Combining the Classification Results of Independent Classifiers Based on the Dempster/Shafer Theory of Evidence

Abstract. In this paper we investigate combination of neural net based classifiers using DempsterShafer Theory of Evidence. Under some assumptions, combination rule resembles a product of errors rule observed in human speech perception. Different combination are tested in ASR experiments both in matched and mismatched conditions and compared with more conventional probability combination rules. Proposed techniques are particularly effective in mismatched conditions.

show abstract

“…see [9]). Main weakness of DS theory is the fact that results are strongly sensitive on the choice of the Basic Probability Function.…”

Section: Product Of Errorsmentioning

confidence: 96%

Combination of Acoustic Classifiers Based on Dempster-Shafer Theory of Evidence

Valente

Heřmanský

2007

2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07

View full text Add to dashboard Cite

show abstract

“…The theory of multi-classifier systems can be traced back at least as far as 1965 (Nilsson, 1965). Previous multiple classifiers combination algorithm include the voting (Lam & Suen, 1994), Bayes rule (Xu et al, 1992), Dempster-Shafer theory (Mandler & Schuermann, 1988), decision tree and other methods. Based on classifier outputs, the multiple classifiers combination methods can be classified into three different levels: abstract level, ranked list of classes, and measurements (Suen & Lam, 2000, Xu et al, 1992.…”

Section: Multiple Classifiers Combinationmentioning

confidence: 99%

Hyperspectral Remote Sensing Data Mining Using Multiple Classifiers Combination

Wen¹,

Yang²,

Hu³

2009

Data Mining and Knowledge Discovery in Real Life Applications

View full text Add to dashboard Cite

“…One of the first works in this area is dated back to 1988 when a method was proposed to transform distance measures of the different base-level classifiers into evidence (Mandler and Schurmann 1988). Once the distances between learning data points and a number of reference points have been calculated, they were used for evaluation of basic probability assignment values, and later the Dempster's combination rule (Eq.…”

Section: Ensemble Systems and Dempster's Combination Rulementioning

confidence: 99%

Building ensemble classifiers using belief functions and OWA operators

Reformat

Yager

2007

Soft Comput

View full text Add to dashboard Cite

A pervasive task in many forms of human activity is classification. Recent interest in the classification process has focused on ensemble classifier systems. These types of systems are based on a paradigm of combining the outputs of a number of individual classifiers. In this paper we propose a new approach for obtaining the final output of ensemble classifiers. The method presented here uses the Dempster-Shafer concept of belief functions to represent the confidence in the outputs of the individual classifiers. The combing of the outputs of the individual classifiers is based on an aggregation process which can be seen as a fusion of the Dempster rule of combination with a generalized form of OWA operator. The use of the OWA operator provides an added degree of flexibility in expressing the way the aggregation of the individual classifiers is performed.

show abstract

Combining the Classification Results of Independent Classifiers Based on the Dempster/Shafer Theory of Evidence

Cited by 80 publications

References 1 publication

Combination of Acoustic Classifiers Based on Dempster-Shafer Theory of Evidence

Combination of Acoustic Classifiers Based on Dempster-Shafer Theory of Evidence

Hyperspectral Remote Sensing Data Mining Using Multiple Classifiers Combination

Building ensemble classifiers using belief functions and OWA operators

Contact Info

Product

Resources

About