Pattern Classification

With the booming of cyber attacks and cyber criminals against cyber-physical systems (CPSs), detecting these attacks remains challenging. It might be the worst of times, but it might be the best of times because of opportunities brought by machine learning (ML), in particular deep learning (DL). In general, DL delivers superior performance to ML because of its layered setting and its effective algorithm for extract useful information from training data. DL models are adopted quickly to cyber attacks against CPS systems. In this survey, a holistic view of recently proposed DL solutions is provided to cyber attack detection in the CPS context. A six-step DL driven methodology is provided to summarize and analyze the surveyed literature for applying DL methods to detect cyber attacks against CPS systems. The methodology includes CPS scenario analysis, cyber attack identification, ML problem formulation, DL model customization, data acquisition for training, and performance evaluation. The reviewed works indicate great potential to detect cyber attacks against CPS through DL modules. Moreover, excellent performance is achieved partly because of several high-quality datasets that are readily available for public use. Furthermore, challenges, opportunities, and research trends are pointed out for future research.

This paper presents the results of handwritten digit recognition on well-known image databases using state-of-the-art feature extraction and classiÿcation techniques. The tested databases are CENPARMI, CEDAR, and MNIST. On the test data set of each database, 80 recognition accuracies are given by combining eight classiÿers with ten feature vectors. The features include chaincode feature, gradient feature, proÿle structure feature, and peripheral direction contributivity. The gradient feature is extracted from either binary image or gray-scale image. The classiÿers include the k-nearest neighbor classiÿer, three neural classiÿers, a learning vector quantization classiÿer, a discriminative learning quadratic discriminant function (DLQDF) classiÿer, and two support vector classiÿers (SVCs). All the classiÿers and feature vectors give high recognition accuracies. Relatively, the chaincode feature and the gradient feature show advantage over other features, and the proÿle structure feature shows e ciency as a complementary feature. The SVC with RBF kernel (SVC-rbf) gives the highest accuracy in most cases but is extremely expensive in storage and computation. Among the non-SV classiÿers, the polynomial classiÿer and DLQDF give the highest accuracies. The results of non-SV classiÿers are competitive to the best ones previously reported on the same databases.

Document Image processing and Optical Character Recognition (OCR) have been a frontline research area in the field of human-machine interface for the last few decades. Recognition of Indian language characters has been a topic of interest for quite some time. The earlier contributions were reported in [1] and [2]. A more recent work is reported in [3] and [9]. The need for efficient and robust algorithms and systems for recognition is being felt in India, especially in the post and telegraph department where OCR can assist the staff in sorting mail. Character recognition can also form a part in applications like intelligent scanning machines, text to speech converters, and automatic language-to-language translators.

The mixture of gait deviations seen in patients following a stroke is remarkably variable. An objective system for classification of gait patterns for this population could be used to guide treatment planning. Quantitated gait analysis was conducted for 47 individuals at admission to in-patient rehabilitation and again at 6 months post-stroke for 42 subjects. Non-hierarchical cluster analysis was used to classify the gait patterns of patients based on the temporal Á/spatial and kinematic parameters of walking. Four clusters of patients were identified at both assessment intervals. At the admission test walking velocity, peak knee extension in mid stance and peak dorsiflexion in swing were the three factors that best characterized the groups. At 6 months the explanatory variables were velocity, knee extension in terminal stance, and knee flexion in pre swing. Differences in muscle strength and muscle activation patterns during walking were identified between groups. #

Human identification at a distance has recently gained growing interest from computer vision researchers. Gait recognition aims essentially to address this problem by identifying people based on the way they walk. In this paper, a simple but efficient gait recognition algorithm using spatial-temporal silhouette analysis is proposed. For each image sequence, a background subtraction algorithm and a simple correspondence procedure are first used to segment and track the moving silhouettes of a walking figure. Then, eigenspace transformation based on Principal Component Analysis (PCA) is applied to time-varying distance signals derived from a sequence of silhouette images to reduce the dimensionality of the input feature space. Supervised pattern classification techniques are finally performed in the lower-dimensional eigenspace for recognition. This method implicitly captures the structural and transitional characteristics of gait. Extensive experimental results on outdoor image sequences demonstrate that the proposed algorithm has an encouraging recognition performance with relatively low computational cost.

