Knowledge extraction and analysis to evaluate the financial performance of an organization using OLAM

Data mining or the discovery of knowledge out of databases extracts patterns, and useful and non-substantial, implicit, and unknown information from large databases. Searching for associative rules is a data mining method where relationships and dependencies Interactions between a large set of data items are shown. In large organizations, most data are created at the passage of time. Relational tables in a variety of business or scientific domains have rich information, with high quantative and nominal data types. Therefore, in order to gather information, measure performance and increase business efficiency, it is necessary to isolate relational database and store data in a data warehouse. Using the multidimensional data mining that integrates online analytical processing with data mining, knowledge can be found in multidimensional databases. In this article, to avoid inappropriate rules in exploring associative rules, the online analytical processing technique is combined with the Apriori algorithm and explores associative rules on multi-dimensional and multilevel data using the data cubic and Apriori algorithm. In addition, the criteria of accuracy, tool performance, and runtime are defined for comparing techniques for exploring associative rules. The metrics mentioned in the scope of this paper have been compared, and data from the Central Insurance Agency have been analyzed using data cubes and the Apriori algorithm, and useful association rules have been generated. Results of the research displays improvement of the response time compared to the other methods.