Reward–Penalty Assignments and Genetic Algorithms for Ordinal Interval Number Group Decision Making

Ordinal interval number group decision making (GDM) is implemented using the closed interval preferences provided by the decision makers (DMs) for the potential courses of action. Two reward–penalty assignments (RPAs) are introduced, which express the compatibility between the ordinal intervals provided by the DMs in a purely binary and a fully graded manner, respectively. Genetic algorithms are employed for establishing a collectively preferred ranking, with chromosome‐validity‐enforcing crossover and gene‐swapping mutation being appropriately combined for efficient convergence. The effect that the consensus function, as derived from two RPAs, has on the fitness landscape and, consequently, on the preferred course(s) of action is illustrated in a number of GDM problems derived from the relevant literature covering (a) complete and incomplete problems, (b) perturbations of the relative importance values assigned to the DMs, (c) full and partial preferences, and (d) increasing numbers of DMs and/or available courses of action (scaling potential).     

Using geometrical information for accurate scene understanding in an artificial vision system

The scene understanding stage of an artificial vision system, which is constructed on the principles of the line-based approach of artificial intelligence, is presented. This receives a line-drawing representation of the viewed 3-D scene and produces its structural and depth analysis (e.g., characterization of the surfaces as background or foreground regions and of the edges as terminating, convex or concave parts of the objects, discovery of surface orientation, relative depth and tilt of the edges). The scene understanding stage comprises three processes: (1) Line-drawing encoding—The nontrivial tasks of vertex classification and line discrimination of the line-drawing are performed, whereas the line-drawing is represented in an accurate, compact, and uniform manner. The clockwise order is assigned to the lines connected to each vertex. Encoding employs geometrical information of the lines in the line-drawing and is realized in an elegant (flexible and computationally economic) fashion which can be easily applied to any domain. (2) Initial analysis—Region segmentation, interior/exterior line categorization, discovery of the inside/outside relations of the regions separated along contours of exterior lines as well as region grouping are accomplished employing a combination of the enhancement technique and the boundary stroll. The initial analysis reduces the complexity and ambiguity of the ensuing line-drawing analysis. (3) Line-drawing analysis—A Harmony Theory artificial neural network implementation of a combination of artificial intelligence line-labeling schemes is utilized. This executes parallel constraint propagation and outputs the labels for the lines of the line-drawing that produce the most globally coherent interpretation of the structure and depth relations in the 3-D scene. © 1996 John Wiley & Sons, Inc.     

Non-invasive on-line two-phase flow regime identification employing artificial neural networks

A novel non-invasive approach to the on-line identification of BWR two-phase flow regimes is investigated. The proposed approach receives neutron radiography images of coolant flow recordings as its input and performs feature extraction on each image via simple and directly computable statistical operators. The extracted features are subsequently used as inputs to an ensemble of self-organizing maps whose outputs demonstrate swift and accurate classification of each image into its corresponding flow regime. The novelty of the approach lies in the use of the self-organizing map which generates the different classes by itself, according to feature similarity of the corresponding images; this contrasts traditional artificial neural networks where the user has to define both the number of distinct classes as well as to supply separate training vectors for each class.     

A Methodology for Creating a Segment Inventory for Greek Time Domain Speech Synthesis

This article focuses on the systematic design of a segment database which has been used to support a time-domain speech synthesis system for the Greek language. Thus, a methodology is presented for the generation of a corpus containing all possible instances of the segments for the specific language. Issues such as the phonetic coverage, the sentence selection and iterative evaluation techniques employing custom-built tools, are examined. Emphasis is placed on the comparison of the process-derived corpus to naturally-occurring corpora with respect to their suitability for use in time-domain speech synthesis. The proposed methodology generates a corpus characterised by a near-minimal size and which provides a complete coverage of the Greek language. Furthermore, within this corpus, the distribution of segmental units is similar to that of natural corpora, allowing for the extraction of multiple units in the case of the most frequently-occurring segments. The corpus creation algorithm incorporates mechanisms that enable the fine-tuning of the segment database's language-dependent characteristics and thus assists in the generation of high-quality text-to-speech synthesis.     

