scholarly journals Adapting Neural Turing Machines for linguistic assessments aggregation in neural-symbolic decision support systems

2021 ◽  
pp. 40-50
Author(s):  
Alexander Demidovskij ◽  
Eduard Babkin
Keyword(s):  
Neural Network ◽  
Decision Support ◽  
Decision Support Systems ◽  
Network Architecture ◽  
Support Systems ◽  
Distributed Representation ◽  
Numerical Representation ◽  
Problem Situation ◽  
Tensor Representations ◽  
Linguistic Assessment

Introduction: The construction of integrated neurosymbolic systems is an urgent and challenging task. Building neurosymbolic decision support systems requires new approaches to represent knowledge about a problem situation and to express symbolic reasoning at the subsymbolic level.  Purpose: Development of neural network architectures and methods for effective distributed knowledge representation and subsymbolic reasoning in decision support systems in terms of algorithms for aggregation of fuzzy expert evaluations to select alternative solutions. Methods: Representation of fuzzy and uncertain estimators in a distributed form using tensor representations; construction of a trainable neural network architecture for subsymbolic aggregation of linguistic estimators. Results: The study proposes two new methods of representation of linguistic assessments in a distributed form. The first approach is based on the possibility of converting an arbitrary linguistic assessment into a numerical representation and consists in converting this numerical representation into a distributed one by converting the number itself into a bit string and further forming a matrix storing the distributed representation of the whole expression for aggregating the assessments. The second approach to translating linguistic assessments to a distributed representation is based on representing the linguistic assessment as a tree and coding this tree using the method of tensor representations, thus avoiding the step of translating the linguistic assessment into a numerical form and ensuring the transition between symbolic and subsymbolic representations of linguistic assessments without any loss of information. The structural elements of the linguistic assessment are treated as fillers with their respective positional roles. A new subsymbolic method of aggregation of linguistic assessments is proposed, which consists in creating a trainable neural network module in the form of a Neural Turing Machine. Practical relevance: The results of the study demonstrate how a symbolic algorithm for aggregation of linguistic evaluations can be implemented by connectionist (or subsymbolic) mechanisms, which is an essential requirement for building distributed neurosymbolic decision support systems.

scholarly journals Optimization of parameters of neuro-fuzzy model

2020 ◽  
Vol 17 (3) ◽  
pp. 1206
Author(s):  
Lyalya Bakievna Khuzyatova ◽  
Lenar Ajratovich Galiullin
Keyword(s):  
Neural Network ◽  
Decision Support ◽  
Decision Support Systems ◽  
Fuzzy Neural Network ◽  
Support Systems ◽  
Fuzzy Model ◽  
Knowledge Bases ◽  
Fuzzy Neural Networks ◽  
Neural Network Models ◽  
Fuzzy Neural

<p>The questions and problems of the formation of knowledge bases of intelligent man-machine decision support systems are considered. The neuron-fuzzy model used in the work is described. The need for increasing the efficiency of the neuron-fuzzy model in the formation of knowledge bases is being updated. The task is to develop methods and algorithms for presetting and optimizing the parameters of a fuzzy neural network. To solve difficult formalized tasks, it is necessary to develop decision support systems - expert systems based on a knowledge base. ES developers are constantly faced with the problems of “extraction” and formalization of knowledge, as well as the search for new ways to obtain it. To do this, use the extraction, acquisition and formation of knowledge. Currently, the formation of knowledge bases is relevant for the creation of hybrid technologies - fuzzy neural networks that combine the advantages of neural network models and fuzzy systems. The analysis of the efficiency of the fuzzy neural network carried out in the work showed that the quality of training of the NN largely depends on the choice of the number of fuzzy granules for input drugs. In addition, to use fuzzy information formalized by the mathematical apparatus of fuzzy logic, procedures are required for selecting optimal forms and presetting the parameters of the corresponding membership functions (MF).</p>

scholarly journals Improvement of the method of estimation and forecasting of the state of the monitoring object in intelligent decision support systems

2021 ◽  
Vol 4 (3(112)) ◽  
pp. 43-55
Author(s):  
Areej Adnan Abed ◽  
Iurii Repilo ◽  
Ruslan Zhyvotovskyi ◽  
Andrii Shyshatskyi ◽  
Spartak Hohoniants ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Decision Support ◽  
Decision Support Systems ◽  
Support Systems ◽  
The State ◽  
Intelligent Decision ◽  
Intelligent Decision Support Systems ◽  
Artificial Neural ◽  
Intelligent Decision Support

In order to objectively and completely analyze the state of the monitored object with the required level of efficiency, the method for estimating and forecasting the state of the monitored object in intelligent decision support systems was improved. The essence of the method is to provide an analysis of the current state of the monitored object and short-term forecasting of the state of the monitored object. Objective and complete analysis is achieved using advanced fuzzy temporal models of the object state, taking into account the type of uncertainty and noise of initial data. The novelty of the method is the use of an improved procedure for processing initial data in conditions of uncertainty, an improved procedure for training artificial neural networks and an improved procedure for topological analysis of the structure of fuzzy cognitive models. The essence of the training procedure is the training of synaptic weights of the artificial neural network, the type and parameters of the membership function and the architecture of individual elements and the architecture of the artificial neural network as a whole. The procedure of forecasting the state of the monitored object allows for multidimensional analysis, accounting and indirect influence of all components of the multidimensional time series with their different time shifts relative to each other in conditions of uncertainty. The method allows increasing the efficiency of data processing at the level of 12–18 % using additional advanced procedures. The proposed method can be used in decision support systems of automated control systems (ACS DSS) for artillery units, special-purpose geographic information systems. It can also be used in ACS DSS for aviation and air defense and ACS DSS for logistics of the Armed Forces of Ukraine

