Association Rules Mining for a Specific Time Period in a Day in Large Transactional Database

Market basket analysis is a trendy topic in the field of data mining. Analyzing the daily transactions, several types of associations were noticed in the transacted item sets. The transactional database used in retail marketing has a large number of itemsets that are frequently purchased exclusively during specific hours of the day and are not usually considered to be frequently purchased. In some instances, association rule mining at a lower data level with uniform support doesn't reveal any notable patterns, yet it may be suppressing crucial data. For instance, the morning session sees the biggest demand for items like bread and cake, while the afternoon and evening sessions find higher demand for items like drinks and noodles. In the transaction at a supermarket, this study discovered the frequent item and the link among the frequent item sets. In this study, we used the tree-based FP-growth idea to locate the pertinent rules in the transaction based on the time of day, specifically the morning, afternoon, and evening. Our empirical analysis demonstrates that the suggested approach is computationally effective in a different time period of the day, enabling the Bangladeshi retailer Super Shop to make the best decisions and boost sales.