Published Paper Details:

Abstract: The accuracy of the FPM will be assessed by exploring the interesting associations between gene variables. Exploration of genetic structures like DNA sequence, RNA sequence and protein sequences from gene variables will improve the medical diagnosis process. To do this correlated sequence patterns are considered as very constructive for analyzing these data sets. Correlated sequence pattern mining has become increasingly important recently as an alternative or an augmentation of association rule mining. Though correlated pattern mining discloses the correlation relationships among data objects and reduces significantly the number of patterns produced by the association mining, it still generates quite a large number of patterns. In this paper, a novel approach for DFP-Miner for finding correlated sequence pattern from Biological Datasets is examined to reduce the number of correlated patterns produced without information loss. DFP-Miner effectively discovers confidence closed correlated frequent determinant patterns which are further explored to generate correlated sequence patterns with vector intersection operation. In this approach, a new integrated data structure used which is a combination of one-dimensional array pair set and a virtual data matrix to discover determinate frequent patterns from biological datasets and is called hyperstructure(H-struct). Hyper structure has a variety of feature to facilitate and rapidly keeps running in memory-based limitations which are that it has amazingly constrained and precisely unsurprising primary memory. The newly composed algorithm DFP-Miner and it takes just a single scan over the database to find a large pattern by iteratively specifying H-struct framework. The obvious investigation on DFP-Miner demonstrates and attains better mining efficiency with a superior mining algorithm CARPENTER on different biological datasets on various settings. The execution of DFP-Miner is evaluated with Frequency and Accuracy measures.