DocumentCode :
2434598
Title :
Application of improved Fisher Linear Discriminant Analysis approaches
Author :
Zhang, Lishi ; Wang, Dehong ; Gao, Shengzhe
Author_Institution :
Sch. of Sci., Dalian Ocean Univ., Dalian, China
fYear :
2011
fDate :
8-11 Jan. 2011
Firstpage :
1311
Lastpage :
1314
Abstract :
The objective of discriminant analysis is to use information from the independent variables to achieve the clearest possible separation or discrimination between or among groups, the main task of the present work is to modify the objective functions and make the separation as large as possible. Fisher Linear Discriminant Analysis approach is modified to meet the requirement that maximizes the objective function, some computer simulations show that the modified ones can achieve the same results as the previous one.
Keywords :
pattern recognition; statistical analysis; Fisher linear discriminant analysis; computer simulations; objective function; Accuracy; Classification algorithms; Computer simulation; Iris; Linear discriminant analysis; Oceans; Pattern recognition; fuzzy c-means; iris data; linear discriminant analysis; wine classification data;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Management Science and Industrial Engineering (MSIE), 2011 International Conference on
Conference_Location :
Harbin
Print_ISBN :
978-1-4244-8383-9
Type :
conf
DOI :
10.1109/MSIE.2011.5707664
Filename :
5707664
Link To Document :
بازگشت