Developing Customer Vulnerability Models Using Data Mining Techniques

	  Evangelos Simoudis    George John    Randy Kerber
		     Brian Livezey   Peter Miller
			 Lockheed AI Center
			 3251 Hanover Street
			 Palo Alto, CA 94304
     {simoudis, gjohn, kerber, livezey, pmiller}@aic.lockheed.com

  Data mining is the process of finding previously unknown and
  potentially interesting patterns and relationships in large databases.
  Consumer vulnerability analysis refers to the process of mining
  various types of consumer data to extract models, called vulnerability
  models, that predict consumer loyalty level to a particular product,
  e.g., orange juice, or class of products, e.g., frozen fruit juices.
  In this paper we use Recon*, a data mining system, to develop
  vulnerability models for frozen orange juice from a database
  containing the supermarket purchases of 15,000 households over a
  three-year period.

Recon is a trademark and service mark of Lockheed-Martin Corporation.

Citation:

International Symposium on Intelligent Data Analysis (IDA-95), August
1995, Baden-Baden, Germany.