In this study, a hybrid genetic algorithm is adopted to find a subset of features that are most relevant to the classification task. Two stages of optimization are involved. The outer optimization stage completes the global search for the best subset of features in a wrapper way, in which the mutual information between the predictive labels of a trained classifier and the true classes serves as the fitness function for the genetic algorithm. The inner optimization performs the local search in a filter manner, in which an improved estimation of the conditional mutual information acts as an independent measure for feature ranking taking account of not only the relevance of the candidate feature to the output classes but also the redundancy to the already-selected features. The inner and outer optimizations cooperate with each other and achieve the high global predictive accuracy as well as the high local search efficiency. Experimental results demonstrate both parsimonious feature selection and excellent classification accuracy of the method on a range of benchmark data sets.

Resumo -Este artigo apresenta resultados referentes ao módulo de diagnóstico automático da plataforma SINPATCO -Sistema INteligente de diagnóstico de PATologias da COluna vertebral. Este módulo é composto por uma unidade de pré-processamento, cujo objetivo é normalizar os dados e eliminar instâncias discrepantes (outliers) de um banco de dados de casos clínicos, e por um classificador neural baseado na rede Perceptron Multicamadas (MLP). O módulo de diagnóstico é avaliado quanto à capacidade de categorizar casos clínicos em uma das seguintes classes: Normal, Hérnia de Disco e Espondilolistese. Cada caso é representado por 6 atributos biomecânicos extraídos por um médico ortopedista a partir das imagens de Raio-X da coluna do paciente. A rede MLP foi comparada a dois outros classificadores (Naive Bayes e GRNN), tendo apresentado maiores taxas de acerto, maior robustez a outliers e menor número de falsos positivos e falsos negativos.

Support vector machine (SVM) is a novel pattern classification method that is valuable in many applications. Kernel parameter setting in the SVM training process, along with the feature selection, significantly affects classification accuracy. The objective of this study is to obtain the better parameter values while also finding a subset of features that does not degrade the SVM classification accuracy. This study develops a simulated annealing (SA) approach for parameter determination and feature selection in the SVM, termed SA-SVM.

The present study addressed autonomic nervous system (ANS) patterning during experimentally manipulated emotion. Film clips previously shown to induce amusement, anger, contentment, disgust, fear and sadness, in addition to a neutral control film, were presented to 34 college-aged subjects while skin conductance, blood pressure and the electrocardiogram (ECG) were recorded, as was self-reported affect. Both mean of and mean successive difference of heart period were derived from the ECG. Pattern classification analyses revealed emotion-specific autonomic patterning for all emotion conditions except disgust; all emotion conditions exhibited significant patterning using selfreport. Discriminant function analysis was used to describe the location of discrete emotions within dimensional affective space using both self-report and ANS variables. Findings suggest that the dimensions of valence and activation portray the structure of self-reported emotion, but that valence is more accurately described as approachwithdrawal when applied to autonomic responses during discrete emotions. The findings provide further support for the existence of emotion-specific ANS activity, and are consistent with a hybrid discrete-dimensional model of affective space. ᮊ

In this paper, our goal is to (a) survey some of the legal contexts within which violence risk assessment already plays a prominent role, (b) explore whether developments in neuroscience could potentially be used to improve our ability to predict violence, and (c) discuss whether neuropredictive models of violence create any unique legal or moral problems above and beyond the well worn problems already associated with prediction more generally. In "Violence Risk Assessment and the Law", we briefly examine the role currently played by predictions of violence in three high stakes legal contexts: capital Neuroethics (2012) 5:67-99

