![Close [x]](http://dw1.s81c.com/i/c.gif){kind=small}
Introducing this tutorial
This tutorial shows you how to apply data mining techniques using IBM DB2 Intelligent Miner to generate automatically product recommendations for customers in a possible e-commerce shop environment. Moreover, we can characterize the customers in terms of the products they purchase and the frequency of their shoppings. This customer profiling will be key to determine targets for an outbound cross-sell campaign. We will be able to ensure that the recommendations will be appropriate to each customer because other customers with a similar behaviour bought what we are now recommending.
To fulfill these objectives, this article shows you how to use IBM DB2 Intelligent Miner Modeling to create a mining model and then IBM DB2 Intelligent Miner Visualization to evaluate it and to display its results. Furthermore, this tutorial gives you some examples of how you could integrate all with the existing business applications, as the integration of the data mining results with channels and reporting is key for a successful marketing strategy.
You will learn how to do this using the Easy Mining Procedures for DB2 Intelligent Miner, an easy-to-use SQL interface for the main steps of the data-mining process. In particular, they contain stored procedures for the basic mining functions such as classification, clustering, or discovery of association rules and procedures for general, higher-level, common mining tasks such as the discovery of a prediction model or the detection of outliers in a table.
This tutorial takes you through the entire process of extracting product recommendations and customer's profiling step by step, using real data taken from the retail business. By following the examples, you can modify various parameters to observe the resulting outcome. By completing the exercises, you can check what you have learned.
When you have completed this tutorial, you can:
- Understand the concepts of data mining:
- What is data mining?
- What can you achieve with data mining?
- How does the data mining process look like?
- Apply the techniques that you have learned to your own corporate problems and put each of the theoretical steps of the Data Mining process into practice.
- Generate business explanations and deploy them.
NOTE: In the future we may refer to the IBM DB2 Intelligent Miner tools as IM Visualization/IM modeling/ IM Scoring, where IM means Intelligent Miner.
Who should read this tutorial?
This tutorial is addressed to Information Technology (IT) professionals. To complete this tutorial, you should be familiar with databases. You need not be familiar with data mining concepts. If you do not have domain knowledge of the business area, you should work together with people that can provide this knowledge.
Knowing the business area is mandatory. With data mining, you can solve business problems by analyzing the corporate data. However, you will not come up with a proper answer or explanation if there is not a concise definition and understanding of the business problem.
Nowadays internet shops offer product recommendations every time the customer adds a product into the cart. In an early future, there will be little computer devices in each cart of a physical supermarket. Each time a customer puts a new product in the cart, its code will be scanned and a display will show some product recommendations in real time, taking into account the kind of customer that is purchasing and the item he/she has put in the cart.
Right now companies can apply product recommendation to improve their cross-selling campaigns, for example. The cross-selling strategy is based in pushing new products to current customers based on their past purchases. Cross-selling is designed to widen the customer's reliance on the company and decrease the likelihood of the customer switching to a competitor. Indeed, companies are worried in losing their clients as it is very expensive to attract new ones with promotion campaigns, publicity, etc.
This is our scenario:
"A retail company wants to have a better knowledge of their clients and their behaviour
so that it can offer a better product recommendation in its next cross selling campaign."
To complete the steps in this tutorial, the following software must be installed on your computer:
- IBM DB2 Version 8.2 or higher (Download the trial version from http://www.ibm.com/software/data/db2/udb)
- IBM DB2 Data Warehouse Edition Version 9.1 with Intelligent Miner Modeling, Scoring and Visualization
As operating systems, this tutorial supports Windows and Linux.
To complete this tutorial, you need about 3 to 3.5 hours in average.
However, you might want to read this tutorial more than once when you want to translate the data mining techniques into your own business environment.
Retrieving the required files
The data from the retail company is stored in the retail-case folder or in this zip file.
We assume that you have installed and configured DB2 Intelligent Miner Modeling, Scoring and Visualization according to the installation instructions.
Creating and configuring the database and importing the data
Before you can build or call a model, you must create and configure the retail database, and import the data.
Windows:
- To display a DB2 command window, click Start -> Programs -> IBM DB2 -> Command Line Tools -> Command Window.
- Go to the directory where you have saved the retail-case folder, for example,
:e\retail_tutorial\retail-case\, and type the following command on the command line for creating and populating the retail database:
[tutorial path]:\> setup
Linux:
- Log on as the owner of the DB2 instance (e.g., db2inst1) and execute, if necessary, db2start.
- Go to the directory where you have saved the retail-case folder, for example,
/retail_tutorial/retail-case/, and type the following command on the command line for creating and populating the retail database:
/[tutorial path]> sh setup.sh
The following terms are trademarks of the IBM Corporation in the United States, other countries, or both:
- DB2
- DB2 Universal Database
- Websphere
- IBM
- Intelligent Miner
- Redbooks
Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both.
Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both.
Other company, product, and service names may be trademarks or service marks of others.
Table of contents
Next steps from IBM
Tags
Use the slider bar to see more or fewer tags.
Popular tags shows the top tags for this particular content zone (for example, Java technology, Linux, WebSphere).
My tags shows your tags for this particular content zone (for example, Java technology, Linux, WebSphere).
Popular tutorial tags |
My tutorial tagsSkip to tags list
Popular tutorial tags |
My tutorial tags
