Pattern Generation for Complex Data Using Hybrid Mining

Combined mining is a hybrid mining approach for mining informative patterns from single or multiple data-sources, multiple-features extraction and applying multiple-methods as per the requirements. Data mining applications often involve complex data like multiple heterogeneous data sources, different user preference and create decision-making actions. The complete useful information may not be obtained by using single data mining method in the form of informative patterns as that would consume more time and space. This paper implements hybrid or combined mining approach that applies Lossy-counting algorithm on each data-source to get the frequent data item-sets and then generates the combined association rules. Applying multi-feature approach, we generate incremental pair patterns and incremental cluster patterns. In multi-method combined mining approach, FP-growth and Bayesian Belief Network are combined to generate classifier to get more informative knowledge. This paper uses two different data-sets to get more useful knowledge and compare the results.