In this presentation, former BizTalk Premier Field Engineer and BizTastic founder Ricardo Torre shows how to apply machine learning to hybrid integrations using BizTalk, Azure and Logic Apps.
3. from Wikipedia
• Machine learning is a field of computer science that
gives computers the ability to learn without being explicitly
programmed.
4. Agenda
• State of Azure Machine Learning
• How can we use ML today in your integration scenarios
• Applying ML to BizTalk operational data
• Applying ML to Logic Apps operational data
• Use high level ML (cognitive services)
5. Azure Machine Learning Studio
Azure Machine Learning Studio is a GUI-based integrated development
environment for constructing and operationalizing Machine
Learning workflow on Azure.
• Cloud predictive analytics service
• Ready to use algorithms
• Quick deploy
• Easy drag and drop development
• Extend with custom R and Python scripts
9. Azure Machine Learning Workbench
• Visual Studio Code and the AI Extension with Azure Machine Learning
Work Bench
10. How to: ML in your integration scenarios
• Actions that can be helped but using ML in integration
• Operational
• Performance
• Anomaly Identification
• Preemptive operational actions
• Business related actions
• How are my orders doing?
• Any abnormal behavior up/down stream affecting
• Development/Architecture
• Business continuity
11. Applying ML to BizTalk operational data
• Using BizTalk Tracking data
• Collect and prepare data for Azure ML Experiment
Solution
• Manually export tracking data
• Application Insights/Event Hubs and Stream Insights
• Prepare data
• Build and deploy model
12. Anomalies in BizTalk tracking data
• Unexpected long processing times
• Unhandled Exceptions
• Wrong message flow
• Changes in volume of messages
• Changes message type distribution
• Upstream/Downstream inconsistencies
14. Applying ML to Logic Apps operational data
• Logic Apps generate a similar tracking information
• Data needs preparing
• Similar solution can be developed
• Automate data collection and ingestion
15. Use high level ML (cognitive services)
• Intelligent algorithms:
• See
• Hear
• Speak
• understand and interpret natural communication
• Search
Transform your business with AI today
AI, ML, VR, IoT Platforms and SDS- Impact in 2 to 5 years
Talk points:
Tech hype cycle
Machine learning is a data science technique that allows computers to use existing data to forecast future behaviors, outcomes, and trends.
Azure Machine Learning Studio is a GUI-based integrated development environment for constructing and operationalizing Machine Learning workflow on Azure.
Tell a story from my days working with large BizTalk customers where problems happened but were hard to identify until we looked through tracking data to find the culprit
Show Azure ML experiment where:
Collect significant amount of tracking data
Upload to Azure ML Studio
Create experiment that prepares data
Build Model
Deploy Model – show practical example of anomalies being identified and normal data
Retrain Model
Use AI to solve business problems
Vision
Image-processing algorithms to smartly identify, caption and moderate your pictures.
Speech
Convert spoken audio into text, use voice for verification, or add speaker recognition to your app.
Knowledge
Map complex information and data in order to solve tasks such as intelligent recommendations and semantic search.
Search
Add Bing Search APIs to your apps and harness the ability to comb billions of webpages, images, videos, and news with a single API call.
Language
Allow your apps to process natural language with pre-built scripts, evaluate sentiment and learn how to recognize what users want.
Microsoft AI Platform
Uber example
Fun demo using Cognitive Services using the Logic Apps