MULTI-JOIN OPERATION USING GENETIC ALGORITHM IN ACTIVE DATA WAREHOUSE

Abstract

In dynamic heterogeneous application, data format transformation and data updation are handled because of fast changing of data content in a conventional data warehousing concepts. In order to meet a high demand of maintaining a dynamic large multi-format data warehouse, many techniques have been used. For an active data warehouse, multi-way join is complex, so, to improve the multiple relations of joins being generated by streams from different direction in active data warehousing, the previous work presented an optimization technique to have a more cohesive multi-relation joins of the stream tuples. But the drawback is that the efficiency of multi-join relation using an optimization technique in active data warehouse is less and consumes more time to perform the operation by deriving the optimal threshold value. To improve the reliability and efficiency of multi-join operation in active data warehouse, in this paper, we are going to present a genetic algorithm technique to perform a multi-join operational data in active data warehousing retrieval of data based on multiple queries. A multiple query operations are combined in active data warehouse, and the selection of more appropriate combination of multiple relations are done by using genetic algorithm. A crossover and mutation chooses the best combination of multiple relations of joins for by retrieving the data and produces the output in active data warehouse. Experimental evaluations are carried out with both synthetic and real datasets to estimate the performance of the proposed genetic algorithm for multi-join relation in active data warehouse in terms of data retrieval, efficiency of multi-join operation and scalability.