A novel artificial neural network for sorting

An artificial neural network (ANN) is employed for sorting a sequence of real elements in monotonic (descending or ascending) order. Although inspired by harmony theory (HT), whereby the same construction as for the HT ANN is followed, the proposed ANN differs in the mode of operation, namely the obliteration of the consensus (harmony) function, the circumvention of simulated annealing as a means of settling to a solution, the simplification of the activation updating of the nodes of the upper layer, the clamping of the nodes of the lower layer, the gradual shrinking of the ANN and the use of an automatic termination criterion. The creation of the sorted sequence is progressive, whereby at most as many network updates are required as there are elements in the sequence. Ties between elements are resolved by simultaneous activation of the corresponding nodes. Finally, the min and max problems are solved in a single network update.     

Artificial neural networks for neutron source localization within sealed tanks

A modular back-propagation ANN has been implemented for the non-destructive localization of a source of Even Plutonium Isotopes (EPI) contained in sealed tanks. The ANN has been trained on data obtained from a simulation of a well counter (filtered and Fourier transformed signals of the neutron detectors surrounding the well counter) for known positions of the EPI. After training, the ANN can predict the position of EPI within sealed tanks from the corresponding detector signals. The introduction of median and majority ANNs has been found to significantly improve the accuracy of prediction. Furthermore, these ANNs perform in a satisfactory manner when noise is injected to the detector signals; prediction is corrupted in a manner which is directly related to the extent and amount of noise.

The motivation for using back-propagation ANNs is twofold: on one hand (theoretical importance), they are capable of learning to approximate complex functions such as the strongly non-linear relation that exists between the neutron detector signals and the EPI position; on the other hand, they accomplish on-line localization which is of practical interest.     

A simulated annealing artificial neural network implementation of the n-queens problem

A Harmony Theory artificial neural network implementation of the n-queens problem is presented in this piece of research. The problem is encoded in the two layers of the artificial neural network in such a manner that the inherent constraints of the problem are made directly available. Subsequently, during the simulated annealing procedure of Harmony Theory, maximal constraint satisfaction is accomplished in parallel and an optimal solution of the n-queens problem is produced. This solution indicates the appropriate locations of the greatest possible number of queens that can be placed on the n × n chessboard in a valid configuration, i.e., so that no queen threatens or is threatened by another queen. The proposed parallel implementation of the n-queens problem, combined with the application of the simulated annealing procedure, offers an interesting alternative to existing techniques (e.g., search, constraint propagation) in terms of optimality as well as computational and time efficiency. © 1997 John Wiley & Sons, Inc.     

Improving the clustering performance of the scanning n-tuple methodby using self-supervised algorithms to introduce subclasses

The scanning n-tuple technique (as introduced by Lucas and Amiri, 1996) is studied in pattern recognition tasks, with emphasis placed on methods that improve its recognition performance. We remove potential edge effect problems and optimize the parameters of the scanning n-tuple method with respect to memory requirements, processing speed and recognition accuracy for a case study task. Next, we report an investigation of self-supervised algorithms designed to improve the performance of the scanning n-tuple method by focusing on the characteristics of the pattern space. The most promising algorithm is studied in detail to determine its performance improvement and the consequential increase in the memory requirements. Experimental results using both small-scale and real-world tasks indicate that this algorithm results in an improvement of the scanning n-tuple classification performance     

An artificial neural network satisfiability tester

An artificial neural network tester for the satisfiability problem of propositional calculus is presented. Satisfiability is treated as a constraint satisfaction optimization problem and, contrary to most of the existing satisfiability testers, the expressions are converted into disjunctive normal form before testing. The artificial neural network is based on the principles of harmony theory. Its basic characteristics are the simulated annealing procedure and the harmony function; the latter constitutes a measure of the satisfiability of the expression under the current truth assignment to its variables. The tester is such that: (a) the satisfiability of any expression is determined; (b) a truth assignment to the variables of the expression is output which renders true the greatest possible number of clauses; (c) all the truth assignments which render true the maximum number of clauses can be produced. © 2001 John Wiley & Sons, Inc.