SlideShare a Scribd company logo
1 of 16
Download to read offline
- 1 / 26 - 방갈로르연구소(SISO)11
Smart Analytics in
Smartphones
Satnam Singh, PhD
Samsung Research India -Bangalore
Fifth Elephant Conference, Bangalore
July 13, 2012
Disclaimer: Talk is based on my personal views and knowledge gathered from open sources
- 2 / 26 - 방갈로르연구소(SISO)22
• What is Smart Analytics?
• Trends in Smart Analytics
• Why to do Analytics in Device?
• Case Study: Sensory Data Analytics
Outline
- 3 / 26 - 방갈로르연구소(SISO)33
Smart Analytics
- Analytics keeping end-user in mind
- Enable use cases to bring new experience, ease and
benefits to end-user
Buying habits
Location and time
Activity
Entertainment
User Presence
Sensor
Data
User
Data
Social
Data
SNS Data, RSS feeds
Images, Videos, Music,
Call logs, SMS data
Browser data
- 4 / 26 - 방갈로르연구소(SISO)44
Smart Analytics in Smartphones
Sensor
Data
- Enhance User Experience
- Recommendations
- Personalization
Social
data
User data…
Analytics (Text Mining, Machine
Learning, Signal Processing)
Sensor
Data
User
Data
Social
Data
3rd Party Applications, Native
Applications
- 5 / 26 - 방갈로르연구소(SISO)55
User Data Analytics- Trends
Breadcrumbs
• A Simple Timeline of your Day
• Everything happening at your places
• Offers and Deals for your favorite places
Radii
• Connecting Personality to Places
• Match the place's personality with users
personality to give the best recommendations
• Deliver movie-like game experiences,
videos, images and wallpapers
• Bring users into the film's story and world
Paramount Pictures - Star Trek Into Darkness
Qualcomm’s Gimbal Platform Applications
- 6 / 26 - 방갈로르연구소(SISO)66
Sensory Data Analytics - Trends
Galaxy S4 Sensors  Multiple sensors,
Environment sensing
Activity Recognition [Sensor Platforms, Alohar
mobile, ActiServ]
- 7 / 26 - 방갈로르연구소(SISO)77
Analytics in Server vs. Device
Device-based Analytics - Privacy concerns are taken care of..
• It works even if no network !!
• Need predictive models to run close to real-time and
automatically deploy them
• Power and battery consumption should be kept under
control
Server-based Analytics is needed if the application is too
compute intensive for a smart phone
• Latency and data transfer cost
• Data must be communicated securely
• Authentication before any data transfer
- 8 / 26 - 방갈로르연구소(SISO)88
Case Study: Sensory Data Analytics
Activity Recognition: Detect walking, driving, biking, climbing
stairs, standing, etc.
Activity
Recognition
Running Biking
Climbing stairs Walking
Sitting
1. If phone call comes then
Send an automated SMS to
call later
3. Do not refresh
location  Save
battery power
2. If phone
call then
increase ring
tone
- 9 / 26 - 방갈로르연구소(SISO)9
Data Visualization – Raw Data & Activity (Class Variable)
[Ref] Rattle R Data Mining Tool
Bar Plot
Example of Accelerometer data
- 10 / 26 - 방갈로르연구소(SISO)1010
Activity Recognition - Steps
Feature
Extraction
Time Series Data 43 Features
Mean for each
acc. Axis (3)
Std. dev. for each
acc. Axis (3)
200 samples (10 sec)
Avg. Abs. diff. from
Mean for each
acc. Axis (3)
Avg. Resultant Acc. (1)
Histogram (30)
Classifier
CART: Decision Tree
Classify the
Activity
[Ref] Gary M. Weiss and Jeffrey W. Lockhart, Fordham University, Bronx, NY
[Ref] Jordan Frank, McGill University
[Ref] Commercial API Providers: Sensor Platoforms, Movea, Alohar
- 11 / 26 - 방갈로르연구소(SISO)11
[Ref] Rattle R Data Mining Tool
Decision Tree
-Accuracy for general model~75%, >95%
personalized model using 10 seconds
training for each activity
-Accelerometer sensor is low power
consuming sensor
- Use other sensors to figure out where is
smartphone  Enhance accuracy by 5-6%
- 12 / 26 - 방갈로르연구소(SISO)12
Activity Recognition: Engg. Challenges
“Design Considerations for the WISDM Smart Phone-based Sensor Mining Architecture,”
SensorKDD ’11, Fordham University
• Supervised models- problems in collecting user data
• Data sampling rate for each activity:
o High sampling rate than needed  waste CPU cycles,
o While low sampling rate degrade the performance
• App should work even if device is in hibernation mode
• Control SQLite database overheads
• Power consumption and real-time computations
• Benchmarking and user testing is a key challenge
• Global user – support multiple languages for any text
mining application
- 13 / 26 - 방갈로르연구소(SISO)13
• Fusion of data science and domain knowledge
can bring new experiences for end-users
• Getting data analytics-based feature in
product needs intense team effort between
various stakeholders
Summary
Thanks!!
- 14 / 26 - 방갈로르연구소(SISO)1414
Backup Slides
- 15 / 26 - 방갈로르연구소(SISO)15
[Ref] Rattle R Data Mining Tool
…
Σ
Random Forest
Tree1 Tree2
Treen
Random Forest: An Ensemble of Trees
- 16 / 26 - 방갈로르연구소(SISO)16
Another Approach: Activity Recognition
Feature
Extraction
using
PCA
Classification
using
SVM
9 PCs Classify
the
activity
“Activity and Gait Recognition with Time-Delay Embeddings” Jordan Frank, AAAI Conference on
Artificial Intelligence -2010
McGill University

