SlideShare a Scribd company logo
1 of 25
Team DevScope
https://www.kaggle.com/c/kdm-porto-2019/overview
https://www.kaggle.com/c/kdm-porto-2019/leaderboard
• Non-Stationary Time Series
• (obvious trends/change points)
• Small Test Set (744 rows, 31 days, 24 hours)
• 23 months, but only Dec 2017 available
• (hard for models to model proper seasonality for Dec 2018 prediction)
• Hard to have an internal validation strategy
• We used Nov 2018 as validation set, closest to the target month Dec 2018
• Besides that, we also trusted on submission visualization and PLB score
• Time splitted validation for bayesian optimization
• Time!  (lack of!)
• Blend of very different approaches
• (regression, time series, human knowledge)
• Post/processing, human in the loop forecast
• Christmas holidays, model obvious errors
• Obsessive & Constant EDA of Train & Submissions
• LightGBM, Prophet, Bayesian Search
• Ceiling raw predictions (to be confirmed)
• “The time series with the real values will be constructed
using a simple agent headcount average rounded to the next integer”
• Probe the leaderboard (as feedback is important)
• Yes, It’s a competition!
Model Public Score Private Score
Blend of Top 3 below +ceiling 8.53526 8.69781 1st
Prophet Based 10.39806 9.86448 -
31 Models (LGBM regression) 10.44760 10.51587 -
LGBM regression 10.23931 10.34396 -
+Adjustments values for Christmas holidays/special days, based on 2017 Dec
+Other Post Processing fixing specific model limitations
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
• Post/processing & fixing
obvious errors
• (model limitations, not always
time to inspect root causes)
• Negative predictions -> 0 (as
agent head count cannot be
negative)
•
•
• Use other features (we mostly used target only)
• Explore feature engineering with additional features
• Understand the predictive power of other
features/metrics
• Explore Azure AutoML Forecast
• promising results but couldn't make it on time to submit 
• Built-in time split cross validation, holidays & model search
• Used Total Calls as a feature to predict agents.
• But it did not work. Score private: 15.41 and public: 16.37
• Did not work on local validation
• Model with 4 variables (On a Call Time, Total Handle Time, Total Calls, and Agent
Headcount)
• One reason for not improving the score considering additional variables, can
be that, additional variables are calculated based on Agent Headcount of
that hour.
What didn’t work (so far...)
Microsoft
LightGBM
•
•
•
•
•
•
• Non-Gaussian forecasting using fable
• On the Automation of Time Series Forecasting
Models: Technical and Organizational Considerations.
• Forecasting Stock Prices using Exponential Smoothing
• An overview of time series forecasting models
• https://github.com/DevScope/ai-lab
• https://medium.com/devscope-ai
Also reach us at
© 2019 DevScope. All rights reserved.
Rua Passos Manuel Nº 223 – 4º Andar
4000-385 Porto
T. +351 223 751 350/51
F. +351 223 751 352
Av. Sidónio Pais, Nº 2 – 3º Andar
1050-214 5 Lisboa
info@devscope.net
www.devscope.net

More Related Content

What's hot

Using Processes and Timers for Long-Running Asynchronous Tasks
Using Processes and Timers for Long-Running Asynchronous TasksUsing Processes and Timers for Long-Running Asynchronous Tasks
Using Processes and Timers for Long-Running Asynchronous TasksOutSystems
 
PPM Studio: 7 Steps For Projects Portfolio Improvement
PPM Studio: 7 Steps For Projects Portfolio ImprovementPPM Studio: 7 Steps For Projects Portfolio Improvement
PPM Studio: 7 Steps For Projects Portfolio Improvementgoeldg
 
AOTB Accounting for Agile
AOTB Accounting for AgileAOTB Accounting for Agile
AOTB Accounting for AgileBrett Ansley
 
