SlideShare a Scribd company logo
1 of 30
Cooking with AI/ML at SnapLogic
SnapLogic Technology Open House | January 11, 2018
The Story of Iris
Education
• Bachelor of Engineering in Computer Engineering
@ Chulalongkorn University
• Master of Science in Computer Science
@ University of San Francisco
Experience
• Machine Learning Engineer @ SnapLogic
Accomplishment
• Published Research in Machine Learning
• Fulbright Scholar
• Cook Good Foods
• Be Happy
2
Jump Thanawut
Ananpiriyakul
About Me
Agenda
3
Cooking with AI/ML at SnapLogic
Get to Know SnapLogic
• SnapLogic: Enterprise Integration Cloud Platform
• Iris Artificial Intelligence: Machine-Learning-Driven Integration Assistant
The Story of Iris
• Iris's Restaurant
• Survey the Market
• Check our Inventory
• Develop the Recipe
The Future of Iris
Jump's Dream
corporate overview
Get to Know SnapLogic
5
Get to Know SnapLogic
Integration Platform as a Service (iPaaS)
• Data Integration Platform
• Connect Applications and Data
• Self-Service, Fast, Easy
• Reliable, Scalable
• Accelerate Business
Our Product
6
Get to Know SnapLogic
7
Get to Know SnapLogic
Live Demo
Iris Artificial Intelligence
8
Get to Know SnapLogic
What is Iris?
• Artificial Intelligence
• Friendly Assistant
What is good about Iris?
• Accelerate New Joiners
• Improve Efficiency of Experts
• Bring Happiness to Everyone
corporate overview
The Story of Iris
Iris's Restaurant
10
The Story of Iris
What kind of dishes should we serve here?
Survey the Market
Check what people do.
❏ Research Papers
❏ Academic Proven
❏ Informative
❏ Developer Communities
❏ Full of Arguments
❏ Responsive
❏ Market Leaders
❏ Innovative
❏ Competitors
11
The Story of Iris
What's good and how?
❏ Surgery Robot
❏ High Precision
❏ No Emotion
❏ No Fatigue
❏ Self Driving Car
❏ Less Accident
❏ Face ID
❏ More Secure
Innovate new things.
❏ Exclusively for
SnapLogic's Customers
❏ Useful & Reliable
Estimate the cost.
❏ People
❏ Time
❏ Money
Check our Inventory
12
The Story of Iris
Check what ingredients we have.
❏ Raw Data
❏ Log Data
❏ Traffic
What are we going to cook?
❏ Recommendation Engine
❏ Pipeline Generator
❏ Pattern / Event Detector
❏ Automatic Optimizer
❏ Personal Assistant Chatbot
❏ Global Search Engine
Develop the Recipe
13
The Story of Iris
Plating
Make the Final Touch
Turn off the heat. Add Thai chilies, kaffir lime
leaves and lime juice. Make a quick stir.
Pour the soup into the bowl.
Serve with steamed Thai jasmine rice.
Cooking
Listen to the Voice of Ingredients
Boil the water, add all herbs except Thai chilies and
kaffir lime leaves. Wait until the onion is soft.
Add mushrooms. Wait until boiled.
Add Thai chili paste and shrimps. Wait until boiled.
Season with fish sauce and sugar.
Preparation
Clean them All
Wash lemongrass, galangal, Thai onion, Thai chilies and
kaffir lime leaves. Let them dry, beat them with knife.
Wash oyster mushrooms and straw mushrooms.
Clean the shrimps, peel their shells, keep their heads.
Serving
Be Proud
Serve while hot. Reheat if needed.
Be careful when serving hot foods.
Say "Please Enjoy!" with a big smile.
Tom Yum Kung (ต้มยำกุ้ง)
Thai River Prawn Spicy Soup
Develop the Recipe
14
The Story of Iris
Preparation
Develop the Recipe
15
The Story of Iris
Preparation
Previous Snap 3 Previous Snap 2 Previous Snap 1 Current Snap Next Snap
File Reader CSV Parser
File Reader CSV Parser Mapper
File Reader CSV Parser Mapper Copy
File Reader CSV Parser Mapper Copy Mapper
File Reader CSV Parser Mapper Copy Mapper
CSV Parser Mapper Copy Mapper CSV Formatter
CSV Parser Mapper Copy Mapper CSV Formatter
Develop the Recipe
16
The Story of Iris
Preparation
Dataset Extractor and Updater
Develop the Recipe
17
The Story of Iris
Cooking
https://www.bangkokpost.com/lifestyle/social-and-lifestyle/1381587/a-reluctant-star
Iris Development
Process
3 RESULT ANALYSIS
Analyze the result.
Reverse engineer from result to data.
Plan for improvement.
1 ALGORITHM DEVELOPMENT
Choose the right algorithm.
Choose the right configurations.
Choose the right strategies.
2 EVALUATION
Evaluate the algorithms using metrics:
performance, accuracy, and cost.
Develop the Recipe
18
The Story of Iris
Cooking
Most Popular Snap Decision Tree
Plating
Develop the Recipe
19
The Story of Iris
Front-End UIBack-End Service Logging ServiceModel Builder
Serving
Develop the Recipe
20
The Story of Iris
Develop the Recipe
21
The Story of Iris
Plating
Make the Final Touch
Design and implement API on backend side.
Design and implement user interface.
Run load test to estimate the size of servers.
Final test to make sure it is delicious.
Cooking
Listen to the Voice of Ingredients
Try different cooking methods: statistical models,
k-nearest neighbors, decision tree, random forests, etc.
Evaluate with daily backtesting technique.
Make sure the recommendations make sense.
Preparation
Clean them All
Extract successfully executed pipelines.
Transform into directed acyclic graphs.
Chop them into segments of 5 nodes or less.
Preprocess the data.
Serving
Be Proud
Released in 4.9 (Spring, 2017).
Source Snap Recommendation in 4.11 (Fall 2017).
Observe usages and issues.
Continue research and development.
corporate overview
The Future of Iris
Neural Networks
23
The Future of Iris
http://playground.tensorflow.org
Iris X Neural Networks
24
The Future of Iris
Decision Tree
❏ It's interpretable.
❏ It takes minutes to train.
❏ It performs well on smaller dataset.
❏ It's solid and specific.
❏ It's for prototyping and early release.
❏ It's cheaper.
Neural Networks
❏ It's magic.
❏ It takes hours or days to train.
❏ It's more powerful on large dataset.
❏ It's flexible and optimizable.
❏ It's for advancement and future release.
❏ It's a lot more expensive.
Neural Networks and Deep Learning are definitely a present. They are powerful and
super flexible. We are proud to announce that Iris has adopted those technologies.
The Truth
25
The Future of Iris
Large Neural Networks (90%*)
Expected to reach 90% hit rate with additional
layers and wider layers.
1
Improved Decision Tree (81%)
Prune the tree to get the best result.3
Statistical Model (49%)
Always recommend top 5 popular snaps.5
Small Neural Networks (87%*)
Look back upto last 9 snaps. Also learn from user
information to make recommendations personalized.
2
Decision Tree (78%)
Consider last 4 snaps on the canvas as
an input set. Walk down the decision tree
and get recommendations from the leaf.
4
Latest Prototype
26
The Future of Iris
input_layer = Input(shape=(9,))
input_org_layer = Input(shape=(1,))
emb_layer = Embedding(num_snap_type, 128, input_length=seg_len-1)(input_layer)
emb_org_layer = Embedding(num_org, 64, input_length=1)(input_org_layer)
flatten_org = Flatten()(emb_org_layer)
lstm_layer = LSTM(1024)(emb_layer)
drop_layer = Dropout(0.2)(lstm_layer)
merge_layer = Concatenate(axis=1)([drop_layer, flatten_org])
output_layer = Dense(num_snap_type, activation='softmax')(merge_layer)
model = Model(inputs=[input_layer, input_org_layer], outputs=output_layer)
corporate overview
Jump's Dream
Design Concept: The Easier Future of AI
28
Jump's Dream
Design Concept: The Easier Future of AI
29
Jump's Dream
Q & A
Please Enjoy!Thank You for Coming