More Related Content

Similar to Smart Analytics in Smartphones

PIMRC-2012, Sydney, Australia, 28 July, 2012
PIMRC-2012, Sydney, Australia, 28 July, 2012PIMRC-2012, Sydney, Australia, 28 July, 2012
PIMRC-2012, Sydney, Australia, 28 July, 2012Charith Perera
 
GaitProjectProposal
GaitProjectProposalGaitProjectProposal
GaitProjectProposalVivek Kumar
 
Proof of concepts and use cases with IoT technologies
Proof of concepts and use cases with IoT technologiesProof of concepts and use cases with IoT technologies
Proof of concepts and use cases with IoT technologiesHeikki Ailisto
 
Io t analytics panel
Io t   analytics panelIo t   analytics panel
Io t analytics panelMassTLC
 
Io t platform-infotech_arpanpal
Io t platform-infotech_arpanpalIo t platform-infotech_arpanpal
Io t platform-infotech_arpanpalArpan Pal
 
Data Analytics in your IoT Solution Fukiat Julnual, Technical Evangelist, Mic...
Data Analytics in your IoT SolutionFukiat Julnual, Technical Evangelist, Mic...Data Analytics in your IoT SolutionFukiat Julnual, Technical Evangelist, Mic...
Data Analytics in your IoT Solution Fukiat Julnual, Technical Evangelist, Mic...BAINIDA
 
Personalized power saving profiles generation analyzing smart device usage pa...
Personalized power saving profiles generation analyzing smart device usage pa...Personalized power saving profiles generation analyzing smart device usage pa...
Personalized power saving profiles generation analyzing smart device usage pa...Soumya Kanti Datta
 
Arpan pal tac tics2012
Arpan pal tac tics2012Arpan pal tac tics2012
Arpan pal tac tics2012Arpan Pal
 
Your Data Science Journey - Setting Up Analytics Units From Scratch
Your Data Science Journey - Setting Up Analytics Units From ScratchYour Data Science Journey - Setting Up Analytics Units From Scratch
Your Data Science Journey - Setting Up Analytics Units From ScratchNUS-ISS
 
