Applications of Nonclassical Logic Methods for Purposes of Knowledge Discovery and Data Mining

* The work is partially supported by Grant no. NIP917 of the Ministry of Science and Education – Republic of Bulgaria.

Applications of the Sufficiency Principle in Acceleration of Neural Networks Trainig

One of the problems in AI tasks solving by neurocomputing methods is a considerable training time. This problem especially appears when it is needed to reach high quality in forecast reliability or pattern recognition. Some formalised ways for increasing of networks’ training speed without loosing of precision are proposed here. The offered approaches are based on the Sufficiency Principle, which is formal representation of the aim of a concrete task and conditions (limitations) of their solving [1]. This is development of the concept that includes the formal aims’ description to the context of such AI tasks as classification, pattern recognition, estimation etc.

Projective Methods of Image Recognition

We propose a method for image recognition on the base of projections. Radon transform gives an opportunity to map image into space of its projections. Projection properties allow constructing informative features on the base of moments that can be successfully used for invariant recognition. Offered approach gives about 91-97% of correct recognition.

Improved Hybrid Model of HMM/GMM for Speech Recognition

In this paper, we propose a speech recognition engine using hybrid model of Hidden Markov Model (HMM) and Gaussian Mixture Model (GMM). Both the models have been trained independently and the respective likelihood values have been considered jointly and input to a decision logic which provides net likelihood as the output. This hybrid model has been compared with the HMM model. Training and testing has been done by using a database of 20 Hindi words spoken by 80 different speakers. Recognition rates achieved by normal HMM are 83.5% and it gets increased to 85% by using the hybrid approach of HMM and GMM.

Automatic ECG Analysis for Preliminary and Detailed Diagnostics Based on Scale-space Representation

A novel approach of automatic ECG analysis based on scale-scale signal representation is proposed. The approach uses curvature scale-space representation to locate main ECG waveform limits and peaks and may be used to correct results of other ECG analysis techniques or independently. Moreover dynamic matching of ECG CSS representations provides robust preliminary recognition of ECG abnormalities which has been proven by experimental results.

Basic Principles of Organization of the Medium and Thinking Process of the Human in its Conceptual Presentation

Given in the report conceptual presentation of the main principles of fractal-complexity Ration of the media and thinking processes of the human was formulated on the bases of the cybernetic interpretation of scientific information (basically from neurophysiology and neuropsychology, containing the interpretation giving the best fit to the authors point of view) and plausible hypothesis's, filling the lack of knowledge.

Оптимизация Оценки Вероятности Ошибочной Классификации в Дискретном Случае

* Работа выполнена при поддержке РФФИ, гранты 07-01-00331-a и 08-01-00944-a

"AVO-Polynom" Recognition Algorithm

Estimates Calculating Algorithms have a long story of application to recognition problems. Furthermore they have formed a basis for algebraic recognition theory. Yet use of ECA polynomials was limited to theoretical reasoning because of complexity of their construction and optimization. The new recognition method “AVO- polynom” based upon ECA polynomial of simple structure is described.

Построение Логико-Вероятностных Моделей Временных Рядов с Использованием Цепей Маркова

The method of logic and probabilistic models constructing for multivariate heterogeneous time series is offered. There are some important properties of these models, e.g. universality. In this paper also discussed the logic and probabilistic models distinctive features in comparison with hidden Markov processes. The early proposed time series forecasting algorithm is tested on applied task.

The Usage of Neural Networks for the Medical Diagnosis

The problem of cancer diagnosis from multi-channel images using the neural networks is investigated. The goal of this work is to classify the different tissue types which are used to determine the cancer risk. The radial basis function networks and backpropagation neural networks are used for classification. The results of experiments are presented.

Self Evolving Character Recognition Using Genetic Operators

In this paper, a novel approach for character recognition has been presented with the help of genetic operators which have evolved from biological genetics and help us to achieve highly accurate results. A genetic algorithm approach has been described in which the biological haploid chromosomes have been implemented using a single row bit pattern of 315 values which have been operated upon by various genetic operators. A set of characters are taken as an initial population from which various new generations of characters are generated with the help of selection, crossover and mutation. Variations of population of characters are evolved from which the fittest solution is found by subjecting the various populations to a new fitness function developed. The methodology works and reduces the dissimilarity coefficient found by the fitness function between the character to be recognized and members of the populations and on reaching threshold limit of the error found from dissimilarity, it recognizes the character. As the new population is being generated from the older population, traits are passed on from one generation to another. We present a methodology with the help of which we are able to achieve highly efficient character recognition.

A Method for Classification of Doubly Resolvable Designs and Its Application

This article presents the principal results of the Ph.D. thesis Investigation and classification of doubly resolvable designs by Stela Zhelezova (Institute of Mathematics and Informatics, BAS), successfully defended at the Specialized Academic Council for Informatics and Mathematical Modeling on 22 February 2010.

Hough Transform Approach to Identification of Flare Stars in Multi-exposure Plate Images

ACM Computing Classification System (1998): I.2.8 , I.2.10, I.5.1, J.2.

How different are the visual representations used for object recognition in middle childhood and adulthood?

