Abstract
Ambient systems weave computing and communication aspects into everyday life. To provide self-adaptive services, it is necessary to acquire context information using sensors and to leverage the collected information for reasoning and classification of situations. To enable self-learning systems, we propose to depart from static rule-based decisions and first-order logic to define situations from basic context, but to build on machine-learning techniques. However, existing learning algorithms show substantial weaknesses if applied in highly dynamic environments, where we expect accurate decisions in realtime while the user is in-the-loop to give feedback to the system’s recommendations. To address ambient and pervasive computing environments, we propose the FLORA—multiple classification (FLORA-MC) online learning algorithm. In particular, we enhance the FLORA algorithm to allow for (1) multiple classification and (2) numerical input values, while improving its concept drift handling capabilities; thus, making it an excellent choice for use in the area of ambient computing. The multiple classification allows context-aware systems to differentiate between multiple categories instead of taking binary decisions. Support for numerical input values enables the processing of arbitrary sensor inputs beyond nominal data. To provide the capability of concept drift handling, we propose the use of an advanced window adjustment heuristic, which allows FLORA-MC to continuously adapt to the user’s behavior, even if her/his preferences change abruptly over time. In combination with the inherent characteristics of online learning algorithms, our scheme is very well suited for realtime application in the area of ambient and pervasive computing. We describe the design and implementation of FLORA-MC and evaluate its performance vs. state-of-the-art learning algorithms. We are able to show the superior performance of our algorithm with respect to reaction time and concept drift handling, while maintaining an excellent accuracy. Our implementation is available to the research community as a WEKA module.
Similar content being viewed by others
Notes
Please note that the original FLORA algorithm performs the generalization differently: nominal values are directly removed from the DIs (we keep these nominal values and extend the intervals of the numerical values only). A weakness of FLORA’s simple removal strategy is the easy over-generalization of concepts, which leads to inferior classification performance.
References
Tsymbal A (2004) The problem of concept drift: definitions and related work. Department of Computer Science, Trinity College Dublin, Ireland, Technical Report TCD-CS-2004-15. https://www.cs.tcd.ie/publications/tech-reports/reports.04/TCD-CS-2004-15.pdf, April
Widmer G (1997) Tracking context changes through meta-learning. Mach Learn 27(3):259–286, June
Widmer G, Kubat M (1996) Learning in the presence of concept drift and hidden contexts. Mach Learn 23(1):69–101, April
The University of Waikato (2006) University of Waikato Weka Data Mining Software. http://www.cs.waikato.ac.nz/~ml/weka/
Görtz M (2005) Effiziente Echtzeit-Kommunikationsdienste durch Einbeziehung von Kontexten. Ph D dissertation, Technische Universität Darmstadt
Schmitt J, Hollick M, Steinmetz R (2007) Der Assistent im Hintergrund: Adaptives Kommunikationsmanagement durch Lernen vom Nutzer. PIK - Praxis der Informationsverarbeitung und Kommunikation 30(2):100–105, April–June
Schilit WN (1995) A system architecture for context-aware mobile computing. Ph D dissertation, Columbia University
Dey A, Abowd G, Salber D (1999) A context-based infrastructure for smart environments. In: Proceedings of the 1st international workshop on managing interactions in smart environments (MANSE ’99), pp 114–128
Dey A (2000) Providing architectural support for building context-aware applications. Ph.D. dissertation, Georgia Institute of Technology
Schmidt A, Aidoo KA, Takaluomai A, Tuomelai U, Laerhoven KV, de Velde WV (1999) Advanced interaction in context. In: Lecture notes in computer science 1707
Schmidt A (2002) Ubiquitous computing—computing in context. PhD dissertation, Lancaster University
Amigo (2008) Amigo—Ambient intelligence for the networked home environment. European Project, Information Society Technologies. http://www.hitech-projects.com/euprojects/amigo/
Ambient Networks (2006) Ambient Networks European Project, Information Society Technologies. http://www.ambient-networks.org/
Littlestone N (1988) Learning quickly when irrelevant attributes abound: a new linear-threshold algorithm. Mach Learn 2(4):285–318, April
Maloof MA, Michalski RS (2000) Selecting examples for partial memory learning. Mach Learn 41(1):27–52
Elio R, Watanabe L (1991) An incremental deductive strategy for controlling constructive induction in learning from examples. Mach Learn 7(1):7–44, July
Iba W, Woogulis J, Langley P (1988) Trading simplicity and coverage in incremental concept learning. In: Proceedings of the fifth international conference on machine learning. Morgan Kaufmann, San Francisco, pp 73–79
Flora-MC - Implementation as WEKA Module (2007) http://www.kom.tu-darmstadt.de/en/downloads/software/flora-mc/
Roos C (2006) Adaptation and extension of FLORA for application within a self-learning communication service. Student Thesis, Technische Universität Darmstadt
Schlimmer JC, Granger RH (1986) Beyond incremental processing: tracking concept drift. In: Proceedings of fifth national conference on artificial intelligence—AAAI, pp 502–507
Kolter JZ, Maloof MA (2003) Dynamic weighted majority: a new ensemble method for tracking concept drift. In: Proceedings of third IEEE international conference on data mining (ICDM 2003), Washington, DC, USA, pp 123–130, November
Acknowledgements
The authors would like to thank the guest editors and the anonymous reviewers for providing valuable comments that aided to improve the quality of the manuscript.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Schmitt, J., Hollick, M., Roos, C. et al. Adapting the User Context in Realtime: Tailoring Online Machine Learning Algorithms to Ambient Computing. Mobile Netw Appl 13, 583–598 (2008). https://doi.org/10.1007/s11036-008-0095-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11036-008-0095-8