Journal ArticleDOI
Methods of combining multiple classifiers and their applications to handwriting recognition
Lei Xu,Adam Krzyżak,Ching Y. Suen +2 more
- Vol. 22, Iss: 3, pp 418-435
Reads0
Chats0
TLDR
On applying these methods to combine several classifiers for recognizing totally unconstrained handwritten numerals, the experimental results show that the performance of individual classifiers can be improved significantly.Abstract:
Possible solutions to the problem of combining classifiers can be divided into three categories according to the levels of information available from the various classifiers. Four approaches based on different methodologies are proposed for solving this problem. One is suitable for combining individual classifiers such as Bayesian, k-nearest-neighbor, and various distance classifiers. The other three could be used for combining any kind of individual classifiers. On applying these methods to combine several classifiers for recognizing totally unconstrained handwritten numerals, the experimental results show that the performance of individual classifiers can be improved significantly. For example, on the US zipcode database, 98.9% recognition with 0.90% substitution and 0.2% rejection can be obtained, as well as high reliability with 95% recognition, 0% substitution, and 5% rejection. >read more
Citations
More filters
Journal ArticleDOI
Generating classifier outputs of fixed accuracy and diversity
TL;DR: An algorithm for random generation of classifier outputs with specified individual accuracies and pairwise dependencies for a hypothetical data set and the generated team output can be used to study the majority vote over multiple dependent classifiers.
Book ChapterDOI
Feature Subsets for Classifier Combination: An Enumerative Experiment
TL;DR: The potential for improvement in classifier teams designed by the feature subspace method, where the set of features is partitioned and each subset is used by one classifier in the team, is studied.
Proceedings ArticleDOI
Fusion of multiple experts in multimodal biometric personal identity verification systems
Josef Kittler,Kieron Messer +1 more
TL;DR: Two trainable methods of classifier fusion in the context of multimodal personal identity verification involving eight experts which exploit voice characteristics and frontal face biometrics are investigated.
Journal ArticleDOI
Combining heterogeneous classifiers for stock selection
George T. Albanis,Roy Batchelor +1 more
TL;DR: It is found that simple “Majority Voting” improves accuracy and profitability only marginally, and much greater gains come from applying the “Unanimity Principle”, whereby a share is not held in the high-performing portfolio unless all classifiers agree.
Proceedings ArticleDOI
Comparative study of information fusion methods for sonar images classification
TL;DR: A comparative study of information fusion methods for sonar images classification using the weighted vote approach, or coming from the possibility theory and evidence theory, have been employed.
References
More filters
Book
Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference
TL;DR: Probabilistic Reasoning in Intelligent Systems as mentioned in this paper is a complete and accessible account of the theoretical foundations and computational methods that underlie plausible reasoning under uncertainty, and provides a coherent explication of probability as a language for reasoning with partial belief.
Book
A mathematical theory of evidence
TL;DR: This book develops an alternative to the additive set functions and the rule of conditioning of the Bayesian theory: set functions that need only be what Choquet called "monotone of order of infinity." and Dempster's rule for combining such set functions.
Journal ArticleDOI
Statistical and structural approaches to texture
TL;DR: This survey reviews the image processing literature on the various approaches and models investigators have used for texture, including statistical approaches of autocorrelation function, optical transforms, digital transforms, textural edgeness, structural element, gray tone cooccurrence, run lengths, and autoregressive models.
Journal ArticleDOI
An introduction to hidden Markov models
TL;DR: The purpose of this tutorial paper is to give an introduction to the theory of Markov models, and to illustrate how they have been applied to problems in speech recognition.