Structured vs Unstructured Data
❑Accordingtosomeestimates,around80%ofabusiness’sdatais
unstructured.
❑Becauseofthisfact,dataanalyticsmethodsthatcanbeappliedto
unstructureddatacanbecognitivecomputingandmachinelearning.
[cognitivecomputing:hardwareand/orsoftwarethatmimicsthefunctioningofthe
humanbrainandhelpstoimprovehumandecision-making]

14. Data in Motion vs Data at Rest
❑DatainIoTnetworksiseitherintransit(“datainmotion”)orbeingheldor
stored(“dataatrest”).
❑Examplesofdatainmotionincludetraditionalclient/serverexchanges,
suchaswebbrowsingandfiletransfersandemail.
❑datasavedtoaharddrive,storagearrayorUSBdriveisdataatrest.

15. Data in Motion vs Data at Rest
❑FromIoTperspective,thedatafromsmartobjectisconsidereddatain
motionasitpassesthroughthenetworkenroutetoitsfinaldestination.
❑thisoftenprocessedattheedgeusingfogcomputing.
❑Attheedge,datamaybefilteredanddeletedorforwardedonfurther
processingandpossiblestorageatafognodeorindatacenter.
❑DataatrestinIoTnetworkscanbetypicallyfoundinIoTbrokersorinsome
sortofstoragearrayatthedatacenter

19. What is Data Analytics?
Dataanalyticsisthescienceandart!Applyingthestatisticaltechniquesto
largedatasetstoobtainactionableinsightsformakingsmartdecisions.
Itistheprocessuncoverhiddenpatterns,unknowncorrelations,trends,any
otherusefulbusinessinformation.

20. Type of
Data
Analysis
•What is happening?
Descriptive
•Why did it happen?
Diagnostic
•What is likely to happen?
Predictive
•What should I do about it?
Prescriptive

21. •Descriptivedataanalysistellsyouwhatishappening,either
noworinthepast.
•Forexample,athermometerinatruckenginereports
temperaturevalueseverysecond.
•Fromadescriptiveanalysisperspective,youcanpullthis
dataatanymomenttogaininsightintothecurrent
operatingconditionofthetruckengine.
•Ifthetemperaturevalueistoohigh,thentheremaybea
coolingproblem,ortheenginemaybeexperiencingtoo
muchload
Descriptive

22. •Whenyouareinterestedinthe“why,”diagnosticdataanalysis
canprovidetheanswer.
•Continuingwiththeexampleofthetemperaturesensorinthe
truckengine,youmightwonderwhythetruckenginefailed.
•Diagnosticanalysismightshowthatthetemperatureofthe
enginewastoohigh,andtheengineoverheated.
•Applyingdiagnosticanalysisacrossthedatageneratedbya
wide
•rangeofsmartobjectscanprovideaclearpictureofwhya
•problemoraneventoccurred
Diagnostic

23. •Predictiveanalysisaimstoforetellproblemsorissues
beforetheyoccur.
•Forexample,withhistoricalvaluesoftemperaturesforthe
truckengine,predictiveanalysiscouldprovideanestimate
ontheremaininglifeofcertaincomponentsintheengine.
•Thesecomponentscouldthenbeproactivelyreplaced
beforefailureoccurs.
•Orperhapsiftemperaturevaluesofthetruckenginestartto
riseslowlyovertime,thiscouldindicatetheneedforanoil
changeorsomeothersortofenginecoolingmaintenance.
Predictive

24. •Prescriptiveanalysisgoesastepbeyondpredictiveand
recommendssolutionsforupcomingproblems.
•Aprescriptiveanalysisofthetemperaturedatafromatruck
enginemightcalculatevariousalternativestocost-
effectivelymaintainourtruck
•Thesecalculationscouldrangefromthecostnecessaryfor
morefrequentoilchangesandcoolingmaintenanceto
installingnewcoolingequipmentontheengineorupgrading
toaleaseonamodelwithamorepowerfulengine.
•Prescriptiveanalysislooksatavarietyoffactorsandmakes
theappropriaterecommendation
Prescriptive

25. Both predictive and prescriptive analyses are more resource intensive and increase
complexity, but the value they provide is much greater than the value from descriptive
and diagnostic analysis

26. IoT Analytics Challenges
Too much
data
Security
Misbehaving
devices

27. IoT Analytics Challenges
Too
much
data
The total amount of data
being collected may be so
large that it may not be
possible to move it over the
network to a central location

28. IoT Analytics Challenges
SecurityIf the security on a specific vendor’s
outdoor sensor is weak, and the sensor is
connected to other devices, the likelihood
of ‘indirect’ critical impact is high.
Attackers can compromise the sensor and
modify its data or exploit the connection to
other devices to cause damage.

29. IoT Analytics Challenges
Misbehaving
device These are devices or sensors that go
bad and begin sending false readings
to the system. For example, a low
battery, a software bug, or a hardware
failure, could cause such readings.
This could ruin the inventory of the
warehouse.

30. Conclusion

31. Class Activities
1.Take your Phone Camera, walk around in the campus and take any picture that you think can be
capture any data of it.
2.Present your finding in the class. The slide presentation should include what type of
categorization of data is your finding
and what is the challenge to do the analysis of it.