Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent Processing Applied toStreams of...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerFrom this
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerTo this
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerRequest/ReplyClient ServerRequestReply
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerTo this
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerand this
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerPublish/SubscribePublisher Subscriber...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerIS-Home
Altibox at homeINTERNETFrom 10 to 400Mbit/s, both waysDIGITAL TVChoose between150 TV channels,film rental and more.ALARM S...
TvFILM RENTALChoose from1000 titlesNow in HD tooFOOTBALLThe whole PremierLeague andTippeligaen (Norwegianfootball pools le...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerResearch ContributionsTwo main topics...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerMotivationSensorsandactuators
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSenseWrap Features• IP-enabling senso...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSenseWrap ArchitectureVirtualSensorPh...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent-Driven ArchitectureEvent Proces...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerComponents• Esper• HornetQ• Custom Ja...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent Flow and InterfacesEventProcess...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent Processing Language (EPL)• Decl...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerContinuous Queries
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSliding Windowswin:time(10 sec)time0 ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent PatternsSimpleAEventComplexCEve...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerCurrent State of TV Viewer Statistics...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerMotivation• IP-based Set Top Boxes (S...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerThe Future of Media Measurement• Curr...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerAltibox Deployment of IPTV• Over 320,...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOur Scenario• Two-way communication• ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerObserved Events• Channel change event...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerNumber of zap events/day over a 15-da...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerZapping Activity0 5 10 15 2000.511.5·...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEPL Viewer StatisticsEsper EngineZaps...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerTrude/18Sykkel:Girod`Italia14.etappeS...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorer3000035000400004500050000550006000065...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerAnnoyance Detection• Aimed at detecti...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerAnnoyance DetectorAnnoyance detectorZ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerDeployment Details• Processes over 40...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerCPU load - ZapCollector
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerCPU load - Message bus
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerCPU load - EventCaster instance
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEPL and Java Implementations
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerExperimental Setup for Performace Eva...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerProcessing capacityMemory Disk UDP Ho...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerMemory consumptionFrom memory From di...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerPerformance Evaluation - Observations...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerHalstead’s Metrics• Metrics that desc...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerComplexity Metrics0 100 200 300 400Vo...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerComplexity Evaluation - Observations•...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSummary• The Java implementation is t...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerDeploymentTV NetworkAdDetectorMuteAdS...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerDefining an advertisement as a context...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerExperimental Setup for AdScorer Evalu...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerViewer numbersFigure: Viewership (in ...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerViewer numbers12	  13	  14	  15	  16	...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerRetained Viewers
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerGained Viewers
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerRaw output for a single ad1 Event: {2...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSummary• AdScorer is capable of deliv...
Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerThank You for listening!
Upcoming SlideShare
Loading in …5
×

Event Processing Applied to Streams of TV Channel Zaps and Sensor Middleware with Virtualization

297 views
238 views

Published on

My PhD thesis defense from 23.04.2013

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
297
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
6
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Event Processing Applied to Streams of TV Channel Zaps and Sensor Middleware with Virtualization