More Related Content

What's hot

SnapLogic Live: Enabling the Citizen Integrator
SnapLogic Live: Enabling the Citizen IntegratorSnapLogic Live: Enabling the Citizen Integrator
SnapLogic Live: Enabling the Citizen IntegratorSnapLogic
 
Pivotal Digital Transformation Forum: Data Science Technical Overview
Pivotal Digital Transformation Forum: Data Science Technical OverviewPivotal Digital Transformation Forum: Data Science Technical Overview
Pivotal Digital Transformation Forum: Data Science Technical OverviewVMware Tanzu
 
Connect Faster with SnapLogic at Workday Rising
Connect Faster with SnapLogic at Workday RisingConnect Faster with SnapLogic at Workday Rising
Connect Faster with SnapLogic at Workday RisingSnapLogic
 
The SnapLogic Integration Cloud for ServiceNow
The SnapLogic Integration Cloud for ServiceNowThe SnapLogic Integration Cloud for ServiceNow
The SnapLogic Integration Cloud for ServiceNowSnapLogic
 
Appplications – Driving Expansion In The Cloud
Appplications – Driving Expansion In The CloudAppplications – Driving Expansion In The Cloud
Appplications – Driving Expansion In The CloudNetAppUK
 
NetApp at Gartner Symposium Show Guide
NetApp at Gartner Symposium Show GuideNetApp at Gartner Symposium Show Guide
NetApp at Gartner Symposium Show GuideNetAppUK
 
Bringing AIOps to Hybrid Cloud Monitoring and Management
Bringing AIOps to Hybrid Cloud Monitoring and ManagementBringing AIOps to Hybrid Cloud Monitoring and Management
Bringing AIOps to Hybrid Cloud Monitoring and ManagementOpsRamp
 
PgConf 2018 - Postgres in a World of DevOps
PgConf 2018 - Postgres in a World of DevOpsPgConf 2018 - Postgres in a World of DevOps
PgConf 2018 - Postgres in a World of DevOpsEDB
 
Discover how the cloud will empower every organization on the planet to achie...
Discover how the cloud will empower every organization on the planet to achie...Discover how the cloud will empower every organization on the planet to achie...
Discover how the cloud will empower every organization on the planet to achie...Codit
 
Webinar: Attaining Excellence in Big Data Integration
Webinar: Attaining Excellence in Big Data IntegrationWebinar: Attaining Excellence in Big Data Integration
Webinar: Attaining Excellence in Big Data IntegrationSnapLogic
 
Data Warehousing in the Cloud: Practical Migration Strategies
Data Warehousing in the Cloud: Practical Migration Strategies Data Warehousing in the Cloud: Practical Migration Strategies
Data Warehousing in the Cloud: Practical Migration Strategies SnapLogic
 
