In this paper, we propose energy efficient big data mining scheme for forest cover type and gas drift classification. Efficient machine learning and data mining techniques provide unprecedented opportunity to monitor and characterize physical environments, such as forest cover type, using low cost wireless sensor networks. The experimental validation on two different sensor network datasets, forest cover type and gas sensor array drift dataset from publicly available UCI machine learning repository. Coupled with an appropriate feature selection, the complete scheme leads towards an energy efficient protocol for intelligent monitoring of large physical environments instrumented with wireless sensor networks.
|Procedia Computer Science
|International Conference on Information and Communication Technologies
|3/12/14 → 5/12/14