Fuzzy Logic and Applications

When representing knowledge, it may be fruitful to distinguish between negative and positive information in the following sense. There are pieces of information ruling out what is known as impossible on the one hand, and pieces of evidence pointing out things that are guaranteed to be possible. But what is not impossible is not necessarily guaranteed to be possible. This applies as well to the modelling of the preferences of an agent when some potential choices are rejected since they are rather unacceptable, while others are indeed really satisfactory if they are available, leaving room for alternatives to which the agent is indifferent. The combination of negative information is basically conjunctive (as done classically in logic), while it is disjunctive in the case of positive information, which is cumulative by nature. This second type of information has been largely neglected by the logical tradition. Both types may be pervaded with uncertainty when modelling knowledge, or may be a matter of degree when handling preferences. The presentation will first describe how the two types of information can be accommodated in the framework of possibility theory. The existence of the two types of information can shed new light on the revision of a knowledge / preference base when receiving new information. It is also highly relevant when reasoning with (fuzzy) if-then rules, or for improving the expressivity of flexible queries.

Binary fingerprints are binary vectors used to represent chemical molecules by recording the presence or absence of particular substructures, such as labeled paths in the 2D graph of bonds. Complete fingerprints are often reduced to a compressed format–of typical dimension

n

= 512 or

n

= 1024–by using a simple congruence operation. The statistical properties of complete or compressed fingerprints representations are important since fingerprints are used to rapidly search large databases and to develop statistical machine learning methods in chemoinformatics. Here we present an empirical and mathematical analysis of the distribution of complete and compressed fingerprints. In particular, we derive formulas that provide good approximation for the expected number of bits set to one in a compressed fingerprint, given its uncompressed version, and vice versa.

The technique of direct and inverse fuzzy (F-)transforms of three different types is introduced and approximating properties of the inverse F-transforms are described. A method of lossy image compression and reconstruction on the basis of the F-transform is presented.

We can get much knowledge from images. This process can be done in the mind by a human, and implementation of this mind processing by a system is very difficult. This project attempt to image mining for a simple case. In this paper we develop designed neuro-fuzzy system and it is used for accident prediction in two vehicles scenario. The results show better performance respect to previous version.

Fuzzy Q–learning extends the Q–learning algorithm to work in presence of continuous state and action spaces. A Takagi–Sugeno Fuzzy Inference System (FIS) is used to infer the continuous executed action and its action–value, by means of cooperation of several rules. Different kinds of evolution of the parameters of the FIS are possible, depending on different strategies of distribution of the reinforcement signal. In this paper, we compare two strategies: the classical one, focusing on rewarding the rules that have proposed the actions composed to produce the actual action, and a new one we are introducing, where reward goes to the rules proposing actions closest the ones actually executed.

In this paper several combinatorial optimization problems on fuzzy weighted matroid are considered. It is shown how to characterize the degrees of optimality of elements in the setting of the possibility theory.

This paper is a first step in the direction of extending possibilistic planning to take advantage of the expressive power and reasoning capabilities of fuzzy description logics. Fuzzy description logics are used to describe knowledge about the world and about actions. Fundamental definitions are given and the possibilistic planning problem is recast in this new setting.

A prospective study of the use of ordered multi-lattices as underlying sets of truth-values for a generalised framework of logic programming is presented. Specifically, we investigate the possibility of using multi-lattice-valued interpretations of logic programs and the theoretical problems that this generates with regard to its fixed point semantics.

Allen’s interval algebra allows one to formulate problems that are, in the general case, intractable; for this reason several tractable sub-algebrae have been proposed. In this paper the attention is focused on the fuzzy counterparts of those sub-algebrae and a different method to identify their relations is shown: rules for identifying fuzzy tractable relations starting from the knowledge of the classic tractable relations. Enumeration is used to verify the rules and quantify expressiveness, and algebraic considerations adopted to bind the enumeration itself.