Lightning-Fast, Interactive Business Intelligence Performance with MicroStrat...
Lightning-Fast, Interactive Business Intelligence Performance with MicroStrat...Lightning-Fast, Interactive Business Intelligence Performance with MicroStrat...
Lightning-Fast, Interactive Business Intelligence Performance with MicroStrat...Tyler Wishnoff
 
Postgres Vision 2018: The Pragmatic Cloud
Postgres Vision 2018:  The Pragmatic CloudPostgres Vision 2018:  The Pragmatic Cloud
Postgres Vision 2018: The Pragmatic CloudEDB
 
10 Good Reasons: NetApp for Healthcare
10 Good Reasons: NetApp for Healthcare10 Good Reasons: NetApp for Healthcare
10 Good Reasons: NetApp for HealthcareNetApp
 
Embracing Cloud Agility to Maximize Flexibility & Performance
Embracing Cloud Agility to Maximize Flexibility & Performance Embracing Cloud Agility to Maximize Flexibility & Performance
Embracing Cloud Agility to Maximize Flexibility & Performance Talend
 
How to accelerate Splunk analytics
How to accelerate Splunk analyticsHow to accelerate Splunk analytics
How to accelerate Splunk analyticsClearSky Data
 
Postgres Vision 2018: Taking Postgres Everywhere
Postgres Vision 2018: Taking Postgres EverywherePostgres Vision 2018: Taking Postgres Everywhere
Postgres Vision 2018: Taking Postgres EverywhereEDB
 
Downsizing Data Centers by NetApp IT
Downsizing Data Centers by NetApp ITDownsizing Data Centers by NetApp IT
Downsizing Data Centers by NetApp ITNetApp
 
Hybrid Cloud: The Cloud on Your Terms
Hybrid Cloud: The Cloud on Your TermsHybrid Cloud: The Cloud on Your Terms
Hybrid Cloud: The Cloud on Your TermsNetApp
 

What's hot (19)

SnapLogic Live: Enabling the Citizen Integrator
SnapLogic Live: Enabling the Citizen IntegratorSnapLogic Live: Enabling the Citizen Integrator
SnapLogic Live: Enabling the Citizen Integrator
 
Pivotal Digital Transformation Forum: Data Science Technical Overview
Pivotal Digital Transformation Forum: Data Science Technical OverviewPivotal Digital Transformation Forum: Data Science Technical Overview
Pivotal Digital Transformation Forum: Data Science Technical Overview
 
Connect Faster with SnapLogic at Workday Rising
Connect Faster with SnapLogic at Workday RisingConnect Faster with SnapLogic at Workday Rising
Connect Faster with SnapLogic at Workday Rising
 
The SnapLogic Integration Cloud for ServiceNow
The SnapLogic Integration Cloud for ServiceNowThe SnapLogic Integration Cloud for ServiceNow
The SnapLogic Integration Cloud for ServiceNow
 
Appplications – Driving Expansion In The Cloud
Appplications – Driving Expansion In The CloudAppplications – Driving Expansion In The Cloud
Appplications – Driving Expansion In The Cloud
 
NetApp at Gartner Symposium Show Guide
NetApp at Gartner Symposium Show GuideNetApp at Gartner Symposium Show Guide
NetApp at Gartner Symposium Show Guide
 
Bringing AIOps to Hybrid Cloud Monitoring and Management
Bringing AIOps to Hybrid Cloud Monitoring and ManagementBringing AIOps to Hybrid Cloud Monitoring and Management
Bringing AIOps to Hybrid Cloud Monitoring and Management
 
PgConf 2018 - Postgres in a World of DevOps
PgConf 2018 - Postgres in a World of DevOpsPgConf 2018 - Postgres in a World of DevOps
PgConf 2018 - Postgres in a World of DevOps
 
Discover how the cloud will empower every organization on the planet to achie...
Discover how the cloud will empower every organization on the planet to achie...Discover how the cloud will empower every organization on the planet to achie...
Discover how the cloud will empower every organization on the planet to achie...
 
Webinar: Attaining Excellence in Big Data Integration
Webinar: Attaining Excellence in Big Data IntegrationWebinar: Attaining Excellence in Big Data Integration
Webinar: Attaining Excellence in Big Data Integration
 
Data Warehousing in the Cloud: Practical Migration Strategies
Data Warehousing in the Cloud: Practical Migration Strategies Data Warehousing in the Cloud: Practical Migration Strategies
Data Warehousing in the Cloud: Practical Migration Strategies
 
Lightning-Fast, Interactive Business Intelligence Performance with MicroStrat...
Lightning-Fast, Interactive Business Intelligence Performance with MicroStrat...Lightning-Fast, Interactive Business Intelligence Performance with MicroStrat...
Lightning-Fast, Interactive Business Intelligence Performance with MicroStrat...
 