This paper introduces the concept of distributed representation of fuzzy rules and applies it to classification problems. Distributed representation is implemented by superimposing many fuzzy rules corresponding to different fuzzy partitions of a pattern space. This means that we simultaneously employ many fuzzy rule tables corresponding to different fuzzy partitions in fuzzy inference. In order to apply distributed representation of fuzzy rules to pattern classification problems, we first propose an algorithm to generate fuzzy rules from numerical data. Next we propose a fuzzy inference method using the generated fuzzy rules. The classification power of distributed representation is compared with that of ordinary fuzzy rules which can be viewed as local representation.

This paper shows how the rule weight of each fuzzy rule can be specified in fuzzy rule-based classification systems. First, we propose two heuristic methods for rule weight specification. Next, the proposed methods are compared with existing ones through computer simulations on artificial numerical examples and real-world pattern classification problems. Simulation results show that the proposed methods outperform the existing

The correct diagnosis of breast cancer is one of the major problems in the medical field. From the literature it has been found that different pattern recognition techniques can help them to improve in this domain. These techniques can help doctors form a second opinion and make a better diagnosis. In this paper we present a novel improvement in neural network training for pattern classification. The proposed training algorithm is inspired by the biological metaplasticity property of neurons and Shannon's information theory. During the training phase the Artificial metaplasticity Multilayer Perceptron (AMMLP) algorithm gives priority to updating the weights for the less frequent activations over the more frequent ones. In this way metaplasticity is modeled artificially. AMMLP achieves a more effcient training, while maintaining MLP performance. To test the proposed algorithm we used the Wisconsin Breast Cancer Database (WBCD). AMMLP performance is tested using classification accuracy, sensitivity and specificity analysis, and confusion matrix. The obtained AMMLP classification accuracy of 99.26%, a very promising result compared to the Backpropagation Algorithm (BPA) and recent classification techniques applied to the same database.

A new channel pattern classification is presented based on theoretically derived channel pattern discriminant functions. The thresholds are formulated as power laws that relate the critical slope associated with a change in channel pattern to dimensionless discharge and relative bank strength. One threshold demarcates the boundary between stable single-thread channels (both straight and meandering) and stable multiple-thread channels (anabranching). Another threshold separates anabranching from braided channels, where braided channels are considered to be fundamentally unstable. The exponents of the thresholds are nearly identical to that in the threshold equation originally proposed by Leopold and Wolman (1957), and the coefficients are quite similar as well. An analysis of their data set using our dimensionless approach reveals that no fundamental difference exists between meandering and straight patterns, and thus data from both types are grouped together under the more general heading of single-thread channels. Furthermore, we demonstrate that over a limited range of conditions, an unstable single-thread channel can form stable multiple-threads; but that for systems far from the threshold bounding the single-thread channels, the number of divisions required to produce stable anabranches grows geometrically: this motivates a separation of multiple-thread channels into anabranching and braided types. Our theoretical thresholds are then compared against several large data sets of field data, and the results broadly confirm our proposed thresholds.

Over the last few decades pattern classification has been one of the most challenging area of research. In the present-age pattern classification problems, the support vector machines (SVMs) have been extensively adopted as machine learning tools. SVM achieves higher generalization performance, as it utilizes an induction principle called structural risk minimization (SRM) principle. The SRM principle seeks to minimize the upper bound of the generalization error consisting of the sum of the training error and a confidence interval. SVMs are basically designed for binary classification problems and employs supervised learning to find the optimal separating hyperplane between the two classes of data. The main objective of this paper is to introduce a most promising pattern recognition technique called cross-correlation aided SVM based classifier. The idea of using cross-correlation for feature extraction is relatively new in the domain of pattern recognition. In this paper, the proposed technique has been utilized for binary classification of EEG signals. The binary classifiers employ suitable features extracted from crosscorrelograms of EEG signals. These cross-correlation aided SVM classifiers have been employed for some benchmark EEG signals and the proposed method could achieve classification accuracy as high as 95.96% compared to a recently proposed method where the reported accuracy was 94.5%.