Using Processes and Timers for Long-Running Asynchronous Tasks
Using Processes and Timers for Long-Running Asynchronous TasksUsing Processes and Timers for Long-Running Asynchronous Tasks
Using Processes and Timers for Long-Running Asynchronous TasksOutSystems
 
Digital Transformation & Solvency II Simulations for L&G: Optimizing, Acceler...
Digital Transformation & Solvency II Simulations for L&G: Optimizing, Acceler...Digital Transformation & Solvency II Simulations for L&G: Optimizing, Acceler...
Digital Transformation & Solvency II Simulations for L&G: Optimizing, Acceler...OW2
 
From zero to test in 60 seconds
From zero to test in 60  secondsFrom zero to test in 60  seconds
From zero to test in 60 secondsHugh McCamphill
 
Bogdan molocea scaling up using automation and performance testing
Bogdan molocea   scaling up using automation and performance testingBogdan molocea   scaling up using automation and performance testing
Bogdan molocea scaling up using automation and performance testingCodecamp Romania
 
JOSA TechTalk - Lambda architecture and real-time processing
JOSA TechTalk - Lambda architecture and real-time processingJOSA TechTalk - Lambda architecture and real-time processing
JOSA TechTalk - Lambda architecture and real-time processingMahmoud Jalajel
 
Introduction to Graph QL
Introduction to Graph QLIntroduction to Graph QL
Introduction to Graph QLDeepak More
 
Serverless Apps on Google Cloud: more dev, less ops
Serverless Apps on Google Cloud:  more dev, less opsServerless Apps on Google Cloud:  more dev, less ops
Serverless Apps on Google Cloud: more dev, less opsJoseph Lust
 
Tool it up #5 new relic
Tool it up #5   new relicTool it up #5   new relic
Tool it up #5 new relictoolitup
 
ERP Software for Contractors construction-erp
ERP Software for Contractors construction-erpERP Software for Contractors construction-erp
ERP Software for Contractors construction-erpconstructionerp
 
Troubleshooting Dashboard Performance
Troubleshooting Dashboard PerformanceTroubleshooting Dashboard Performance
Troubleshooting Dashboard PerformanceOutSystems
 
Voxeo Summit Day 2 - Using CXP hotspot analytics
Voxeo Summit Day 2 - Using CXP hotspot analyticsVoxeo Summit Day 2 - Using CXP hotspot analytics
Voxeo Summit Day 2 - Using CXP hotspot analyticsVoxeo Corp
 
Xenon - Test Automation Tool
Xenon - Test Automation Tool Xenon - Test Automation Tool
Xenon - Test Automation Tool Jade Global
 
Sulph(on)ation Network Optimization rev1
Sulph(on)ation Network Optimization rev1Sulph(on)ation Network Optimization rev1
Sulph(on)ation Network Optimization rev1Andrew Gardner
 
Xenon Automation Testing Tool - Test Anything, Anywhere, Anytime
Xenon Automation Testing Tool - Test Anything, Anywhere, AnytimeXenon Automation Testing Tool - Test Anything, Anywhere, Anytime
Xenon Automation Testing Tool - Test Anything, Anywhere, AnytimeJade Global
 

What's hot (20)

Using Processes and Timers for Long-Running Asynchronous Tasks
Using Processes and Timers for Long-Running Asynchronous TasksUsing Processes and Timers for Long-Running Asynchronous Tasks
Using Processes and Timers for Long-Running Asynchronous Tasks
 
PPM Studio: 7 Steps For Projects Portfolio Improvement
PPM Studio: 7 Steps For Projects Portfolio ImprovementPPM Studio: 7 Steps For Projects Portfolio Improvement
PPM Studio: 7 Steps For Projects Portfolio Improvement
 
AOTB Accounting for Agile
AOTB Accounting for AgileAOTB Accounting for Agile
AOTB Accounting for Agile
 
Using Processes and Timers for Long-Running Asynchronous Tasks
Using Processes and Timers for Long-Running Asynchronous TasksUsing Processes and Timers for Long-Running Asynchronous Tasks
Using Processes and Timers for Long-Running Asynchronous Tasks
 
Digital Transformation & Solvency II Simulations for L&G: Optimizing, Acceler...
Digital Transformation & Solvency II Simulations for L&G: Optimizing, Acceler...Digital Transformation & Solvency II Simulations for L&G: Optimizing, Acceler...
Digital Transformation & Solvency II Simulations for L&G: Optimizing, Acceler...
 