  1. 1. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent Processing Applied toStreams of TV Channel Zaps andSensor Middleware with VirtualizationPhD DefenseP˚al EvensenDepartment of Electrical Engineering and Computer ScienceUniversity of Stavanger, NorwayApril 23rd, 2013
  2. 2. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ContextResearch ContributionsSenseWrapSenseWrapEventCasterSystem ArchitectureEsper and EPLTelevision StatisticsBackgroundImplementationStatisticsDeploymentParadigm ComparisonPerformanceComplexityConclusionsAdScorerOverviewExperimentsLive Scoring ResultsSummary
  3. 3. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerFrom this
  4. 4. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerTo this
  5. 5. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerRequest/ReplyClient ServerRequestReply
  6. 6. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerTo this
  7. 7. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerand this
  8. 8. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerPublish/SubscribePublisher SubscriberPublisherSubscriberSubscriberSubscriberEventServiceNotificationpublishnotifysubscribe/unsubscribe
  9. 9. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerIS-Home
  10. 10. Altibox at homeINTERNETFrom 10 to 400Mbit/s, both waysDIGITAL TVChoose between150 TV channels,film rental and more.ALARM SERVICESAlarm monitoringcentre. Direct alarmconnection to the firebrigade. MOBILE PHONESERVICESSame rate, nomatter who you callIP TELEPHONYNo charges toLandlines inScandinavia
  11. 11. TvFILM RENTALChoose from1000 titlesNow in HD tooFOOTBALLThe whole PremierLeague andTippeligaen (Norwegianfootball pools league)NEWSUp to date local newsTV-GUIDEA complete overviewfor your TV evening
  12. 12. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerResearch ContributionsTwo main topics:1. Sensors in smart homes• Middleware for hiding the heterogeneity of devices2. Event processing applied to TV channel zaps• Architecture for efficient processing of high volumes of events• Evaluating programming models for stateful event processing
  13. 13. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ContextResearch ContributionsSenseWrapSenseWrapEventCasterSystem ArchitectureEsper and EPLTelevision StatisticsBackgroundImplementationStatisticsDeploymentParadigm ComparisonPerformanceComplexityConclusionsAdScorerOverviewExperimentsLive Scoring ResultsSummary
  14. 14. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerMotivationSensorsandactuators
  15. 15. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSenseWrap Features• IP-enabling sensor devices• Uniform interface to hetereogeneous devices• A “blueprint” for developers of smart home applications• Automatic network configuration and service discovery
  16. 16. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSenseWrap ArchitectureVirtualSensorPhysical sensorsSensewrapVirtualSensorSensor protocol (ZigBee, Bluetooth, etc)zeroconfDriver DriverGatewayServiceUDP TCPServiceTCPServiceHTTPClient Client Client
  17. 17. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ContextResearch ContributionsSenseWrapSenseWrapEventCasterSystem ArchitectureEsper and EPLTelevision StatisticsBackgroundImplementationStatisticsDeploymentParadigm ComparisonPerformanceComplexityConclusionsAdScorerOverviewExperimentsLive Scoring ResultsSummary
  18. 18. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent-Driven ArchitectureEvent ProcessingNetworkEventProcessingAgentsEvent ConsumersEvent Producers
  19. 19. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerComponents• Esper• HornetQ• Custom Java glue
  20. 20. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent Flow and InterfacesEventProcessorAdapterEventPublisherEventProcessorEventPublisherMessage busProducer CoreEventEPLEPLUpdateListener
  21. 21. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent Processing Language (EPL)• Declarative query language derived from SQL• Operates on stream data as opposed to relational data• Looks for event patterns that matches the query, andproduces an output event• Includes additional operators, such as sliding windows• Part of the Esper framework (Open Source)
  22. 22. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerContinuous Queries
  23. 23. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSliding Windowswin:time(10 sec)time0 10 20nowe1 e2 e3 e4 e51 select ∗ from ChannelWin.win:time(10 sec)
  24. 24. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEvent PatternsSimpleAEventComplexCEventSimpleBEvent1 select ∗ from pattern [every a=A −> b=B]
  25. 25. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ContextResearch ContributionsSenseWrapSenseWrapEventCasterSystem ArchitectureEsper and EPLTelevision StatisticsBackgroundImplementationStatisticsDeploymentParadigm ComparisonPerformanceComplexityConclusionsAdScorerOverviewExperimentsLive Scoring ResultsSummary
  26. 26. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerCurrent State of TV Viewer Statistics• Sample size is 0.045 % of Norwegian television households• Only 0.022 % of American households are sampled• Data collection requires specialized equipment• Transferred once a day
  27. 27. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerMotivation• IP-based Set Top Boxes (STBs) allows for more accuratestatistics• Ads can be evaluated on an individual basis• Additional behavioral markers such as mute and volumechanges allows for better understanding
  28. 28. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerThe Future of Media Measurement• Current measurement methods for TV not in line with newmodels of media consumption• Gartner: The online/offline division of media will be replacedby measured/unmeasured by 2015• STB data increasingly being used to augment traditional TVratings
  29. 29. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerAltibox Deployment of IPTV• Over 320,000 STBs deployed in Norway and Denmark
  30. 30. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOur Scenario• Two-way communication• Immediate transfer of zap, mute, hdmi and volume events
  31. 31. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerObserved Events• Channel change event (also called a zap event)• HDMI status event: TV set on/off• STB audio on/off event (mute)• STB volume change event
  32. 32. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerNumber of zap events/day over a 15-day period31 1 2 3 4 5 6 7 8 9 101112131422.22.42.6·106Avg=2212097
  33. 33. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerZapping Activity0 5 10 15 2000.511.5·105Time (hrs)30 Minutes15 Minutes5 Minutes
  34. 34. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEPL Viewer StatisticsEsper EngineZaps ChannelWinchannelName,viewersselect(...)channelName,viewers1 select channelName, viewers2 from ChannelWin3 output snapshot every 1 sec
