This document provides an overview of using Amazon Machine Learning to perform targeted marketing with machine learning. It describes downloading banking data from a public dataset, creating a machine learning model to predict customer subscriptions, evaluating the model, generating predictions for new customer data, and cleaning up resources. The process involves creating a datasource in Amazon ML, training a default and custom model, evaluating model performance, generating batch predictions, and deleting input and output data from S3.

1. Targeted Marketing with Amazon Machine Learning
Barbara Pogorzelska,
Technical Program Manager

2. Agenda
• Problem description
• Downloading, editing and uploading the data
• Datasource creation
• ML model creation
• Model evaluation
• Batch prediction
• Clean up

3. Problem Description

4. Machine learning & the use case
Machine learning is the technology that automatically finds
patterns in your data and uses them to make predictions for
new data points as they become available
Problem
How to identify potential customers for targeted marketing
campaigns?
Data available
Publicly available banking and marketing dataset from the
University of California at Irvine (UCI) repository

5. Amazon Machine Learning tutorial

6. Amazon Machine Learning tutorial

7. Data (see http://archive.ics.uci.edu/ml/datasets/Bank+Marketing)
Training data
41188 data points
20 attributes
binary output
Batch predictions
4119 data points
Bank client data
1 - age
2 - job (admin., blue-collar, entrepreneur, …)
3 - marital (single, divorces, married, …)
4 - education (basic.4y, basic.6y, university.degree, …)
5 - default: has credit in default?
6 - housing: has housing loan?
7 - loan: has personal loan?
Related with the last contact of the current campaign
8 - contact: communication type: (cellular, telephone)
9 - month: last contact month of year
10 - day_of_week: last contact day of the week
11 - duration: last contact duration, in seconds
Other attributes
12 - campaign: number of contacts performed during this
campaign and for this client
13 - pdays: number of days that passed by after the client
was last contacted from a previous campaign
14 - previous: number of contacts performed before this
campaign and for this client
15 - poutcome: outcome of the previous marketing
campaign
Social and economic context attributes
16 - emp.var.rate: employment variation rate
17 - cons.price.idx: consumer price index
18 - cons.conf.idx: consumer confidence index
19 - euribor3m: euribor 3 month rate - daily indicator
20 - nr.employed: number of employees
Output variable (desired target)
21 - y - has the client subscribed a term deposit?

8. Three steps to create a prediction

9. Downloading, editing and uploading the data

10. Storing the data on S3
• Download from https://s3.amazonaws.com/aml-sample-data/banking.csv and
https://s3.amazonaws.com/aml-sample-data/banking-batch.csv
– Replaced yes/no with 1/0
• Store data on S3

11. Datasource creation

12. Datasource creation

13. Datasource creation

14. Datasource creation

15. Datasource creation

16. Datasource creation

17. Three steps to create a prediction

18. ML model creation

19. Model creation – default model

20. Model creation – default model

21. Model creation – default model

22. Model creation – default model

23. Model evaluation

24. Model evaluation – default model

25. Model evaluation – default model

26. Model evaluation – default model

27. Data Insights

28. Data Insights

29. Data Insights

30. Data Insights

31. Data Insights

32. Data Insights

33. Model creation – custom model

34. Model creation – custom model

35. Model creation – custom model

36. Model creation – custom model

37. Three steps to create a prediction

38. Batch predictions

39. Batch predictions - default model

40. Batch predictions - default model

41. Batch predictions - default model

42. Batch predictions - default model

43. Batch predictions - default model

44. Batch predictions - default model

45. Batch predictions - default model

46. Batch predictions - default model

47. Batch predictions - custom model

48. Batch predictions - custom model

49. Batch predictions - custom model

50. Batch predictions - custom model

51. Clean up

52. Clean up your account
To delete the input data used for training, evaluation, and batch prediction steps
1. Open the Amazon S3 console.
2. Navigate to the S3 bucket where you stored the banking.csv and banking-batch.csv.
3. Select the two files and the .writePermissionCheck.tmp file.
4. Choose Actions, Delete.
5. When prompted for confirmation, choose OK.
To delete the predictions generated from the batch prediction step
1. Open the Amazon S3 console.
2. Navigate to the bucket where you stored the output of the batch predictions.
3. Select the batch-prediction folder.
4. Choose Actions, Delete.
5. When prompted for confirmation, click OK.

53. Try out machine-learning-samples from github

54. https://aws.amazon.com/de/machine-learning/
Get Started on AWS with
Amazon Machine Learning