SlideShare a Scribd company logo
1 of 26
Download to read offline
A METHOD TO
AI MADNESS
Vishwa Kolla
Head, Advanced Analytics
John Hancock Insurance
TOPICS
Background Framework Case Studies
2
BUILD ME A MODEL TO …
3
REDUCE
COMPLAINTS
GROW
WALLET-
SHARE
GROW
CSAT
REDUCE
CHURN
REDUCE
COST TO
TARGET
GROW
BOTTOM-LINE
GROW
TOP-LINE
REDUCE
COST TO
ACQUIRE
MODEL BUILD IS THE PATH OF LEAST RESISTANCE
4
Platforms
R
P
H20
SPARK
TENSOR
FLOW
TBD
SUPERVISED
UN
SUPERVISED
NLTK
NUM
PY
PAN
DAS
PLOT
LY
PLATFORMS
ALGORITHMS
PACKAGES
PYO
DBC
CRY
PTO
PYPD
F
SCI
KIT
TOR
NAD
O
ZICT
BAB
EL
BLA
ZE
A THOUGHTFUL APPROACH CAN YIELD BETTER OUTCOMES
BUSINESS
USE CASES
DATA MATH
TECHNICAL
IMPL.
BUSINESS
IMPL.
FEED
BACK
5
TOPICS
Background Framework Case Studies
6
“A” MODEL BUILD FRAMEWORK
7
DATA TARGET CONSTRUCTION EVALUATION PERFORMANCE
SOURCES
DISTANCE FROM
SIGNAL
SAMPLING
METHOD
SAMPLE
SIZE
SIGNAL SIZE
PREDICTION
HORIZON
UNIT OF
ANALYSIS
ONE MODEL vs.
STRATIFIED
ONE MODEL vs.
SEVERAL MODELS
TARGET
DEFINITION
PRESENCE OR
ABSENCE
BLACK BOX vs.
CLEAR BOX
RECENCY
FREQUENCY
SEVERITY
FEATURE
SELECTION
MODELING
STRATEGY
MODEL
STRENGTH
EXPLANATORY vs.
IMPORTANCE
ACCURACY vs.
SENSITIVITY vs.
SPECIFICITY
ECONOMIES OF
SCOPE
MODEL
FIT
BAGGING
ENSEMBLE
SINGLE vs.
MULTIPLE STAGES
PREDICTION &
OPTIMIZATION
BOOSTING
TOPICS
Background Framework Case Studies
8
Business
9
IT (ALWAYS) STARTS WITH A BUSINESS PROBLEM
PROSPECTING NURTUREACQUISITION
MARKET
SEGMENTS
CUSTOMER
SEGMENTS
LIKELY TO [*]
MEDIA
MIX
CHANNEL
SURVEY
ANALYTICS
CROSS / UP-
SELL
OCR
MISREP
LIKELIHOOD
MORTALITY
APS
SUMMARY
FLUIDLESS
SMOKER
LIKELIHOOD
MORBIDITY
CHURN
NEXT BEST
OFFER
CLAIM
LIKELI-
HOOD
JOURNEY
CLAIM
SEVERITY
NEXT BEST
ACTION
FRAUD
>>
TEXT
ANALYTICS
OPTIMIZE
NEXT LIKELY
ACTION
WELLNESS
IOT
ANALYTICS
NPS
ANOMALY
>>
10
FOCUS ON INCREMENTAL VALUE KEPT US GROUNDED
BUSINESS CASE OPTICAL REALIZABLE SHARED INCREMENTAL
IN PROSPECTING, TARGET OPTIMIZAITON IS A JOURNEY
12
… LOWER CUSTOMER TARGETING COSTSA SERIES OF OPTIMIZATION TARGETS …
Prospects
Leads
Apps
Issued
Placed
CPL
CPA
CPP
CP[*]
CHANNEL
MIX
Data
13
PLANS ARE NOTHING ; PLANNING IS EVERYTHING
14
EDA
USEABLE
USEFUL
DERIVATIVES
BI-VARIATE CROSSTAB PRINCOMP JOURNEY
A DATA STITCH IN TIME SAVES NINE
15
CLAIM
TERMIN
ATION
CLAIM
ACTV.
DEMOS
CALLS
INTERA
CTION
CLAIM
INIT.
CUSTOMER
MONTH
FRAUD
DETECTION
UNDERSTANDING DATA SAVES (NOT WASTES) TIME
16
Signal
Distribution
Pop. Incidence Rate
Skews Model Inclusion
Math
17
FLEXIBILITY IN TARGET DEFINITION IMPROVED ACTIONABILITY
18
20172014
Predict incidence
In next 3 years
2007
20172014
Predict incidence
3 years out
2007
Vs.
RIGHT SIZING SIGNAL CAN YIELD BETTER OUTCOMES
19
SIGNAL
DILUTION
SIGNAL
AMPLIFICATION
1%
99%
40%
60%
SIMPLE MODELS CAN HELP US EXPLAIN BIG DRIVERS
20
PREDICTORS
PRESENCE
RECENT
FREQUENT
SEVERE
QUANTIFICATION OF INFORMATION GAP IS A GOOD FIRST STEP
© Andrew Ng
INFORMATION GAP
WINNING
MODEL
CHALLENGING CHAMPIONS HELPS US UP THE ANTE
22
DATA TARGET
SOURCES
DISTANCE FROM
SIGNAL
SAMPLING
METHOD
SAMPLE
SIZE
SIGNAL SIZE
PREDICTION
HORIZON
UNIT OF
ANALYSIS
ONE MODEL vs.
STRATIFIED
ONE MODEL vs.
SEVERAL MODELS
TARGET
DEFINITION
METHODS
LINEAR
TREES
DEEP-
LEARNING
EVALUATION
MODEL
STRENGTH
EXPLANATORY vs.
IMPORTANCE
ACCURACY vs.
SENSITIVITY vs.
SPECIFICITY
ECONOMIES OF
SCOPE
MODEL
FIT
Technical Implementation
23
MULTI-STAGED MODELS PROVIDED IMPLEMENTATION FLEXIBILITY
24
9-101-8
1-7 8-10
Likely to
Qualify
Likely to
Respond
Sweet
Spot
DESIRED
SIGNAL
MODEL
MIS-
CLASSIFICATION
MODEL
EXCLUDE NOISE1
INCLUDE MIS-CLASSIFIERS2
STAGES
Business Implementation
25
A CULTURE OF MEASUREMENT, TEST AND LEARN IMPROVES VALUE
26
Measure
TestLearn
Build
CONTINUOUS
IMPROVEMENT

