Generalization, discrimination, and multiple categorization using adaptive resonance theory
نویسندگان
چکیده
منابع مشابه
Generalization, discrimination, and multiple categorization using adaptive resonance theory
The internal competition between categories in the adaptive resonance theory (ART) neural model can be biased by replacing the original choice function by one that contains an attentional tuning parameter under external control. For the same input but different values of the attentional tuning parameter, the network can learn and recall different categories with different degrees of generality,...
متن کاملOdor discrimination using adaptive resonance theory
Ž . The paper presents two neural networks based on the adaptive resonance theory ART for the recognition of several odors subjected Ž . to drift. The neural networks developed by Grossberg supervised and unsupervised have been used for two different drift behaviors. One in which the clusters end up to overlap each other and the other when they do not. The latter case is solved by unsupervision...
متن کاملCustomer Relationship Management using Adaptive Resonance Theory
CRM is a kind of implemented model for managing a company’s interactions with their customers. CRM involves the customer classification to understand the behavior of the customer. There is a vital role of the data mining techniques for the classification. This paper presents the concept of one of the data mining technique ART for the customer classification for CRM.
متن کاملLithofacies identi®cation using multiple adaptive resonance theory neural networks and group decision expert system
Lithofacies identi®cation supplies qualitative information about rocks. Lithofacies represent rock textures and are important components of hydrocarbon reservoir description. Traditional techniques of lithofacies identi®cation from core data are costly and dierent geologists may provide dierent interpretations. In this paper, we present a lowcost intelligent system consisting of three adaptiv...
متن کاملAdaptive Resonance Theory Microchips
Abstract. Recently, a real-time clustering microchip based on the ART1 algorithm has been reported. That chip was able to classify 100-bit input patterns into up to 18 categories. However, its high area comsumption ( lcm 2) caused a very poor yield (6%). In this paper, an improved prototype is presented. In this chip, a different approach has been used to implement the most area consuming eleme...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Transactions on Neural Networks
سال: 1999
ISSN: 1045-9227
DOI: 10.1109/72.774213