BUTLER project presentation
BUTLER project presentationBUTLER project presentation
BUTLER project presentationbutler-iot
 
Future-Proof Your Streaming Analytics Architecture- StreamAnalytix Webinar
Future-Proof Your Streaming Analytics Architecture- StreamAnalytix WebinarFuture-Proof Your Streaming Analytics Architecture- StreamAnalytix Webinar
Future-Proof Your Streaming Analytics Architecture- StreamAnalytix WebinarImpetus Technologies
 
Employment Performance Management Using Machine Learning
Employment Performance Management Using Machine LearningEmployment Performance Management Using Machine Learning
Employment Performance Management Using Machine LearningIRJET Journal
 
Emerging engineering issues for building large scale AI systems By Srinivas P...
Emerging engineering issues for building large scale AI systems By Srinivas P...Emerging engineering issues for building large scale AI systems By Srinivas P...
Emerging engineering issues for building large scale AI systems By Srinivas P...Analytics India Magazine
 
Analytics as-a-service-io t-asia-arpanpal
Analytics as-a-service-io t-asia-arpanpalAnalytics as-a-service-io t-asia-arpanpal
Analytics as-a-service-io t-asia-arpanpalArpan Pal
 

Similar to Smart Analytics in Smartphones (20)

PIMRC-2012, Sydney, Australia, 28 July, 2012
PIMRC-2012, Sydney, Australia, 28 July, 2012PIMRC-2012, Sydney, Australia, 28 July, 2012
PIMRC-2012, Sydney, Australia, 28 July, 2012
 
Analytics&IoT
Analytics&IoTAnalytics&IoT
Analytics&IoT
 
GaitProjectProposal
GaitProjectProposalGaitProjectProposal
GaitProjectProposal
 
Proof of concepts and use cases with IoT technologies
Proof of concepts and use cases with IoT technologiesProof of concepts and use cases with IoT technologies
Proof of concepts and use cases with IoT technologies
 
Io t analytics panel
Io t   analytics panelIo t   analytics panel
Io t analytics panel
 
IJET-V3I2P15
IJET-V3I2P15IJET-V3I2P15
IJET-V3I2P15
 
Secure you
Secure you Secure you
Secure you
 
Io t platform-infotech_arpanpal
Io t platform-infotech_arpanpalIo t platform-infotech_arpanpal
Io t platform-infotech_arpanpal
 
Machine Data Analytics
Machine Data AnalyticsMachine Data Analytics
Machine Data Analytics
 
Data Analytics in your IoT Solution Fukiat Julnual, Technical Evangelist, Mic...
Data Analytics in your IoT SolutionFukiat Julnual, Technical Evangelist, Mic...Data Analytics in your IoT SolutionFukiat Julnual, Technical Evangelist, Mic...
Data Analytics in your IoT Solution Fukiat Julnual, Technical Evangelist, Mic...
 
Personalized power saving profiles generation analyzing smart device usage pa...
Personalized power saving profiles generation analyzing smart device usage pa...Personalized power saving profiles generation analyzing smart device usage pa...
Personalized power saving profiles generation analyzing smart device usage pa...
 
Automated Analytics at Scale
Automated Analytics at ScaleAutomated Analytics at Scale
Automated Analytics at Scale
 
Arpan pal tac tics2012
Arpan pal tac tics2012Arpan pal tac tics2012
Arpan pal tac tics2012
 
Your Data Science Journey - Setting Up Analytics Units From Scratch
Your Data Science Journey - Setting Up Analytics Units From ScratchYour Data Science Journey - Setting Up Analytics Units From Scratch
Your Data Science Journey - Setting Up Analytics Units From Scratch
 
BUTLER project presentation
BUTLER project presentationBUTLER project presentation
BUTLER project presentation
 
