Analysis and Evaluation of Using Neural Network and Genetic Algorithm

Due to the increasing volume of information and the complexity of engineering, social and economic systems, it has become difficult to evaluate input data and manage these systems correctly. Current decision support systems (DSS) strive to achieve optimal outcomes while minimizing the risks of serious casualties. The purpose of this DSS is to help the decision maker who is faced with the problem of huge amount of data and ambiguous reactions of complex systems depending on external factors. By using detailed and in-depth analysis, DSSs are expected to provide users with accurate predictive indicators and foolproof decisions. In this paper, we propose a new DSS structure that can be used in a wide range of difficult to formal tasks and achieve high speed of deliberation and decision making. We evaluate different approaches to determining the dependence of a target variable on input data and review the most common methods of statistical prediction. The advantages of using neural networks for this purpose are described. We propose the use of intermediate neural networks for computations with intermediate data, which allows the user to use our DSS in a wide range of complex tasks. We also developed a corresponding learning algorithm for intermediate neural networks. The advantages of using a genetic algorithm (GA) to select the most meaningful inputs are shown. We justify the use of general-purpose computing on graphics processing units (GPGPUs) to achieve high-speed computing with a high-speed support system in the query. A functional diagram of the system is also presented and described. Results and examples of using DSS are shown.

کلیدواژه ها:

Support System ، Neural Network ، Genetic Algorithm