Discovering Fuzzy Periodic-Frequent Patterns in Quantitative Temporal Databases
R. Uday Kiran, C. Saideep, Penugonda Ravikumar, Koji Zettsu, Masashi Toyoda, Masaru Kitsuregawa, P. Krishna Reddy
Abstract
Periodic-frequent pattern mining is a challenging problem of great importance in many applications. Most previous works focused on finding these patterns in binary temporal databases and did not take into account the quantities of items within the data. This paper proposes a novel model of fuzzy periodic-frequent pattern (FPFP) that may exist in a quantitative temporal database (QTD). Finding FPFPs in QTD is a non-trivial and challenging task due to its huge search space. A novel pruning technique, called improved maximum scalar cardinality, has been introduced to effectively reduce the search space and the computational cost of finding the desired itemsets. This technique facilitates the mining of FPFPs in real-world very large databases practicable. An efficient algorithm has also been presented to find all FPFPs in a QTD. Experimental results demonstrate that the proposed algorithm is efficient. We also present a case study in which we apply our model to find useful information in air pollution database.