Open AccessDesign and Implementation of Retail Product Assortment System with Private Label Considerations and Weighted Apriori
Author(s) -
Auliasa Enlina,
Jaka Sembiring
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/803/1/012036
Subject(s) - database transaction , profit (economics) , computer science , association rule learning , apriori algorithm , transaction data , a priori and a posteriori , data mining , product (mathematics) , loyalty , database , business , marketing , mathematics , economics , philosophy , geometry , epistemology , microeconomics
Increasing product variety raise the needs for analytical model to determine an efficient and optimal assortment based on consumer preference considerations. However, most of retailers have not considered the private label factor that can generate profit differentiation while building store loyalty. Apriori algorithm, which is often used to find consumer shopping habit patterns in retail, still has many drawbacks that can increase the burden of input, output, and processing time. Thus, this study proposes a specific procedure to establishing basic product based on private label, improving association rules by reducing transaction through Size of Transaction (SoT) attribute, and adopting maximum weighted value concept in apriori algorithm, therefore the result of association rules will be reflecting user considerations. This research produces a model to recommend assortment by considering private label and user reference, and building a system to integrate the whole process. Based on evaluation result, this study was able to reduce number of scanning databases, number of candidate and frequent itemsets generated, as well as strengthen the store image (qualitative) and increase company profit (quantitative).