Soft Computing Paradigms and Regression Trees in Decision Support Systems

2011 ◽  
pp. 1011-1035 ◽  
Author(s):  
Cong Tran ◽  
Ajith Abraham ◽  
Lakhmi Jain
Keyword(s):  
Neural Network ◽  
Decision Support ◽  
Decision Support Systems ◽  
Soft Computing ◽  
Intelligent Systems ◽  
Support Systems ◽  
Regression Trees ◽  
Gradient Algorithm ◽  
Neural Network Learning ◽  
Intelligent Decision Support Systems

Decision-making is a process of choosing among alternative courses of action for solving complicated problems where multi-criteria objectives are involved. The past few years have witnessed a growing recognition of soft computing (SC) (Zadeh, 1998) technologies that underlie the conception, design, and utilization of intelligent systems. In this chapter, we present different SC paradigms involving an artificial neural network (Zurada, 1992) trained by using the scaled conjugate gradient algorithm (Moller, 1993), two different fuzzy inference methods (Abraham, 2001) optimised by using neural network learning/evolutionary algorithms (Fogel, 1999), and regression trees (Breiman, Friedman, Olshen, & Stone, 1984) for developing intelligent decision support systems (Tran, Abraham, & Jain, 2004). We demonstrate the efficiency of the different algorithms by developing a decision support system for a tactical air combat environment (TACE) (Tran & Zahid, 2000). Some empirical comparisons between the different algorithms are also provided.

Decision Support and Problem Formulation Activity

2011 ◽  
pp. 192-199 ◽  
Author(s):  
David Paradice
Keyword(s):  
Decision Support ◽  
Decision Support Systems ◽  
Decision Maker ◽  
Support Systems ◽  
Problem Formulation ◽  
Decision Makers ◽  
Problem Situation ◽  
Continue Development ◽  
Decisionmaking Process ◽  
Decision Choices

While decision choices are certainly important and warrant appropriate attention, early stages of the decisionmaking process may be even more critical in terms of needing adequate support. The alternatives from which a decision maker may be able to choose are integrally tied to the assumptions made about the problem situation. Consequently, decision support systems (DSSs) may be more effective in helping decision makers to make good choices when support for problem formulation is provided. Research validates the notion that support for problem formulation and structuring leads to better decisions. This article explores this concept and looks at opportunities in emerging software trends to continue development of problem formulation support in DSS-type settings.

Soft Computing Paradigms and Regression Trees in Decision Support Systems

2006 ◽  
pp. 1-28 ◽  
Author(s):  
Cong Tran ◽  
Ajith Abraham ◽  
Lakhmi Jain
Keyword(s):  
Neural Network ◽  
Decision Support ◽  
Decision Support Systems ◽  
Soft Computing ◽  
Intelligent Systems ◽  
Support Systems ◽  
Regression Trees ◽  
Gradient Algorithm ◽  
Neural Network Learning ◽  
Intelligent Decision Support Systems

Decision-making is a process of choosing among alternative courses of action for solving complicated problems where multi-criteria objectives are involved. The past few years have witnessed a growing recognition of soft computing (SC) (Zadeh, 1998) technologies that underlie the conception, design, and utilization of intelligent systems. In this chapter, we present different SC paradigms involving an artificial neural network (Zurada, 1992) trained by using the scaled conjugate gradient algorithm (Moller, 1993), two different fuzzy inference methods (Abraham, 2001) optimised by using neural network learning/evolutionary algorithms (Fogel, 1999), and regression trees (Breiman, Friedman, Olshen, & Stone, 1984) for developing intelligent decision support systems (Tran, Abraham, & Jain, 2004). We demonstrate the efficiency of the different algorithms by developing a decision support system for a tactical air combat environment (TACE) (Tran & Zahid, 2000). Some empirical comparisons between the different algorithms are also provided.

scholarly journals DEVELOPING AND TRAINING MODEL OF ARTIFICIAL NEURAL NETWORK FOR CREATING DECISION SUPPORT SYSTEMS

2020 ◽  
Vol 2020 (3) ◽  
pp. 61-73
Author(s):  
Anna Olegovna Chupakova ◽  
Sergey Vital'evich Gudin ◽  
Renat Shamil'evich Khabibulin
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Decision Support ◽  
Decision Support Systems ◽  
Support Systems ◽  
Training Data ◽  
Advantages And Disadvantages ◽  
Artificial Neural ◽  
And Training

The article highlights the significant increase of industrial capacities and automation of production, which requires taking effective management decisions by a responsible person. There have been outlined the important achievements of the scientists in application of the artificial neural networks in the various fields of activity and decision support systems involving the information analysis and processing with the results obtained. There has been proposed a review of publications on training artificial neural networks and on their efficient application in solving problems of classification, prediction and control. The most common structures of neural networks, their advantages and disadvantages, as well as the methods used to create training data arrays have been studied. A comparative analysis of using various structures of artificial neural networks and the effectiveness of existing teaching methods and the prospects for their use has been carried out. There has been defined the most preferred neural network topology for solving problems of fire safety management at the production facilities as an active decision support system. Using the analysis results, the most common and effective training methods have been identified, application of which is appropriate for developing and training various types of neural networks. The use of the technology is well grounded for reducing the errors in data processing, the financial costs for ensuring security, as well as for possible using the neural networks in the decision support systems to optimize these systems.

Sign in / Sign up

Export Citation Format

Share Document