Extracting dynamic adaptations from the context through reinforcement learning

Context-aware dynamic adaptive systems have the peculiarity of adapting their behavior according to situations gathered from their surrounding environment, for example, information gathered from user actions. However, the larger the system is, the higher the likelihood of situations with multiple possible adaptations to the system base behavior, for such systems foreseeing all possible situations is unfeasible, especially if user interaction is involved. In this thesis, we explore a reinforcement learning approach, to extract these situations, where we validate the thesis with the development of a prototype of a web dynamic public urban transport system.

• Autores: Castro Villamizar, Jorge Humberto
• Fecha de publicación: 2018
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: eng
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/34625
• Acceso en línea: http://hdl.handle.net/1992/34625
• Palabra clave: Sistemas autoadaptativos; Aprendizaje por refuerzo (Aprendizaje automático); Ingeniería de software; Ingeniería
• Rights: openAccess
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/