Multilayered feedforward neural networks possess a number of properties which make them particularly suited to complex pattern classification problems. However, their application to some realworld problems has been hampered by the lack of a training algonthm which reliably finds a nearly globally optimal set of weights in a relatively short time. Genetic algorithms are a class of optimization procedures which are good at exploring a large and complex space in an intelligent way to find values close to the global optimum. Hence, they are well suited to the problem of training feedforward networks. In this paper, we describe a set of experiments performed on data from a sonar image classification problem. These experiments both 1) illustrate the improvements gained by using a genetic algorithm rather than backpropagation and 2) chronicle the evolution of the performance of the genetic algorithm as we added more and more domain-specific knowledge into it.

The notion of a random graph is formally defined. It deals with both the probabilistic and the structural aspects of relational data. By interpreting an ensemble of attributed graphs as the outcomes of a random graph, we can use its lower order distribution to characterize the ensemble. To reflect the variability of a random graph, Shannon's entropy measure is used. To synthesize an ensemble of attributed graphs into the distribution of a random graph (or a set of distributions), we propose a distance measure between random graphs based on the minimum change of entropy before and after their merging. When the ensemble contains more than one class of pattern graphs, the synthesis process yields distributions corresponding to various classes. This process corresponds to unsupervised learning in pattern classification. Using the maximum likelihood rule and the probability computed for the pattern graph, based on its matching with the random graph distributions of different classes, we can classify the pattern graph to a class.

In this review, we have discussed the class-prediction and discovery methods that are applied to gene expression data, along with the implications of the findings. We attempted to present a unified approach that considers both class-prediction and class-discovery. We devoted a substantial part of this review to an overview of pattern classification/recognition methods and discussed important issues such as preprocessing of gene expression data, curse of dimensionality, feature extraction/selection, and measuring or estimating classifier performance. We discussed and summarized important properties such as generalizability (sensitivity to overtraining), built-in feature selection, ability to report prediction strength, and transparency (ease of understanding of the operation) of different class-predictor design approaches to provide a quick and concise reference. We have also covered the topic of biclustering, which is an emerging clustering method that processes the entries of the gene expression data matrix in both gene and sample directions simultaneously, in detail.

The basic question of how to optimally make use of a finite number of available samples in designing pattern recognition systems is considered. This has several components: optimal use of the samples for design and testing; and the relationship between the number of measurements and the number of samples for various probability structural constraints. A spectrum of possibilities has been demonstrated, placing several apparently conflicting recent results in perspective.

This paper shows how a small number of simple fuzzy if-then rules can be selected for pattern classification problems with many continuous attributes. Our approach consists of two phases: Candidate rule generation by rule evaluation measures in data mining and rule selection by multi-objective evolutionary algorithms. In our approach, first candidate fuzzy if-then rules are generated from numerical data and

Automated fault classification has been an important pattern recognition problem for decades. In the performance of all motor driven systems, bearings play an important role. The purpose of condition monitoring and fault diagnostics are to detect and distinguish faults occurring in machinery, in order to provide a significant improvement in plant economy, reduce operational and maintenance costs and improve the level of safety. This paper addresses a recent study to employ Wavelet decomposition to process the accelerometer signals and identify band patterns features. Selected features are extracted from the vibration signatures so obtained and these are used as inputs to three types of artificial networks trained to identify the bearing conditions at three different rotational speeds. Vibration signals for normal bearings, bearing with inner race fault, outer race faults and ball faults were acquired from a motor-driven experimental system. The experimental results are presented and compared with those of currently best-performing on this field. Later sections explain some of the artificial intelligence methods design considerations such as network architecture, performance and implementation.The results demonstrate that the developed diagnostic method can reliably detect and classify four different bearing fault conditions into distinct groups.

