Electrocardiograms are generated by electrical signals, and therefore individual electrocardiogram signals can vary according to the measurement environment that is created by the subjects' behavioral characteristics. Generally, post-exercise electrocardiograms that are used as biometric identification data do not match pre-exercise electrocardiograms owing to temporarily occurring tachycardia, and this reduces user identification performance. Research is being conducted on normalization techniques that make the post-exercise electrocardiograms match the pre-exercise electrocardiograms, but there have been problems caused by the distortion of morphological features such as P waves, QRS complexes, and T waves. To address problem, this study looked at the measurements of electrocardiograms before and after exercise and performed linear interpolation normalization on the P and T waves of a post-exercise tachycardia electrocardiogram cycle to make it match the pre-exercise electrocardiogram cycle. This study also proposes a user recognition system based on this technique. To analyze the proposed system's performance, the lead-I electrocardiogram signals of 20 subjects were measured three times at 2- to 3-day intervals before and after exercise, and a database was constructed. The experiment results showed that the conventional algorithm's maximum recognition performance was 88.33%, and the proposed normalization method's performance was 91.67%. A combined method was found to be both excellent and similar to the proposed method, with a performance of 92.5%.
Information granules are formed to reduce the complexity of the description of real-world systems. The improved generality of information granules is attained through sacrificing some of the numerical precision of point-data. In this study we consider a hyperbox-based clustering and classification of granular data, and discuss detailed criteria for the assessment of the quality of the combined classification and clustering. The robustness of the criteria is assessed on both synthetic data and real-life data from the domain of traffic control.
In this paper, we propose reinforced fuzzy clustering-based ensemble neural networks (FCENNs) classifier. The objective of this paper is focused on the development of the design methodologies of ensemble neural networks classifier for constructing the network structure and enhancing the learning methods of fuzzy clustering-based neural networks through the combination of the probabilistic model and its learning mechanism. The proposed FCENNs classifier takes into consideration a cross-entropy error function to improve learning while L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> norm regularization is used to reduce overfitting as well as enhance generalization abilities. The essential points of the proposed reinforced FCENNs classifier can be enumerated as follows: First, in the proposed classifier, the cross-entropy error function is used as a cost function; to do this, a softmax function is applied to represent a categorical distribution located at the nodes of the output layer. Second, the learning mechanism is composed of two parts. First, fuzzy C-means clustering forms the connections (weights) of the hidden layer while the connections of the output layer are adjusted with the aid of the nonlinear least squares method using Newton's method-based learning. Third, L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> norm-regularization is considered to avoid the degradation of generalization ability caused by overfitting. The learning mechanism similar to ridge regression is realized by adding L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> penalty term to the cross-entropy error function. From the viewpoint of performance improvement achieved through the proposed novel learning method, the design methodology for the ensemble neural networks classifier is discussed and analyzed with the aid of a diversity of two-dimensional synthetic data and machine learning datasets.
We introduce a model of granular data emerging through a summarization and processing of numeric data. It supports data analysis and casts it in the setting of data mining. The structure of data is revealed through the FCM equipped with the Tchebyschev (l/sub /spl infin//) metric. The study offers a novel contribution to a gradient-based learning of the prototypes developed in the l/sub /spl infin//-based data space. The l/sub /spl infin// metric promotes a development of easily interpretable information granules, namely hyperboxes. A detailed discussion of their geometry is provided. In particular, we discuss a deformation effect of the hyperbox-shape of granules due to an interaction between the granules. We also show how the clustering gives rise to a two-level topology of information granules. A core part of the topology comes in the form of hyperbox information granules. A residual structure is expressed through detailed, yet difficult to interpret, membership grades. Illustrative examples including synthetic data are studied.