From zero to test in 60 seconds
From zero to test in 60  secondsFrom zero to test in 60  seconds
From zero to test in 60 seconds
 
Bogdan molocea scaling up using automation and performance testing
Bogdan molocea   scaling up using automation and performance testingBogdan molocea   scaling up using automation and performance testing
Bogdan molocea scaling up using automation and performance testing
 
Supervise your Akka actors
Supervise your Akka actorsSupervise your Akka actors
Supervise your Akka actors
 
Digital project process
Digital project processDigital project process
Digital project process
 
JOSA TechTalk - Lambda architecture and real-time processing
JOSA TechTalk - Lambda architecture and real-time processingJOSA TechTalk - Lambda architecture and real-time processing
JOSA TechTalk - Lambda architecture and real-time processing
 
Introduction to Graph QL
Introduction to Graph QLIntroduction to Graph QL
Introduction to Graph QL
 
Serverless Apps on Google Cloud: more dev, less ops
Serverless Apps on Google Cloud:  more dev, less opsServerless Apps on Google Cloud:  more dev, less ops
Serverless Apps on Google Cloud: more dev, less ops
 
Tool it up #5 new relic
Tool it up #5   new relicTool it up #5   new relic
Tool it up #5 new relic
 
ERP Software for Contractors construction-erp
ERP Software for Contractors construction-erpERP Software for Contractors construction-erp
ERP Software for Contractors construction-erp
 
Troubleshooting Dashboard Performance
Troubleshooting Dashboard PerformanceTroubleshooting Dashboard Performance
Troubleshooting Dashboard Performance
 
Voxeo Summit Day 2 - Using CXP hotspot analytics
Voxeo Summit Day 2 - Using CXP hotspot analyticsVoxeo Summit Day 2 - Using CXP hotspot analytics
Voxeo Summit Day 2 - Using CXP hotspot analytics
 
Xenon - Test Automation Tool
Xenon - Test Automation Tool Xenon - Test Automation Tool
Xenon - Test Automation Tool
 
Sulph(on)ation Network Optimization rev1
Sulph(on)ation Network Optimization rev1Sulph(on)ation Network Optimization rev1
Sulph(on)ation Network Optimization rev1
 
Xenon Automation Testing Tool - Test Anything, Anywhere, Anytime
Xenon Automation Testing Tool - Test Anything, Anywhere, AnytimeXenon Automation Testing Tool - Test Anything, Anywhere, Anytime
Xenon Automation Testing Tool - Test Anything, Anywhere, Anytime
 
Data Analytics Domain
Data Analytics DomainData Analytics Domain
Data Analytics Domain
 

Similar to Kaggle Days Porto 2019 - 1st place presentation by team DevScope

Wbs, estimation and scheduling
Wbs, estimation and schedulingWbs, estimation and scheduling
Wbs, estimation and schedulingSulman Ahmed
 
Scrum Crash Course - Anatoli Iliev and Lyubomir Cholakov, Infragistics
Scrum Crash Course - Anatoli Iliev and Lyubomir Cholakov, InfragisticsScrum Crash Course - Anatoli Iliev and Lyubomir Cholakov, Infragistics
Scrum Crash Course - Anatoli Iliev and Lyubomir Cholakov, InfragisticsbeITconference
 
The Dashlane Agile Journey
The Dashlane Agile JourneyThe Dashlane Agile Journey
The Dashlane Agile JourneyDashlane
 
