Market basket analysis: Analysis of association rules in different time periods

Teng, Wen Foong (2024) Market basket analysis: Analysis of association rules in different time periods. Final Year Project, UTAR.

Preview

PDF Download (4Mb) | Preview

Abstract

Transaction data is a set of records associated with the sales-purchase activities of a business. As a result of digitalization, the transaction data stored in databases have been used by the analyst to extract valuable information. The most widely used method to discover information from transaction data is market basket analysis. Market basket analysis is a technique that determines the relationship between items and the relationship is known as the association rule. The association rules could be further studied to identify commonly purchased items and gain a deeper understanding of customer purchasing behaviours. In this study, performing market basket analysis is not the only objective but the variation of association rules over time will also be focused on. Therefore, market basket analysis will be conducted in different time periods to study the regular cyclic variation of association rules over time and determine the interaction between association rules and time. After analysing the trend of the customers purchasing behaviours from the association rules, the interesting association relationships between large quantities of business transaction data could assist the businesses to design marketing strategies, promote cross-selling, and other business decision-making to boost their profit. An optimal period to conduct market basket analysis will be recommended to ensure the effectiveness of the association rules is maximized all the time and allow the businesses to adjust their marketing strategies from time to time.

Actions (login required)