A Better Approach for Horizontal Aggregations in SQL Using Data Sets for Data Mining Analysis

To analyzing the data efficiently in Data mining systems are widely using datasets with columns in horizontal tabular layout. Generally preparing a data set is the more complex task in a data mining project, require many complex SQL queries, aggregating columns and joining tables. Conventional RDBMS usually manage tables with vertical form. Aggregated columns in a horizontal tabular layout returns set of numbers, instead of one number per row. This new class of function is called horizontal aggregations. The system uses one parent table and different child tables, operations are then performed on the data loaded from multiple tables. We proposed three fundamental methods .They are SPJ (select-project-join-Aggregation), CASE, and PIVOT. SPJ based on standard relational algebra operators. CASE is useful to exploiting the programming case construct. PIVOT is a built-in operator in a commercial DBMS, PIVOT operator, offered by RDBMS is used to calculate aggregate operations. PIVOT methods are much faster methods and offer much scalability. Partitioning large set of data, obtained from the result of horizontal aggregation. Key Terms: Aggregation; Data Mining; Data preparation; Structured Query Language (SQL); Pivot Full Text: http://www.ijcsmc.com/docs/papers/August2013/V2I8201353.pdf