This is to understand the possible solutions for autonomic computing algorithm in Maths. At the moment , according to my own experiences , binomial heap really great in this approach compared to others. Bayesian Network , linear regression , N-Gram requires extra parameters in solving false and right alarms. Any suggestion how the algorithm can really inline with K-MAPE will be really grateful. Thanks in advanced.
I think Reinforcement learning (Q-learning control) is a cognitive algorithm which is inline with IMB MAPE idea.
Not sure if this helps but I think that Utility functions and Policy Autonomics are great tools for implementing IBM's MAPE-based control.
This paper (http://inderscience.metapress.com/content/n0155x8t4w0173l6/fulltext.pdf) by Richard Anthony was helpful for me.
- Badges
- Science topic