  35. 35. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerTrude/18Sykkel:Girod`Italia14.etappeSport&SpillReisemålLørdagsmagasinetSportsnyheteneVæretVinnpåminuttet9:12NyheteneGodkveldNorgeSportsnyheteneFilm:MordpåkredittSykkel:Girod`Italia-oppsummeringFarmen30:30NyheteneAkvarietTV2hjelperdeg11010000200003000040000500006000012:00 14:00 16:00 18:00 20:00 22:00ViewersTimeViewers for TV2 Norge + TV2 HD, Saturday 21.05.2011, Separated by Zip CodesTotal viewers0001-39994000-59996000-79998000-9999
  36. 36. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorer30000350004000045000500005500060000650007000010:00 12:00 14:00 16:00 18:00 20:00 22:00SeereTidspunktSeerstatistikk for TV2 Norge og NRK1, Lørdag 21.05.2011TV2 Norge + HDNRK1 + HD
  37. 37. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerAnnoyance Detection• Aimed at detecting ads/programs that is causing viewers tochange channel• Triggers an output event on rapid drop in viewers on aparticular channel• Sliding window algorithm, continously comparing viewernumber with the last minute average
  38. 38. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerAnnoyance DetectorAnnoyance detectorZapSnap.win.time(1 min)channelName,viewers,avg(viewers)avg(viewers)having(...)channelName,viewers1 select channelName, viewers, avg(viewers)2 from ChannelWin(viewers > 2000).win:time(1 min)3 group by channelName4 having viewers < avg(viewers) ∗ 0.85
  39. 39. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerDeployment Details• Processes over 40,000 events/min at peak hours• Deployed in the Altibox network across four servers:ZapCollector MessagebusDatabaseCore+Manager
  40. 40. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerCPU load - ZapCollector
  41. 41. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerCPU load - Message bus
  42. 42. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerCPU load - EventCaster instance
  43. 43. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ContextResearch ContributionsSenseWrapSenseWrapEventCasterSystem ArchitectureEsper and EPLTelevision StatisticsBackgroundImplementationStatisticsDeploymentParadigm ComparisonPerformanceComplexityConclusionsAdScorerOverviewExperimentsLive Scoring ResultsSummary
  44. 44. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerEPL and Java Implementations
  45. 45. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerExperimental Setup for Performace Evaluation• 2,117,897 zap events (one day)• RHEL6,Quad Core 2.4GHz CPU, 14GB RAM• Processing capacity tested by loading events from:• Memory• Disk• Network: UDP and HornetQ• Memory consumption tested by loading events from:• Memory• Disk
  46. 46. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerProcessing capacityMemory Disk UDP HornetQ02468·105Events/secEsper Java
  47. 47. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerMemory consumptionFrom memory From disk6008001,0001,2001,4001,600 1,5701,401590 577Memoryused(MB)Esper Java
  48. 48. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerPerformance Evaluation - Observations• The Java implementation outperforms the Esper version• Can be accredited to abstraction overhead imposed by theEsper engine• Performance margin decrease as network overhead isintroduced, but is still significant• UDP benchmarks offer 45% higher throughput with the Javaversion
  49. 49. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerHalstead’s Metrics• Metrics that describe the complexity and effort required towrite a program• Sees a program as a series of tokens, classified as eitheroperands or operators
  50. 50. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerComplexity Metrics0 100 200 300 400Vocabulary (n1+n2)Program length (N1+N2)Total operators (N1)Total operands (N2)Unique operators (n1)Unique operands (n2)Lines of code(Java)(EPL)Viewer StatisticsAnnoyance DetectorParsing and Query SetupUtility Functions
  51. 51. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerComplexity Evaluation - Observations• The EPL implementation scores slightly better for the viewerstatistics application• EPL scores significantly better for the annoyance detector• Initial effort was similar for both applications• Expanding EPL applications is significantly easier than Java
  52. 52. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSummary• The Java implementation is the most performant• The Esper implementation is easier to expand• Can be attributed to different degrees of abstraction
  53. 53. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerOutlineIntroductionBackgroundProject ContextResearch ContributionsSenseWrapSenseWrapEventCasterSystem ArchitectureEsper and EPLTelevision StatisticsBackgroundImplementationStatisticsDeploymentParadigm ComparisonPerformanceComplexityConclusionsAdScorerOverviewExperimentsLive Scoring ResultsSummary
  54. 54. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerDeploymentTV NetworkAdDetectorMuteAdSuccess-EvaluatorSTBsInputadapters StatsAd startChannel zap events Filtering,transformationZapAd success scoreHistoricalstatsProcessingQueueingInput
  55. 55. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerDefining an advertisement as a context in EPLListing 1: Context declaration1 create context AdBreakCtx as2 initiated by3 AdIdentified(begin=true) as ad4 terminated by5 AdIdentified(detectId=ad.detectId, begin=false) as endAdListing 2: Populating the context1 context AdBreakCtx2 create window STBsnapshots.win:keepall() as STBWin3 insert where4 channel in (context.ad.channel) and hdmi=true and mute=false
  56. 56. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerExperimental Setup for AdScorer Evaluation• 1.5 hour of prime time TV was recorded• Advertisement times were manually recorded• 23 days of STB data was sent through the system in order toreach a correct state• Logged STB data and AdIdentified-events was then pushed tothe system• Simulation run
  57. 57. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerViewer numbersFigure: Viewership (in thousands) for the three largest channels, NRK1,TV2, and TVN.
  58. 58. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerViewer numbers12  13  14  15  16  17  18  19  20  21  S E S E S ETVN  Figure: Viewership (in thousands) for a single channel (TVN).
  59. 59. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerRetained Viewers
  60. 60. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerGained Viewers
  61. 61. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerRaw output for a single ad1 Event: {2 AdId: Mitsubishi ASX,3 channel: TV2 Norge,4 startTime: 22:55:12,5 stopTime: 22:55:32,6 viewersBegin: 83846,7 retained: 79244,8 IAR: 94.51,9 mutes: 141,10 viewersLost: 4602,11 average volume: 49.6412 }
  62. 62. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerSummary• AdScorer is capable of delivering an unprecedented level ofdetail, not possible through the current measurement regime• The results indicate that our implementation is capable ofscoring advertisements on multiple channels simultaneously innear real-time with consistent results
  63. 63. Introduction SenseWrap EventCaster Television Statistics Paradigm Comparison AdScorerThank You for listening!

×