In this paper we introduce reasoning procedures for

$\mathcal{ALCQ}^{+}_{F}$

, a fuzzy description logic with extended qualified quantification. The language allows for the definition of fuzzy quantifiers of the absolute and relative kind by means of piecewise linear functions on ℕ and ℚ ∩ [0,1] respectively. In order to reason about instances, the semantics of quantified expressions is defined based on recently developed measures of the cardinality of fuzzy sets. A procedure is described to calculate the fuzzy satisfiability of a fuzzy assertion, which is a very important reasoning task. The procedure considers several different cases and provides direct solutions for the most frequent types of fuzzy assertions.

This paper focuses on the integration of the (also integrated) declarative paradigms of functional logic and fuzzy logic programming, in order to obtain a richer and much more expressive framework where mathematical functions cohabit with fuzzy logic features. In this sense, this paper must be seen as a first stage in the development of this new research line. Starting with two representative languages from both settings, namely Curry and Likelog, we propose an hybrid dialect where a set of rewriting rules associated to the functional logic dimension of the language, are accompanied with a set of similarity equations between symbols of the same nature and arity, which represents the fuzzy counterpart of the new environment. We directly act inside the kernel of the operational mechanism of the language, thus obtaining a fuzzy variant of

needed narrowing

which fully exploits the similarities collected in a given program. A key point in the design of this last operational method is that, apart from computing at least the same elements of the crisp case, all similar terms of a given goal are granted to be completely treated too while avoiding the risk of infinite loops associated to the intrinsic (reflexive, symmetric and transitive) properties of similarity relations.

In this paper some effective methods for calculating the exact degrees of possible and necessary optimality of an element in matroids with ill-known weights modeled by fuzzy intervals are presented.

When the time span of an event is imprecise, it can be represented by a fuzzy set, called a fuzzy time interval. In this paper we propose a representation for 13 relations that can hold between intervals. Since our model is based on fuzzy orderings of time points, it is not only suitable to express precise relationships between imprecise events (“the mid 1930’s came

before

the late 1930’s) but also imprecise relationships (“the late 1930’s came

long before

the early 1990’s). Furthermore we show that our model preserves many of the properties of the 13 relations Allen introduced for crisp time intervals.

As the amount of information on the web grows, users may find increasing challenges in trusting and sometimes distrusting sources. One possible aid is to maintain a network of trust between sources. In this paper, we propose to model such a network as an intuitionistic fuzzy relation. This allows to elegantly handle together the problem of ignorance, i.e. not knowing whether to trust or not, and vagueness, i.e. trust as a matter of degree. We pay special attention to deriving trust information through a trusted third party, which becomes especially challenging when distrust is involved.

This paper reports experiments that explore performance differences in two previous studies that investigated SVM classification of neonatal pain expressions using the Infant COPE database. This database contains 204 photographs of 26 neonates (age 18-36 hours) experiencing the pain of heel lancing and three nonpain stressors. In our first study, we reported experiments where representative expressions of all subjects were included in the training and testing sets, an experimental protocol suitable for intensive care situations. A second study used an experimental protocol more suitable for short-term stays: the SVMs were trained on one sample and then evaluated on an unknown sample. Whereas SVM with polynomial kernel of degree 3 obtained the best classification score (88.00%) using the first evaluation protocol, SVM with a linear kernel obtained the best classification score (82.35%) using the second protocol. However, experiments reported here indicate no significant difference in performance between linear and nonlinear kernels.

In this paper, we implement hand gesture recognition system using fuzzy algorithm and neural network for Post PC (the embedded-ubiquitous environment using blue-tooth module, embedded i.MX21 board and smart gate-notebook computer). Also, we propose most efficient and reasonable hand gesture recognition interface for Post PC through evaluation and analysis of performance about each gesture recognition system. The proposed gesture recognition system consists of three modules: 1) gesture input module that processes motion of dynamic hand to input data, 2) Relational Database Management System (hereafter, RDBMS) module to segment significant gestures from input data and 3) 2 each different recognition module: fuzzy max-min and neural network function recognition module to recognize significant gesture of continuous / dynamic gestures. Experimental result shows the average recognition rate of 98.8% in fuzzy max-min module and 96.7% in neural network recognition module about significantly dynamic gestures.

