© Springer International Publishing Switzerland 2015. Data mining techniques have been developed to automatically learn consumption behaviours of households from smart meter data. In this paper, recurrent routine behaviours are introduced to characterize regular consumption activities in smart meter time series. A novel algorithm is proposed to efficiently discover recurrent routine behaviours in smart meter time series by growing subsequences. We evaluate the proposed algorithm on synthetic data and demonstrate the recurrent routine behaviours extracted on a real-world dataset from the city of Kalgoorlie-Boulder in Western Australia.
|Title of host publication||Advances in Knowledge Discovery and Data Mining|
|Publisher||Springer International Publishing|
|Publication status||Published - 2015|
|Event||Efficient discovery of recurrent routine behaviours in smart meter time series by growing subsequences - Ho Chi Minh City, Vietnam|
Duration: 1 Jan 2015 → …
|Conference||Efficient discovery of recurrent routine behaviours in smart meter time series by growing subsequences|
|Period||1/01/15 → …|