Combining SVM classifiers for handwritten digit recognition
Date Issued
2002
Author(s)
Gjorgjevikj, Dejan
Chakmakov, Dushan
Abstract
In this paper, we investigate the advantages and weaknesses of various decision fusion schemes using statistical
and rule-based reasoning. The cooperation schemes are
applied on two SVM (Support Vector Machine) classifiers
performing classification task on two feature families
referenced as structural and statistical features. The obtained results show that it is difficult to exceed the recognition rate of a single classifier applied straightforwardly
on both feature families as one set. The rule based cooperation schemes enable an easy and efficient implementation of various rejection criteria. On the other hand, the
statistical cooperation schemes provide higher recognition rates and offer possibility for fine-tuning of the recognition versus the reliability tradeoff.
and rule-based reasoning. The cooperation schemes are
applied on two SVM (Support Vector Machine) classifiers
performing classification task on two feature families
referenced as structural and statistical features. The obtained results show that it is difficult to exceed the recognition rate of a single classifier applied straightforwardly
on both feature families as one set. The rule based cooperation schemes enable an easy and efficient implementation of various rejection criteria. On the other hand, the
statistical cooperation schemes provide higher recognition rates and offer possibility for fine-tuning of the recognition versus the reliability tradeoff.
File(s)![Thumbnail Image]()
Loading...
Name
download.pdf
Size
384.6 KB
Format
Adobe PDF
Checksum
(MD5):28ea68ab90b389a0913b44e9ea509a02