More Related Content

What's hot

My Life with Data
My Life with DataMy Life with Data
My Life with DataGramener
 
IBM presentation at the Chief Analytics Officer Forum East Coast USA (#CAOForum)
IBM presentation at the Chief Analytics Officer Forum East Coast USA (#CAOForum)IBM presentation at the Chief Analytics Officer Forum East Coast USA (#CAOForum)
IBM presentation at the Chief Analytics Officer Forum East Coast USA (#CAOForum)Chief Analytics Officer Forum
 
Cognizant Making AI Real with Microsoft
Cognizant Making AI Real with MicrosoftCognizant Making AI Real with Microsoft
Cognizant Making AI Real with MicrosoftSteve Lennon
 
HLEG thematic workshop on measuring economic, social and environmental resili...
HLEG thematic workshop on measuring economic, social and environmental resili...HLEG thematic workshop on measuring economic, social and environmental resili...
HLEG thematic workshop on measuring economic, social and environmental resili...StatsCommunications
 
Analytics in financial services prez behavioral finance + data visualizatio...
Analytics in financial services prez   behavioral finance + data visualizatio...Analytics in financial services prez   behavioral finance + data visualizatio...
Analytics in financial services prez behavioral finance + data visualizatio...Fitzgerald Analytics, Inc.
 
Exploring the Tech Specific Purchase Process
Exploring the Tech Specific Purchase ProcessExploring the Tech Specific Purchase Process
Exploring the Tech Specific Purchase ProcessIDG
 
Does big data = big insights?
Does big data = big insights?Does big data = big insights?
Does big data = big insights?Colin Strong
 
Accenture Public Service Citizen Survey: Public Administration
Accenture Public Service Citizen Survey: Public AdministrationAccenture Public Service Citizen Survey: Public Administration
Accenture Public Service Citizen Survey: Public Administrationaccenture
 
Business model innovation by experimentation
Business model innovation by experimentationBusiness model innovation by experimentation
Business model innovation by experimentationEnergized Work
 
What people analytics can not capture.
What people analytics can not capture.What people analytics can not capture.
What people analytics can not capture.Tanya Gupta
 
Doing Less with Less
Doing Less with LessDoing Less with Less
Doing Less with LessBruce Janda
 
Beeline Analytics Workshop
Beeline Analytics WorkshopBeeline Analytics Workshop
Beeline Analytics WorkshopBeeline
 
Algorithms and the technology of personalisation final
Algorithms and the technology of personalisation finalAlgorithms and the technology of personalisation final
Algorithms and the technology of personalisation finalColin Strong
 
