Abstract
The explosive growth of commercial and scientific databases has outpaced our ability to manually analyze and interpret this data. The newly emerging interdisciplinary field of knowledge discovery in databases (KDD) provides methodologies for seeking valuable and useful information from these databases. In this paper we describe a methodology for identifying high fault modules in software metrics databases. It employs radial basis function model for the data mining phase of the KDD process based on our newly developed algorithm. We use the well known bootstrap method for model validation and accuracy estimation of the classification task. As an example, a genuine problem from NASA software database is explored.
| Original language | English |
|---|---|
| Pages (from-to) | 226-233 |
| Number of pages | 8 |
| Journal | Proceedings of SPIE - The International Society for Optical Engineering |
| Volume | 3695 |
| State | Published - 1999 |
| Event | Proceedings of the 1999 Data Mining and Knowledge Discovery: Theory, Tools, and Technology - Orlando, FL, USA Duration: 5 Apr 1999 → 6 Apr 1999 |