Apriori와 FP-growth 실험을 통한 효과적인 장바구니 분석

Effective basket analysis through Apriori and FP-growth experiments

The past from data mining is known to the database did not find the new data in the data model derived Extract the actionable information on future refers to the process used in decision-making. In other words, by finding the hidden patterns and relationships in data to find the lode naedeutyi discover that information. Information found here is the data in applying advanced statistical analysis and modeling techniques to the process to find useful patterns and relationships. So compare what better way than to find experimentally useful information to evaluate it. Accordingly, in the data mining Classify, Cluster, Associate such as based on the performance and results of the experimental verification to use different techniques had appeared optionally optimal algorithm for a particular situation the step of evaluating the suitability of the algorithm for each situation was required. Thus, association of the paper, the method of analyzing using the rule (Associate) using Apriori Algorithm and FP-growth method described algorithm the differences between the two methods, and the suitability of the algorithm using the difference in the specific technology Item Set forming method, data analysis of efficiency were evaluated difference between each set of items. Conformity of the items set by this, the efficiency, the correlation of experimental data were analyzed to know the performance difference.