Recent experimental studies have shown that development towards adult performance levels in configural processing in object recognition is delayed through middle childhood. Whilst partchanges to animal and artefact stimuli are processed with similar to adult levels of accuracy from 7 years of age, relative size changes to stimuli result in a significant decrease in relative performance for participants aged between 7 and 10. Two sets of computational experiments were run using the JIM3 artificial neural network with adult and 'immature' versions to simulate these results. One set progressively decreased the number of neurons involved in the representation of view-independent metric relations within multi-geon objects. A second set of computational experiments involved decreasing the number of neurons that represent view-dependent (nonrelational) object attributes in JIM3's Surface Map. The simulation results which show the best qualitative match to empirical data occurred when artificial neurons representing metric-precision relations were entirely eliminated. These results therefore provide further evidence for the late development of relational processing in object recognition and suggest that children in middle childhood may recognise objects without forming structural description representations.

The structure and dynamics of saxicolous lichen communities

The saxicolous lichen vegetation on Ordovician slate rock at the mouth of the River Dovey, South Merionethshire, Wales was described in relation to several environmental variables which include aspect, slope angle, light intensity, rock porosity, rock microtopography and rock stability. Each of the measured environmental variables was shown to influence the lichen vegetation. A number of groups of species which were characteristic of certain environments were described. The data from the saxicolous lichen communities were analysed using multivariate analysis. Qualitative and quantitative data were ordinated, the qualitative data being easier to interpret ecologically, and site number (which reflects distance from the sea and altitude), rock porosity and light intensity were shown to be important environmental variables. A classification of the data was also carried out. The results of the ordination and classification were combined together and a model constructed which describes saxicolous lichen vegetation. A method which uses the model as an aid to the design and interpretation of field experiments is described. The model is applied to an experiment which investigates the effect on growth of transplanting four saxicolous lichens to different aspects. Growth was inhibited in Physcia orbicularis and Parmelia conspersa on rock surfaces of northwest aspect compared with growth on rock surfaces of southeast aspect. Growth was inhibited in Parmelia glabratula ssp. fuliginosa on rock surfaces of southeast aspect compared with rock surfaces of northwesr aspect. The growth of Parmelia saxatilis was similar at both southeast and northwesr aspects. Growth inhibition or stimulation in thalli of Physcia orbicularis, Parmelia conspersa and Parmelia glabratula ssp. fuliginosa after transplantation was consistent with the predictions of the model while the results for Parmelia saxatilis were not as expected. There was evidence that the frequency of Parmelia conspersa and Parmelia glabratula at a site is related to an effect of the environment on the growth of the thalli. There was also evidence that the frequency of Physcia orbicularis at a site is related to an effect of the environment on the establishment phase of the thalli and for the competitive exclusion of Parmelia saxatilis thalli from southeast facing rock surfaces. The distribution of lichens in relation to height on nine rock surfaces was investigated. It was suggested that the distribution of the lichens was influenced by microclimatic factors which are related to height on the rock, environmental variables which are associated with the rock substratum (e.g. rock porosity and rock microtopography) and by historical factors. The pattern of one crustose and one foliose lichen on four rock surfaces of different aspect and slope was investigated. On the vertically inclined surface the density of small thalli of Buellia aethalea and Parmelia glabratula ssp fuliginosa was correlated with the microtopography of the surface in transects horizontally across the rock surface but not in transects vertically down the rock surface. there were consitent differences in the scale and intensity of pattern horizontally and vertically and also a decrease in the intensity of pattern vertically as the slope of the rock surface decreased. These results were consistent with the suggestion of a gradient of microclimatic factors up the rock. The differences in the scale and intensity of pattern in different size classes in the population were consistent with the changes in pattern with time which have been shown to occur during succession in sand dune and salt marsh vegetation. The relationship between thallus size and height on a rock surface and between the radial growth rate and location of a thallus on a rock surface were investigated. Thalli of Parmelia glabratula ssp. fuliginosa were larger at the top of the rock surface than at the bottom and the data were consistent with the suggestion that the colonisation of the rock surface began at the top and, in time, spread downwards. The radial growth rate of the thalli could not be related to variation in slope, porosity, microtopography or directly to height on the rock but could be related to the horizontal location of the thalli on the rock. These results were consistent with the suggestion that here is a gradient of microclimatic factors across the rock surface which is also modified by height on the rock surface. The succession of lichen communities was described by relating the vegetation to rock porosity, rock microtopography, species diversity and rock stability. An initial stage dominated by crustose lichens leads to communities dominated by crustose, foliose and fruticose species. In the late stages of the succession on some rock surfaces crustose species again become dominant. The occurrence of the climax state and cyclic vegetation change in lichen communities are discussed. A mthod of estimating the age structure of a lichen population by relating thallus size to growth rate is described. The sources of error in the method are discussed in some detail and several refinements suggested to increase the accuracy of the method. The population dynamics of Parmelia glabratula ssp. fuliginosa was investigated by applying life tables to the age structures of eight different populations. The data were consistent with a period of relatively constant recruitment of thalli into the populations. Mortality in lichen populations was divided into deaths which occur after fragmentation of the thallus and deaths which occur after catastrophic environmental events. THe data suggest that the rate of fragmenting death is dependent on the age of the thallus while the rate of catastrophic death is dependent on the number of thalli established in an age class. A comparison of the numbers of thalli in each age class in the eight populations suggested that population density is controlled firstly, by climate and secondly, by variables related to the local rock surface environment. The rate of fragmenting death is related to the diversity of the community and the influence of diversity together with environmental variables in fluctuating or cyclic changes in population number.