| dc.description.abstract | The field of data mining seeks to recognize the regularities, patterns and behaviours of large data collections. Association mining is used to discover elements that occur frequently within a dataset consisting of multiple independent selections of elements and to discover rules. This mining approach can find rules which predicts the occurrence of an item, based on the occurrences of other items in a particular transaction. Apriori algorithm is an influential algorithm designed to operate on data collections enclosing transactions such as in market basket analysis. To address various issues Apriori algorithm has been extended in different perspectives. In real world scenario, one of the major objectives in performing a market basket analysis is to maximize the profit. In Apriori algorithm, Support value and Confidence value are the dominant factors in generating association rules which seems to be insufficient to achieve the said objective as the algorithm does not consist a variable to maximize the profit gain. Moreover, consideration of frequent items, rather than rare items, significantly impact the profit maximization. Therefore, this research was focused to develop a new algorithm based on an extended Apriori approach which maximize the profit of a transaction using frequent items as well as rare items in a market basket analysis. The developed new algorithm and the extended Apriori algorithm were applied to a real world data set and the results were compared focusing the profit gain from each algorithm separately. Finally, the results conclude that the proposed algorithm derives association rules which significantly increase the profit gain, disregard of the number of items involving in the transaction. | en_US |