Ubiquitous computing is a new era in the evolution of computers. After the mainframe and PC (personal computers) phases, the phase of ubiquitous computing device begins. In this paper, we implement and evaluate glove-based HCI (Human Computer Interaction) methods using fuzzy algorithm and neural network for post PC in the ubiquitous computing. Using glove, we implement hand gesture recognition systems for the wearable PC. One system uses combination of fuzzy algorithm and RDBMS (Relational Database Management System) module, the other system uses neural network. Both systems are implemented on the platform of minimized wearable computer (based on i.MX21). After implementation, we conduct some performance evaluation in the mobile condition. And then we discuss strength and weakness of each method. Finally, we suggest possible improvements methods for HCI based on the wearable computers in the mobile condition.

The method of speaker normalization has been known as the successful method for improving the speech recognition at speaker independent speech recognition system. This paper propose a new power spectrum warping approach to making improvement of speaker normalization better than a frequency warping. The power spectrum warping uses Mel-frequency cepstral of Mel filter bank in MFCC. Also, this paper proposes the hybrid VTN combined the power spectrum warping and a frequency warping. Experiment of this paper did a comparative analysis about the recognition performance of the SKKU PBW DB applied each the power spectrum is 3.06%, and hybrid VTN is 4.07% word error rate reduction as word recognition performance of baseline system.

In the context of inductive inference Solomonoff complexity plays a key role in correctly predicting the behavior of a given phenomenon. Unfortunately, Solomonoff complexity is not algorithmically computable. This paper deals with a Genetic Programming approach to inductive inference of chaotic series, with reference to Solomonoff complexity, that consists in evolving a population of mathematical expressions looking for the ‘optimal’ one that generates a given series of chaotic data. Validation is performed on the Logistic, the Henon and the Mackey–Glass series. The results show that the method is effective in obtaining the analytical expression of the first two series, and in achieving a very good approximation and forecasting of the Mackey–Glass series.

Particle Swarm Optimization (PSO) is a heuristic optimization technique showing relationship with Evolutionary Algorithms and strongly based on the concept of swarm. It is used in this paper to face the problem of classification of instances in multiclass databases. Only a few papers exist in literature in which PSO is tested on this problem and there are no papers showing a thorough comparison for it against a wide set of techniques typically used in the field. Therefore in this paper PSO performance is compared on nine typical test databases against those of nine classification techniques widely used for classification purposes. PSO is used to find the optimal positions of class centroids in the database attribute space, via the examples contained in the training set. Performance of a run, instead, is computed as the percentage of instances of testing set which are incorrectly classified by the best individual achieved in the run. Results show the effectiveness of PSO, which turns out to be the best on three out of the nine challenged problems.

In this paper we exploit multi-objective genetic algorithms to identify Takagi-Sugeno (TS) fuzzy systems that show simultaneously high accuracy and low complexity. Using this approach, we approximate the Pareto optimal front by first identifying TS models with different structures (i.e., different number of rules and input variables), and then performing a local optimization of these models using an ANFIS learning approach. The results obtained allow determining a posteriori the optimal TS system for the specific application. Main features of our approach are selection of the input variables and automatic determination of the number of rules.

Modern pharmacology, combining pharmacokinetic, pharmacodynamic, and pharmacogenomic data, is dealing with high dimensional, nonlinear, stiff systems. Mathematical modeling of these systems is very difficult, but important for understanding them. At least as important is to adequately control them through inputs – drugs’ dosage regimens. Genetic programming (GP) and neural networks (NN) are alternative techniques for these tasks. We use GP to automatically write the model structure in C++ and optimize the model’s constants. This gives insights into the subjacent molecular mechanisms. We also show that NN feedback linearization (FBL) can adequately control these systems, with or without a mathematical model. The drug dosage regimen will determine the output of the system to track very well a therapeutic objective. To our knowledge, this is the first time when a very large class of complex pharmacological problems are formulated and solved in terms of GP modeling and NN modeling and control.