Future-Proof Your Streaming Analytics Architecture- StreamAnalytix Webinar
Future-Proof Your Streaming Analytics Architecture- StreamAnalytix WebinarFuture-Proof Your Streaming Analytics Architecture- StreamAnalytix Webinar
Future-Proof Your Streaming Analytics Architecture- StreamAnalytix Webinar
 
Employment Performance Management Using Machine Learning
Employment Performance Management Using Machine LearningEmployment Performance Management Using Machine Learning
Employment Performance Management Using Machine Learning
 
LEGaTO: Use cases
LEGaTO: Use casesLEGaTO: Use cases
LEGaTO: Use cases
 
Emerging engineering issues for building large scale AI systems By Srinivas P...
Emerging engineering issues for building large scale AI systems By Srinivas P...Emerging engineering issues for building large scale AI systems By Srinivas P...
Emerging engineering issues for building large scale AI systems By Srinivas P...
 
Analytics as-a-service-io t-asia-arpanpal
Analytics as-a-service-io t-asia-arpanpalAnalytics as-a-service-io t-asia-arpanpal
Analytics as-a-service-io t-asia-arpanpal
 

More from Satnam Singh

InfoSec Deep Learning in Action
InfoSec Deep Learning in ActionInfoSec Deep Learning in Action
InfoSec Deep Learning in ActionSatnam Singh
 
Probabilistic signals and systems satnam singh
Probabilistic signals and systems satnam singhProbabilistic signals and systems satnam singh
Probabilistic signals and systems satnam singhSatnam Singh
 
Threat Hunting with Deceptive Defense and Splunk Enterprise Security
Threat Hunting with Deceptive Defense and Splunk Enterprise SecurityThreat Hunting with Deceptive Defense and Splunk Enterprise Security
Threat Hunting with Deceptive Defense and Splunk Enterprise SecuritySatnam Singh
 
A Game between Adversary and AI Scientist
A Game between Adversary and AI ScientistA Game between Adversary and AI Scientist
A Game between Adversary and AI ScientistSatnam Singh
 
Deep learning fundamentals workshop
Deep learning fundamentals workshopDeep learning fundamentals workshop
Deep learning fundamentals workshopSatnam Singh
 
Deception-Triggered Security Data Science to Detect Adversary Movements
Deception-Triggered Security Data Science to Detect Adversary MovementsDeception-Triggered Security Data Science to Detect Adversary Movements
Deception-Triggered Security Data Science to Detect Adversary MovementsSatnam Singh
 
AI for CyberSecurity
AI for CyberSecurityAI for CyberSecurity
AI for CyberSecuritySatnam Singh
 
Using Deception to Detect and Profile Hidden Threats
Using Deception to Detect and Profile Hidden ThreatsUsing Deception to Detect and Profile Hidden Threats
Using Deception to Detect and Profile Hidden ThreatsSatnam Singh
 
HawkEye : A Real-time Anomaly Detection System
HawkEye : A Real-time Anomaly Detection SystemHawkEye : A Real-time Anomaly Detection System
HawkEye : A Real-time Anomaly Detection SystemSatnam Singh
 
India software developers conference 2013 Bangalore
India software developers conference 2013 BangaloreIndia software developers conference 2013 Bangalore
India software developers conference 2013 BangaloreSatnam Singh
 
Big Data Analytics Insights Conference- Satnam
Big Data Analytics Insights Conference- SatnamBig Data Analytics Insights Conference- Satnam
Big Data Analytics Insights Conference- SatnamSatnam Singh
 

More from Satnam Singh (11)

InfoSec Deep Learning in Action
InfoSec Deep Learning in ActionInfoSec Deep Learning in Action
InfoSec Deep Learning in Action
 
Probabilistic signals and systems satnam singh
Probabilistic signals and systems satnam singhProbabilistic signals and systems satnam singh
Probabilistic signals and systems satnam singh
 
