Title :
Applying Distributed Classification Algorithms to Wireless Sensor Networks – A Brief View into the Application of the SPRINT Algorithm Family
Author_Institution :
Univ. of Rostock, Rostock
Abstract :
The SPRINT algorithm describes a distributed way to construct a decision tree for classification in large data sets. It can be applied to in-network classification tree construction. The costly data transfer of sensor data to the sink can be avoided while execution time is still acceptable. The SPRINT algorithm and its extensions are introduced. Furthermore, different scenarios that implement classification tree construction based on WSN data are evaluated. A WSN-Model is given. Special characteristics of WSN are implied. Tools for a task specific selection of WSN data processing organisation are sketched.
Keywords :
data mining; pattern classification; tree data structures; wireless sensor networks; data processing organisation; decision tree; distributed classification algorithms; innetwork classification tree construction; large data sets classification; wireless sensor networks; Classification algorithms; Classification tree analysis; Costs; Data mining; Data processing; Decision trees; Neural networks; Partitioning algorithms; Sensor phenomena and characterization; Wireless sensor networks; SPRINT; WSN; classification; tree construction;
Conference_Titel :
Networking, 2008. ICN 2008. Seventh International Conference on
Conference_Location :
Cancun
Print_ISBN :
978-0-7695-3106-9
Electronic_ISBN :
978-0-7695-3106-9
DOI :
10.1109/ICN.2008.73
