A hybrid supervised ANN for classification and data visualization
Supervised ANNs such as Learning Vector Quantization (LVQs) and Multi-Layer Perceptrons (MLPs) usually do not support data visualization beside classification. Unsupervised visualization focused ANNs such as Self-organizing Maps (SOM) and its variants such as Visualization induced SOM (ViSOM) on the...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2008
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/16550/1/Chee%20Siong%20Teh.pdf http://ir.unimas.my/id/eprint/16550/ http://ieeexplore.ieee.org/document/4633848/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Universiti Malaysia Sarawak |
| Language: | English |
| Summary: | Supervised ANNs such as Learning Vector Quantization (LVQs) and Multi-Layer Perceptrons (MLPs) usually do not support data visualization beside classification. Unsupervised visualization focused ANNs such as Self-organizing Maps (SOM) and its variants such as Visualization induced SOM (ViSOM) on the other hand, usually do not optimize data classification as compared with supervised ANNs such as LVQ. Thus to provide supervised classification and data visualization simultaneously, this work is motivated to propose a novel hybrid supervised ANN of LVQwithAC by hybridizing LVQ and modified Adaptive Coordinate (AC) approach. Empirical studies on benchmark data sets proven that, LVQwithAC was able to provide superior classification accuracy than SOM and ViSOM. Beside LVQwithAC was able to provide data topology, data structure, and inter-neuron distance preserve visualization. LVQwithAC was also proven able to perform promising classification among other supervised classifiers besides its additional data visualization ability over them. Thus, for applications requiring data visualization and classification LVQwithAC demonstrated its potential if supervised learning is all possible. |
|---|