Postgres Vision 2018: The Pragmatic Cloud
Postgres Vision 2018:  The Pragmatic CloudPostgres Vision 2018:  The Pragmatic Cloud
Postgres Vision 2018: The Pragmatic Cloud
 
10 Good Reasons: NetApp for Healthcare
10 Good Reasons: NetApp for Healthcare10 Good Reasons: NetApp for Healthcare
10 Good Reasons: NetApp for Healthcare
 
Embracing Cloud Agility to Maximize Flexibility & Performance
Embracing Cloud Agility to Maximize Flexibility & Performance Embracing Cloud Agility to Maximize Flexibility & Performance
Embracing Cloud Agility to Maximize Flexibility & Performance
 
How to accelerate Splunk analytics
How to accelerate Splunk analyticsHow to accelerate Splunk analytics
How to accelerate Splunk analytics
 
Postgres Vision 2018: Taking Postgres Everywhere
Postgres Vision 2018: Taking Postgres EverywherePostgres Vision 2018: Taking Postgres Everywhere
Postgres Vision 2018: Taking Postgres Everywhere
 
Downsizing Data Centers by NetApp IT
Downsizing Data Centers by NetApp ITDownsizing Data Centers by NetApp IT
Downsizing Data Centers by NetApp IT
 
Hybrid Cloud: The Cloud on Your Terms
Hybrid Cloud: The Cloud on Your TermsHybrid Cloud: The Cloud on Your Terms
Hybrid Cloud: The Cloud on Your Terms
 

Similar to SnapLogic Technology Open House – January 2018

R for SAS Users Complement or Replace Two Strategies
R for SAS Users Complement or Replace Two StrategiesR for SAS Users Complement or Replace Two Strategies
R for SAS Users Complement or Replace Two StrategiesRevolution Analytics
 
SAP HANA - The Foundation of Real Time, Now on the AWS Cloud Computing Platform
SAP HANA - The Foundation of Real Time, Now on the AWS Cloud Computing PlatformSAP HANA - The Foundation of Real Time, Now on the AWS Cloud Computing Platform
SAP HANA - The Foundation of Real Time, Now on the AWS Cloud Computing PlatformAmazon Web Services
 
Augmented OLAP for Big Data Analytics
Augmented OLAP for Big Data AnalyticsAugmented OLAP for Big Data Analytics
Augmented OLAP for Big Data AnalyticsTyler Wishnoff
 
Build and maintain large Ruby apps 0.0.1
Build and maintain large Ruby apps 0.0.1Build and maintain large Ruby apps 0.0.1
Build and maintain large Ruby apps 0.0.1Enrico Teotti
 
six-sigma-and-minitab-13 (1).ppt
six-sigma-and-minitab-13 (1).pptsix-sigma-and-minitab-13 (1).ppt
six-sigma-and-minitab-13 (1).pptPTD QUYCOCTU
 
How Gousto is moving to just-in-time personalization with Snowplow
How Gousto is moving to just-in-time personalization with SnowplowHow Gousto is moving to just-in-time personalization with Snowplow
How Gousto is moving to just-in-time personalization with SnowplowGiuseppe Gaviani
 
Initiate Edinburgh 2019 - Big Data Meets AI
Initiate Edinburgh 2019 - Big Data Meets AIInitiate Edinburgh 2019 - Big Data Meets AI
Initiate Edinburgh 2019 - Big Data Meets AIAmazon Web Services
 
What Managers Need to Know about Data Science
What Managers Need to Know about Data ScienceWhat Managers Need to Know about Data Science
What Managers Need to Know about Data ScienceAnnie Flippo
 
When Data is Your Product: Empowering Business Users
When Data is Your Product: Empowering Business UsersWhen Data is Your Product: Empowering Business Users
When Data is Your Product: Empowering Business UsersIntuit Inc.
 
Webinar on Big Data Challenges : Presented by Raj Kasturi
Webinar on Big Data Challenges : Presented by Raj KasturiWebinar on Big Data Challenges : Presented by Raj Kasturi
Webinar on Big Data Challenges : Presented by Raj KasturioGuild .
 
Turn Data Into Actionable Insights - StampedeCon 2016
Turn Data Into Actionable Insights - StampedeCon 2016Turn Data Into Actionable Insights - StampedeCon 2016
Turn Data Into Actionable Insights - StampedeCon 2016StampedeCon
 
Big Data Day LA 2015 - Applications of the Apriori Algorithm on Open Data by ...
Big Data Day LA 2015 - Applications of the Apriori Algorithm on Open Data by ...Big Data Day LA 2015 - Applications of the Apriori Algorithm on Open Data by ...
Big Data Day LA 2015 - Applications of the Apriori Algorithm on Open Data by ...Data Con LA
 
Empower customer success at LinkedIn with advanced analytics and great visual...
Empower customer success at LinkedIn with advanced analytics and great visual...Empower customer success at LinkedIn with advanced analytics and great visual...
Empower customer success at LinkedIn with advanced analytics and great visual...Michael Li
 
Hadoop 2015: what we larned -Think Big, A Teradata Company
Hadoop 2015: what we larned -Think Big, A Teradata CompanyHadoop 2015: what we larned -Think Big, A Teradata Company
Hadoop 2015: what we larned -Think Big, A Teradata CompanyDataWorks Summit
 
Strata+hadoop data kitchen-seven-steps-to-high-velocity-data-analytics-with d...
Strata+hadoop data kitchen-seven-steps-to-high-velocity-data-analytics-with d...Strata+hadoop data kitchen-seven-steps-to-high-velocity-data-analytics-with d...
Strata+hadoop data kitchen-seven-steps-to-high-velocity-data-analytics-with d...DataKitchen
 
Fri benghiat gil-odsc-data-kitchen-data science to dataops
Fri benghiat gil-odsc-data-kitchen-data science to dataopsFri benghiat gil-odsc-data-kitchen-data science to dataops
Fri benghiat gil-odsc-data-kitchen-data science to dataopsDataKitchen
 
ODSC data science to DataOps
ODSC data science to DataOpsODSC data science to DataOps
ODSC data science to DataOpsChristopher Bergh
 
20190219 Telling a story with your data and power bi
20190219 Telling a story with your data and power bi20190219 Telling a story with your data and power bi
20190219 Telling a story with your data and power biAroh Shukla
 
