在分析frequent pattern 時通常我們使用關聯規則做資料探勘，在這過程中，經由許多次的交易紀錄的審視，最後得到我們所要的frequent items。傳統的關聯規則像是Apriori 方法，沒有考慮到產品購買頻率這項因素，例如牛奶這項產品，可能在某些消費者的消費習性裡，每隔一週或每隔一段期間就會被購買。在以前的文獻，單一產品的重複購買已經有相關研究，但銷售者有興趣的不僅是單一產品的購買週期，因為在許多的促銷方案裡，會有特惠產品組合，所以多種產品組合的購買週期也會是銷售決策時的重點。因而我們的研究對於具有相同購買週期的兩個產品是否具有關聯再去分析這兩個產品組合的購買頻率，最後得到的是由具有相同購買週期的單一產品所組成的相關聯產品組合可能的購買週期。

Frequent patterns mining usually use association rule, the process is to mined items that appear frequently together in a transaction data set. But the traditional association rule method is lack of the factor of frequency of repeating purchase. In this research, we consider the frequency factor to mine correlated repeated-buying items with the same purchased period items. The first step of the experiment is to determined single item frequency, and we use statistics of data to calculate the frequency. Then items with the same frequency are put together and analyzed with the statistics method. The results we get some correlated items purchased period .

Reference
[1]. Chiang, D. A., Wang, Y. H., & Chen, S. P. (2010). Analysis on repeat-buying patterns. Knowledge-Based Systems, 23(8), 757-768.
[2]. Cheung, C. F., & Li, F. L. (2012). A quantitative correlation coefficient mining method for business intelligence in small and medium enterprises of trading business. Expert Systems with Applications, 39(7), 6279-6291.
[3]. Lassnig, M., Fahringer, T., Garonne, V., Molfetas, A., & Barisits, M. (2011, November). A similarity measure for time, frequency, and dependencies in large-scale workloads. In Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis (p. 43). ACM.
[4]. Kiran, R. U., & Reddy, P. K. (2011, January). An alternative interestingness measure for mining periodic-frequent patterns. In Database Systems for Advanced Applications (pp. 183-192). Springer Berlin Heidelberg.
[5]. Sridevi, S., Rajaram, S., & Swadhikar, C. (2010, December). An intelligent prediction system for time series data using periodic pattern mining in temporal databases. In Proceedings of the First International Conference on Intelligent Interactive Technologies and Multimedia (pp. 163-171). ACM.
[6]. Miao, R., & Shen, X. J. (2010, August). Construction of periodic temporal association rules in data mining. In Fuzzy Systems and Knowledge Discovery (FSKD), 2010 Seventh International Conference on (Vol. 5, pp. 2133-2137). IEEE.
[7]. Babicz-Zielinska, E. (1999). Food preferences among the Polish young adults. Food Quality and Preference, 10(2), 139-145.
[8]. de Mendonça, S. N. T. G., Brandão, H. C. A. D. N. T. M., Brandão, W. A. P. L. N. T. M., Quintino, C. A. A., De Francisco, A., & Teixeira, E. (2013). Food preferences of middle aged and elderly subjects in a Brazilian city. The journal of nutrition, health & aging, 17(2), 130-135.
[9]. Agrawal, R., & Srikant, R. (1995, March). Mining sequential patterns. In Data Engineering, 1995. Proceedings of the Eleventh International Conference on (pp. 3-14). IEEE.
[10]. Han, J., Pei, J., Mortazavi-Asl, B., Pinto, H., Chen, Q., Dayal, U., & Hsu, M. C. (2001, April). PrefixSpan: Mining sequential patterns efficiently by prefix-projected pattern growth. In Proceedings of the 17th International Conference on Data Engineering (pp. 215-224).