Predictive modelling with azure ml
1 like423 views
This document provides an overview of predictive modelling with Azure Machine Learning. It discusses trends in internet of things and big data that are driving growth in machine learning. It introduces machine learning concepts and how Azure ML can be used to build predictive models with strengths like a visual interface and support for collaborative work. The document outlines the Azure ML workflow from exploring data in the studio to operationalizing models with API services.
Predictive modelling with azure ml
- 1. Predictive Modelling with Azure ML Koray Kocabas
- 3. #sqlsatistanbul Agenda 3 Trends about Future What’s IoT Introduction to Big Data Welcome to Data World! Machine Learning Overview AzureML Strengths Microsoft Azure Overview Setting Up An AzureML Workspace Exploring AzureML Studio Azure ML API Services
- 4. #sqlsatistanbul Target Audience Data Professional Developer Business Intelligence Trend Hunter Programming Probability / Statistics Calculus Database Minimal Prerequsities
- 5. #sqlsatistanbul Internet Of Things Big Data Predictive Analysis 3 Trends about Future
- 6. #sqlsatistanbul "Internet Of Thing" connects the physical world to the Internet Within 6 years, 60.000 new "things" will be added to the Internet every 1 second!!! 212 billion devices will be part of the Internet of Things by 2020 (IDC) IDC predicts that IoT will generate nearly $9 Trillion in annual sales by 2020
- 7. #sqlsatistanbul IoT starts to be more visible in our everyday lives. A Car produced in 2014 has an average of 60 - 100 built- in sensors The Internet of Things isn't coming soon. It's already here... IoT will soon be everywhere. Like that: • Home (Consumer) • Fashion (Style) • Transport (Traffic) • Retail (Inventory) • Logistic (Real Time) • Manufacturing (M2M) • Cities (Industry) • Health (Body) • Environment (Prevention) • Building (Infrastructure) • Agriculture (Control) • Security (Detection)
- 8. #sqlsatistanbul IoT data is meaningful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
- 9. #sqlsatistanbul What is Big Data? Data that is too large or complex for analysis in traditional relational databases Typified by the “3 V’s”: Volume – Huge amounts of data to process Variety – A mixture of structured and unstructured data Velocity – New data generated extremely frequently Web server log reporting Social media sentiment analysis Sensor anomaly detection
- 10. #sqlsatistanbul Big Data Technologies Hadoop Open source distributed data processing cluster Data processed in Hadoop Distributed File System (HDFS) Related projects Hive Pig HCatalog Oozie Sqoop Others HDFS Name Node Data Nodes Hadoop Cluster
- 11. #sqlsatistanbul Map Reduce 1. Source data is divided among data nodes 2. Map phase generates key/value pairs 3. Reduce phase aggregates values for each key Lorem ipsum sit amet magma sit elit Fusce magna sed sit amet magna Key Value Lorem 1 ipsum 1 sit 1 amet 1 magma 1 sit 1 elit 1 Key Value Fusce 1 magma 1 sed 1 sit 1 amet 1 magma 1 Key Value Lorem 1 ipsum 1 sit 3 amet 2 magma 3 elit 1 Fusce 1 sed 1 MAPREDUCE
- 12. #sqlsatistanbul
- 14. #sqlsatistanbul 1980 («Neuro») 1990 («Symbolic») 2000 («Kernel Machines») 2005 («Graphical Models») 2011 («Big Data, DNN») Neural Networks Artificial Intelligence Learning = Adaptation of Neurons based on External Stimuli Expert Systems Decision Tree Learning Learning = Methods to automaticaly build Expert Systems Statistical Learning Theory Scoring Systems Learning = Optimization of Convex Functions Wide application in products Statistical Modeling of Data Learning = Parameter Estimation or Inference Distributed computing and storage Deep Neural Networks Learning = Scalable, Adaptive, Computation for Various Big Data History
- 15. #sqlsatistanbul Azure Machine Learning allows us to Solve extremely hard problems better Extract more value from Big Data Drive a shift in Data Analytics The goal of machine learning is to program computers to use example data or past experience to solve a given problem - Introduction to Machine Learning, 2nd Edition, MIT Press
- 16. #sqlsatistanbul
- 17. #sqlsatistanbul
- 18. #sqlsatistanbul • Accessible through a web browser, no software to install • Collaborative, work with anyone, anywhere via Azure workspace • Visual composition with end2end support for data science workflow • Extensible, support for R OSS
- 20. #sqlsatistanbul ML Studio Data Scientist Azure Portal & ML API Service Azure Ops Team ML API Services Developer
- 21. #sqlsatistanbul Machine Learning & Predictive Analytics are core capabilities that are needed throughout your business
- 22. #sqlsatistanbul Buraya Seatte görseli gelecek
- 23. #sqlsatistanbul