抽象的

Horizontal Aggregations in SQL to Generate Query Sets for Data Mining and OLAP Cube Exploration

Rekha S. Nyaykhor , Nilesh T. Deotale

Data mining is the domain which has utility in real world applications. Data sets are prepared from regular transactional databases for the purpose of data mining. A huge amount of time is needed for making the dataset for the data mining analysis because data mining practitioners required to write complex SQL queries and many tables are to be joined to get the aggregated result. We recommend simple, however powerful, techniques to generate SQL code to return aggregated columns in a very horizontal tabular page layout, returning a few numbers as opposed to one variety per short period. This new class involving functions is named horizontal aggregations. Horizontal aggregations build data sets with a horizontal denormalized layout (e.g., point-dimension, observation- variable, instance-feature), which is the standard layout required by most data mining algorithms. This paper focuses on building user-defined horizontal aggregations such as PIVOT, SPJ (SELECT PROJECT JOIN) and CASE whose underlying logic uses SQL queries, which prepares a query set consists of procedures and this query set will be stored in the database. One can easily access the query set and get the desired output to analyze the datasets. The queries which are not available in the query set can be generated dynamically using dynamic query generation. Finally we will see the experimental evaluation of data sets using SPJ and CASE method.