Open AccessImplementation and comparison analysis of apriori and fp-growth algorithm performance to determine market basket analysis in Breiliant shop
Author(s) -
A A Hidayat,
Ali Rahman,
Resi Mustifa Wangi,
Rachmat Jaenal Abidin,
Rifqi Syamsul Fuadi,
W Budiawan
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1402/7/077031
Subject(s) - apriori algorithm , database transaction , association rule learning , transaction data , affinity analysis , marketing , order (exchange) , value (mathematics) , computer science , algorithm , cashew nut , competitive analysis , business , advertising , database , data mining , mathematics , agricultural science , finance , environmental science , machine learning , mathematical analysis , upper and lower bounds
The creativity of an online store owner who develops a business in a retail business in determining a marketing strategy will affect his competitive ability with other online stores. However, services that are in accordance with costumers attitude is a stuff that should be awarded by shop owners in order to improve not only customer satisfaction but also store income. The method used in this study is Market Basket Analysis to know customers attitude by analysing data from sales transactions to show customers attitude toward one sold item and the others. Research aimed that it is necessary to have a system that can help to form a combination among products using the association rule method, the algorithm used to analyse this method is apriori and fp-growth algorithm. Result from data analysis test using monthly sales transaction data of cosmetics in Breiliant Store during November 2018 with 34 number of data sales transaction. It could be concluded that a combination of products which have strong support and confidence were Original Liquid Bleaching Seeds, Harva Peeling Gel and Castor oil. It had support value of 8.8% and 30% confidence value, with a filtering time of 0.036 seconds.