A Mobile- Cloud Framework for Context-aware and Mobility-driven Recommender for Smart Markets

Abstract

Context and challenges: In recent years, smart city systems have emerged as solutions that transform the conventional cities and societies into information and communication (ICT) driven cities. Smart city systems o er improved and digitized urban services such as smart health and smart shopping to the stakeholders such as citizens, business entities and organizations. However, in mobile-cloud based smart city systems, challenges such as context-awareness, performance of cloud servers and the resource poverty of mobile devices must be addressed. Solution and implications : In this thesis, we focus on the integration of the mobile computing and cloud computing technologies to develop a system that o ers its users with context- aware and mobility-driven recommender system for smart markets. In the proposed solution, a mobile device represents the front-end (mobility and context-aware) interface for recommendations, while; cloud-based server represents the back-end (computation-intensive) processor of the system to enable digital match making between potential customer and business entities.. Evaluations and conclusions : We have utilized the ISO-IEC-9126 quality model to evaluate the accuracy and e ciency of the proposed recommender systems. Considering the resource poverty of a mobile device, the evaluation results suggest the accuracy, along with computational and energy e ciency of the recommender system. The future research is focused on the application of machine learning approaches for an intelligent, context-aware recommendation system.