You may not need big data after all
You may not need big data after allYou may not need big data after all
You may not need big data after allParul Verma
 
Age Friendly Economy - The Future of Big Data
Age Friendly Economy  - The Future of Big DataAge Friendly Economy  - The Future of Big Data
Age Friendly Economy - The Future of Big DataAgeFriendlyEconomy
 
2020 IDG Role & Influence of the Technology Decision-Maker
2020 IDG Role & Influence of the Technology Decision-Maker2020 IDG Role & Influence of the Technology Decision-Maker
2020 IDG Role & Influence of the Technology Decision-MakerIDG
 
First Round State of Startups 2018
First Round State of Startups 2018First Round State of Startups 2018
First Round State of Startups 2018First Round Capital
 
What people analytics can't capture
What people analytics can't captureWhat people analytics can't capture
What people analytics can't captureSushant Kumar
 
An Evaluation of Investment Models within Information Security
An Evaluation of Investment Models within Information SecurityAn Evaluation of Investment Models within Information Security
An Evaluation of Investment Models within Information SecurityTodd Nelson
 

What's hot (20)

My Life with Data
My Life with DataMy Life with Data
My Life with Data
 
IBM presentation at the Chief Analytics Officer Forum East Coast USA (#CAOForum)
IBM presentation at the Chief Analytics Officer Forum East Coast USA (#CAOForum)IBM presentation at the Chief Analytics Officer Forum East Coast USA (#CAOForum)
IBM presentation at the Chief Analytics Officer Forum East Coast USA (#CAOForum)
 
Cognizant Making AI Real with Microsoft
Cognizant Making AI Real with MicrosoftCognizant Making AI Real with Microsoft
Cognizant Making AI Real with Microsoft
 
HLEG thematic workshop on measuring economic, social and environmental resili...
HLEG thematic workshop on measuring economic, social and environmental resili...HLEG thematic workshop on measuring economic, social and environmental resili...
HLEG thematic workshop on measuring economic, social and environmental resili...
 
Ai in Society
Ai in SocietyAi in Society
Ai in Society
 
Analytics in financial services prez behavioral finance + data visualizatio...
Analytics in financial services prez   behavioral finance + data visualizatio...Analytics in financial services prez   behavioral finance + data visualizatio...
Analytics in financial services prez behavioral finance + data visualizatio...
 
Exploring the Tech Specific Purchase Process
Exploring the Tech Specific Purchase ProcessExploring the Tech Specific Purchase Process
Exploring the Tech Specific Purchase Process
 
Does big data = big insights?
Does big data = big insights?Does big data = big insights?
Does big data = big insights?
 
Accenture Public Service Citizen Survey: Public Administration
Accenture Public Service Citizen Survey: Public AdministrationAccenture Public Service Citizen Survey: Public Administration
Accenture Public Service Citizen Survey: Public Administration
 
Business model innovation by experimentation
Business model innovation by experimentationBusiness model innovation by experimentation
Business model innovation by experimentation
 
What people analytics can not capture.
What people analytics can not capture.What people analytics can not capture.
What people analytics can not capture.
 
Doing Less with Less
Doing Less with LessDoing Less with Less
Doing Less with Less
 
Beeline Analytics Workshop
Beeline Analytics WorkshopBeeline Analytics Workshop
Beeline Analytics Workshop
 
Algorithms and the technology of personalisation final
Algorithms and the technology of personalisation finalAlgorithms and the technology of personalisation final
Algorithms and the technology of personalisation final
 
You may not need big data after all
You may not need big data after allYou may not need big data after all
You may not need big data after all
 
Age Friendly Economy - The Future of Big Data
Age Friendly Economy  - The Future of Big DataAge Friendly Economy  - The Future of Big Data
Age Friendly Economy - The Future of Big Data
 
2020 IDG Role & Influence of the Technology Decision-Maker
2020 IDG Role & Influence of the Technology Decision-Maker2020 IDG Role & Influence of the Technology Decision-Maker
2020 IDG Role & Influence of the Technology Decision-Maker
 
First Round State of Startups 2018
First Round State of Startups 2018First Round State of Startups 2018
First Round State of Startups 2018
 
What people analytics can't capture
What people analytics can't captureWhat people analytics can't capture
What people analytics can't capture
 
An Evaluation of Investment Models within Information Security
An Evaluation of Investment Models within Information SecurityAn Evaluation of Investment Models within Information Security
An Evaluation of Investment Models within Information Security
 

Viewers also liked

The Essentials of PowerPoint Color Theme
The Essentials of PowerPoint Color ThemeThe Essentials of PowerPoint Color Theme
The Essentials of PowerPoint Color Theme24Slides
 
