An Approach for Incremental Data Mining of Multi objective association Rules Using Soft Computing Techniques

Abstract

In Data mining, important data is extracted from a large database. It is the process of picking out relevant information from a large amount of data by using certain sophisticated algorithms. It transforms data into useful information which helps in decision making. After extensive research and inventions, data mining techniques evolve, and the journey is continuously going on. Massive amount of data is available in the data warehouses. Therefore, mining association rules help in numerous decision-making processes. newline newlineMost of the researchers have considered the association rule mining problems as a single objective problem and validated on the static database. Whereas association rule mining problems is a multi-objective problem and it uses measures like support count, comprehensibility and interestingness for mining the strong association rule. The database is being updated periodically due to daily business requirement. Hence, incremental mining deals with generating association rules from the updated database. Soft Computing is a promising collection of intelligent methodologies. Soft computing methodologies have been helpful in many applications. It is used where inexact solutions are computationally-hard tasks. newline newlineIn this research identify the various objectives for association rule mining, examine the limitations of the various objectives and propose two new objectives as High and Low Correlation objectives for 2-variables and 3-variables. newline newline