Page 63 -

2.6 Thc Dimensionality Ratio Prohlem 49

Figure 2.26. Percentage of normally distributed samples lying within one standard
deviation neighboiirhood (black bars) and two standard deviations neighbourhood
(grey bars) for several values of d.

Bibliography

Andrews HC (1972) Introduction to Mathematical Techniques in Pattern Recognition. John
Wiley Sr Sons. Inc.
Bishop CM (1995) Neural Networks for Pattern Recognition. Clarcndon Press, Oxford.
Bronson R (1991) Matrix Methods. An Introduction. Academic Prcss, Inc.
Duda R, Hart P (1973) Pattern Classification and Scenc Analysis. Wilry, New Yosk.
Friedrnan M, Kandel A (1999) Introduction to Pattern Recognition. Statistical, Structural,
Neural and Fuzzy Logic Approaches. World Scientific Co. Pte. Ltd., Singapore.
Fukunaga K (1990) Introduction to Statistical Pattern Recognition. Academic Press.
Hoel PG (1975) Elementary Statistics. John Wiley & Sons.
Looney CG (1997) Pattern Recognition Using Neural Networks. 'Theory arid Algorithms Ihr
Engineers and Scientists. Orford University Press.
Pipes LA. Hovanessiari SA (1996) Matrix-Computer Methods in Engineering. Kohert
Krieger Pub. Co.. Florida.
Schalkoff RJ (1992) Pattern Recognition. Statistical, Structural and Neural Approaches.
John Wiley & Sons, Inc.
Siege1 S, Castellan Jr NJ (1998) Nonpara~netric Statistics for the Behavioral Sciences.
McGraw Hill Book Co., Sidney.
Spath H (1980) Cluster Arialysis Algorithms. Ellis Horwood, Ltd., England.
Milton JS, Mc'reer PM, Corhct JJ (17000) lntrociuctiori to Statistics. Mctiraw Hill College
Div.

Exercises

2.1 Consider the two-dimcns~orial Glubirlur data included in the Cluster.xls file. With the
help of a scatter plot determine thc linear decision fuiiction that separates the two