Training Data Correlation
by
Osama Sammodi
—
last modified
Apr 27, 2012 10:36
—
filed under:
KnowledgeModel
Definitions
Term: Training Data Correlation |
Domain: Cross-cutting issues | ||||
---|---|---|---|---|---|
Engineering and Design (KM-ED) |
Adaptation and Monitoring (KM-AM) |
Quality Definition, Negotiation and
Assurance (KM-QA) |
Generic (domain independent) |
||
D o m a i n : L a y e r s |
Business Process Management (KM-BPM) |
||||
Service Composition and
Coordination (KM-SC) |
|||||
Service Infrastructure (KM-SI) |
|||||
Generic (domain independent) |
Training Data Correlation is a standard machine
learning approach to evaluating regression models. It is defined as the
statistical correlation between all training instance outcomes and the
predictions that the model would deliver for these training instances.
[Leitner et al. 2009] |
References
- [Leitner et al. 2009] Leitner, P., Wetzstein, B., Rosenberg, F., Michlmayr, A., Dustdar, S., & Leymann, F. (2009). Runtime Prediction of Service Level Agreement Violations for Composite Services. In 3rd Workshop on Non-Functional Properties and SLA Management in Service-Oriented Computing, co- located with ICSOC 2009.