Fast Algorithms for Mining Association Rules in Large Databases
VLDB '94 Proceedings of the 20th International Conference on Very Large Data Bases
Hi-index | 0.00 |
The relationship between streamflow q and electrical conductivity k is explored in this paper, using data from Hollin Cave Spring in New South Wales, Australia. A temporal rule extraction algorithm is used to identify frequent patterns in each time series. The frequent patterns are then refined using the concept of profile convexity, and parametrised for compactness of representation, before the coupling between flow and conductivity is examined. Results show that two frequent peak patterns occur in flow and two troughs in electrical conductivity, and that the shapes of all these can be characterised with a single magnitude parameter. The coupling between events in the two series is investigated, and reveals that the depth of k troughs depend heavily on the initial state of k, and more weakly on the magnitude of the flow peak.