20190219 Telling a story with your data and power bi
20190219 Telling a story with your data and power bi20190219 Telling a story with your data and power bi
20190219 Telling a story with your data and power biAroh Shukla
 

Similar to SnapLogic Technology Open House – January 2018 (20)

R for SAS Users Complement or Replace Two Strategies
R for SAS Users Complement or Replace Two StrategiesR for SAS Users Complement or Replace Two Strategies
R for SAS Users Complement or Replace Two Strategies
 
SAP HANA - The Foundation of Real Time, Now on the AWS Cloud Computing Platform
SAP HANA - The Foundation of Real Time, Now on the AWS Cloud Computing PlatformSAP HANA - The Foundation of Real Time, Now on the AWS Cloud Computing Platform
SAP HANA - The Foundation of Real Time, Now on the AWS Cloud Computing Platform
 
Augmented OLAP for Big Data Analytics
Augmented OLAP for Big Data AnalyticsAugmented OLAP for Big Data Analytics
Augmented OLAP for Big Data Analytics
 
Build and maintain large Ruby apps 0.0.1
Build and maintain large Ruby apps 0.0.1Build and maintain large Ruby apps 0.0.1
Build and maintain large Ruby apps 0.0.1
 
six-sigma-and-minitab-13 (1).ppt
six-sigma-and-minitab-13 (1).pptsix-sigma-and-minitab-13 (1).ppt
six-sigma-and-minitab-13 (1).ppt
 
How Gousto is moving to just-in-time personalization with Snowplow
How Gousto is moving to just-in-time personalization with SnowplowHow Gousto is moving to just-in-time personalization with Snowplow
How Gousto is moving to just-in-time personalization with Snowplow
 
Initiate Edinburgh 2019 - Big Data Meets AI
Initiate Edinburgh 2019 - Big Data Meets AIInitiate Edinburgh 2019 - Big Data Meets AI
Initiate Edinburgh 2019 - Big Data Meets AI
 
What Managers Need to Know about Data Science
What Managers Need to Know about Data ScienceWhat Managers Need to Know about Data Science
What Managers Need to Know about Data Science
 
When Data is Your Product: Empowering Business Users
When Data is Your Product: Empowering Business UsersWhen Data is Your Product: Empowering Business Users
When Data is Your Product: Empowering Business Users
 
Webinar on Big Data Challenges : Presented by Raj Kasturi
Webinar on Big Data Challenges : Presented by Raj KasturiWebinar on Big Data Challenges : Presented by Raj Kasturi
Webinar on Big Data Challenges : Presented by Raj Kasturi
 
Turn Data Into Actionable Insights - StampedeCon 2016
Turn Data Into Actionable Insights - StampedeCon 2016Turn Data Into Actionable Insights - StampedeCon 2016
Turn Data Into Actionable Insights - StampedeCon 2016
 
Big Data Day LA 2015 - Applications of the Apriori Algorithm on Open Data by ...
Big Data Day LA 2015 - Applications of the Apriori Algorithm on Open Data by ...Big Data Day LA 2015 - Applications of the Apriori Algorithm on Open Data by ...
Big Data Day LA 2015 - Applications of the Apriori Algorithm on Open Data by ...
 
Empower customer success at LinkedIn with advanced analytics and great visual...
Empower customer success at LinkedIn with advanced analytics and great visual...Empower customer success at LinkedIn with advanced analytics and great visual...
Empower customer success at LinkedIn with advanced analytics and great visual...
 
Hadoop 2015: what we larned -Think Big, A Teradata Company
Hadoop 2015: what we larned -Think Big, A Teradata CompanyHadoop 2015: what we larned -Think Big, A Teradata Company
Hadoop 2015: what we larned -Think Big, A Teradata Company
 
Strata+hadoop data kitchen-seven-steps-to-high-velocity-data-analytics-with d...
Strata+hadoop data kitchen-seven-steps-to-high-velocity-data-analytics-with d...Strata+hadoop data kitchen-seven-steps-to-high-velocity-data-analytics-with d...
Strata+hadoop data kitchen-seven-steps-to-high-velocity-data-analytics-with d...
 
The Container Shame Spiral
The Container Shame SpiralThe Container Shame Spiral
The Container Shame Spiral
 
Fri benghiat gil-odsc-data-kitchen-data science to dataops
Fri benghiat gil-odsc-data-kitchen-data science to dataopsFri benghiat gil-odsc-data-kitchen-data science to dataops
Fri benghiat gil-odsc-data-kitchen-data science to dataops
 
ODSC data science to DataOps
ODSC data science to DataOpsODSC data science to DataOps
ODSC data science to DataOps
 
20190219 Telling a story with your data and power bi
20190219 Telling a story with your data and power bi20190219 Telling a story with your data and power bi
20190219 Telling a story with your data and power bi
 
20190219 Telling a story with your data and power bi
20190219 Telling a story with your data and power bi20190219 Telling a story with your data and power bi
20190219 Telling a story with your data and power bi
 

More from SnapLogic

The AI Mindset: Bridging Industry and Academic Perspectives
The AI Mindset: Bridging Industry and Academic PerspectivesThe AI Mindset: Bridging Industry and Academic Perspectives
The AI Mindset: Bridging Industry and Academic PerspectivesSnapLogic
 
SnapLogic Culture Deck
SnapLogic Culture DeckSnapLogic Culture Deck
SnapLogic Culture DeckSnapLogic
 
