BibTeX:

@article{IJIRSTV3I2071,
     title={Enhanced SWASP Algorithm for Mining Associated Patterns from Wireless Sensor Networks Dataset },
     author={Surya S, Mary Mareena P V and Dr. S Dhanabal},
     journal={International Journal for Innovative Research in Science & Technology},
     volume={3},
     number={2},
     pages={209--214},
     year={},
     url={http://www.ijirst.org/articles/IJIRSTV3I2071.pdf},
     publisher={IJIRST (International Journal for Innovative Research in Science & Technology)},
}

A Wireless Sensor Network are successfully deployed for various applications such as low cost area monitoring, environment monitoring, industrial and machine health monitoring, and military surveillance and they are spatially distributed autonomous sensors to monitor conditions such as physical and environmental. WSNs generate a large amount of data streams. Mining useful information from these data stream is a challenging task. Many algorithms have been proposed to extract the useful knowledge from sensor data and the widely used algorithm is Associated Sensor Pattern and compact tree structure, called Associated Sensor Pattern tree which uses pattern growth-based approach to generate all associated patterns with only one database scan over dataset. But when data stream flows through associated sensor pattern may fail to capture the significance of recent data. To overcome this limitation Associated Sensor Pattern tree is further enhanced to Sliding Window Associated Sensor Pattern tree by adopting sliding observation window and updating the tree structure accordingly and a mining algorithm Sliding Window Associated Sensor Pattern is used to mine recent associated patterns. But the limitation of Sliding Window Associated Sensor Pattern is that it takes too much time to mine associated patterns for large dataset. To enhance the performance of this algorithm, the dataset is partitioned into different sub parts and by using this partitioned dataset the Sliding Window Associated Sensor Pattern algorithm are run parallel by using multithread concept. And this enhanced algorithm reduces the total execution time and also increases the memory efficiency.

Wireless Sensor Network, Sensor Data Stream, Behavioral Patterns, Data Mining, Knowledge Discovery