Several pattern classifiers give high classification accuracy but their storage requirements and processing time are severely expensive. On the other hand, some classifiers require very low storage requirement and processing time but their classification accuracy is not satisfactory. In either of the cases the performance of the classifier is poor. In this paper, we have presented a technique based on the combination of minimum distance classifier (MDC), class-dependent principal component analysis (PCA) and linear discriminant analysis (LDA) which gives improved performance as compared with other standard techniques when experimented on several machine learning corpuses.

We describe an approach of automatic feature extraction for shape characterization of seven distinct species of Eimeria, a protozoan parasite of domestic fowl. We used digital images of oocysts, a round-shaped stage presenting inter-specific variability. Three groups of features were used: curvature characterization, size and symmetry, and internal structure quantification. Species discrimination was performed with a Bayesian classifier using Gaussian distribution. A database comprising 3891 micrographs was constructed and samples of each species were employed for the training process. The classifier presented an overall correct classification of 85.75%. Finally, we implemented a real-time diagnostic tool through a web interface, providing a remote diagnosis front-end.

Ensemble learning for improving weak classifiers is one important direction in the current research of machine learning, and thereinto bagging, boosting and random subspace are three powerful and popular representatives. They have so far shown efficacies in many practical classification problems. However, for electroencephalogram (EEG) signal classification with application to brain-computer interfaces (BCIs), there are almost no studies investigating their feasibilities. The present study systematically evaluates the performance of the three ensemble methods for EEG signal classification of mental imagery tasks. With the base classifiers of k-nearest-neighbor, decision tree and support vector machine, classification experiments are carried out upon real EEG recordings. Experimental results suggest the feasibilities of ensemble classification methods, and we also derive some valuable conclusions on the performance of ensemble methods for EEG signal classification.

An accurate and computationally efficient means of classifying surface myoelectric signal patterns has been the subject of considerable research effort in recent years. Effective feature extraction is crucial to reliable classification and, in the quest to improve the accuracy of transient myoelectric signal pattern classification, an ensemble of time-frequency based representations are proposed. It is shown that feature sets based upon the short-time Fourier transform, the wavelet transform, and the wavelet packet transform provide an effective representation for classification, provided that they are subject to an appropriate form of dimensionality reduction.

Over the years, the management of municipal solid waste (MSW) has been improved to some extent through installation of various schemes, development of new treatment technologies and implementation of economic instruments. Despite such progress, solid waste problems still impose an increasing pressure on cities and remain one of the major challenges in urban environmental management. Although approximating of waste generation in its management is important, the prediction of its production is a difficult job due to the effect of various factors on it. Artificial intelligence is an exciting and relatively new application of computers. It provides new opportunities for harnessing the scarce and often scattered pieces of valuable knowledge and experience in solid waste management which at present is in the possession of the privileged few. While conventional algorithmic programming replaced much of the sophisticated and repetitive analytical work of the solid waste practitioner, artific...

Several studies on structural MRI in children with autism spectrum disorders (ASD) have mainly focused on samples prevailingly consisting of males. Sex differences in brain structure are observable since infancy and therefore caution is required in transferring to females the results obtained for males. The neuroanatomical phenotype of female children with ASD (ASDf) represents indeed a neglected area of research. In this study, we investigated for the first time the anatomic brain structures of a sample entirely composed of ASDf (n = 38; 2-7 years of age; mean = 53 months; SD = 18) with respect to 38 female age and non verbal IQ matched controls, using both mass-univariate and pattern classification approaches. The whole brain volumes of each group were compared using voxel-based morphometry (VBM) with diffeomorphic anatomical registration through exponentiated lie algebra (DARTEL) procedure, allowing us to build a study-specific template. Significantly more gray matter (GM) was found in the left superior frontal gyrus (SFG) in ASDf subjects compared to controls. The GM segments obtained in the VBM-DARTEL preprocessing are also classified with a support vector machine (SVM), using the leave-pair-out cross-validation protocol. Then, the recursive feature elimination (SVM-RFE) approach allows for the identification of the most discriminating voxels in the GM segments and these prove extremely consistent with the SFG region identified by the VBM analysis. Furthermore, the SVM-RFE map obtained with the most discriminating set of voxels corresponding to the maximum Area Under the Receiver Operating Characteristic Curve (AUC max = 0.80) highlighted a more complex circuitry of increased cortical volume in ASDf, involving bilaterally the SFG and the right temporo-parietal junction (TPJ). The SFG and TPJ abnormalities may be relevant to the pathophysiology of ASDf, since these structures participate in some core atypical features of autism.