Tracxn Research — Enterprise Collaboration Landscape, November 2016
Tracxn Research — Enterprise Collaboration Landscape, November 2016Tracxn Research — Enterprise Collaboration Landscape, November 2016
Tracxn Research — Enterprise Collaboration Landscape, November 2016Tracxn
 
Introduction to Machine Learning for Heathcare, Finance, Energy, and Commerce...
Introduction to Machine Learning for Heathcare, Finance, Energy, and Commerce...Introduction to Machine Learning for Heathcare, Finance, Energy, and Commerce...
Introduction to Machine Learning for Heathcare, Finance, Energy, and Commerce...Sam Putnam [Deep Learning]
 
Python for Image Understanding: Deep Learning with Convolutional Neural Nets
Python for Image Understanding: Deep Learning with Convolutional Neural NetsPython for Image Understanding: Deep Learning with Convolutional Neural Nets
Python for Image Understanding: Deep Learning with Convolutional Neural NetsRoelof Pieters
 
Artificial Intelligence (AI) for Financial Services
Artificial Intelligence (AI) for Financial Services Artificial Intelligence (AI) for Financial Services
Artificial Intelligence (AI) for Financial Services NVIDIA
 
14 Startups Leading the Artificial Intelligence (AI) Revolution
14 Startups Leading the Artificial Intelligence (AI) Revolution14 Startups Leading the Artificial Intelligence (AI) Revolution
14 Startups Leading the Artificial Intelligence (AI) RevolutionNVIDIA
 
Transform your Business with AI, Deep Learning and Machine Learning
Transform your Business with AI, Deep Learning and Machine LearningTransform your Business with AI, Deep Learning and Machine Learning
Transform your Business with AI, Deep Learning and Machine LearningSri Ambati
 
The Spring of AI : Leveraging Collective Disruptor Insights
The Spring of AI : Leveraging Collective Disruptor InsightsThe Spring of AI : Leveraging Collective Disruptor Insights
The Spring of AI : Leveraging Collective Disruptor InsightsZinnov
 
Deep neural networks
Deep neural networksDeep neural networks
Deep neural networksSi Haem
 
[DSC 2016] 系列活動:李宏毅 / 一天搞懂深度學習
[DSC 2016] 系列活動:李宏毅 / 一天搞懂深度學習[DSC 2016] 系列活動:李宏毅 / 一天搞懂深度學習
[DSC 2016] 系列活動:李宏毅 / 一天搞懂深度學習台灣資料科學年會
 
Deep Learning - The Past, Present and Future of Artificial Intelligence
Deep Learning - The Past, Present and Future of Artificial IntelligenceDeep Learning - The Past, Present and Future of Artificial Intelligence
Deep Learning - The Past, Present and Future of Artificial IntelligenceLukas Masuch
 
How to Become a Thought Leader in Your Niche
How to Become a Thought Leader in Your NicheHow to Become a Thought Leader in Your Niche
How to Become a Thought Leader in Your NicheLeslie Samuel
 

Viewers also liked (13)

The Essentials of PowerPoint Color Theme
The Essentials of PowerPoint Color ThemeThe Essentials of PowerPoint Color Theme
The Essentials of PowerPoint Color Theme
 
Deep Learning for Sales Professionals
Deep Learning for Sales ProfessionalsDeep Learning for Sales Professionals
Deep Learning for Sales Professionals
 
Tracxn Research — Enterprise Collaboration Landscape, November 2016
Tracxn Research — Enterprise Collaboration Landscape, November 2016Tracxn Research — Enterprise Collaboration Landscape, November 2016
Tracxn Research — Enterprise Collaboration Landscape, November 2016
 
Introduction to Machine Learning for Heathcare, Finance, Energy, and Commerce...
Introduction to Machine Learning for Heathcare, Finance, Energy, and Commerce...Introduction to Machine Learning for Heathcare, Finance, Energy, and Commerce...
Introduction to Machine Learning for Heathcare, Finance, Energy, and Commerce...
 