Digital Transformation is Cloud-Powered
Digital Transformation is Cloud-PoweredDigital Transformation is Cloud-Powered
Digital Transformation is Cloud-PoweredSnapLogic
 
How to Build a Winning Data Culture
How to Build a Winning Data CultureHow to Build a Winning Data Culture
How to Build a Winning Data CultureSnapLogic
 
Overcoming the challenge of multiple data frameworks in a multiple cloud envi...
Overcoming the challenge of multiple data frameworks in a multiple cloud envi...Overcoming the challenge of multiple data frameworks in a multiple cloud envi...
Overcoming the challenge of multiple data frameworks in a multiple cloud envi...SnapLogic
 
Self-Service Integration in the Age of Digital Transformation at Box
Self-Service Integration in the Age of Digital Transformation at BoxSelf-Service Integration in the Age of Digital Transformation at Box
Self-Service Integration in the Age of Digital Transformation at BoxSnapLogic
 
Live Demo: Accelerate the integration of workday applications
Live Demo: Accelerate the integration of workday applicationsLive Demo: Accelerate the integration of workday applications
Live Demo: Accelerate the integration of workday applicationsSnapLogic
 
The new dominant companies are running on data
The new dominant companies are running on data The new dominant companies are running on data
The new dominant companies are running on data SnapLogic
 
Spring 2017 release customer webinar
Spring 2017 release customer webinarSpring 2017 release customer webinar
Spring 2017 release customer webinarSnapLogic
 
SnapLogic unveils machine-learning-driven integration assistant
SnapLogic unveils machine-learning-driven integration assistantSnapLogic unveils machine-learning-driven integration assistant
SnapLogic unveils machine-learning-driven integration assistantSnapLogic
 
Webinar: Evolution of Data Management for the IoT
Webinar: Evolution of Data Management for the IoTWebinar: Evolution of Data Management for the IoT
Webinar: Evolution of Data Management for the IoTSnapLogic
 
SnapLogic Culture
SnapLogic CultureSnapLogic Culture
SnapLogic CultureSnapLogic
 
Big Data Management: What's New, What's Different, and What You Need To Know
Big Data Management: What's New, What's Different, and What You Need To KnowBig Data Management: What's New, What's Different, and What You Need To Know
Big Data Management: What's New, What's Different, and What You Need To KnowSnapLogic
 
SnapLogic Live: Workday Integration
SnapLogic Live: Workday IntegrationSnapLogic Live: Workday Integration
SnapLogic Live: Workday IntegrationSnapLogic
 
SnapLogic Live: Big Data Integration
SnapLogic Live: Big Data IntegrationSnapLogic Live: Big Data Integration
SnapLogic Live: Big Data IntegrationSnapLogic
 
SnapLogic Live: IoT Integration
SnapLogic Live: IoT IntegrationSnapLogic Live: IoT Integration
SnapLogic Live: IoT IntegrationSnapLogic
 
SnapLogic Live: Powering Cloud Analytics
SnapLogic Live: Powering Cloud AnalyticsSnapLogic Live: Powering Cloud Analytics
SnapLogic Live: Powering Cloud AnalyticsSnapLogic
 
SnapLogic Live: ServiceNow Integration
SnapLogic Live: ServiceNow IntegrationSnapLogic Live: ServiceNow Integration
SnapLogic Live: ServiceNow IntegrationSnapLogic
 
SnapLogic Live: Salesforce Integration
SnapLogic Live: Salesforce IntegrationSnapLogic Live: Salesforce Integration
SnapLogic Live: Salesforce IntegrationSnapLogic
 

More from SnapLogic (20)

The AI Mindset: Bridging Industry and Academic Perspectives
The AI Mindset: Bridging Industry and Academic PerspectivesThe AI Mindset: Bridging Industry and Academic Perspectives
The AI Mindset: Bridging Industry and Academic Perspectives
 
SnapLogic Culture Deck
SnapLogic Culture DeckSnapLogic Culture Deck
SnapLogic Culture Deck
 
Digital Transformation is Cloud-Powered
Digital Transformation is Cloud-PoweredDigital Transformation is Cloud-Powered
Digital Transformation is Cloud-Powered
 
How to Build a Winning Data Culture
How to Build a Winning Data CultureHow to Build a Winning Data Culture
How to Build a Winning Data Culture
 
Overcoming the challenge of multiple data frameworks in a multiple cloud envi...
Overcoming the challenge of multiple data frameworks in a multiple cloud envi...Overcoming the challenge of multiple data frameworks in a multiple cloud envi...
Overcoming the challenge of multiple data frameworks in a multiple cloud envi...
 
Self-Service Integration in the Age of Digital Transformation at Box
Self-Service Integration in the Age of Digital Transformation at BoxSelf-Service Integration in the Age of Digital Transformation at Box
Self-Service Integration in the Age of Digital Transformation at Box
 
Live Demo: Accelerate the integration of workday applications
Live Demo: Accelerate the integration of workday applicationsLive Demo: Accelerate the integration of workday applications
Live Demo: Accelerate the integration of workday applications
 
The new dominant companies are running on data
The new dominant companies are running on data The new dominant companies are running on data
The new dominant companies are running on data
 
Spring 2017 release customer webinar
Spring 2017 release customer webinarSpring 2017 release customer webinar
Spring 2017 release customer webinar
 
SnapLogic unveils machine-learning-driven integration assistant
SnapLogic unveils machine-learning-driven integration assistantSnapLogic unveils machine-learning-driven integration assistant
SnapLogic unveils machine-learning-driven integration assistant
 
