As databases grow larger, analysts are turning to computers to help them analyze the massive amounts of data their computers have collected. As the difference between having data and having useful information becomes more clear, different methods of using computers to analyze data are becoming available. Knowledge Discovery in Databases (KDD) is a general methodology for preparing the data, using software algorithms to discover new patterns or relationships in the data, and integrating the results back into the system. The KDD methodology is explained and hypothetically applied to usage statistics generated by the CSB/SJU Libraries Internet resources. Examples are drawn from that source and from other industries to clearly illustrate the properties of Knowledge Discovery and decide if KDD is an appropriate methodology for the Libraries to use in this situation.
Available by permission of the author. Reproduction or retransmission of this material in any form is prohibited without expressed written permission of the author.
Lindquist, Peter J., "Data Mining in Electronic Media Usage Statistics: A Case Study of Knowledge Discovery in Databases" (1998). Honors Theses, 1963-2015. 660.