Estimating time-tracking
Estimating time-trackingEstimating time-tracking
Estimating time-trackingLeigh White
 
Software development planning and essentials
Software development planning and essentialsSoftware development planning and essentials
Software development planning and essentialsRajesh P
 
Software development planning and essentials
Software development planning and essentialsSoftware development planning and essentials
Software development planning and essentialsRajesh P
 
Introduction to scrum
Introduction to scrumIntroduction to scrum
Introduction to scrumSemen Arslan
 
Htf2014 managing share point projects with agile and tfs andy
Htf2014 managing share point projects with agile and tfs   andyHtf2014 managing share point projects with agile and tfs   andy
Htf2014 managing share point projects with agile and tfs andySparkhound Inc.
 

Similar to Kaggle Days Porto 2019 - 1st place presentation by team DevScope (20)

Oracle Compensation Workbench Webinar
Oracle Compensation Workbench WebinarOracle Compensation Workbench Webinar
Oracle Compensation Workbench Webinar
 
5. PMP Training - Time management
5. PMP Training - Time management5. PMP Training - Time management
5. PMP Training - Time management
 
Wbs
WbsWbs
Wbs
 
Wbs, estimation and scheduling
Wbs, estimation and schedulingWbs, estimation and scheduling
Wbs, estimation and scheduling
 
Agile Methodologies
Agile MethodologiesAgile Methodologies
Agile Methodologies
 
Scrum Crash Course - Anatoli Iliev and Lyubomir Cholakov, Infragistics
Scrum Crash Course - Anatoli Iliev and Lyubomir Cholakov, InfragisticsScrum Crash Course - Anatoli Iliev and Lyubomir Cholakov, Infragistics
Scrum Crash Course - Anatoli Iliev and Lyubomir Cholakov, Infragistics
 
The Dashlane Agile Journey
The Dashlane Agile JourneyThe Dashlane Agile Journey
The Dashlane Agile Journey
 
Estimating time-tracking
Estimating time-trackingEstimating time-tracking
Estimating time-tracking
 
Software development planning and essentials
Software development planning and essentialsSoftware development planning and essentials
Software development planning and essentials
 
Software development planning and essentials
Software development planning and essentialsSoftware development planning and essentials
Software development planning and essentials
 
Agile by KD
Agile by KDAgile by KD
Agile by KD
 
Agile by KD
Agile by KDAgile by KD
Agile by KD
 
Scrum toufiq
Scrum toufiqScrum toufiq
Scrum toufiq
 
Implementing Scrum with Kanban
Implementing Scrum with KanbanImplementing Scrum with Kanban
Implementing Scrum with Kanban
 
Scrum 101
Scrum 101 Scrum 101
Scrum 101
 
Introduction to scrum
Introduction to scrumIntroduction to scrum
Introduction to scrum
 
Adamson "Blueprint for Managing Your Project"
Adamson "Blueprint for Managing Your Project"Adamson "Blueprint for Managing Your Project"
Adamson "Blueprint for Managing Your Project"
 
Htf2014 managing share point projects with agile and tfs andy
Htf2014 managing share point projects with agile and tfs   andyHtf2014 managing share point projects with agile and tfs   andy
Htf2014 managing share point projects with agile and tfs andy
 
Agile India 2014 - Venkatraman L on Scaling Agile
Agile India 2014 - Venkatraman L on Scaling AgileAgile India 2014 - Venkatraman L on Scaling Agile
Agile India 2014 - Venkatraman L on Scaling Agile
 
SCRUM Intro
SCRUM IntroSCRUM Intro
SCRUM Intro
 

More from Rui Quintino

“Houston, we have a model...” Introduction to MLOps
“Houston, we have a model...” Introduction to MLOps“Houston, we have a model...” Introduction to MLOps
“Houston, we have a model...” Introduction to MLOpsRui Quintino
 
