scholarly journals Application of a Multi-Agent System for Risk Assessment with Fuzzy Background Information

The article offers the results of a study of various multiagent systems on the example of a number of models and multiagent implementations for risk assessment with fuzzy initial information. General construction methods and issues are related to their behavior, criteria for the quality of system performance that highlighted. The regularities, interrelations between the properties and parameters used when specifying a multi-agent intelligent system are defined. Developed approaches for processing complexly structured information. Algorithms for constructing a multi-agent intelligent risk assessment system have been developed.

The article offers the results of a study of various multiagent systems on the example of a number of models and multiagent implementations for risk assessment with fuzzy initial information. General construction methods and issues are related to their behavior, criteria for the quality of system performance that highlighted. The regularities, interrelations between the properties and parameters used when specifying a multi-agent intelligent system are defined. Developed approaches for processing complexly structured information. Algorithms for constructing a multi-agent intelligent risk assessment system have been developed.


2008 ◽  
Vol 144 ◽  
pp. 232-237
Author(s):  
Durmus Karayel ◽  
Sinan Serdar Ozkan ◽  
Fahri Vatansever

In this study, an intelligent system model that can evaluate experimental material properties and safety factors is developed. The model contains Artificial Intelligence Technologies such as Artificial Neural Network (ANN) and Fuzzy Logic. It consists of sub modules into interaction. Also, the model can obtain more precision values than interpolation techniques used to classical design. The study contributes to define safety factors, design criterions and safety stress according to a new approach based on information technologies. So, this study can be seen as one of the sub modules of Intelligence Multi Agent System and it can be integrated with Multi Agent System Model for design. Also, it can be used for classical design studies so that results can be quickly obtained. It is expected that this approach will be widely used by designers.


2020 ◽  
Vol 17 (5) ◽  
pp. 2035-2038
Author(s):  
E. Ajith Jubilson ◽  
Ravi Sankar Sangam

Metrics are the essential building blocks for any evaluation process. They establish specific goals for improvement. Multi agent system (MAS) is complex in nature, due to the increase in complexity of developing a multi agent system, the existing metrics are less sufficient for evaluating the quality of an MAS. This is due to the fact that agent react in an unpredictable manner. Existing metrics for measuring MAS quality fails to addresses potential communication, initiative behaviour and learn-ability. In this work we have proposed additional metrics for measuring the software agent. A software agent for online shopping system is developed and the metrics values are obtained from it and the quality of the multi agent system is analysed.


Ergodesign ◽  
2021 ◽  
Vol 2021 (1) ◽  
pp. 36-40
Author(s):  
Nataliya Sukhanova

The purpose of this work is to assess the quality of functioning intelligent systems. Tasks to be solved: synthesis of an intelligent system based on unified modules, system quality assessment, system reconfiguration at a quality decrease. The research method is system analysis. A new flexible programmable architecture of intelligent systems has been developed. The flexible architecture of an intelligent system allows you to change the mutual relationships between subsystems, components and modules. The intelligent system is implemented on the basis of the unified modules that contain programmable switches. Switches are connected to the system inputs and outputs and are networked to transmit information.


Author(s):  
Ольга Владимировна Шаталова ◽  
Дмитрий Андреевич Медников ◽  
Зейнаб Усама Протасова

Цель исследования заключается в повышении качества прогнозирования ишемической болезни сердца путем учета синергетического эффекта наличия сопутствующих заболеваний и факторов профессиональной среды посредством многоагентных интеллектуальных систем. Методы исследования. Для прогнозирования ишемической болезни сердца предложена базовая структура многоагентной интеллектуальной системы, содержащая «сильные» и «слабые» классификаторы. При этом «слабые» классификаторы разделены на четыре группы, первая из которых осуществляет анализ данных, полученных на основе традиционных факторов риска ишемической болезни сердца, вторая - на основе анализа электрокардиологических исследований, третья группа «слабых» классификаторов предназначена для диагностики сопутствующих заболеваний и синдромов по предикторам, используемых первыми двумя группами агентов, а четвертая - анализирует факторы риска окружающей среды. Мультиагентная система позволяет управлять процессом принятия решений посредством сочетания экспертных оценок, статистических данных и текущей информации. Результаты. Проведены экспериментальные исследования различных модификаций предложенной модели классификатора, заключающихся в последовательном исключении из агрегатора решений «слабых» классификаторов на различных иерархических уровнях. В ходе экспериментального оценивания и в результате математического моделирования было показано, что при использовании всех информативных признаков уверенность в правильном прогнозе по риску ишемической болезни сердца превышает величину 0,8. Показатели качества прогнозирования выше, чем у известной системы прогнозирования ишемической болезни сердца - превышает SCORE, в среднем, на 14%. Выводы. Анализ показателей качества классификации в экспериментальной группе обследуемых с различным показателем ишемического риска и в контрольной группе, составленной из машинистов электролокомотивов, для которых релевантными показателями ишемических рисков являются вибрационная болезнь и пребывание в электромагнитных полях, показал, что учет влияния этих факторов риска в контрольной группе повышает диагностическую эффективность на семь процентов по сравнению с экспериментальной группой, выступающей как фоновая The aim of the study is to improve the quality of predicting coronary heart disease by taking into account the synergistic effect of the presence of concomitant diseases and occupational factors through multi-agent intelligent systems. Research methods. To predict coronary heart disease, a basic structure of a multi-agent intelligent system is proposed, which contains “strong” and “weak” classifiers. At the same time, the "weak" classifiers are divided into four groups, the first of which analyzes data obtained on the basis of traditional risk factors for coronary heart disease, the second - based on the analysis of electrocardiological studies, the third group of "weak" classifiers is intended for the diagnosis of concomitant diseases and syndromes based on predictors used by the first two groups of agents, and the fourth analyzes environmental risk factors. The mobile system allows you to manage the decision-making process through a combination of expert assessments, statistical data and current information. Results. Experimental studies of various modifications of the proposed model of the classifier, consisting in the sequential exclusion from the aggregator of decisions of "weak" classifiers at various hierarchical levels, have been carried out. In the course of experimental evaluation and as a result of mathematical modeling, it was shown that when using all informative signs, the confidence in the correct forecast for the risk of coronary heart disease exceeds 0.8. The indicators of the quality of prediction are higher than those of the known predictive system for coronary heart disease - they exceed SCORE, on average, by 14%. Conclusions. Analysis of the classification quality indicators in the experimental group of subjects with different ischemic risk indicators and in the control group made up of electric locomotive drivers, for whom vibration sickness and exposure to electromagnetic fields are relevant indicators of ischemic risks, showed that taking into account the influence of these risk factors in the control group increases diagnostic efficiency by seven percent compared with the experimental group serving as background


Sign in / Sign up

Export Citation Format

Share Document