Over the last two decades, there has been a considerably increase in the number of publications of research projects for the detection and classification of welding defects in radiographs using image processing and pattern recognition tools. All these research projects aim to set up an automatic or semi-automatic classification system for weld joint defects detected by the radiographic method. A classification system as such would allow a reduction in some inherent inexactnesses that occur in the conventional method, consequently increasing the precision of the report. This work is a study to estimate the accuracy of classification of the main classes of weld defects, such as: undercut, lack of penetration, porosity, slag inclusion, crack or lack of fusion. To carry out this work nonlinear pattern classifiers were developed, using neural networks. Also the largest number of radiographic patterns as possible was used as well as statistical inference techniques of random selection of data with (Bootstrap) and without repositioning in order to estimate the accuracy of the classification. The results pointed to an estimated accuracy of around 80% for the classes of defects analyzed.

An image analysis based pattern classification method is proposed to authentic the printing process used in printing different texts on currency notes. Features suitable for doing this are selected and then studied to detect fraudulent samples based on the printing method. This classification is done by using Support Vector Machines and Neural Nets. The discriminatory power of the selected features in authenticating the printing process is tested using the Linear Discriminate Analysis. Experimental results show that the proposed framework provides a highly accurate framework for authenticating the printing process in bank notes.

Partial discharge (PD) measurement is a proven flaw detection technique for finding cavities that are defects in the insulating material. In this paper, a novel approach for the classification of cavity sizes, based on their maximum PD charge transfer-applied voltage (1Q-V) characteristics using a fuzzy decision tree system, is proposed. The (1Q-V) partial discharge patterns for different cavity sizes are represented by features extracted from their pulse shapes, and the classification rules are directly extracted from the data using the decision tree. The decision rules obtained from the decision tree are then converted to the fuzzy IF-then rules, and the back-propagation algorithm is utilized to tune the parameters of the membership functions employed in the fuzzy classifier. The neuro-fuzzy classification technique is shown to provide successful classification of void sizes in an easily interpretive fashion.

A number of mining and environmental related problems have been approached using ANN technology. These problems commonly relate to pattern classification, prediction and optimisation. ANNs have been successfully applied to these areas and are therefore suitable for similar mining and environmental problems. The general trend in the mining industry for automation to the greatest degree calls for technologies such as the ANNs that can utilise large amounts of data for the development of models which otherwise are very difficult or sometimes even impossible to identify. The examples presented in this paper support the choice of ANNs as the basis for developing solutions to mining and environmental problems were conventional techniques fail in one way or another.

MAHNOB-HCI is a multimodal database recorded in response to affective stimuli with the goal of emotion recognition and implicit tagging research. A multimodal setup was arranged for synchronized recording of face videos, audio signals, eye gaze data, and peripheral/central nervous system physiological signals. Twenty-seven participants from both genders and different cultural backgrounds participated in two experiments. In the first experiment, they watched 20 emotional videos and self-reported their felt emotions using arousal, valence, dominance, and predictability as well as emotional keywords. In the second experiment, short videos and images were shown once without any tag and then with correct or incorrect tags. Agreement or disagreement with the displayed tags was assessed by the participants. The recorded videos and bodily responses were segmented and stored in a database. The database is made available to the academic community via a web-based system. The collected data were analyzed and single modality and modality fusion results for both emotion recognition and implicit tagging experiments are reported. These results show the potential uses of the recorded modalities and the significance of the emotion elicitation protocol.