Threat Hunting with Deceptive Defense and Splunk Enterprise Security
Threat Hunting with Deceptive Defense and Splunk Enterprise SecurityThreat Hunting with Deceptive Defense and Splunk Enterprise Security
Threat Hunting with Deceptive Defense and Splunk Enterprise Security
 
A Game between Adversary and AI Scientist
A Game between Adversary and AI ScientistA Game between Adversary and AI Scientist
A Game between Adversary and AI Scientist
 
Deep learning fundamentals workshop
Deep learning fundamentals workshopDeep learning fundamentals workshop
Deep learning fundamentals workshop
 
Deception-Triggered Security Data Science to Detect Adversary Movements
Deception-Triggered Security Data Science to Detect Adversary MovementsDeception-Triggered Security Data Science to Detect Adversary Movements
Deception-Triggered Security Data Science to Detect Adversary Movements
 
AI for CyberSecurity
AI for CyberSecurityAI for CyberSecurity
AI for CyberSecurity
 
Using Deception to Detect and Profile Hidden Threats
Using Deception to Detect and Profile Hidden ThreatsUsing Deception to Detect and Profile Hidden Threats
Using Deception to Detect and Profile Hidden Threats
 
HawkEye : A Real-time Anomaly Detection System
HawkEye : A Real-time Anomaly Detection SystemHawkEye : A Real-time Anomaly Detection System
HawkEye : A Real-time Anomaly Detection System
 
India software developers conference 2013 Bangalore
India software developers conference 2013 BangaloreIndia software developers conference 2013 Bangalore
India software developers conference 2013 Bangalore
 
Big Data Analytics Insights Conference- Satnam
Big Data Analytics Insights Conference- SatnamBig Data Analytics Insights Conference- Satnam
Big Data Analytics Insights Conference- Satnam
 

Recently uploaded

Scanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsScanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsRizwan Syed
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024Stephanie Beckett
 
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptx
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptxPasskey Providers and Enabling Portability: FIDO Paris Seminar.pptx
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptxLoriGlavin3
 
Gen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfGen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfAddepto
 
Advanced Computer Architecture – An Introduction
Advanced Computer Architecture – An IntroductionAdvanced Computer Architecture – An Introduction
Advanced Computer Architecture – An IntroductionDilum Bandara
 
Generative AI for Technical Writer or Information Developers
Generative AI for Technical Writer or Information DevelopersGenerative AI for Technical Writer or Information Developers
Generative AI for Technical Writer or Information DevelopersRaghuram Pandurangan
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxNavinnSomaal
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 3652toLead Limited
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLScyllaDB
 
What is DBT - The Ultimate Data Build Tool.pdf
What is DBT - The Ultimate Data Build Tool.pdfWhat is DBT - The Ultimate Data Build Tool.pdf
What is DBT - The Ultimate Data Build Tool.pdfMounikaPolabathina
 
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek SchlawackFwdays
 
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptxUse of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptxLoriGlavin3
 
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdfHyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdfPrecisely
 
WordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your BrandWordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your Brandgvaughan
 
TrustArc Webinar - How to Build Consumer Trust Through Data Privacy
TrustArc Webinar - How to Build Consumer Trust Through Data PrivacyTrustArc Webinar - How to Build Consumer Trust Through Data Privacy
TrustArc Webinar - How to Build Consumer Trust Through Data PrivacyTrustArc
 
How to write a Business Continuity Plan
How to write a Business Continuity PlanHow to write a Business Continuity Plan
How to write a Business Continuity PlanDatabarracks
 
Take control of your SAP testing with UiPath Test Suite
Take control of your SAP testing with UiPath Test SuiteTake control of your SAP testing with UiPath Test Suite
Take control of your SAP testing with UiPath Test SuiteDianaGray10
 
DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsSergiu Bodiu
 
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptxThe Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptxLoriGlavin3
 

Recently uploaded (20)

Scanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsScanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL Certs
 