Webinar: Evolution of Data Management for the IoT
Webinar: Evolution of Data Management for the IoTWebinar: Evolution of Data Management for the IoT
Webinar: Evolution of Data Management for the IoT
 
The API Lie
The API LieThe API Lie
The API Lie
 
SnapLogic Culture
SnapLogic CultureSnapLogic Culture
SnapLogic Culture
 
Big Data Management: What's New, What's Different, and What You Need To Know
Big Data Management: What's New, What's Different, and What You Need To KnowBig Data Management: What's New, What's Different, and What You Need To Know
Big Data Management: What's New, What's Different, and What You Need To Know
 
SnapLogic Live: Workday Integration
SnapLogic Live: Workday IntegrationSnapLogic Live: Workday Integration
SnapLogic Live: Workday Integration
 
SnapLogic Live: Big Data Integration
SnapLogic Live: Big Data IntegrationSnapLogic Live: Big Data Integration
SnapLogic Live: Big Data Integration
 
SnapLogic Live: IoT Integration
SnapLogic Live: IoT IntegrationSnapLogic Live: IoT Integration
SnapLogic Live: IoT Integration
 
SnapLogic Live: Powering Cloud Analytics
SnapLogic Live: Powering Cloud AnalyticsSnapLogic Live: Powering Cloud Analytics
SnapLogic Live: Powering Cloud Analytics
 
SnapLogic Live: ServiceNow Integration
SnapLogic Live: ServiceNow IntegrationSnapLogic Live: ServiceNow Integration
SnapLogic Live: ServiceNow Integration
 
SnapLogic Live: Salesforce Integration
SnapLogic Live: Salesforce IntegrationSnapLogic Live: Salesforce Integration
SnapLogic Live: Salesforce Integration
 

Recently uploaded

Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostZilliz
 
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
 
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
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubKalema Edgar
 
From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .Alan Dix
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfAlex Barbosa Coqueiro
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024Lorenzo Miniero
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
Connect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck PresentationConnect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck PresentationSlibray Presentation
 
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Mark Simos
 
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
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...Fwdays
 
Search Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfSearch Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfRankYa
 
How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.Curtis Poe
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Commit University
 
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo DayH2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo DaySri Ambati
 
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
 
"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
 
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
 

Recently uploaded (20)

Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
 
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
 
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
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding Club
 
From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .From Family Reminiscence to Scholarly Archive .
From Family Reminiscence to Scholarly Archive .
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdf
 
SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024SIP trunking in Janus @ Kamailio World 2024
SIP trunking in Janus @ Kamailio World 2024
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
Connect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck PresentationConnect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck Presentation
 
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
Tampa BSides - Chef's Tour of Microsoft Security Adoption Framework (SAF)
 
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
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
 
Search Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfSearch Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdf
 
How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.How AI, OpenAI, and ChatGPT impact business and software.
How AI, OpenAI, and ChatGPT impact business and software.
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!
 
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptxE-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
 
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo DayH2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
 