Power BI for Data Science and Machine Learning - Data Science Portugal meetup
Power BI for Data Science and Machine Learning  - Data Science Portugal meetupPower BI for Data Science and Machine Learning  - Data Science Portugal meetup
Power BI for Data Science and Machine Learning - Data Science Portugal meetupRui Quintino
 
Empowering you - Power BI, Power Platform & AI Builder
Empowering you  -  Power BI, Power Platform & AI BuilderEmpowering you  -  Power BI, Power Platform & AI Builder
Empowering you - Power BI, Power Platform & AI BuilderRui Quintino
 
Jupyter Notebooks: Introduction, Tips & Tools
Jupyter Notebooks: Introduction, Tips & ToolsJupyter Notebooks: Introduction, Tips & Tools
Jupyter Notebooks: Introduction, Tips & ToolsRui Quintino
 
DataSciencePT #27 - Fifty Shades of Automated Machine Learning
DataSciencePT #27 - Fifty Shades of Automated Machine LearningDataSciencePT #27 - Fifty Shades of Automated Machine Learning
DataSciencePT #27 - Fifty Shades of Automated Machine LearningRui Quintino
 
Docker & Containers for Big Data, Data Science, Machine Learning & Deep Learning
Docker & Containers for Big Data, Data Science, Machine Learning & Deep LearningDocker & Containers for Big Data, Data Science, Machine Learning & Deep Learning
Docker & Containers for Big Data, Data Science, Machine Learning & Deep LearningRui Quintino
 
Microsoft Cognitive Services & Bot Framework - Universidade Fernando Pessoa
Microsoft Cognitive Services & Bot Framework - Universidade Fernando PessoaMicrosoft Cognitive Services & Bot Framework - Universidade Fernando Pessoa
Microsoft Cognitive Services & Bot Framework - Universidade Fernando PessoaRui Quintino
 
Open Source Deep Learning & Machine Learning with Microsoft CNTK & LightGBM
Open Source Deep Learning & Machine Learning with Microsoft CNTK & LightGBMOpen Source Deep Learning & Machine Learning with Microsoft CNTK & LightGBM
Open Source Deep Learning & Machine Learning with Microsoft CNTK & LightGBMRui Quintino
 
Data Science Portugal Meetup 7 - Machine Learning & Data Science Safety Remi...
Data Science Portugal  Meetup 7 - Machine Learning & Data Science Safety Remi...Data Science Portugal  Meetup 7 - Machine Learning & Data Science Safety Remi...
Data Science Portugal Meetup 7 - Machine Learning & Data Science Safety Remi...Rui Quintino
 
Microsoft Data Platform Airlift 2017 Rui Quintino Machine Learning with SQL S...
Microsoft Data Platform Airlift 2017 Rui Quintino Machine Learning with SQL S...Microsoft Data Platform Airlift 2017 Rui Quintino Machine Learning with SQL S...
Microsoft Data Platform Airlift 2017 Rui Quintino Machine Learning with SQL S...Rui Quintino
 
Sql Saturday Lisbon 2017 Rui Quintino -R first steps for sql devs & dbas
Sql Saturday Lisbon 2017 Rui Quintino -R first steps for sql devs & dbasSql Saturday Lisbon 2017 Rui Quintino -R first steps for sql devs & dbas
Sql Saturday Lisbon 2017 Rui Quintino -R first steps for sql devs & dbasRui Quintino
 
The Power of Now! Azure Stream Analytics - Microsoft ITPro AirLift
The Power of Now! Azure Stream Analytics - Microsoft ITPro AirLiftThe Power of Now! Azure Stream Analytics - Microsoft ITPro AirLift
The Power of Now! Azure Stream Analytics - Microsoft ITPro AirLiftRui Quintino
 
SQL Saturday #188 Portugal - "Faster than the speed of light"... with Microso...
SQL Saturday #188 Portugal - "Faster than the speed of light"... with Microso...SQL Saturday #188 Portugal - "Faster than the speed of light"... with Microso...
SQL Saturday #188 Portugal - "Faster than the speed of light"... with Microso...Rui Quintino
 
