# PROF. WAGACHA PETER WAIGANJO

## BSc.(Hons)Physics (UoN),MSc.Computer Science & Applications(UoN), PhD. Computer Science(UoN)

Tel: +254 20 4447870, Email: waiganjo@uonbi.ac.ke

Tel: +254 20 4447870, Email: waiganjo@uonbi.ac.ke

- Citation:
- Wagacha PW. "Machine Learning Notes on: I. Classifier Learning and Generalization, II. Data Preparation, III. Validation Methods." Institute of Computer Science, University of Nairobi. 2002.

The goal of classifier model training is not to learn an exact representation of the training

data itself, but rather to build a statistical model of the process which generates the data.

This is important if the classifier is to exhibit good generalization, that is, to make good

predictions for new inputs. Using the simple analogy of curve fitting using polynomials, we

should that a polynomial with too few coefficients gives poor predictions for new data, ie

poor generalization, since the polynomial function has too little flexibility.