Marek Kurzynski

A probabilistic model of classifier competence for dynamic ensemble selection

The concept of a classifier competence is fundamental to multiple classifier systems (MCSs). In this study, a method for calculating the classifier competence is developed using a probabilistic model. In the method, first a randomised reference classifier (RRC) whose class supports are realisations of the random variables with beta probability distributions is constructed. The parameters of the distributions are chosen in such a way that, for each feature vector in a validation set, the expected values of the class supports produced by the RRC and the class supports produced by a modelled classifier are equal. This allows for using the probability of correct classification of the RRC as the competence of the modelled classifier. The competences calculated for a validation set are then generalised to an entire feature space by constructing a competence function based on a potential function model or regression. Three systems based on a dynamic classifier selection and a dynamic ensemble selection (DES) were constructed using the method developed. The DES based system had statistically significant higher average rank than the ones of eight benchmark MCSs for 22 data sets and a heterogeneous ensemble. The results obtained indicate that the full vector of class supports should be used for evaluating the classifier competence as this potentially improves performance of MCSs.

A measure of competence based on random classification for dynamic ensemble selection

In this paper, a measure of competence based on random classification (MCR) for classifier ensembles is presented. The measure selects dynamically (i.e. for each test example) a subset of classifiers from the ensemble that perform better than a random classifier. Therefore, weak (incompetent) classifiers that would adversely affect the performance of a classification system are eliminated. When all classifiers in the ensemble are evaluated as incompetent, the classification accuracy of the system can be increased by using the random classifier instead. Theoretical justification for using the measure with the majority voting rule is given. Two MCR based systems were developed and their performance was compared against six multiple classifier systems using data sets taken from the UCI Machine Learning Repository and Ludmila Kuncheva Collection. The systems developed had typically the highest classification accuracies regardless of the ensemble type used (homogeneous or heterogeneous).

The optimal strategy of a tree classifier

Abstract This paper deals with the decision rules of a tree classifier for performing the classification at each nonterminal node, under the assumption of complete probabilistic information. For given tree structure and feature subsets to be used, the optimal decision rules (strategy) are derived which minimize the overall probability of misclassification. The primary result is illustrated by an example.

Prediction of progression of radiographic knee osteoarthritis using tibial trabecular bone texture

OBJECTIVE To develop a system for predicting the progression of radiographic knee osteoarthritis (OA) using tibial trabecular bone texture. METHODS We studied 203 knees with (n = 68) or without (n = 135) radiographic tibiofemoral OA in 105 subjects (90 men and 15 women with a mean age of 54 years) in whom 2 sets of knee radiographs were obtained 4 years apart. We determined medial and lateral compartment tibial trabecular bone texture using an automated region selection method. Three texture parameters were calculated: roughness, degree of anisotropy, and direction of anisotropy based on a signature dissimilarity measure method. We evaluated tibiofemoral OA progression using a radiographic semiquantitative outcome: an increase in the medial joint space narrowing (JSN) grade. We examined the predictive ability of trabecular bone texture in knees with and those without preexisting radiographic OA, with adjustment for age, sex, and body mass index, using logistic regression (generalized estimating equations) and receiver operating characteristic curves. RESULTS The prediction of increased medial JSN in knees with or without preexisting radiographic OA was the most accurate for medial trabecular bone texture; the area under the curve (AUC) was 0.77 and 0.75, respectively. For lateral trabecular bone texture, the AUC was 0.71 in knees with preexisting OA and 0.72 in knees without preexisting OA. CONCLUSION We have developed a system, based on analyzing tibial trabecular bone texture, which yields good prediction of loss of tibiofemoral joint space. The predictive ability of the system needs to be further validated.

Optimal selection of ensemble classifiers using measures of competence and diversity of base classifiers

In this paper, a new probabilistic model using measures of classifier competence and diversity is proposed. The multiple classifier system (MCS) based on the dynamic ensemble selection scheme was constructed using both developed measures. Two different optimization problems of ensemble selection are defined and a solution based on the simulated annealing algorithm is presented. The influence of minimum value of competence and diversity in the ensemble on classification performance was investigated. The effectiveness of the proposed dynamic selection methods and the influence of both measures were tested using seven databases taken from the UCI Machine Learning Repository and the StatLib statistical dataset. Two types of ensembles were used: homogeneous or heterogeneous. The results show that the use of diversity positively affects the quality of classification. In addition, cases have been identified in which the use of this measure has the greatest impact on quality.

