Amrita International Conference of Women in Computing - 2013 |
Foundation of Computer Science USA |
AICWIC - Number 4 |
January 2013 |
Authors: K. S. Kalaivani, S. Kuppuswami |
41ebf8a5-5972-4e1c-9779-58fa26e17f75 |
K. S. Kalaivani, S. Kuppuswami . A Novel Acceleration Technique to Improve the Speed of Mining Frequent U2 Patterns. Amrita International Conference of Women in Computing - 2013. AICWIC, 4 (January 2013), 5-9.
Frequent pattern mining is the method of finding patterns like itemsets, subsequences and substructures that repeatedly occur in a dataset. In Univariate Uncertain data, each attribute present in a transaction is represented by a quantitative interval and a probability value. U2P-Miner algorithm is used to mine frequent patterns from U2 data. The number of intervals has a great impact on the time taken for mining frequent patterns. A novel acceleration technique which compares the expected support with the user specified threshold is introduced to minimize the number of intervals thereby improving the speed of the mining process. The runtime of the modified U2P-Miner algorithm is compared with the existing U2P-Miner algorithm.