Python for Image Understanding: Deep Learning with Convolutional Neural Nets
Python for Image Understanding: Deep Learning with Convolutional Neural NetsPython for Image Understanding: Deep Learning with Convolutional Neural Nets
Python for Image Understanding: Deep Learning with Convolutional Neural Nets
 
Artificial Intelligence (AI) for Financial Services
Artificial Intelligence (AI) for Financial Services Artificial Intelligence (AI) for Financial Services
Artificial Intelligence (AI) for Financial Services
 
14 Startups Leading the Artificial Intelligence (AI) Revolution
14 Startups Leading the Artificial Intelligence (AI) Revolution14 Startups Leading the Artificial Intelligence (AI) Revolution
14 Startups Leading the Artificial Intelligence (AI) Revolution
 
Transform your Business with AI, Deep Learning and Machine Learning
Transform your Business with AI, Deep Learning and Machine LearningTransform your Business with AI, Deep Learning and Machine Learning
Transform your Business with AI, Deep Learning and Machine Learning
 
The Spring of AI : Leveraging Collective Disruptor Insights
The Spring of AI : Leveraging Collective Disruptor InsightsThe Spring of AI : Leveraging Collective Disruptor Insights
The Spring of AI : Leveraging Collective Disruptor Insights
 
Deep neural networks
Deep neural networksDeep neural networks
Deep neural networks
 
[DSC 2016] 系列活動:李宏毅 / 一天搞懂深度學習
[DSC 2016] 系列活動:李宏毅 / 一天搞懂深度學習[DSC 2016] 系列活動:李宏毅 / 一天搞懂深度學習
[DSC 2016] 系列活動:李宏毅 / 一天搞懂深度學習
 
Deep Learning - The Past, Present and Future of Artificial Intelligence
Deep Learning - The Past, Present and Future of Artificial IntelligenceDeep Learning - The Past, Present and Future of Artificial Intelligence
Deep Learning - The Past, Present and Future of Artificial Intelligence
 
How to Become a Thought Leader in Your Niche
How to Become a Thought Leader in Your NicheHow to Become a Thought Leader in Your Niche
How to Become a Thought Leader in Your Niche
 

Similar to P 01 paw_methods_2017_10_30_v4

Reason and Passion of XML (J Gollner)
Reason and Passion of XML (J Gollner)Reason and Passion of XML (J Gollner)
Reason and Passion of XML (J Gollner)Joe Gollner
 
SAP SD demo ppt - Introduction - for freshers
SAP SD demo ppt - Introduction - for freshersSAP SD demo ppt - Introduction - for freshers
SAP SD demo ppt - Introduction - for freshersSaravanan Manoharan
 
SAP SD Introduction.pdf
SAP SD Introduction.pdfSAP SD Introduction.pdf
SAP SD Introduction.pdfRohitAlawane
 
For Beginners: How to plan for a Successful Online Branding Campaign? -Summar...
For Beginners: How to plan for a Successful Online Branding Campaign? -Summar...For Beginners: How to plan for a Successful Online Branding Campaign? -Summar...
For Beginners: How to plan for a Successful Online Branding Campaign? -Summar...Muhammad Omar
 
Advanced Analytics - Frameworks, Platforms and Metholodologies v 1.0
Advanced Analytics - Frameworks, Platforms and Metholodologies v 1.0Advanced Analytics - Frameworks, Platforms and Metholodologies v 1.0
Advanced Analytics - Frameworks, Platforms and Metholodologies v 1.0Dr. Mohan K. Bavirisetty
 
Seven Reasons to Implement CSQ3
Seven Reasons to Implement CSQ3Seven Reasons to Implement CSQ3
Seven Reasons to Implement CSQ3Frank Koczwara
 
How to build shared understanding with example mapping
How to build shared understanding with example mappingHow to build shared understanding with example mapping
How to build shared understanding with example mappingKent McDonald
 
Data Science 101
Data Science 101Data Science 101
Data Science 101odsc
 
Case Interview Prep - Crack the Case Workshop
Case Interview Prep - Crack the Case Workshop Case Interview Prep - Crack the Case Workshop
Case Interview Prep - Crack the Case Workshop PrepLounge
 
Tom Martens - Cube Ware - The big data challenge - bo
Tom Martens - Cube Ware - The big data challenge - boTom Martens - Cube Ware - The big data challenge - bo
Tom Martens - Cube Ware - The big data challenge - boSogeti Nederland B.V.
 
Why is Financial Modelling important in doing a mining project?
Why is Financial Modelling important in doing a mining project?Why is Financial Modelling important in doing a mining project?
Why is Financial Modelling important in doing a mining project?Ian du Plessis
 

Similar to P 01 paw_methods_2017_10_30_v4 (20)

Benchmarking
BenchmarkingBenchmarking
Benchmarking
 
Reason and Passion of XML (J Gollner)
Reason and Passion of XML (J Gollner)Reason and Passion of XML (J Gollner)
Reason and Passion of XML (J Gollner)
 
Data Science for BI
Data Science for BIData Science for BI
Data Science for BI
 
SAP SD demo ppt - Introduction - for freshers
SAP SD demo ppt - Introduction - for freshersSAP SD demo ppt - Introduction - for freshers
SAP SD demo ppt - Introduction - for freshers
 
SAP SD Introduction.pdf
SAP SD Introduction.pdfSAP SD Introduction.pdf
SAP SD Introduction.pdf
 
Branding session
Branding sessionBranding session
Branding session
 
For Beginners: How to plan for a Successful Online Branding Campaign? -Summar...
For Beginners: How to plan for a Successful Online Branding Campaign? -Summar...For Beginners: How to plan for a Successful Online Branding Campaign? -Summar...
For Beginners: How to plan for a Successful Online Branding Campaign? -Summar...
 
