An approach for self-optimization of multimedia content adaptation

Over the past few years, researchers have developed adaptive systems to overcome the various constraints of mobile environments and to offer web content tailored to the characteristics of a user’s device. However, most of the existing research has produced systems that prepare different versions of web content in advance by considering the client types that may access the server or by describing the rules for dynamic content conversion. These methods require much effort from the content creator or the system developer, and prospecting the appearance of a new device is difficult work in today’s rapidly changing computing environment. This paper proposes a system that intelligently adapts web content by dynamically converting it into a form suitable for the specific characteristics of a user’s device while conversion rules are adjusted according to user feedback. The system can reduce the workload for content creators or system managers, and it can also handle unforeseen situations where no rule has been defined. We implemented a prototype and we conducted experiments with 120 users to evaluate the proposed system. As a result, the negative feedback decreased when the proposed system was in use. Our results confirm that the proposed method is effective and has potential for wider use.