A holistic approach for resource-aware adaptive data stream mining
New Generation Computing
| Hi-index | 0.00 |
It has been established experimentally that in-network processing in wireless sensor networks is the acceptable mode of operation. However, this solution is faced by resource constraints of the sensor nodes, especially when running traditional data mining techniques that tend to consume the resources rapidly. On the other hand, data stream mining algorithms still fall short with the limited computational capabilities of the nodes. These algorithms need real-time adaptation to availability of resources. Distributed processing is also essential to produce a global model of the data streams emanated from the network. In this paper, we propose a novel distributed data stream classification technique that is able to adapt to availability of resources in wireless sensor networks.