What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024What's New in Teams Calling, Meetings and Devices March 2024
What's New in Teams Calling, Meetings and Devices March 2024
 
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptx
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptxPasskey Providers and Enabling Portability: FIDO Paris Seminar.pptx
Passkey Providers and Enabling Portability: FIDO Paris Seminar.pptx
 
Gen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdfGen AI in Business - Global Trends Report 2024.pdf
Gen AI in Business - Global Trends Report 2024.pdf
 
Advanced Computer Architecture – An Introduction
Advanced Computer Architecture – An IntroductionAdvanced Computer Architecture – An Introduction
Advanced Computer Architecture – An Introduction
 
Generative AI for Technical Writer or Information Developers
Generative AI for Technical Writer or Information DevelopersGenerative AI for Technical Writer or Information Developers
Generative AI for Technical Writer or Information Developers
 
DMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special EditionDMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special Edition
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptx
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQL
 
What is DBT - The Ultimate Data Build Tool.pdf
What is DBT - The Ultimate Data Build Tool.pdfWhat is DBT - The Ultimate Data Build Tool.pdf
What is DBT - The Ultimate Data Build Tool.pdf
 
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
"Subclassing and Composition – A Pythonic Tour of Trade-Offs", Hynek Schlawack
 
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptxUse of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
Use of FIDO in the Payments and Identity Landscape: FIDO Paris Seminar.pptx
 
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdfHyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
 
WordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your BrandWordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your Brand
 
TrustArc Webinar - How to Build Consumer Trust Through Data Privacy
TrustArc Webinar - How to Build Consumer Trust Through Data PrivacyTrustArc Webinar - How to Build Consumer Trust Through Data Privacy
TrustArc Webinar - How to Build Consumer Trust Through Data Privacy
 
How to write a Business Continuity Plan
How to write a Business Continuity PlanHow to write a Business Continuity Plan
How to write a Business Continuity Plan
 
Take control of your SAP testing with UiPath Test Suite
Take control of your SAP testing with UiPath Test SuiteTake control of your SAP testing with UiPath Test Suite
Take control of your SAP testing with UiPath Test Suite
 
DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platforms
 
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptxThe Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
The Fit for Passkeys for Employee and Consumer Sign-ins: FIDO Paris Seminar.pptx
 

