The property recommendation system is an application that is developed using Apriori and Topsis methods. this application will have the first 2 recommendations that will appear based on the results of user clicks on advertisements that are seen using the a priori method, this method will determine frequent itemsets based on the number of ads chosen by the user and will be eliminated with support values after which association rules will be searched produce a value of confidence, the highest confidence value displayed is a recommendation. The second is a property recommendation based on the criteria selected by the user using the topsis method, this method will look for normalized matrices, then multiplied by the weight of each criterion, followed by determining the positive and negative ideal solutions then looking for min max from each criterion, followed by calculating the distance of positive and negative ideal solutions, after we can find the preference value of each ad, the ad with the highest preference value that will be displayed followed by the order of 2 to sequence 5. The test is carried out using 50 property data obtained from PT. Ditra Manunggal Jaya.
Copyrights © 2019