Enhanced Self-Organizing Map with Particle Swarm Optimization for Classification
Author Information
Author(s): S. Hasan, S. M. Shamsuddin
Primary Institution: Universiti Teknologi Malaysia
Hypothesis
Can the hybridization of Self-Organizing Map (SOM) and Particle Swarm Optimization (PSO) improve classification accuracy?
Conclusion
The proposed ESOMPSO method significantly improves classification accuracy and reduces quantization errors compared to traditional methods.
Supporting Evidence
- ESOMPSO achieved a classification accuracy of 95.22% on the dataset.
- The quantization error for ESOMPSO was significantly lower than that of traditional methods.
- The study validated the results using the Kruskal-Wallis test, showing significant differences among methods.
Takeaway
This study shows that combining two smart techniques, ESOM and PSO, helps computers better understand and classify data.
Methodology
The study used a hybrid approach combining Self-Organizing Maps and Particle Swarm Optimization to enhance classification performance on various datasets.
Limitations
The study's performance may vary with different datasets and the choice of parameters.
Statistical Information
P-Value
0.004
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website