Reduce Time Process Using Apriori Algorithm On K-Wayjoin Based To Find Retail Business Data Relationship Pattern

The process of running a trading business, businesses must always update information on market competition that occurs. Running a trading business is not just opening a business place and waiting for consumers to shop. Consumers will lightly come to shopping places to shop for various reasons. From cheap prices, attractive arrangements, large parking lots, to the ease of finding items to buy. From these problems, business people will compete on how to easily attract customers to their place of business. One way to solve these problems is by structuring merchandise with the aim of customers to easily get the items they are looking for. One way to solve these problems is by structuring merchandise with the aim of customers to easily get the items they are looking for. Arrangements that are made do not originate from arranging the location of goods according to taste but are carried out on the basis of trends or trends in goods purchased by consumers when shopping. This process is often referred to as the "Data Mining" process is one of the effective methods for finding consumers' preferences to choose the items they buy. This process can be completed using the Apriori algorithm. The principle used by the Apriori algorithm is that if an itemset appears frequently, then all subset of itemset must also appear frequently. This results in repeated checking and will require a short time. Problems that require a short time, a method is proposed, namely by developing to be able to reduce the travel time of the process. The length of time that occurs in the process of calculating the value of support and confidance and the repetition process to find the value. The method used is to manipulate the use of query languages with a k-way join research approach so that the optimal query language arrangement can be obtained. The results obtained in this study are that execution times are relatively faster, with the results of the same association rules as those produced by the Priori method without any development or modification.