Application of deep reinforced learning for long-term dynamic composition of WEB services

• Дата: 2025
• Автор: Moroz, H.B.
• Формат: Стаття
• Мова: Ukrainian
• Опубліковано: PROBLEMS IN PROGRAMMING 2025
• Теми: service-oriented computing; quality of service; web service composition; composite web service; deep reinforcement learning; UDC 004.41
• Онлайн доступ: https://pp.isofts.kiev.ua/index.php/ojs1/article/view/836
• Назва журналу: Problems in programming

In the service oriented architecture, software and systems are abstracted as web services to be invoked by other systems. Service composition is a technology, which builds a complex system by combining existing simple services. With the development of service oriented architecture and web service technology, massive web ser vices with the same function begin to spring up. These services are maintained by different organizations and have different quality of service. Thus, how to choose the appropriate service to make the whole system to deliver the best overall quality of service has become a key problem in service composition research. Furthermore, because of the complexity and dynamics of the network environment, quality of service may change over time. Therefore, how to dynamically configure the composition system to adapt to the changing environment and ensure the quality of the composed web service is another problem. To address the above challenges, we propose a service composition approach based on quality of web service rediction and reinforcement learning. Specifi cally, we use long short-term memory neural network to predict the quality of web service hrough reinforcement learning. This approach can be well adapted to a dynamic network environment.Prombles in programming 2025; 2: 40-53