A statistical unified framework for rank-based multiple classifier decision combination

This study presents a theoretical investigation of the rank-based multiple classifier decision combination problem, with the aim of providing a unified framework to understand a variety of such systems. The combination of the decisions of more than one classifiers with the aim of improving overall system performance is a concept of general interest in pattern recognition, as a viable alternative to designing a single sophisticated classifier. The problem of combining the classifier decisions in the raw form of candidate class rankings is formulated as a discrete optimization problem. The objective function to be maximized is selected as the overall probability of correct decision. This formulation introduces a set of observation statistics about the joint behavior of the classifiers which are to be estimated by observing the classifiers operated on a cross-validation test set. The resulting binary programming problem is shown to have a simple and global optimum solution but which also necessitates a prohibitive number of observation statistics. From the objective Function expansion, the problem observation space is defined and a method based on partitioning is introduced to reduce its prohibitive dimensionality. Within this partitioning formalism called as the Partitioned Observation Space (POS) theory, the number of behavior observation statistics can be reduced to levels which are feasible to estimate From the available cross-validation test data. It is shown by examples that such specific partitionings can be defined when reasonable assumptions or prior knowledge about the classifiers are incorporated into the problem domain. It is also demonstrated that certain specific partitionings of the classifier observation space effectively lead to the highest rank, Borda count and logistic regression rank-based decision combination methods from the literature. The analysis presented is general and promises to lead to a class of algorithms for rank-based decision combination. The potential of the theory and practical issues in implementation are illustrated by applying it in a real-life phonetic discrimination problem from speech pattern classification with encouraging results. (C) 2001 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.


A unified view of rank-based decision combination
Saranlı, Afşar (2000-09-07)
This study presents a theoretical investigation of the rank-based multiple classifier decision problem for closed-set pattern classification. The case with classifier raw outputs in the form of candidate class rankings is considered and formulated as a discrete optimization problem with the objective function being the total probability of correct decision. The problem has a global optimum solution but is of prohibitive dimensionality. We present a partitioning formalism under which this dimensionality can ...
An information theoretic framework for weight estimation in the combination of probabilistic classifiers for speaker identification
Altincay, H; Demirekler, Mübeccel (2000-04-01)
In this paper, we describe a relation between classification systems and information transmission systems. By looking at the classification systems from this perspective, we propose a method of classifier weight estimation for the linear (LIN-OP) and logarithmic opinion pool (LOG-OP) type classifier combination schemes for which some tools from information theory are used. These weights provide contextual information about the classifiers such as class dependent classifier reliability and global classifier ...
A Counting-Based Heuristic for ILP-Based Concept Discovery Systems
Mutlu, Alev; Karagöz, Pınar; Kavurucu, Yusuf (2013-09-13)
Concept discovery systems are concerned with learning definitions of a specific relation in terms of other relations provided as background knowledge. Although such systems have a history of more than 20 years and successful applications in various domains, they are still vulnerable to scalability and efficiency issues - mainly due to large search spaces they build. In this study we propose a heuristic to select a target instance that will lead to smaller search space without sacrificing the accuracy. The p...
An Approach for determining process economy parameters of multivariate loss functions
Özkan, Gökçe; Köksal, Gülser; Department of Industrial Engineering (2016)
The aim of this study is to provide an effective method for determining parameters of multivariate loss functions, which are related with process economics. The loss functions are widely used in product and process design and other quality engineering applications. Although there are several studies about different types of loss functions, there is a lack of studies on determining cost matrix parameters of these functions. For this purpose, we propose a method based on multi-objective decision making tools....
A sequential classification algorithm for autoregressive processes
Otlu, Güneş; Candan, Çağatay; Çiloğlu, Tolga; Department of Electrical and Electronics Engineering (2011)
This study aims to present a sequential method for the classification of the autoregressive processes. Different from the conventional detectors having fixed sample size, the method uses Wald’s sequential probability ratio test and has a variable sample size. It is shown that the suggested method produces the classification decisions much earlier than fixed sample size alternative on the average. The proposed method is extended to the case when processes have unknown variance. The effects of the unknown pro...
Citation Formats
A. Saranlı, “A statistical unified framework for rank-based multiple classifier decision combination,” PATTERN RECOGNITION, pp. 865–884, 2001, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/39071.