An Efficient Hybrid by Partitioning Approach for Extracting
Maximal Gradual Patterns in Large Databases (MPSGrite)
BOHR International Journal of Data Mining and Big Data, 1 (1):
10-25 (February 2021)DOI: https://doi.org/10.54646/bijdmbd.003
Abstract
Since automatic knowledge extraction must be performed in large volume databases, empirical studies already show that at the level of generalized patterns, association rules and frequent graded patterns, there is an exponential increase in the search space and in the number of relevant patterns extracted. Faced with this problem, many approaches have been proposed, with the aim of reducing the size of the search space and the waiting time in order to offer users a sufficiently small number of relevant patterns to make decisions or refine their analyses in a reasonable and realistic time. Incremental frequency extraction algorithms in large databases are CPU intensive. This paper presents a new technique to improve the performance of maximal frequent gradual pattern extraction algorithms. The exploitation of this technique leads to a new and more efficient hybrid algorithm called MSPGrite. Experiments performed confirm the interest of the proposed approach.