Advanced Analytics - Frameworks, Platforms and Metholodologies v 1.0
Advanced Analytics - Frameworks, Platforms and Metholodologies v 1.0Advanced Analytics - Frameworks, Platforms and Metholodologies v 1.0
Advanced Analytics - Frameworks, Platforms and Metholodologies v 1.0
 
Sap Sddemo
Sap SddemoSap Sddemo
Sap Sddemo
 
3 Key Messages for the CEO
3 Key Messages for the CEO3 Key Messages for the CEO
3 Key Messages for the CEO
 
Seven Reasons to Implement CSQ3
Seven Reasons to Implement CSQ3Seven Reasons to Implement CSQ3
Seven Reasons to Implement CSQ3
 
Business Plan
Business PlanBusiness Plan
Business Plan
 
Build a great pitch deck for your startup.
Build a great pitch deck for your startup.  Build a great pitch deck for your startup.
Build a great pitch deck for your startup.
 
Build a Better Entrepreneur Pitch Deck
Build a Better Entrepreneur Pitch DeckBuild a Better Entrepreneur Pitch Deck
Build a Better Entrepreneur Pitch Deck
 
pitchdeck
pitchdeckpitchdeck
pitchdeck
 
How to build shared understanding with example mapping
How to build shared understanding with example mappingHow to build shared understanding with example mapping
How to build shared understanding with example mapping
 
Data Science 101
Data Science 101Data Science 101
Data Science 101
 
Case Interview Prep - Crack the Case Workshop
Case Interview Prep - Crack the Case Workshop Case Interview Prep - Crack the Case Workshop
Case Interview Prep - Crack the Case Workshop
 
Tom Martens - Cube Ware - The big data challenge - bo
Tom Martens - Cube Ware - The big data challenge - boTom Martens - Cube Ware - The big data challenge - bo
Tom Martens - Cube Ware - The big data challenge - bo
 
Why is Financial Modelling important in doing a mining project?
Why is Financial Modelling important in doing a mining project?Why is Financial Modelling important in doing a mining project?
Why is Financial Modelling important in doing a mining project?
 

Recently uploaded

Amazon TQM (2) Amazon TQM (2)Amazon TQM (2).pptx
Amazon TQM (2) Amazon TQM (2)Amazon TQM (2).pptxAmazon TQM (2) Amazon TQM (2)Amazon TQM (2).pptx
Amazon TQM (2) Amazon TQM (2)Amazon TQM (2).pptxAbdelrhman abooda
 
vip Sarai Rohilla Call Girls 9999965857 Call or WhatsApp Now Book
vip Sarai Rohilla Call Girls 9999965857 Call or WhatsApp Now Bookvip Sarai Rohilla Call Girls 9999965857 Call or WhatsApp Now Book
vip Sarai Rohilla Call Girls 9999965857 Call or WhatsApp Now Bookmanojkuma9823
 
Call Girls In Mahipalpur O9654467111 Escorts Service
Call Girls In Mahipalpur O9654467111  Escorts ServiceCall Girls In Mahipalpur O9654467111  Escorts Service
Call Girls In Mahipalpur O9654467111 Escorts ServiceSapana Sha
 
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)jennyeacort
 
ASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel CanterASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel Cantervoginip
 
9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home Service9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home ServiceSapana Sha
 
04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationships04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationshipsccctableauusergroup
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfgstagge
 
Brighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data StorytellingBrighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data StorytellingNeil Barnes
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsappssapnasaifi408
 
办理学位证纽约大学毕业证(NYU毕业证书)原版一比一
办理学位证纽约大学毕业证(NYU毕业证书)原版一比一办理学位证纽约大学毕业证(NYU毕业证书)原版一比一
办理学位证纽约大学毕业证(NYU毕业证书)原版一比一fhwihughh
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDRafezzaman
 
Industrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdfIndustrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdfLars Albertsson
 
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024thyngster
 
Customer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptxCustomer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptxEmmanuel Dauda
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改yuu sss
 
