Technical seminar for btech student for the presentation which can be present

Technical seminars report for btech students

1. BhojReddy Engineering College for Women
(Sponsored by Sangam LaxmibaiVidyapeet, approved by AICTE and affiliated to JNTUH)
Vinaynagar, IS SadanCrossroads, Saidabad, Hyderabad –500 059, Telangana. www.brecw.ac.in
Department of
Electronics
and Communication Engineering
Seminar on
Application of Artificial Intelligence to
Network Forensics
Internal Guide: Incharge: PranathiiSubburu
SVMG PhaniKumar C Shafia Tasneem 20321A0492
Assistant Professor Assistant Professor IV ECE-B

2. CONTENTS
Introduction.
Datasets for Network Forensics
State of the Art
Advantages
Disadvantages
Applications
Conclusion
Future scope
References
13-10-2023
2

3. INTRODUCTION
ArtificialIntelligenceiswidelyadoptedbyvariousorganizationsandtechnologies.
ArtificialIntelligencetechnologiesareplayingcrucialincybersecurityanddigitalforensics.
TheGrowthincybercrimeandtherelevanceofdigitaldevicesincrimeinvestigationshas
drivendemandfordigitalforensics.
26-09-2023 3

4. 
Digitalforensicsisusedtogatherevidenceforlegalpurposes.Ithelpsuncoverdigital
evidencetosupportinvestigationsandlegalproceedings.
Digitalforensicsinvolveslegallyacceptableprocessincludes
•Identification
•Verifications
•Analysis
•Presentationofdigitalevidence
NetworkForensicsinvolvesanalyzingnetworktraffictoinvestigatesecurityincidents,data
breachesandpolicyviolationsandhelpsinunderstandingamitigatingvariousattacks.
13-10-2023
4

5. Networkforensicsfocusesontheidentificationandinvestigationofinternalandexternal
networkattacksandtheuninstrumentedinvestigationofnetworkeddevices.Itliesatthe
intersectionofdigitalforensics,incidentresponseandnetworksecurity
13-10-2023
5

6. DATASETS FOR NETWORK FORENSICS

Innetworkforensics,datasetsarecollectionsofnetworktrafficdatathatareusedfor
analysisandinvestigation.
Thesedatasetstypicallycontaininformationsuchaspacketcaptures,networklogs,and
othernetwork-relateddata.
Theyarevaluableresourcesforstudyingnetworkbehavior,identifyingsecurityincidents,
andconductingresearchinthefieldofnetworkforensics.
13-10-2023
6

7. There are various publicly available datasets for network forensics, such as
•DARPAIntrusionDetectionDataSets:Thisdatasetisoneofthemostcommonlyusedfor
networkintrusiondetectionresearch.Itcontainsalargeamountofsimulatednetworktraffic
datawithdifferenttypesofattacks,includingDoSattacks,probing,andU2Rattacks.
•UNSW-NB15:TheUniversityofNewSouthWales(UNSW)providesthisdataset,which
includesadiverserangeofnetworktrafficdata.Itcontainsnormaltrafficaswellasvarious
typesofattacks,includingDoS,probing,andmalware.
•ISCXNSL-KDDDataset:ThisdatasetisanimprovedversionoftheKDDCup1999dataset,
addressingsomeofitslimitations.Itincludeslabelednetworktrafficdatawithattacksand
normaltraffic.
13-10-2023 7

8. •
CIC-IDS2017Dataset:Thisdatasetcontainslabelednetworktrafficdatafromthe
CanadianInstituteforCybersecurity(CIC).Itincludesavarietyofattacksandnormaltraffic.
•AWID(AarhusWirelessDataset):Thisdatasetfocusesonwirelessnetworktraffic
analysis.Itincludesdatafromdifferentwirelessprotocolsandcapturesvarioustypesof
networkactivities.
•CTU-13Dataset:TheCzechTechnicalUniversityinPragueprovidesthisdataset,which
containsdatarelatedtorealbotnetattacks.Itincludesnetworktrafficfrom13different
scenarios.
•SCXVPN-nonVPNDataset :ThisdatasetfocusesondistinguishingbetweenVPNand
non-VPNnetworktraffic.Itcontainsdatafrombothtypesoftraffic.
13-10-2023
8

