The article discusses the problem of automating the process of forwarding information objects and choosing the optimal executor in adaptive information systems. To solve it, a neural network method of automatic data forwarding is proposed, characterized by the use of neural networks to determine the optimal executor of an operation based on a forecast of time and the probability of its successful completion. Within the framework of the method, priority matrixes of operations and competencies of executors are formulated, which allowed ranking the operations and comparing them with the level of training of executors. The method is formalized in the notation of set theory. An algorithm for the practical implementation of the method in the form of a service application based on the Keras and Flask libraries is proposed, which will integrate it into various information systems.
Paper
Full text
Neural Network Method of Automatic Data Forwarding in Adaptive Information Systems
Semantic Scholar · Computer Science · 2020
Abstract
The article discusses the problem of automating the process of forwarding information objects and choosing the optimal executor in adaptive information systems. To solve it, a neural network method of automatic data forwarding is proposed, characterized by the use of neural networks to determine the optimal executor of an operation based on a forecast of time and the probability of its successful completion. Within the framework of the method, priority matrixes of operations and competencies of executors are formulated, which allowed ranking the operations and comparing them with the level of training of executors. The method is formalized in the notation of set theory. An algorithm for the practical implementation of the method in the form of a service application based on the Keras and Flask libraries is proposed, which will integrate it into various information systems.