Association rules for supporting hoarding in mobile computing environments
Author
Saygın, Yücel
Ulusoy, Özgür
Elmagarmid, A. K.
Date
2000Source Title
Proceedings Tenth International Workshop on Research Issues in Data Engineering. RIDE 2000
Publisher
IEEE
Language
English
Type
Conference PaperItem Usage Stats
162
views
views
113
downloads
downloads
Abstract
One of the features that a mobile computer should provide is disconnected operation which is performed by hoarding. The process of hoarding can be described as loading the data items needed in the future to the client cache prior to disconnection. Automated hoarding is the process of predicting the hoard set without any user intervention. In this paper, we describe an application independent and generic technique for determining what should be hoarded prior to disconnection. Our method utilizes association rules that are extracted by data mining techniques for determining the set of items that should be hoarded to a mobile computer prior to disconnection. The proposed method was implemented and tested on synthetic data to estimate its effectiveness. Performance experiments determined that the proposed rule-based methods are effective in improving the system performance in terms of the cache hit ratio of mobile clients especially for small cache sizes.
Keywords
AlgorithmsBuffer storage
Computational linguistics
Computer simulation
Data processing
Mobile computing
Association rules
Hoarding
Data mining