This paper describes the implementation of a distributed agent architecture for intrusion detection and response in networked computers. Unlike conventional intrusion detection systems (IDS), this security system attempts to emulate mechanisms of the natural immune system using Java-based mobile software agents. These security agents monitor multiple levels (packet, process, system, and user) of networked computers to determine correlation among the observed anomalous patterns, reporting such abnormal behavior to the network administrator and/or possibly taking some action to counter a suspected security violation. The paper focuses on the design aspects of such an intrusion detection system by integrating different artificial intelligence techniques and a mobile agent architecture. Specifically, IBM's Aglets TM Software Development Kit (ASDK) is used as the base agent architecture, along with Adaptive Resonance Theory (ART-2) neural networks for network pattern classification, and a fuzzy logic controller for decision/action resolution.

Pattern classification using neural networks and statistical methods is discussed. We give a tutorial overview in which popular classifiers are grouped into distinct categories according to their underlying mathematical principles; also, we assess what makes a classifier neural. The overview is complemented by two case studies using handwritten digit and phoneme data that test the performance of a number of most typical neural-network and statistical classifiers. Four methods of our own are included: Reduced kernel discriminant analysis, the learning -nearest neighbors classifier, the averaged learning subspace method, and a version of kernel discriminant analysis. , in 1993. His current research interests include neural networks, pattern recognition, as well as computational and mathematical statistics, in particular nonparametric function estimation. His earlier work has also included functional analysis, geometric modeling, and computer aided design.

Pattern recognition based myoelectric control systems rely on detecting repeatable patterns at given electrode locations. This work describes an experiment to determine the effect of electrode displacements on pattern classification accuracy, and a classifier training strategy to accommodate this degradation. The results show that electrode displacements adversely affect classification accuracy, but training the system to recognize plausible displacement locations mitigates the effect. Furthermore, a combination of time-domain and autoregressive features appears to yield the best classification accuracy and is least affected by electrode displacements. #

This paper empirically compares nine image dissimilarity measures that are based on distributions of color and texture features summarizing over 1,000 CPU hours of computational experiments. Ground truth is collected via a novel random sampling scheme for color, and via an image partitioning method for texture. Quantitative performance evaluations are given for classification, image retrieval, and segmentation tasks, and for a wide variety of dissimilarity measures. It is demonstrated how the selection of a measure, based on large scale evaluation, substantially improves the quality of classification, retrieval, and unsupervised segmentation of color and texture images. Computational complexity medium low medium medium medium medium high high Exploits ground distance no no yes no no no yes yes Individual binning no yes no no no no no yes Multiple dimensions yes yes no yes yes yes yes yes Partial matches see text no no no no no no yes Non-parametric

The k-nearest-neighbor rule is one of the most attractive pattern classification algorithms. In practice, the choice of k is determined by the cross-validation method. In this work, we propose a new method for neighborhood size selection that is based on the concept of statistical confidence. We define the confidence associated with a decision that is made by the majority rule from a finite number of observations and use it as a criterion to determine the number of nearest neighbors needed. The new algorithm is tested on several real-world datasets and yields results comparable to the k-nearest-neighbor rule. However, in contrast to the k-nearest-neighbor rule that uses a fixed number of nearest neighbors throughout the feature space, our method locally adjusts the number of nearest neighbors until a satisfactory level of confidence is reached. In addition, the statistical confidence provides a natural way to balance the trade-off between the reject rate and the error rate by excluding patterns that have low confidence levels. We believe that this property of our method can be of great importance in applications where the confidence with which a decision is made is equally or more important than the overall error rate.