"De histórias mal contadas..."
"De histórias mal contadas...""De histórias mal contadas..."
"De histórias mal contadas..."Rui Quintino
 

More from Rui Quintino (14)

“Houston, we have a model...” Introduction to MLOps
“Houston, we have a model...” Introduction to MLOps“Houston, we have a model...” Introduction to MLOps
“Houston, we have a model...” Introduction to MLOps
 
Power BI for Data Science and Machine Learning - Data Science Portugal meetup
Power BI for Data Science and Machine Learning  - Data Science Portugal meetupPower BI for Data Science and Machine Learning  - Data Science Portugal meetup
Power BI for Data Science and Machine Learning - Data Science Portugal meetup
 
Empowering you - Power BI, Power Platform & AI Builder
Empowering you  -  Power BI, Power Platform & AI BuilderEmpowering you  -  Power BI, Power Platform & AI Builder
Empowering you - Power BI, Power Platform & AI Builder
 
Jupyter Notebooks: Introduction, Tips & Tools
Jupyter Notebooks: Introduction, Tips & ToolsJupyter Notebooks: Introduction, Tips & Tools
Jupyter Notebooks: Introduction, Tips & Tools
 
DataSciencePT #27 - Fifty Shades of Automated Machine Learning
DataSciencePT #27 - Fifty Shades of Automated Machine LearningDataSciencePT #27 - Fifty Shades of Automated Machine Learning
DataSciencePT #27 - Fifty Shades of Automated Machine Learning
 
Docker & Containers for Big Data, Data Science, Machine Learning & Deep Learning
Docker & Containers for Big Data, Data Science, Machine Learning & Deep LearningDocker & Containers for Big Data, Data Science, Machine Learning & Deep Learning
Docker & Containers for Big Data, Data Science, Machine Learning & Deep Learning
 
Microsoft Cognitive Services & Bot Framework - Universidade Fernando Pessoa
Microsoft Cognitive Services & Bot Framework - Universidade Fernando PessoaMicrosoft Cognitive Services & Bot Framework - Universidade Fernando Pessoa
Microsoft Cognitive Services & Bot Framework - Universidade Fernando Pessoa
 
Open Source Deep Learning & Machine Learning with Microsoft CNTK & LightGBM
Open Source Deep Learning & Machine Learning with Microsoft CNTK & LightGBMOpen Source Deep Learning & Machine Learning with Microsoft CNTK & LightGBM
Open Source Deep Learning & Machine Learning with Microsoft CNTK & LightGBM
 
Data Science Portugal Meetup 7 - Machine Learning & Data Science Safety Remi...
Data Science Portugal  Meetup 7 - Machine Learning & Data Science Safety Remi...Data Science Portugal  Meetup 7 - Machine Learning & Data Science Safety Remi...
Data Science Portugal Meetup 7 - Machine Learning & Data Science Safety Remi...
 
Microsoft Data Platform Airlift 2017 Rui Quintino Machine Learning with SQL S...
Microsoft Data Platform Airlift 2017 Rui Quintino Machine Learning with SQL S...Microsoft Data Platform Airlift 2017 Rui Quintino Machine Learning with SQL S...
Microsoft Data Platform Airlift 2017 Rui Quintino Machine Learning with SQL S...
 
Sql Saturday Lisbon 2017 Rui Quintino -R first steps for sql devs & dbas
Sql Saturday Lisbon 2017 Rui Quintino -R first steps for sql devs & dbasSql Saturday Lisbon 2017 Rui Quintino -R first steps for sql devs & dbas
Sql Saturday Lisbon 2017 Rui Quintino -R first steps for sql devs & dbas
 
The Power of Now! Azure Stream Analytics - Microsoft ITPro AirLift
The Power of Now! Azure Stream Analytics - Microsoft ITPro AirLiftThe Power of Now! Azure Stream Analytics - Microsoft ITPro AirLift
The Power of Now! Azure Stream Analytics - Microsoft ITPro AirLift
 
SQL Saturday #188 Portugal - "Faster than the speed of light"... with Microso...
SQL Saturday #188 Portugal - "Faster than the speed of light"... with Microso...SQL Saturday #188 Portugal - "Faster than the speed of light"... with Microso...
SQL Saturday #188 Portugal - "Faster than the speed of light"... with Microso...
 