From idea to production in a day – Leveraging Azure ML and Streamlit to build...
From idea to production in a day – Leveraging Azure ML and Streamlit to build...From idea to production in a day – Leveraging Azure ML and Streamlit to build...
From idea to production in a day – Leveraging Azure ML and Streamlit to build...Florian Roscheck
 
Data Science Jobs and Salaries Analysis.pptx
Data Science Jobs and Salaries Analysis.pptxData Science Jobs and Salaries Analysis.pptx
Data Science Jobs and Salaries Analysis.pptxFurkanTasci3
 

Recently uploaded (20)

Amazon TQM (2) Amazon TQM (2)Amazon TQM (2).pptx
Amazon TQM (2) Amazon TQM (2)Amazon TQM (2).pptxAmazon TQM (2) Amazon TQM (2)Amazon TQM (2).pptx
Amazon TQM (2) Amazon TQM (2)Amazon TQM (2).pptx
 
vip Sarai Rohilla Call Girls 9999965857 Call or WhatsApp Now Book
vip Sarai Rohilla Call Girls 9999965857 Call or WhatsApp Now Bookvip Sarai Rohilla Call Girls 9999965857 Call or WhatsApp Now Book
vip Sarai Rohilla Call Girls 9999965857 Call or WhatsApp Now Book
 
Call Girls in Saket 99530🔝 56974 Escort Service
Call Girls in Saket 99530🔝 56974 Escort ServiceCall Girls in Saket 99530🔝 56974 Escort Service
Call Girls in Saket 99530🔝 56974 Escort Service
 
Call Girls In Mahipalpur O9654467111 Escorts Service
Call Girls In Mahipalpur O9654467111  Escorts ServiceCall Girls In Mahipalpur O9654467111  Escorts Service
Call Girls In Mahipalpur O9654467111 Escorts Service
 
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
 
ASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel CanterASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel Canter
 
9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home Service9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home Service
 
04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationships04242024_CCC TUG_Joins and Relationships
04242024_CCC TUG_Joins and Relationships
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdf
 
Brighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data StorytellingBrighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data Storytelling
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
 
办理学位证纽约大学毕业证(NYU毕业证书)原版一比一
办理学位证纽约大学毕业证(NYU毕业证书)原版一比一办理学位证纽约大学毕业证(NYU毕业证书)原版一比一
办理学位证纽约大学毕业证(NYU毕业证书)原版一比一
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
 
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
 
Industrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdfIndustrialised data - the key to AI success.pdf
Industrialised data - the key to AI success.pdf
 
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
Consent & Privacy Signals on Google *Pixels* - MeasureCamp Amsterdam 2024
 
Customer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptxCustomer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptx
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
 
From idea to production in a day – Leveraging Azure ML and Streamlit to build...
From idea to production in a day – Leveraging Azure ML and Streamlit to build...From idea to production in a day – Leveraging Azure ML and Streamlit to build...
From idea to production in a day – Leveraging Azure ML and Streamlit to build...
 
Data Science Jobs and Salaries Analysis.pptx
Data Science Jobs and Salaries Analysis.pptxData Science Jobs and Salaries Analysis.pptx
Data Science Jobs and Salaries Analysis.pptx
 