Over the years, the management of municipal solid waste (MSW) has been improved to some extent through installation of various schemes, development of new treatment technologies and implementation of economic instruments. Despite such progress, solid waste problems still impose an increasing pressure on cities and remain one of the major challenges in urban environmental management. Although approximating of waste generation in its management is important, the prediction of its production is a difficult job due to the effect of various factors on it. Artificial intelligence is an exciting and relatively new application of computers. It provides new opportunities for harnessing the scarce and often scattered pieces of valuable knowledge and experience in solid waste management which at present is in the possession of the privileged few. While conventional algorithmic programming replaced much of the sophisticated and repetitive analytical work of the solid waste practitioner, artificial intelligence is poised to take over the no-less important tasks of the ill-structured and lessdeterministic parts of the planning, design and management processes. In this research with application of feed forward artificial neural network, we proposed an appropriate model to predict weight of waste generation in Saqqez city of Iran. For this purpose, we used time series of generated waste of Saqqez which have been arranged weekly, from 2004 to 2007. After performing of the mentioned model, determination coefficient (R ) and mean absolute relative error (MARE) in neural network for test have been achieved to be 2 equal to 0.648 and 2.17% respectively.

Tools of sensor-data-driven anomaly detection facilitate condition monitoring of dynamical systems especially if the physics-based models are either inadequate or unavailable. Along this line, symbolic dynamic filtering (SDF) has been reported in literature as a real-time data-driven tool of feature extraction for pattern identification from sensor time series. However, an inherent difficulty for a data-driven tool is that the quality of detection may drastically suffer in the event of sensor degradation. This paper proposes an anomaly detection algorithm for condition monitoring of nuclear power plants, where symbolic feature extraction and the associated pattern classification are optimized by appropriate partitioning of (possibly noise-contaminated) sensor time series. In this process, the system anomaly signatures are identified by masking the sensor degradation signatures. The proposed anomaly detection methodology is validated on the International Reactor Innovative & Secure (IRIS) simulator of nuclear power plants, and its performance is evaluated by comparison with that of principal component analysis (PCA).

This paper shows a comparison between two clustering algorithms that use divergence measures to aid the clustering task. Both algorithms take a N-dimensional data set and uses competitive neural networks to separate them into isotropic clusters. Those clusters are then grouped based on a divergence measure. In this paper we compare that procedure using two different divergence measures, the Mahalanobis distance and the Kullback-Leibler divergence. The paper shows some results of clustering on both algorithms and make a few comments about the choice of the free parameters in both situations.

Since microcalcifications in X-ray mammograms are the primary indicator of breast cancer, detection of microcalcifications is central to the development of an effective diagnostic system. This paper proposes a two-stage detection procedure. In the first stage, a data driven, closed form mathematical model is used to calculate the location and shape of suspected microcalcifications. When tested on the Nijmegen University Hospital (Netherlands) database, data analysis shows that the proposed model can effectively detect the occurrence of microcalcifications. The proposed mathematical model not only eliminates the need for system training, but also provides information on the borders of suspected microcalcifications for further feature extraction. In the second stage, 61 features are extracted for each suspected microcalcification, representing texture, the spatial domain and the spectral domain. From these features, a sequential forward search (SFS) algorithm selects the classification input vector, which consists of features sensitive only to microcalcifications. Two types of classifiers-a general regression neural network (GRNN) and a support vector machine (SVM)-are applied, and their classification performance is compared using the Az value of the Receiver Operating Characteristic curve. For all 61 features used as input vectors, the test data set yielded Az values of 97.01% for the SVM and 96.00% for the GRNN. With input features selected by SFS, the corresponding Az values were 98.00% for the SVM and 97.80% for the GRNN. The SVM outperformed the GRNN, whether or not the input vectors first underwent SFS feature selection. In both cases, feature selection dramatically reduced the dimension of the input vectors (82% for the SVM and 59% for the GRNN). Moreover, SFS feature selection improved the classification performance, increasing the Az value from 97.01 to 98.00% for the SVM and from 96.00 to 97.80% for the GRNN.