"De histórias mal contadas..."
"De histórias mal contadas...""De histórias mal contadas..."
"De histórias mal contadas..."
 

Recently uploaded

Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessPixlogix Infotech
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processorsdebabhi2
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUK Journal
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobeapidays
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024The Digital Insurer
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Enterprise Knowledge
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Drew Madelung
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
Developing An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of BrazilDeveloping An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of BrazilV3cube
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Scriptwesley chun
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherRemote DBA Services
 

Recently uploaded (20)

Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
Developing An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of BrazilDeveloping An App To Navigate The Roads of Brazil
Developing An App To Navigate The Roads of Brazil
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 

Kaggle Days Porto 2019 - 1st place presentation by team DevScope

  • 4. • Non-Stationary Time Series • (obvious trends/change points) • Small Test Set (744 rows, 31 days, 24 hours) • 23 months, but only Dec 2017 available • (hard for models to model proper seasonality for Dec 2018 prediction) • Hard to have an internal validation strategy • We used Nov 2018 as validation set, closest to the target month Dec 2018 • Besides that, we also trusted on submission visualization and PLB score • Time splitted validation for bayesian optimization • Time!  (lack of!)
  • 5. • Blend of very different approaches • (regression, time series, human knowledge) • Post/processing, human in the loop forecast • Christmas holidays, model obvious errors • Obsessive & Constant EDA of Train & Submissions • LightGBM, Prophet, Bayesian Search • Ceiling raw predictions (to be confirmed) • “The time series with the real values will be constructed using a simple agent headcount average rounded to the next integer” • Probe the leaderboard (as feedback is important) • Yes, It’s a competition!
  • 6. Model Public Score Private Score Blend of Top 3 below +ceiling 8.53526 8.69781 1st Prophet Based 10.39806 9.86448 - 31 Models (LGBM regression) 10.44760 10.51587 - LGBM regression 10.23931 10.34396 - +Adjustments values for Christmas holidays/special days, based on 2017 Dec +Other Post Processing fixing specific model limitations
  • 8.
  • 9.
  • 10.
  • 15. • Post/processing & fixing obvious errors • (model limitations, not always time to inspect root causes) • Negative predictions -> 0 (as agent head count cannot be negative)
  • 17.
  • 18. • Use other features (we mostly used target only) • Explore feature engineering with additional features • Understand the predictive power of other features/metrics • Explore Azure AutoML Forecast • promising results but couldn't make it on time to submit  • Built-in time split cross validation, holidays & model search
  • 19. • Used Total Calls as a feature to predict agents. • But it did not work. Score private: 15.41 and public: 16.37 • Did not work on local validation • Model with 4 variables (On a Call Time, Total Handle Time, Total Calls, and Agent Headcount) • One reason for not improving the score considering additional variables, can be that, additional variables are calculated based on Agent Headcount of that hour. What didn’t work (so far...)
  • 21.
  • 23. • Non-Gaussian forecasting using fable • On the Automation of Time Series Forecasting Models: Technical and Organizational Considerations. • Forecasting Stock Prices using Exponential Smoothing • An overview of time series forecasting models
  • 25. © 2019 DevScope. All rights reserved. Rua Passos Manuel Nº 223 – 4º Andar 4000-385 Porto T. +351 223 751 350/51 F. +351 223 751 352 Av. Sidónio Pais, Nº 2 – 3º Andar 1050-214 5 Lisboa info@devscope.net www.devscope.net