Association Rule Mining Analyzation Using Column Oriented Database

The logical view of data is a two dimensional table and the physical storage is a single dimensional. Two approaches exist to map two dimensional data on to a single dimensional storage: Row oriented and Column oriented. Common database applications are developed using traditional roworiented database systems. Data Mining (DM) is a promising research area, deals with huge data with large numbers of attributes and records. DM algorithms are more analytical in nature with the goal of reading through the data to gain new insight and use it for planning make Column oriented database systems more preferable. The Column oriented database systems show better performance than traditional database on analytical workloads such as those found in data warehouses, decision support, and business intelligence applications. The Column oriented databases like MonetDB is utilized for performance analysis of SQL queries. This paper is focused on the utilization of Column oriented databases like MonetDB with Oracle 11g the famous Row oriented database for execution time analysis for famous DM algorithm: APRIORI. Experiment results show the faster execution time of MonetDB compare to Oracle for different supports and justifies the suitability of the Column oriented database for such data mining algorithm.