International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems

This paper reviews the Hybrid Location-Content Addressable Memories (HyLCAMs) and applies them to the printed character recognition problem in order to illustrate their performance in real world problems. Three indirect code generation methods which are the key of the HyLCAM encoding is described to determine a set of proper intermediate states that are linearly separable with inputs. Rejection mechanisms based on code mismatch and rejection region of activation functions are introduced. Computer simulations are performed in order to compare generalization abilities and performance on the rejection mechanism of the HyLCAM models with those of the multilayer perceptrons.

It has never been demonstrated that a pure semantic analysts of an English sentence can be accomplished without any aid from a syntactic analyzer. It has therefore become interesting to demonstrate semantic systems which can be guided by fast and efficient syntactic methods. We show that non-probabilistic, abductive techniques can be used in a hybrid network to correctly interpret the meaning of an English sentence. We discuss the implementation of an abductive system which uses heuristics working together with a semantic network in an attempt to eliminate uncertainty and ambiguity in natural language text.

A decade ago, Edelsbrunner et. al. introduced the concept of α-hulls for finding the shape of a planar set. But it still suffers from a major deficiency i.e., the single parameter α must be fine-tuned. The present article initially proposes a selection criterion of α for α-hulls in R². Then the concept of α-hulls is extended to define the fuzzy α-hulls. The effectiveness of the fuzzy α-hulls and the proposed selection criterion of α is demonstrated on some artificially generated data sets. The convergence (with sample size) of both the crisp and fuzzy α-hulls, based on the proposed selection criterion of α, to the original pattern class has also been verified.

Nowadays, the digital register process of electrocardiographical signals (ECG) constitutes a common practice for the diagnosis and controlling of patients suffering from cardiac disorders. In this paper we study the usefulness of Artificial Neural Network (ANN) for clinical diagnosis through the detection of arrythmias and the reduction of the large spaces occupied by ECG records.

The segmentation of scenes into perceptually meaningful partitions has been a basic problem in image understanding, especially when unsupervised methodology has been desired. A novel unsupervised segmentation approach based on texture is developed. The texture model is based on sets of gray level cooccurence (GLC) matrices rather than measures extracted from them. The algorithmic constituents for the segmentation scheme: choice of seed regions, normalized match distances between texture models, region homogeneity, and aggregation criteria are systematically developed. The unsupervised algorithm works so that “seed” regions are discovered by an image search process. Initial estimates of the texture model prototypes are automatically computed for each “seed” region, and classification thresholds are based on the variance of the model over the “seed” region. An aggregation process then results in regions being successively classified and segmented “out” of the image. This recursive process of segmentation is continued until all pixels are classified. The segmentation strategy was tested successfully on natural texture mosaics. The results are analytically presented. These experiments demonstrate that the unsupervised process can correctly identify the perceptual constituents of the image based on texture.

This study presents an effective approach to the hitherto little addressed problem of feature assessment and selection for pattern recognition in imprecisely supervised environments. Unlike in classical supervised environments wherein the representative training samples have crisp class labels, here the samples have fuzzy memberships in several of the different pattern classes in the environment. The new methodology reported here is an outgrowth of a recently developed tool CORPS—Class Overlap Region Partitioning Scheme initially designed for operation in supervised environments and extended later for operation in imperfectly supervised environments. The emphasis here has been the development of a computationally efficient scheme capable of evaluating as rapidly as practical a large number of features individually as to their discrimination potential based on which a smaller subset may be selected, if so desired, for more detailed evaluation in different combinations by other tools.

This paper describes a generalized fuzzy covariance matrix based on the notion of fuzzy similitude and its application to discriminant analysis. It is shown that this generalized fuzzy covariance matrix may be useful in the determination of outliers. It is further noted that the generalized fuzzy covariance matrix may help to distinguish outliers from high-leverage points, and both of these from “normal” points, and thus may have general application to sensitivity analysis in discriminant analysis and clustering problems.