Making Sense of Data: A Practical Guide to Exploratory Data Analysis and Data Mining

8.4 DEPLOYMENT SCENARIOS

Exploratory data analysis and data mining has been deployed to a variety of problems. The following illustrates some of the areas where this technology has been deployed:

• Personalized e-commerce: Customers characteristics, based on profiles and historical purchasing information, can be used to personalize e-commerce web sites. Customers can be directed to products and services matching their anticipated needs.
• Churn analysis: Profiles of customers discontinuing a particular product or service can be analyzed and prediction models generated for customers who are likely to switch. These models can be used to identify at risk customers providing an opportunity to target them with a focused marketing campaign in order to retain their business.
• Quality control: Quality is critical to all production systems and exploratory data analysis and data mining approaches are important tools in creating and maintaining a high quality production system. For example, the 6-sigma quality control methodology uses many of the statistical methods described in Chapter 5.
• Experimental design and analysis: Experiments are widely used in all areas of research and development to design, test and assess new products. Exploratory data analysis and data mining are key tools in both the design of these experiments and the analysis of the results. For example, every day biologists are experimentally generating millions of data points concerning genes and it is critical to make use of ...