Human–machine interface in bioprosthesis control using EMG signal classification

: We present a concept of human–machine interface intended for the task of bioprosthesis decision control by means of sequential recognition of the patients intent based on the electromyography (EMG) signal acquired from his/her body. The EMG signal characteristics, the problem of processing the signals including acquisition and feature extraction and their classification are discussed. The contextual (sequential) recognition via fuzzy relations for the classification of the patients intent is considered and the implied decision algorithms are presented. In the proposed method, the fuzzy relation is determined on the basis of the learning set as a solution of an appropriate optimization problem and then this relation is used in the form of a matrix of membership degrees at successive instants of the sequential decision process. Three algorithms of sequential classification which differ from one another in the sets of input data and procedure are described. The proposed algorithms were experimentally tested in the recognition of phases of the grasping process of the hand on the basis of the EMG signal, where the real-coded genetic algorithm was used as an optimization procedure. The concept of the measurement stand which was the source of information exploited in the experimental investigations of the algorithms is also described.

A Measure of Competence Based on Randomized Reference Classifier for Dynamic Ensemble Selection

This paper presents a measure of competence based on a randomized reference classifier (RRC) for classifier ensembles. The RRC can be used to model, in terms of class supports, any classifier in the ensemble. The competence of a modelled classifier is calculated as the probability of correct classification of the respective RRC. A multiple classifier system (MCS) was developed and its performance was compared against five MCSs using eight databases taken from the UCI Machine Learning Repository. The system developed achieved the highest overall classification accuracies for both homogeneous and heterogeneous ensembles.

Brief paper: Hand movement recognition based on biosignal analysis

This paper proposes a methodology that analyses and classifies the electromyographic (EMG) signals using neural networks to control multifunction prostheses. The control of these prostheses can be made using myoelectric signals taken from surface electrodes. Finger motions discrimination is the key problem in this study. Thus the emphasis, in the proposed work, is put on myoelectric signal processing approaches. The EMG signals classification system was established using the linear neural network. The experimental results show a promising performance in classification of motions based on biosignal patterns.

A signature dissimilarity measure for trabecular bone texture in knee radiographs

PURPOSE The purpose of this study is to develop a dissimilarity measure for the classification of trabecular bone (TB) texture in knee radiographs. Problems associated with the traditional extraction and selection of texture features and with the invariance to imaging conditions such as image size, anisotropy, noise, blur, exposure, magnification, and projection angle were addressed. METHODS In the method developed, called a signature dissimilarity measure (SDM), a sum of earth movers distances calculated for roughness and orientation signatures is used to quantify dissimilarities between textures. Scale-space theory was used to ensure scale and rotation invariance. The effects of image size, anisotropy, noise, and blur on the SDM developed were studied using computer generated fractal texture images. The invariance of the measure to image exposure, magnification, and projection angle was studied using x-ray images of human tibia head. For the studies, Mann-Whitney tests with significance level of 0.01 were used. A comparison study between the performances of a SDM based classification system and other two systems in the classification of Brodatz textures and the detection of knee osteoarthritis (OA) were conducted. The other systems are based on weighted neighbor distance using compound hierarchy of algorithms representing morphology (WND-CHARM) and local binary patterns (LBP). RESULTS Results obtained indicate that the SDM developed is invariant to image exposure (2.5-30 mA s), magnification (x1.00 - x1.35), noise associated with film graininess and quantum mottle (< 25%), blur generated by a sharp film screen, and image size (> 64 x 64 pixels). However, the measure is sensitive to changes in projection angle (> 5 degrees), image anisotropy (> 30 degrees), and blur generated by a regular film screen. For the classification of Brodatz textures, the SDM based system produced comparable results to the LBP system. For the detection of knee OA, the SDM based system achieved 78.8% classification accuracy and outperformed the WND-CHARM system (64.2%). CONCLUSIONS The SDM is well suited for the classification of TB texture images in knee OA detection and may be useful for the texture classification of medical images in general.

Proceedings of the 8th International Conference on Computer Recognition Systems CORES 2013

The computer recognition systems are nowadays one of the most promising directions in artificial intelligence. This book is the most comprehensive study of this field. It contains a collection of 86 carefully selected articles contributed by experts of pattern recognition. It reports on current research with respect to both methodology and applications. In particular, it includes the following sections:Biometrics Data Stream Classification and Big Data AnalyticsFeatures, learning, and classifiers Image processing and computer vision Medical applications Miscellaneous applications Pattern recognition and image processing in roboticsSpeech and word recognitionThis book is a great reference tool for scientists who deal with the problems of designing computer pattern recognition systems. Its target readers can be the as well researchers as students of computer science, artificial intelligence or robotics.