Presented By O'Reilly and Cloudera
Make Data Work
December 1–3, 2015 • Singapore

Deploying models with Azure Machine Learning

Danielle Dean (iRobot), Wee Hyong Tok (Microsoft)
1:30pm–5:00pm Tuesday, 12/01/2015
Data Science and Advanced Analytics
Location: 334 Level: Intermediate
Average rating: ****.
(4.57, 7 ratings)

Prerequisite Knowledge

The target audience for the tutorial is a range of individuals with an intermediate to advanced background in data science.


In this tutorial, you will create end-to-end predictive models based on an extensive library of machine learning algorithms included in Microsoft Azure Machine Learning studio with its R and Python language extensibility. You will then deploy and consume the model and use it for making predictions over business data. You will be walking through the typical steps of performing machine learning, which can be summarized as: data ingestion, cleaning and data exploration, feature engineering, model selection, and evaluation of results.

As a part of the tutorial, you will be creating two different models based on a fictitious dataset. This dataset consists of records belonging to the customers of a telecom service provider. The columns of the dataset hold information such as the length of customer account, total day, night, evening, and international minutes used.

The first model you will create is called churn analysis, or customer attrition, which is the problem of identifying the customers who are likely to leave a service or a business. The goal of the analysis is to contact these high-risk individuals and take necessary actions such as providing special offers and discounts to prevent them from leaving the business. You will model the problem using the binary classification technique. You will then create a web service for the model and visualize the classification results.

Additionally, a second model you can choose to create is a segmentation model where the objective is to find natural clusters of customers within the data sets who have similar characteristics. This is also extremely beneficial to understand the customer base for targeted marketing applications, where the goal is to target the right individuals in order to grow the business.

Photo of Danielle Dean

Danielle Dean


Danielle Dean is the technical director of machine learning at iRobot. Previously, she was a principal data science lead at Microsoft. She holds a PhD in quantitative psychology from the University of North Carolina at Chapel Hill.

Photo of Wee Hyong Tok

Wee Hyong Tok


Wee Hyong Tok is a principal data science manager with the AI CTO Office at Microsoft, where he leads the engineering and data science team for the AI for Earth program. Wee Hyong has worn many hats in his career, including developer, program and product manager, data scientist, researcher, and strategist, and his track record of leading successful engineering and data science teams has given him unique superpowers to be a trusted AI advisor to customers. Wee Hyong coauthored several books on artificial intelligence, including Predictive Analytics Using Azure Machine Learning and Doing Data Science with SQL Server. Wee Hyong holds a PhD in computer science from the National University of Singapore.

Comments on this page are now closed.


Picture of Danielle Dean
Danielle Dean
03/11/2016 3:17pm +08

See material here:

Rajesh Haran
02/24/2016 5:59pm +08

do you have the slides posted?