9. STATE OF THE ART
13-10-2023
9
NETWORK TAFFIC ANALYSIS
NetworkTrafficAnalysisdetectsand
analyzessecuritythreatsandoperational
difficultiesusingnetworkcommunications
andprotocols.
NetworkTrafficAnalysismovesthreat
huntingfromsecurityperimetersand
endpointstonetworkflows.

10. NTAcombinesmachinelearning,advancedanalytics,andrule-baseddetectiontocreatea
baselinemodelofnormalnetworkactivity.
Anomalouspatternstriggerhighlycontextualizedwarnings.
IncreasednetworktrafficvolumeshaveledtothedevelopmentofnewNTAtechniques.
Expertsystemswerepopularinthe1990sforNTA,utilizingrule-baseddetectionand
logicalconclusions.
SEMACSisareal-timemonitorandcontrolsystemimplementedonthenetwork.
13-10-2023 10

11. INSTRUSION DETECTION SYSTEM
ANetwork-basedIntrusionDetectionSystem(NIDS)isamechanismthatmonitorsnetwork
trafficforhostileandsuspiciousbehavior.
NIDSplaysavitalroleinenhancingsecuritycontrolsandsecuringnetworkenvironments.
13-10-2023 11

12. 
NIDSmonitorsnetworktrafficforsuspiciousbehavior.
NIDScanstrengthensecuritycontrolsandsecurethenetworkenvironmentoforganizations.
NIDSusestwomainapproaches:signature-basedandanomaly-based.
•Signature-basedNIDSreliesonalibraryofknownattackstocomparenetworktrafficagainst
knownvulnerabilities.
•Anomaly-basedNIDSusesAItechniquestodetectabnormalbehavior.
NIDSmostlyrelyonpubliclyavailabledatasets,raisingquestionsabouttheirrepresentation
ofreal-worldtraffic.
13-10-2023 12

13. IoT FORENSICS
IoT(InternetofThings)isarelativelynewcategory
ofconsumerandindustrialelectronics.
IoTforensicsisanemergingfieldfocusedon
detectingandretrievingdigitalinformationinanethicalandforensicallyreliablemanner.
EvidenceinIoTforensicscanbecollectedfromanIoTdevice'slocalstorage,localnetwork,
orassociatedcloudserviceback-end.
13-10-2023 13

14. AspecializedincidentresponsetechniqueforIoTforensicsisneededtoaddresscybercrime
incidentsintheIoTdomain.
CLOUDFORENSICS
Cloudforensicsinvolvestheexamination
ofcybercrimethatrequiresevidencefrom
cloudcomputingplatformsorservices.
Thechallengeincloudforensicsisthatevidencecanbestoredanywhereinavirtual
environment.
13-10-2023 14

15. Earlyinvestigationsreliedontraditionaldigitalforensicsmethodologiesandtechnologies.
Rapidimprovementsincloudcomputingledtothedevelopmentofnewapproaches,frameworks,
andtoolsforcloudforensics.
Cloudforensicscoverevidencecollection,networkconcerns,privacyissues,andframeworks.
Challengesincloudcomputingincludetrust,threats,risks,andinsiderattacks.
DNSTUNNELING
DNStunnelingisanattacktechniquethatusesDNSrequestsandresponsestotransmitdata.
DNStunnelscancarrymalwarepayloadsandcommandandcontrolmessaging.
13-10-2023 15

16. AItechniqueshavebeendeployedtoaddress
DNStunnelingattacks.
DNStunneldetectionapproachescanberule-
basedormodel-based.
Rule-baseddetectionusesmanuallydefined
rulesbasedonrelevantfeatures.
Signature-baseddetectioncandetectDNStunnelswithhighaccuracy.
DNStunnelingdetectionischallengedbythediversityofmaliciousbehavior.
13-10-2023 16

