Title :
A model for mixed category perception based on absolute feature values
Author :
Basak, Jayanta ; Pal, Sankar K.
Author_Institution :
Machine Intelligence Unit, Indian Stat. Inst., Calcutta, India
Abstract :
Recently, X-tron was developed by Basak et al. (1993, 1995) by considering the task of mixed category perception as a set covering problem where a hypothesis is formed about the presence of a set of objects which would be able to interpret the presence of input features. Here we present a new version of X-tron which is able to accept the absolute values of the features and interpret them even in a mixed form. The range of absolute information of each feature is viewed here as consisting of an unknown number of quantized slots. The degree of presence of a feature corresponding to a category is determined with a membership function. An initial guess is made about the size of the slots. Then the network automatically learns the number of slots and the membership function during the categorisation/self-organisation process
Keywords :
feature extraction; learning (artificial intelligence); neural nets; object recognition; X-tron; absolute feature values; absolute information; categorisation; learning rules; membership function; mixed category perception; object recognition; probability; quantized slots; self-organisation process; Electronic mail; Layout; Logic; Machine intelligence; Monitoring; Predictive models; Transfer functions; Virtual manufacturing;
Conference_Titel :
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location :
Perth, WA
Print_ISBN :
0-7803-2768-3
DOI :
10.1109/ICNN.1995.488966
