The major rise in data collection and storage has raised the necessity for much more powerful data analysis tools. The data collected in huge databases needs to be handled effectively and efficiently. The important and highly critical decisions are made not on the basis of information rich data stored in databases but instead on a decision maker�s instinct merely because of the absence of the tools capable of extracting the valuable knowledge from vast amount of the data. Currently expert systems depends on users to manually input knowledge into knowledge bases. This process is often time consuming, expensive, and bias. The problem with data mining algorithms are their non-capability of dealing with non-static, and unbalanced data. There is a need for constantly updating the models to handle data velocity or new incoming data.

The objectives of the research paper is to implement the three popular data mining algorithms (Apriori algorithm, FP � Growth algorithm, and Map Reduce algorithm) using appropriate programming tool (preferably Java). The paper also perform comparative analysis of the three algorithms under study via measuring efficiency in terms of time. The paper also elaborates on analysis of all three algorithms on the basis of performance evaluation using accuracy metric.