17. SMART GRID FORENSICS
SmartGridforensicsisusedtoidentifysecurityissuesinsmartelectricalgridsystems.
Itcanalsobeusedforcybercrimeinvestigations,includinghackinganddatatheft.
Thesmartgridsystemandfocusesonefficient
energymanagement.
AItechniques,suchasMLandDL,areusedto
resolvecomplexproblemsinthesmartgrid.
Securityinthesmartgridnetworkisbasedonconfidentiality,integrity,andavailability.
13-10-2023 17

18. VEHICULARFORENSICS
Digitalvehicleforensicsinvolvescapturingandanalyzingdigitalevidencefrommotorvehicles.
Thisevidencecanbeusedininvestigatingcrimesinvolvingmotorvehiclesordeterminingthe
causeofautomobileaccidents.
TheDrivingAdHocNetworkingInfrastructure(DAHNI)focusesondeliveringdriver
assistanceusinglocationawareness,adhocnetworking,andaccesstofixedinfrastructure.
Utilizingvehicularnetworkstotracknearbyvehiclesandalertdriverstopotentialrisks.
13-10-2023 18

19. ADVANTAGES
Efficiency
Adaptability
EarlyThreatDetection
ResourceEfficiency
13-10-2023
19

20. DISADVANTAGES
Complexity
ResourceIntensiveness
FalseAlarms
13-10-2023
20

21. APPLICATIONS
AnomalyDetection
MalwareDetectionandAnalysis
IncidentResponseAutomation
ContinuousNetworkMonitoring
13-10-2023
21

22. CONCLUSION
TheapplicationofArtificialIntelligence(AI)tonetworkforensicsofferssignificant
advantages,includingenhancedthreatdetection,real-timeresponsecapabilities,andthe
abilitytohandlevastandcomplexdatastreams.
AIsystemsarebecomingessentialtoolsfororganizationsseekingtoprotecttheirdigital
assetsinthefaceofincreasinglysophisticatedcyberthreats.
13-10-2023
22

23. FUTURE SCOPE
OneofthesignificantfuturedirectionsforAIinnetworkforensicsisthecreationof
autonomoussystemsthatrelyonmachinelearninganddeeplearningmodels.
Thesesystemswillhavetheabilitytoidentifyandmitigatethreatsinreal-timewithout
humanintervention,resultinginfasterresponsetimesandimprovedcyberattackdefense.
ThefutureofAIinnetworkforensicsisbright,offeringthepromiseofmorerobust,
efficient,andproactivecybersecuritystrategiestosafeguardourdigitalworld
13-10-2023
23

24. REFERENCES
J.FlynnandC.Giannetti,‘‘Usingconvolutionalneuralnetworkstomaphousessuitable
forelectricvehiclehomecharging,’’AI,vol.2,no.1,pp.135–149,Mar.2021[10]Cisco.
(2020).CiscoAnnualInternetReport(2018–2023).
L.Taheri,A.F.A.Kadir,andA.H.Lashkari,‘‘ExtensibleAndroidmal-waredetection
andfamilyclassificationusingnetwork-flowsandAPI-calls,’’inProc.Int.CarnahanConf.
Secur.Technol.(ICCST),Oct.2019,pp.1–8.
Q.Chen,‘‘Towardrealizingself-protectinghealthcareinformationsys-tems:Designand
securitychallenges,’’inAdvancesinComputers,vol.114.Amsterdam,TheNetherlands:
Elsevier,2019.
S.Peisert,R.Gentz,J.Boverhof,C.McParland,S.Engle,A.Elbashandy,andD.Gunter,
‘‘LBNLopenpowerdata,’’LBNL,Berkeley,CA,USA,Tech.Rep.,2017
13-10-2023
24

25. 13-10-2023
25