Publication details
Clustering with Various Distance Measures in Adaptive Resonance Theory
| Authors | |
|---|---|
| Year of publication | 2008 |
| Type | Article in Proceedings |
| Conference | 5. Letní škola aplikované informatiky |
| MU Faculty or unit | |
| Citation | |
| Field | Informatics |
| Keywords | Adaptive Resonance Theory; clustering; Machine Learning. |
| Description | The adaptive resonance theory (ART) was developed by psychologists some 20 years ago. Later on, it was adapted by machine learning for the supervised as well as unsupervised type of learning. In this article, we briefly explain ART-based unsupervised algorithms and the implementation of ART-based algorithms. We compare and discuss the performance of ART and some other well-known algorithms on various datasets. The basic features of the ART-based algorithms are explained, such as speed or how to build parameters' net to gain the global maxima (the best solution) - or some very good local maxima - automatically without the necessity of guessing the best input parameters' setting. |
| Related projects: |