Translating High Level End User Intention Using Machine Learning for Triggering Context Aware Services

Abstract

At present, there is an emerging demand for self-managed services that can evolve at run time in response to unpredictable changes, adaptable to changing user requirements and can adapt system to overcome violations of timing and resource constraints at run time. Several approaches have been proposed to introduce such adaptive services but still those solutions are domain specific and only focus on isolated features. Also we can see, many solutions consider software services but failed to integrate things like IOT. For managing the behaviors of service dynamically, it is necessary to have an appropriate adaptable model for specifying a collaboration of services. We propose an architecture for flexible and adaptive execution of selfmanaged service based process integrated with IOT devices. Our model supports both process design and execution. The model can change its behavior according to user requirement and adapt with the run time environment. This model implicitly acquires user requirement data and use these data with explicitly gathered data (from IOT devices) to generate insight of the proper requirement with the help of machine learning approaches to generate problem specific output.