Fuzzy Classification Algorithm for Polymer Characterization
Author Information
Author(s): D. J. Ramsbottom, M. J. Adams, J. Carroll
Primary Institution: School of Applied Sciences, University of Wolverhampton
Hypothesis
Can a fuzzy c-means clustering algorithm effectively classify polymer samples based on their infra-red spectra?
Conclusion
The fuzzy c-means algorithm can classify polymer samples effectively, providing meaningful clustering without the limitations of traditional binary classification.
Supporting Evidence
- The fuzzy c-means algorithm allows for classification of polymer samples that may consist of multiple components.
- Principal component analysis showed that the first three components accounted for over 91% of the variance in the spectral data.
- The method can classify unknown samples by comparing them to known data.
- Fuzzy clustering allows for membership in multiple clusters, unlike traditional methods.
Takeaway
This study shows that a special computer program can help sort different types of plastic by looking at their unique light patterns.
Methodology
The study used a fuzzy c-means clustering algorithm to analyze infra-red spectra of polymer samples.
Limitations
The study's results may be subjective as interpreting clusters can vary based on user selection.
Want to read the original?
Access the complete publication on the publisher's website