The paper introduces a generalization of the fuzzy logic connectives AND and OR. To define the logical connectives different

t

-norms and

t

-conorms are used. To generalize the

t

-norms (

t

-conorms) the Ordinal Sums are introduced. To learn the parameters of the builded Ordinal Sums and the of weights of the connectives the Genetic Algorithms are applied. Two experiments using both synthetic and benchmark data are made. From one hand, a 2-dimensional classification problem to show the behavior of the approach is considered and on the other hand the Zimmermann-Zysno data set to show the capability of the model is analyzed.

We have developed and continue to enhance automated intelligent software that performs the tasks and decision making which now occurs by the personnel manning watch stations in the Combat Direction Center (CDC) and Task Force Combat Center (TFCC), on-board aircraft carriers and other Navy ships. Integrating information from various sources in a combat station is a complex task; disparate sources of information from radars, sonars, and other sensors are obtained by watch station surveillance guards, who must interpret it and relay it up the chain of command. The

Intelligent Identification Software Module

(IISM) alleviates some of the burden placed on battle commanders by automating tasks like management of historical data, disambiguating multiple track targets, assessing threat levels of targets, and rejecting improbable data. We have knowledge engineered current CDC/TFCC experts and designed IISM using C++ and SimBionic, a visual AI development tool. IISM uses multiple soft computing techniques including Baysian inference and fuzzy reasoning. IISM is interfaced to the Advanced Battle Station (ABS) for use on many US Navy sea vessels.

In this work we present a MATLAB implementation of a fuzzy controller with Conditionally Firing Rules (CFR). The performance of Mamdani-Assilian, Takagi-Sugeno-Kang and CFR inferences are compared and analyzed on two test examples.

For high-throughput screening of genetically modified plant cells, a system for the automatic analysis of huge collections of microscope images is needed to decide whether the cells are infected with fungi or not. To study the potential of feature based classification for this application, we compare different classifiers (kNN, SVM, MLP, LVQ) combined with several feature reduction techniques (PCA, LDA, Mutual Information, Fisher Discriminant Ratio, Recursive Feature Elimination). We achieve a significantly higher classification accuracy using a reduced feature vector instead of the full length feature vector.

We present an evolutionary approach to the search for effective vaccination schedules using mathematical computerized model as a fitness evaluator. Our study is based on our previous model that simulates the Cancer – Immune System competition activated by a tumor vaccine. The model reproduces pre-clinical results obtained for an immunoprevention cancer vaccine (Triplex) for mammary carcinoma on HER-2/neu mice. A complete prevention of mammary carcinoma was obtained

in vivo

using a Chronic vaccination schedule. Our genetic algorithm found complete immunoprevention with a much lighter vaccination schedule. The number of injections required is roughly one third of those used in Chronic schedule.

When applied to genomic data, many popular unsupervised explorative data analysis tools based on clustering algorithms often fail due to their small cardinality and high dimensionality. In this paper we propose a wrapper method for gene selection based on simulated annealing and unsupervised clustering. The proposed approach, even if computationally intensive, permits to select the most relevant features (genes), and to rank their relevance, allowing to improve the results of clustering algorithms.

Data produced by mass spectrometer (MS) have been using in proteomics experiments to identify proteins or patterns in clinical samples that may be responsible of human diseases. Nevertheless, MS data are affected by errors and different preprocessing techniques have to be applied to manipulate and gathering information from data. Moreover, MS samples contain a huge amount of data requiring an efficient organization both to reduce access time to data, and to allow efficient knowledge extraction. We present the design and the implementation of a database for managing MS data, integrated in a software system for the loading, preprocessing, storing and managing of mass spectra data.

Aim of this work is to apply a novel comprehensive data mining machine learning tool to preprocess and to interpret gene expression data. Furthermore, some visualization facilities are provided. The data mining framework consists of two main parts: preprocessing and clustering-agglomerating phases. To the first phase belong a noise filtering procedure and a non-linear PCA Neural Network for feature extraction. The second phase is used to accomplish an unsupervised clustering based on a hierarchy of two approaches: a Probabilistic Principal Surfaces to obtain the rough regions of interesting points and a Fisher-Negentropy information based approach to agglomerate the regions previously found in order to discover substructures present in the data. Experiments on gene microarray data are made. Several experiments are shown varying the threshold, needed by the agglomerative clustering, to understand the structure of the analyzed data set.

