Title :
On weighted expectations
Author :
Kandel, Abraham ; Friedman, Menahem
Author_Institution :
Dept. of Comput. Sci. & Eng., Univ. of South Florida, Tampa, FL, USA
Abstract :
The classical `expected value´ frequently fails to represent a `typical value´ of a given multi-cluster data set. In this case the set is clustered and new quantities which replace the expected value and the standard deviation are defined. The `weighted expected value´ (WEV) is computed by solving a nonlinear equation and defines a `most typical value´ of the set. The `weighted deviation´ (WD) is then defined explicitly and provides a measurement to the `grade of typicality´ of the WEV
Keywords :
statistical analysis; expected value; multi-cluster data set; nonlinear equation; standard deviation; weighted expectations; Arithmetic; Computer science; Information retrieval; Measurement standards; Nonlinear equations; Physics; Statistics; Terminology;
Conference_Titel :
Fuzzy Information Processing Society Biannual Conference, 1994. Industrial Fuzzy Control and Intelligent Systems Conference, and the NASA Joint Technology Workshop on Neural Networks and Fuzzy Logic,
Conference_Location :
San Antonio, TX
Print_ISBN :
0-7803-2125-1
DOI :
10.1109/IJCF.1994.375158
