Integration of Artificial Intelligence Techniques for Test Automation in Context-Based Software Systems

Context-aware software systems are capable of perceiving, interpreting, and using information from the environment, the device, or the user to dynamically adjust their behavior. These types of systems are used in areas such as ubiquitous computing, smart environments, augmented reality, and adaptive mobile applications. However, the testing process for these systems faces significant challenges due to context variability, dependence on sensors and networks, and the difficulty of replicating real conditions in controlled environments. These challenges particularly affect context identification, the definition of suitability criteria, the adaptation of test cases, and automated execution.
Given this complexity, the integration of Artificial Intelligence (AI) techniques represents a promising alternative to support testing processes, as it allows for the incorporation of learning, prediction, and adaptation mechanisms in dynamic scenarios. This project aims to develop and implement AI-based strategies for test automation in context-based software systems in order to improve their efficiency, accuracy, and coverage.
As a starting point, a systematic literature review (SLR) is being conducted to identify the main trends, applied AI techniques, and research gaps in test automation for systems.
As a starting point, a systematic literature review (SLR) is being conducted to identify the main trends, applied AI techniques, and research gaps in the automation of context-sensitive system testing. The results of this review will serve as the basis for the construction of a conceptual framework and for the design of a strategy using AI models oriented towards the design, execution, and evaluation of test cases.
Subsequently, the proposed strategies will be implemented in controlled test environments and real-world scenarios in order to analyze their performance through case studies and comparative studies. The evaluation will be based on software quality metrics and user experience, with iterative adjustments to the models according to the experimental results.
It is estimated that the integration of AI into testing processes can increase the efficiency and adaptability of systems by facilitating task automation and dynamic management of variable contexts. This approach seeks to explore the feasibility of applying AI techniques to support stages of the verification and validation process, with the aim of moving towards more intelligent and flexible methodologies that strengthen the quality and reliability of context-based software systems.

Desarrollar una solución tecnológica basada en Inteligencia Artificial para la automatización de pruebas en sistemas de software basados en el contexto, que incremente la eficiencia, precisión y cobertura del proceso de verificación de calidad, contribuyendo a la reducción de costos y tiempos de desarrollo”.

INTELIGENCIA ARTIFICIAL