In Supervised Learning it is assumed that is straightforward to obtained labeled data. However, in reality labeled data can be scarce or expensive to obtain. Active Learning (AL) is a way to deal with the above problem by asking for the labels of the most “informative” data points. We propose a novel AL method based on wavelet analysis, which pertains especially to the large number of dimensions (i.e. examined genes) of microarray experiments. DNA Microarray expression experiments permit the systematic study of the correlation of the expression of thousands of genes. We have applied our method on such data sets with encouraging results. In particular we studied data sets concerning: Small Round Blue Cell Tumours (4 types), Leukemia (2 types) and Lung Cancer (2 types).

Semi-supervised methods use a small amount of labeled data as a guide to unsupervised techniques. Recent literature shows better performance of these methods with respect to totally unsupervised ones even with a small amount of side-information This fact suggests that the use of semi-supervised methods may be useful especially in very difficult and noisy tasks where little

a priori

information is available. This is the case of biological datasets’ classification. The two more frequently used paradigms to include side-information into clustering are

Constrained Clustering

and

Metric Learning

. In this paper we use a

Metric Learning

approach as a preliminary step to fuzzy clustering and we show that

Semi-Supervised Fuzzy Clustering

(SSFC) can be an effective tool for classification of biological datasets. We used three real biological datasets and a generalized version of the Partition Entropy index to validate our results. In all cases tested the metric learning step produced a better highlight of the datasets’ clustering structure.

A combination of gene ranking, dimensional reduction, and recursive feature elimination (RFE) using a BP-MLP artificial neural network (ANN) was used to select genes for DNA microarray classification. Use of k-means cluster analysis for dimensional reduction and maximum sensitivity for RFE resulted in 64-gene models with fewer invariant and correlated features when compared with PCA and mimimum error. In conclusion, k-means cluster analysis and sensitivity may be better suited for classifying diseases for which gene expression is more strongly influenced by pathway heterogeneity.

An open problem in gene expression data analysis is the evaluation of the performance of gene selection methods applied to discover biologically relevant sets of genes. The problem is difficult, as the entire set of genes involved in specific biological processes is usually unknown or only partially known, making unfeasible a correct comparison between different gene selection methods. The natural solution to this problem consists in developing an artificial model to generate gene expression data, in order to know in advance the set of biologically relevant genes. The models proposed in the literature, even if useful for a preliminary evaluation of gene selection methods, did not explicitly consider the biological characteristics of gene expression data. The main aim of this work is to individuate the main biological characteristics that need to be considered to design a model for validating gene selection methods based on the analysis of DNA microarray data.

We present an application of BioDCV, a computational environment for semisupervised profiling with Support Vector Machines, aimed at detecting outliers and deriving informative subtypes of patients with respect to pathological features. First, a sample-tracking curve is extracted for each sample as a by-product of the profiling process. The curves are then clustered according to a distance derived from Dynamic Time Warping. The procedure allows identification of noisy cases, whose removal is shown to improve predictive accuracy and the stability of derived gene profiles. After removal of outliers, the semisupervised process is repeated and subgroups of patients are specified. The procedure is demonstrated through the analysis of a liver cancer dataset of 213 samples described by 1 993 genes and by pathological features.

We propose a new method for the construction of nearest prototype classifiers which is based on a Gaussian mixture approach interpreted as an annealed version of Learning Vector Quantization. Thereby we allow the adaptation of the underling metric which is useful in proteomic research. The algorithm performs a gradient descent on a cost function adapted from soft nearest prototype classification. We investigate the properties of the algorithm and assess its performance on two clinical cancer data sets. Results show that the algorithm performs reliable with respect to alternative state of the art classifiers.

Computing methods that allow the efficient and accurate processing of experimentally gathered data play a crucial role in biological research. The aim of this paper is to present a supervised learning strategy which combines concepts stemming from coding theory and Bayesian networks for classifying and predicting pathological conditions based on gene expression data collected from micro-arrays. Specifically, we propose the adoption of the Minimum Description Length (MDL) principle as a useful heuristic for ranking and selecting relevant features. Our approach has been successfully applied to the Acute Leukemia dataset and compared with different methods proposed by other researchers.