P 01 paw_methods_2017_10_30_v4

  • 1. A METHOD TO AI MADNESS Vishwa Kolla Head, Advanced Analytics John Hancock Insurance
  • 3. BUILD ME A MODEL TO … 3 REDUCE COMPLAINTS GROW WALLET- SHARE GROW CSAT REDUCE CHURN REDUCE COST TO TARGET GROW BOTTOM-LINE GROW TOP-LINE REDUCE COST TO ACQUIRE
  • 4. MODEL BUILD IS THE PATH OF LEAST RESISTANCE 4 Platforms R P H20 SPARK TENSOR FLOW TBD SUPERVISED UN SUPERVISED NLTK NUM PY PAN DAS PLOT LY PLATFORMS ALGORITHMS PACKAGES PYO DBC CRY PTO PYPD F SCI KIT TOR NAD O ZICT BAB EL BLA ZE
  • 5. A THOUGHTFUL APPROACH CAN YIELD BETTER OUTCOMES BUSINESS USE CASES DATA MATH TECHNICAL IMPL. BUSINESS IMPL. FEED BACK 5
  • 7. “A” MODEL BUILD FRAMEWORK 7 DATA TARGET CONSTRUCTION EVALUATION PERFORMANCE SOURCES DISTANCE FROM SIGNAL SAMPLING METHOD SAMPLE SIZE SIGNAL SIZE PREDICTION HORIZON UNIT OF ANALYSIS ONE MODEL vs. STRATIFIED ONE MODEL vs. SEVERAL MODELS TARGET DEFINITION PRESENCE OR ABSENCE BLACK BOX vs. CLEAR BOX RECENCY FREQUENCY SEVERITY FEATURE SELECTION MODELING STRATEGY MODEL STRENGTH EXPLANATORY vs. IMPORTANCE ACCURACY vs. SENSITIVITY vs. SPECIFICITY ECONOMIES OF SCOPE MODEL FIT BAGGING ENSEMBLE SINGLE vs. MULTIPLE STAGES PREDICTION & OPTIMIZATION BOOSTING
  • 10. IT (ALWAYS) STARTS WITH A BUSINESS PROBLEM PROSPECTING NURTUREACQUISITION MARKET SEGMENTS CUSTOMER SEGMENTS LIKELY TO [*] MEDIA MIX CHANNEL SURVEY ANALYTICS CROSS / UP- SELL OCR MISREP LIKELIHOOD MORTALITY APS SUMMARY FLUIDLESS SMOKER LIKELIHOOD MORBIDITY CHURN NEXT BEST OFFER CLAIM LIKELI- HOOD JOURNEY CLAIM SEVERITY NEXT BEST ACTION FRAUD >> TEXT ANALYTICS OPTIMIZE NEXT LIKELY ACTION WELLNESS IOT ANALYTICS NPS ANOMALY >> 10
  • 11. FOCUS ON INCREMENTAL VALUE KEPT US GROUNDED BUSINESS CASE OPTICAL REALIZABLE SHARED INCREMENTAL
  • 12. IN PROSPECTING, TARGET OPTIMIZAITON IS A JOURNEY 12 … LOWER CUSTOMER TARGETING COSTSA SERIES OF OPTIMIZATION TARGETS … Prospects Leads Apps Issued Placed CPL CPA CPP CP[*] CHANNEL MIX
  • 14. PLANS ARE NOTHING ; PLANNING IS EVERYTHING 14 EDA USEABLE USEFUL DERIVATIVES BI-VARIATE CROSSTAB PRINCOMP JOURNEY
  • 15. A DATA STITCH IN TIME SAVES NINE 15 CLAIM TERMIN ATION CLAIM ACTV. DEMOS CALLS INTERA CTION CLAIM INIT. CUSTOMER MONTH FRAUD DETECTION
  • 16. UNDERSTANDING DATA SAVES (NOT WASTES) TIME 16 Signal Distribution Pop. Incidence Rate Skews Model Inclusion
  • 18. FLEXIBILITY IN TARGET DEFINITION IMPROVED ACTIONABILITY 18 20172014 Predict incidence In next 3 years 2007 20172014 Predict incidence 3 years out 2007 Vs.
  • 19. RIGHT SIZING SIGNAL CAN YIELD BETTER OUTCOMES 19 SIGNAL DILUTION SIGNAL AMPLIFICATION 1% 99% 40% 60%
  • 20. SIMPLE MODELS CAN HELP US EXPLAIN BIG DRIVERS 20 PREDICTORS PRESENCE RECENT FREQUENT SEVERE
  • 21. QUANTIFICATION OF INFORMATION GAP IS A GOOD FIRST STEP © Andrew Ng INFORMATION GAP
  • 22. WINNING MODEL CHALLENGING CHAMPIONS HELPS US UP THE ANTE 22 DATA TARGET SOURCES DISTANCE FROM SIGNAL SAMPLING METHOD SAMPLE SIZE SIGNAL SIZE PREDICTION HORIZON UNIT OF ANALYSIS ONE MODEL vs. STRATIFIED ONE MODEL vs. SEVERAL MODELS TARGET DEFINITION METHODS LINEAR TREES DEEP- LEARNING EVALUATION MODEL STRENGTH EXPLANATORY vs. IMPORTANCE ACCURACY vs. SENSITIVITY vs. SPECIFICITY ECONOMIES OF SCOPE MODEL FIT
  • 24. MULTI-STAGED MODELS PROVIDED IMPLEMENTATION FLEXIBILITY 24 9-101-8 1-7 8-10 Likely to Qualify Likely to Respond Sweet Spot DESIRED SIGNAL MODEL MIS- CLASSIFICATION MODEL EXCLUDE NOISE1 INCLUDE MIS-CLASSIFIERS2 STAGES
  • 26. A CULTURE OF MEASUREMENT, TEST AND LEARN IMPROVES VALUE 26 Measure TestLearn Build CONTINUOUS IMPROVEMENT