Smart Analytics in Smartphones

  • 1. - 1 / 26 - 방갈로르연구소(SISO)11 Smart Analytics in Smartphones Satnam Singh, PhD Samsung Research India -Bangalore Fifth Elephant Conference, Bangalore July 13, 2012 Disclaimer: Talk is based on my personal views and knowledge gathered from open sources
  • 2. - 2 / 26 - 방갈로르연구소(SISO)22 • What is Smart Analytics? • Trends in Smart Analytics • Why to do Analytics in Device? • Case Study: Sensory Data Analytics Outline
  • 3. - 3 / 26 - 방갈로르연구소(SISO)33 Smart Analytics - Analytics keeping end-user in mind - Enable use cases to bring new experience, ease and benefits to end-user Buying habits Location and time Activity Entertainment User Presence Sensor Data User Data Social Data SNS Data, RSS feeds Images, Videos, Music, Call logs, SMS data Browser data
  • 4. - 4 / 26 - 방갈로르연구소(SISO)44 Smart Analytics in Smartphones Sensor Data - Enhance User Experience - Recommendations - Personalization Social data User data… Analytics (Text Mining, Machine Learning, Signal Processing) Sensor Data User Data Social Data 3rd Party Applications, Native Applications
  • 5. - 5 / 26 - 방갈로르연구소(SISO)55 User Data Analytics- Trends Breadcrumbs • A Simple Timeline of your Day • Everything happening at your places • Offers and Deals for your favorite places Radii • Connecting Personality to Places • Match the place's personality with users personality to give the best recommendations • Deliver movie-like game experiences, videos, images and wallpapers • Bring users into the film's story and world Paramount Pictures - Star Trek Into Darkness Qualcomm’s Gimbal Platform Applications
  • 6. - 6 / 26 - 방갈로르연구소(SISO)66 Sensory Data Analytics - Trends Galaxy S4 Sensors  Multiple sensors, Environment sensing Activity Recognition [Sensor Platforms, Alohar mobile, ActiServ]
  • 7. - 7 / 26 - 방갈로르연구소(SISO)77 Analytics in Server vs. Device Device-based Analytics - Privacy concerns are taken care of.. • It works even if no network !! • Need predictive models to run close to real-time and automatically deploy them • Power and battery consumption should be kept under control Server-based Analytics is needed if the application is too compute intensive for a smart phone • Latency and data transfer cost • Data must be communicated securely • Authentication before any data transfer
  • 8. - 8 / 26 - 방갈로르연구소(SISO)88 Case Study: Sensory Data Analytics Activity Recognition: Detect walking, driving, biking, climbing stairs, standing, etc. Activity Recognition Running Biking Climbing stairs Walking Sitting 1. If phone call comes then Send an automated SMS to call later 3. Do not refresh location  Save battery power 2. If phone call then increase ring tone
  • 9. - 9 / 26 - 방갈로르연구소(SISO)9 Data Visualization – Raw Data & Activity (Class Variable) [Ref] Rattle R Data Mining Tool Bar Plot Example of Accelerometer data
  • 10. - 10 / 26 - 방갈로르연구소(SISO)1010 Activity Recognition - Steps Feature Extraction Time Series Data 43 Features Mean for each acc. Axis (3) Std. dev. for each acc. Axis (3) 200 samples (10 sec) Avg. Abs. diff. from Mean for each acc. Axis (3) Avg. Resultant Acc. (1) Histogram (30) Classifier CART: Decision Tree Classify the Activity [Ref] Gary M. Weiss and Jeffrey W. Lockhart, Fordham University, Bronx, NY [Ref] Jordan Frank, McGill University [Ref] Commercial API Providers: Sensor Platoforms, Movea, Alohar
  • 11. - 11 / 26 - 방갈로르연구소(SISO)11 [Ref] Rattle R Data Mining Tool Decision Tree -Accuracy for general model~75%, >95% personalized model using 10 seconds training for each activity -Accelerometer sensor is low power consuming sensor - Use other sensors to figure out where is smartphone  Enhance accuracy by 5-6%
  • 12. - 12 / 26 - 방갈로르연구소(SISO)12 Activity Recognition: Engg. Challenges “Design Considerations for the WISDM Smart Phone-based Sensor Mining Architecture,” SensorKDD ’11, Fordham University • Supervised models- problems in collecting user data • Data sampling rate for each activity: o High sampling rate than needed  waste CPU cycles, o While low sampling rate degrade the performance • App should work even if device is in hibernation mode • Control SQLite database overheads • Power consumption and real-time computations • Benchmarking and user testing is a key challenge • Global user – support multiple languages for any text mining application
  • 13. - 13 / 26 - 방갈로르연구소(SISO)13 • Fusion of data science and domain knowledge can bring new experiences for end-users • Getting data analytics-based feature in product needs intense team effort between various stakeholders Summary Thanks!!
  • 14. - 14 / 26 - 방갈로르연구소(SISO)1414 Backup Slides
  • 15. - 15 / 26 - 방갈로르연구소(SISO)15 [Ref] Rattle R Data Mining Tool … Σ Random Forest Tree1 Tree2 Treen Random Forest: An Ensemble of Trees
  • 16. - 16 / 26 - 방갈로르연구소(SISO)16 Another Approach: Activity Recognition Feature Extraction using PCA Classification using SVM 9 PCs Classify the activity “Activity and Gait Recognition with Time-Delay Embeddings” Jordan Frank, AAAI Conference on Artificial Intelligence -2010 McGill University