The inherently multidimensional problem of evaluating the complexity of an image is of a certain relevance in both computer science and cognitive psychology. Computer scientists usually analyze spatial dimensions, to deal with automatic vision problems, such as feature-extraction. Psychologists seem more interested in the temporal dimension of complexity, to explore attentional models. Is it possible, by merging both approaches, to define an more general index of visual complexity? We have defined a fuzzy mathematical model of visual complexity, using a specific entropy function; results obtained by applying this model to pictorial images have a strong correlation with ones from an experiment with human subjects based on variation of subjective temporal estimations associated with changes in visual attentional load, which is also described herein.

A new method that automatically detects and segments brain tumors in 3D MR images is presented. An initial detection is performed by a fuzzy possibilistic clustering technique and morphological operations, while a deformable model is used to achieve a precise segmentation. This method has been successfully applied on five 3D images with tumors of different sizes and different locations, showing that the combination of region-based and contour-based methods improves the segmentation of brain tumors.

Humans are very efficient in the analysis, exploration and representation of their environment. Based on the neurobiological and cognitive principles of human information processing, we develop a system for the automatic identification and exploration of spatial configurations. The system sequentially selects "informative" regions (regions of interest), identifies the local structure, and uses this information for drawing efficient conclusions about the current scene. The selection process involves low-level, bottom-up processes for sensory feature extraction, and cognitive top-down processes for the generation of active motor commands that control the positioning of the sensors towards the most informative regions. Both processing levels have to deal with uncertain data, and have to take into account previous knowledge from statistical properties and learning. We suggest that this can be achieved in a hybrid architecture which integrates a nonlinear filtering stage modelled after the neural computations performed in the early stages of the visual system, and a cognitive reasoning strategy that operates in an adaptive fashion on a belief distribution.

Nowadays, the processing of KANSEI information is very important in intelligent computing field. Particularly, it is very interesting in image retrieval to deal with human’s KANSEI. In this paper, we use natural language for the representation of KANSEI, including the image structure of Human’s idea, which we can not observe. And then, a KANSEI-Adjective is used as a natural language querying method: In other words, this paper presents the image retrieval based on KANSEI. We propose the background image retrieval based on KAC (KANSEI-Adjective of Color) to represent the sensibility of color. Our method for processing of KANSEI information is the measure of similarity by using the adaptive Lesk algorithm in WordNet. In our experimental results, we are able to retrieve background images with the most appropriate color in term of the query’s feeling. Furthermore, the method achieves an average rate of 63% user’s satisfaction.

In this paper we describe a new segmentation scheme to detect masses in breast radiographs.

Our segmentation method relies on the well known

fuzzy c-means

unsupervised clustering technique using an image representation scheme based on the local power spectrum obtained by a bank of Gabor filters.

We tested our method on 200 mammograms from the CALMA database. The detected regions have been validated by comparing them with the radiologist’s hand-sketched boundaries of real masses. The results, evaluated using ROC curve methodology, show that the greater flexibility and effectiveness provided by the fuzzy clustering approach benefit from an image representation that combine both intensity and local frequency information.

In this paper, we propose a new Chaotic MultiAgent System (CMAS) for image segmentation. This CMAS is a distributed system composed of a set of segmentation agents connected to a coordinator agent. Each segmentation agent performs Iterated Conditional Modes (ICM) starting from its own initial image created initially from the observed one by using a chaotic mapping. However, the coordinator agent receives and diversifies these images using a crossover and a chaotic mutation. A chaotic system is successfully used in order to benefit from the special chaotic characteristic features such as ergodic property, stochastic aspect and dependence on initialization. The efficiency of our approach is shown through experimental results.

We establish in this paper the link between the two main approaches for fuzzy mathematical morphology, based on duality with respect to complementation and on the adjunction property, respectively. We also prove that the corresponding definitions of fuzzy dilation and erosion are the most general ones if a set of classical properties is required.

