K-groups: Tractable Group Detection on Large Link Data Sets
- Topics:
- Electrical and Electronic
- Tags:
- Carnegie-Mellon University,
- Engineering,
- Human Resources,
- K-groups,
- Performance Management,
- Workforce Management
- Source:
- Carnegie Mellon University
FREE Registration is required
Overview: This white paper document presents k-groups - an algorithm that uses an approach similar to that of k-means (hard clustering and localized updates) to significantly accelerate the discovery of the underlying groups while retaining GDA's probabilistic model. In addition, the paper shows that k-groups is guaranteed to converge to a local minimum and also compares the performance of GDA and k-groups on several real world and artificial data sets, showing that k-groups' sacrifice in solution quality is significantly offset by its increase in speed.
(Is this item miscategorized? Does it need more tags? Let us know.)
Format: PDF | Size: 126KB | Date: Sep 2003 | Pages: 22
