在處理大量資料分析，利用序列型樣(Sequential Patterns)分析顧客消費資料時，只能得到產品的購買先後順序，卻無法得知產品先後購買的間隔時間，以至於無法了解此產品的消費週期，導致分析師無法在最適當的時間給予最有利的行銷。 本論文將以時間性的資料探勘技術，建立重覆購買序列型樣的數學模型，尋找出序列型樣中各事件的次序、間隔時間，找出實際消費行為中的變化與規律關係，包括：是否具有週期關係、是否具有重覆購買週期等等。透過此模型以利分析師可以更準確的了解各產品的消費特性，在最佳的時間點擬定最有利的行銷策略，以獲得最加收益。 In processing huge transaction data analysis, when we use Sequential Patterns Mining techniques to discover the buying behaviors of customers, we just can only get the order of the items purchased, but we are hard to find out the time intervals of related items purchased.So that we can not know the period of the product, lead to analysts can not give the most advantageous marketing in the most appropriate time.
The aim of the this research is to develop a methodology to detect of the existence of repeat-buying behavior and discover the potential period of repeat-buying behavior. Using this model can facilitate the analysts to understand the product consumption characteristics more accurate, and let the analysts to determine the most advantageous marketing strategy in the best time, then the corresponding actions can be taken to maximize enterprise’s revenue.