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
 
"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
 
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
 

SnapLogic Technology Open House – January 2018

  • 1. Cooking with AI/ML at SnapLogic SnapLogic Technology Open House | January 11, 2018 The Story of Iris
  • 2. Education • Bachelor of Engineering in Computer Engineering @ Chulalongkorn University • Master of Science in Computer Science @ University of San Francisco Experience • Machine Learning Engineer @ SnapLogic Accomplishment • Published Research in Machine Learning • Fulbright Scholar • Cook Good Foods • Be Happy 2 Jump Thanawut Ananpiriyakul About Me
  • 3. Agenda 3 Cooking with AI/ML at SnapLogic Get to Know SnapLogic • SnapLogic: Enterprise Integration Cloud Platform • Iris Artificial Intelligence: Machine-Learning-Driven Integration Assistant The Story of Iris • Iris's Restaurant • Survey the Market • Check our Inventory • Develop the Recipe The Future of Iris Jump's Dream
  • 4. corporate overview Get to Know SnapLogic
  • 5. 5 Get to Know SnapLogic Integration Platform as a Service (iPaaS) • Data Integration Platform • Connect Applications and Data • Self-Service, Fast, Easy • Reliable, Scalable • Accelerate Business
  • 6. Our Product 6 Get to Know SnapLogic
  • 7. 7 Get to Know SnapLogic Live Demo
  • 8. Iris Artificial Intelligence 8 Get to Know SnapLogic What is Iris? • Artificial Intelligence • Friendly Assistant What is good about Iris? • Accelerate New Joiners • Improve Efficiency of Experts • Bring Happiness to Everyone
  • 10. Iris's Restaurant 10 The Story of Iris What kind of dishes should we serve here?
  • 11. Survey the Market Check what people do. ❏ Research Papers ❏ Academic Proven ❏ Informative ❏ Developer Communities ❏ Full of Arguments ❏ Responsive ❏ Market Leaders ❏ Innovative ❏ Competitors 11 The Story of Iris What's good and how? ❏ Surgery Robot ❏ High Precision ❏ No Emotion ❏ No Fatigue ❏ Self Driving Car ❏ Less Accident ❏ Face ID ❏ More Secure Innovate new things. ❏ Exclusively for SnapLogic's Customers ❏ Useful & Reliable Estimate the cost. ❏ People ❏ Time ❏ Money
  • 12. Check our Inventory 12 The Story of Iris Check what ingredients we have. ❏ Raw Data ❏ Log Data ❏ Traffic What are we going to cook? ❏ Recommendation Engine ❏ Pipeline Generator ❏ Pattern / Event Detector ❏ Automatic Optimizer ❏ Personal Assistant Chatbot ❏ Global Search Engine
  • 13. Develop the Recipe 13 The Story of Iris Plating Make the Final Touch Turn off the heat. Add Thai chilies, kaffir lime leaves and lime juice. Make a quick stir. Pour the soup into the bowl. Serve with steamed Thai jasmine rice. Cooking Listen to the Voice of Ingredients Boil the water, add all herbs except Thai chilies and kaffir lime leaves. Wait until the onion is soft. Add mushrooms. Wait until boiled. Add Thai chili paste and shrimps. Wait until boiled. Season with fish sauce and sugar. Preparation Clean them All Wash lemongrass, galangal, Thai onion, Thai chilies and kaffir lime leaves. Let them dry, beat them with knife. Wash oyster mushrooms and straw mushrooms. Clean the shrimps, peel their shells, keep their heads. Serving Be Proud Serve while hot. Reheat if needed. Be careful when serving hot foods. Say "Please Enjoy!" with a big smile. Tom Yum Kung (ต้มยำกุ้ง) Thai River Prawn Spicy Soup
  • 14. Develop the Recipe 14 The Story of Iris Preparation
  • 15. Develop the Recipe 15 The Story of Iris Preparation Previous Snap 3 Previous Snap 2 Previous Snap 1 Current Snap Next Snap File Reader CSV Parser File Reader CSV Parser Mapper File Reader CSV Parser Mapper Copy File Reader CSV Parser Mapper Copy Mapper File Reader CSV Parser Mapper Copy Mapper CSV Parser Mapper Copy Mapper CSV Formatter CSV Parser Mapper Copy Mapper CSV Formatter
  • 16. Develop the Recipe 16 The Story of Iris Preparation Dataset Extractor and Updater
  • 17. Develop the Recipe 17 The Story of Iris Cooking https://www.bangkokpost.com/lifestyle/social-and-lifestyle/1381587/a-reluctant-star Iris Development Process 3 RESULT ANALYSIS Analyze the result. Reverse engineer from result to data. Plan for improvement. 1 ALGORITHM DEVELOPMENT Choose the right algorithm. Choose the right configurations. Choose the right strategies. 2 EVALUATION Evaluate the algorithms using metrics: performance, accuracy, and cost.
  • 18. Develop the Recipe 18 The Story of Iris Cooking Most Popular Snap Decision Tree
  • 19. Plating Develop the Recipe 19 The Story of Iris Front-End UIBack-End Service Logging ServiceModel Builder
  • 21. Develop the Recipe 21 The Story of Iris Plating Make the Final Touch Design and implement API on backend side. Design and implement user interface. Run load test to estimate the size of servers. Final test to make sure it is delicious. Cooking Listen to the Voice of Ingredients Try different cooking methods: statistical models, k-nearest neighbors, decision tree, random forests, etc. Evaluate with daily backtesting technique. Make sure the recommendations make sense. Preparation Clean them All Extract successfully executed pipelines. Transform into directed acyclic graphs. Chop them into segments of 5 nodes or less. Preprocess the data. Serving Be Proud Released in 4.9 (Spring, 2017). Source Snap Recommendation in 4.11 (Fall 2017). Observe usages and issues. Continue research and development.
  • 23. Neural Networks 23 The Future of Iris http://playground.tensorflow.org
  • 24. Iris X Neural Networks 24 The Future of Iris Decision Tree ❏ It's interpretable. ❏ It takes minutes to train. ❏ It performs well on smaller dataset. ❏ It's solid and specific. ❏ It's for prototyping and early release. ❏ It's cheaper. Neural Networks ❏ It's magic. ❏ It takes hours or days to train. ❏ It's more powerful on large dataset. ❏ It's flexible and optimizable. ❏ It's for advancement and future release. ❏ It's a lot more expensive. Neural Networks and Deep Learning are definitely a present. They are powerful and super flexible. We are proud to announce that Iris has adopted those technologies.
  • 25. The Truth 25 The Future of Iris Large Neural Networks (90%*) Expected to reach 90% hit rate with additional layers and wider layers. 1 Improved Decision Tree (81%) Prune the tree to get the best result.3 Statistical Model (49%) Always recommend top 5 popular snaps.5 Small Neural Networks (87%*) Look back upto last 9 snaps. Also learn from user information to make recommendations personalized. 2 Decision Tree (78%) Consider last 4 snaps on the canvas as an input set. Walk down the decision tree and get recommendations from the leaf. 4
  • 26. Latest Prototype 26 The Future of Iris input_layer = Input(shape=(9,)) input_org_layer = Input(shape=(1,)) emb_layer = Embedding(num_snap_type, 128, input_length=seg_len-1)(input_layer) emb_org_layer = Embedding(num_org, 64, input_length=1)(input_org_layer) flatten_org = Flatten()(emb_org_layer) lstm_layer = LSTM(1024)(emb_layer) drop_layer = Dropout(0.2)(lstm_layer) merge_layer = Concatenate(axis=1)([drop_layer, flatten_org]) output_layer = Dense(num_snap_type, activation='softmax')(merge_layer) model = Model(inputs=[input_layer, input_org_layer], outputs=output_layer)
  • 28. Design Concept: The Easier Future of AI 28 Jump's Dream
  • 29. Design Concept: The Easier Future of AI 29 Jump's Dream
  • 30. Q & A Please Enjoy!Thank You for Coming