We propose an innovative segmentation algorithm based on mathematical morphology operators. This definition is based on a morphological and fuzzy pattern-matching approach, and consists in comparing an object to a fuzzy landscape representing the degree of satisfaction of an affinity relationship. It has good formal properties, it is flexible, it fits the intuition, and it can be used for structural pattern recognition under imprecision. Moreover, it also applies in 3D and for fuzzy objects issued from images.

Document image analysis represents one of the most relevant topics in the field of image processing: many research efforts have been devoted to devising automatic strategies for document region classification. In this paper, we present a peculiar strategy to extract numerical features from segmented image regions, and their employment for classification purposes by means of the KERNEL system, a particular neuro-fuzzy framework suitable for application in predictive tasks. The knowledge discovery process performed by KERNEL proved to be effective in solving the problem of distinguishing between textual and graphical components of a document image. The information embedded into sample data is organised in form of a fuzzy rule base, which results to be accurate and comprehensible for human users.

Intelligent Knowledge Capsule was designed for the functions of knowledge acquisition,Memory retention and Knowledgeretrieval. Specially in this paper, focusing on Knowledge Retrieval process Associative Priming Knowledge Extraction mechanism is designed. A hierarchical associative memory including long term memory, short term memory and synonym net for keyword is established and using this structure Associative Priming Knowledge Extraction mechanism is processed. We apply this mechanism to virtual memory and test the retrieving process.

As the information circumstance is getting more complicated, the requirements for implementing the efficient intelligent system adopting human brain functions is getting high. We focus on the function of Reticular Activating System which takes charge of information selection.In this paper we designed Reticular Activating System with Positive/Negative masking in the associative memory and Thinking chain extraction mechanism specially implemented for flexible memory structure. The proposed Reticular Activating system has Knowledge acquisition, selection, storing, reconfiguration and retrieving part. P/N masking mechanism for flexible memory is specially designed and tested with virtual memory.

Biological system has a very efficient immunity system which selects important signals and protects its body. The functions of immunity system can be successfully adopted to design an intelligent system in the information society. Accordingly in this paper Immunity based system which can select the important data from a large amount data is proposed . we define filtering factor as a criterion for reacting and selecting the data. This system is designed to have learning, perception & inference and Data extraction and to have an additive learning mechanism for the new obtained important information. This system is applied to the area for the analysis of customer’s tastes and its performance is analyzed and compared

Meta-learning practices concern the dynamical search of the bias presiding over the behaviour of artificial learning systems. In this paper we present an original meta-learning framework, namely the

Mindful

(Meta INDuctive neuro-FUzzy Learning) system.

Mindful

is based on a neuro-fuzzy learning strategy providing for the inductive processes applicable both to ordinary base-level tasks and to more general cross-task applications. The peculiar organisation of the system allows a suitable meta-knowledge management, in order to carry on meta-learning investigations and to develop life-long learning strategies.

The aim of this work is to prove the efficacy of a Soft Computing approach to the problem of generating the best suited mesh for solving a differential problem with the Finite Element Method. Using Fuzzy Logic, it is possible to introduce a set of linguistic

if-then

rules reproducing the human expert reasoning used for creating the mesh.

The two dimensional systolic array for multiplication in binary field

GF

(2

m

) with LSB (Least Significant Bit) first algorithm proposed by Yeh et al. has the unfavorable property of bidirectional data flows compared with that of Wang and Lin which use MSB (Most Significant Bit) first algorithm. In this paper, by using a polynomial basis with LSB first algorithm, we present an improved bit parallel systolic array over

GF

(2

m

). Our two dimensional systolic array has unidirectional data flows with 7 latches in each basic cell. Therefore our systolic array has a shorter critical path delay and has the same unidirectional data flows to the multipliers with MSB first scheme.

We propose a new linear array for multiplication in

GF

(2

m

) which outperforms most of the existing linear multipliers in terms of the area and time complexity. Moreover we will give a very detailed comparison of our array with other existing architectures for the five binary fields

GF

(2

m

),

m

= 163,233,283,409,571, recommended by NIST for elliptic curve cryptography.