1,721,094 research outputs found
Analysis of error-reject trade-off in linearly combined multiple classifiers
No full textIn this paper, a theoretical and experimental analysis of the error-reject trade-off achievable by linearly combining the outputs of an ensemble of classifiers is presented. To this aim, the theoretical framework previously developed by Tumer and Ghosh for the analysis of the simple average rule without the reject option has been extended. Analytical results that allow to evaluate the improvement of the error-reject trade-off achievable by simple averaging their outputs under different assumptions about the distributions of the estimation errors affecting a posteriori probabilities, are provided. The conditions under which the weighted average can provide a better error-reject trade-off than the simple average are then determined. From the theoretical results obtained under the assumption of unbiased and uncorrelated estimation errors, simple guidelines for the design of multiple classifier systems using linear combiners are given. Finally, an experimental evaluation and comparison of the error-reject trade-off of the simple and weighted averages is reported for five real data sets. The results show the practical relevance of the proposed guidelines in the design of linear combiners
Learning Relational Concepts at Different Levels of Granularity
No full textIn this paper, an alternative approach to the induction of relational concepts is presented.
The underlying framework relies on the concept of exception, an exception being a
counterexample left within the scope of a description devoted to classifying examples of the
given target concept. While trying to characterize the target concept, first an initial description
is searched for. Such a solution must be complete, although not necessarily consistent. This
means that some counterexamples are allowed to be misclassified. As counterexamples (i.e.,
exceptions) must be taken into account in order to properly classify them, the corresponding
learning process is performed in several steps, each step devoted to coping with exceptions
generated during the previous one. Eventually, the process comes to an end, usually leading to a
description that uses a kind of Vere’s counterfactuals to refine, at different levels of granularity,
the underlying concept
Comparison and Combination of Adaptive Query Shifting and Feature Relevance Learning for Content-Based Image Retrieval
No full textBrowse Conference Publications > Image Analysis and Processing ...
Comparison and combination of adaptive query shifting and feature relevance learning for content-based image retrieval
This paper appears in:
Image Analysis and Processing, 2001. Proceedings. 11th International Conference on
Date of Conference: 26-28 Sep 2001
Author(s): Giacinto, G.
Dept. of Electr. & Electron. Eng., Cagliari Univ.
Roli, F. ; Fumera, G.
Page(s): 422 - 427
Product Type: Conference Publications
ABSTRACT
Despite the efforts to reduce the semantic gap between user perception of similarity and feature-based representation of images, user interaction is essential to improve retrieval performance in content-based image retrieval. To this end a number of relevance feedback mechanisms are currently adopted to refine image queries. They are aimed either to locally modify the feature space or to shift the query point towards more promising regions of the feature space. A novel adaptive query shifting mechanism is proposed to improve retrieval performance beyond that provided by other relevance feedback mechanisms. In addition we discuss the extent to which query shifting may provide better performance than feature weighting and provide experimental results on the complementarity of the two approaches. Finally, some combinational approaches are proposed to exploit such complementarities
Design of effective multiple classifier systems by clustering of classifiers
No full textIn the field of pattern recognition, multiple classifier systems based on the combination of outputs of a set of different classifiers have been proposed as a method for the development of high performance classification systems. Previous work clearly showed that multiple classifier. Systems are effective only if the classifiers forming them make independent errors. Therefore, the fundamental need for methods aimed to design “error-independent” classifiers is currently acknowledged. In the paper, an approach to the automatic design of multiple classifier systems is proposed. Given an initial large set of classifiers, our approach is aimed at selecting the subset formed by the most error-independent classifiers. Reported results on the classification of multisensor remote-sensing images show that this approach allows to design effective multiple classifier system
Fast person re-identification based on dissimilarity representations
No full textPerson re-identification is a recently introduced computer vision task that consists of recognising an individual who was previously observed over a video-surveillance camera network. Among the open problems, in this paper we focus on computational complexity. Despite its practical relevance, especially in real-time applications, this issue has been overlooked in the literature so far. In this paper, we address it by exploiting a framework we proposed in a previous work. It allows us to turn any person re-identification method, that uses multiple components and a body part subdivision model, into a dissimilarity-based one. Each individual is represented as a vector of dissimilarity values to a set of visual prototypes, that are drawn from the original non-dissimilarity representation. Experiments on two benchmark datasets provide evidence that a dissimilarity representation provides very fast re-identification methods. We also show that, even if the re-identification accuracy can be lower (especially when the number of candidates is low), the trade-off between processing time and accuracy can nevertheless be advantageous, in real-time application scenarios involving a human operator
F-measure optimisation in multi-label classifiers
No full textWhen a multi-label classifier outputs a real-valued score for each class, a well known design strategy consists of tuning the corresponding decision thresholds by optimising the performance measure of interest on validation data. In this paper we focus on the F-measure, which is widely used in multi-label problems. We derive two properties of the micro-averaged F measure, viewed as a function of the threshold values, which allow its global maximum to be found by an optimisation strategy with an upper bound on computational complexity of O(n2N2), where N and n are respectively the number of classes and of validation samples. So far, only a suboptimal threshold selection rule and a greedy algorithm without any optimality guarantee were known for this task. We then devise a possible optimisation algorithm based on our strategy, and evaluate it on three benchmark, multi-label data sets
Multi-label classification with a reject option
No full textWe consider multi-label classification problems in application scenarios where classifier accuracy is not satisfactory, but manual annotation is too costly. In single-label problems, a well known solution consists of using a reject option, i.e., allowing a classifier to withhold unreliable decisions, leaving them (and only them) to human operators. We argue that this solution can be exploited also in multi-label problems. However, the current theoretical framework for classification with a reject option applies only to single-label problems. We thus develop a specific framework for multi-label ones. In particular, we extend multi-label accuracy measures to take into account rejections, and define manual annotation cost as a cost function. We then formalise the goal of attaining a desired trade-off between classifier accuracy on non-rejected decisions, and the cost of manually handling rejected decisions, as a constrained optimisation problem. We finally develop two possible implementations of our framework, tailored to the widely used F accuracy measure, and to the only cost models proposed so far for multi- label annotation tasks, and experimentally evaluate them on five application domains
Pattern recognition systems under attack: design issues and research challenges
No full textWe analyze the problem of designing pattern recognition systems in adversarial settings, under an engineering viewpoint, motivated by their increasing exploitation in security-sensitive applications like spam and malware detection, despite their vulnerability to potential attacks has not yet been deeply understood. We ̄rst review previous work and report examples of how a complex system may be evaded either by leveraging on trivial vulnerabilities of its untrained components, e.g. parsing errors in the pre-processing steps, or by exploiting more subtle vul- nerabilities of learning algorithms. We then discuss the need of exploiting both reactive and proactive security paradigms complementarily to improve the security by design. Our ultimate goal is to provide some useful guidelines for improving the security of pattern recognition in adversarial settings, and to suggest related open issues to foster research in this area
F-measure optimisation in multi-label classifiers
No full textWhen a multi-label classifier outputs a real-valued score for each class, a well known design strategy consists of tuning the corresponding decision thresholds by optimising the performance measure of interest on validation data. In this paper we focus on the F-measure, which is widely used in multi-label problems. We derive two properties of the micro-averaged F measure, viewed as a function of the threshold values, which allow its global maximum to be found by an optimisation strategy with an upper bound on computational complexity of O(n2N2), where N and n are respectively the number of classes and of validation samples. So far, only a suboptimal threshold selection rule and a greedy algorithm without any optimality guarantee were known for this task. We then devise a possible optimisation algorithm based on our strategy, and evaluate it on three benchmark, multi-label data sets
- …
