Research Challenge of Locally Computed Ubiquitous Data Mining
Aysegul Cayci (Sabanci University, Turkey), João Bártolo Gomes (Universitad Politecnica, Spain), Andrea Zanda (Universitad Politecnica, Spain), Ernestina Menasalvas (Universitad Politecnica, Spain) and Santiago Eibe (Universitad Politecnica, Spain)
Copyright © 2011.
OnDemand Chapter PDF Download
Download link provided immediately after order completion
List Price: $37.50
Instant access upon order completion.
Advances in wireless, sensor, mobile and wearable technologies present new challenges for data mining research on providing mobile applications with intelligence. Autonomy and adaptability requirements are the two most important challenges for data mining in this new environment. In this chapter, in order to encourage the researchers on this area, we analyzed the challenges of designing ubiquitous data mining services by examining the issues and problems while paying special attention to context and resource awareness. We focused on the autonomous execution of a data mining algorithm and analyzed the situational factors that influence the quality of the result. Already existing solutions in this area and future directions of research are also covered in this chapter.
Issues, Controversies And Problems Of Ubiquitous Data Mining
The advances in wireless, sensor, mobile and wearable technologies affected substantially how and where data is accumulated, processed and analyzed. Data, which used to be entered to a central computer for processing from a limited number of end points, is now dominantly collected by incredible number of devices surrounding us. Ubiquitous nature of this new computing platform brings challenges to several information and computing technologies where data mining is one of them.
Key Terms in this Chapter
Ubiquitous: Device: A computing device that moves or is positioned in time and space, reacts in real-time and also can sense its environment and communicate with others.
Situation Awareness: Capability which encompasses resource-awareness and context-awareness.
Data Stream Mining: The process of knowledge extraction from continuous data.
Ubiquitous Data Mining: An in-device, real-time mining of data on a ubiquitous computing environment in accordance to the environment’s requirements by considering resource constraints of the device, exploiting context information, behaving autonomously and applying special privacy preserving methods.
Resource-Awareness: The capability of knowing the necessary resources for accomplishing its goals and deciding the configuration of its execution by considering the circumstances of these resources.
Autonomous Data Mining: The capability of taking data mining decisions independently by the control mechanism embedded.
Context-Awareness: The capability that enables to incorporate information sensed from the environment for self-configuration.