Generating Market Basket Data with Temporal Information

• Main Authors: LI, Yingjiu, NING, Peng, WANG, X. Sean, JAJODIA, Sushil
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2001
• Subjects: Information Security
• Online Access: https://ink.library.smu.edu.sg/sis_research/1051; https://ink.library.smu.edu.sg/context/sis_research/article/2050/viewcontent/GeneratingMarketBasketData_2001.pdf
• Institution: Singapore Management University

This paper presents a synthetic data generator that outputs timestamped transactional data with embedded temporal patterns controlled by a set of input parameters. In particular, calendar schema, which is determined by a hierarchy of input time granularities, is used as a framework of possible temporal patterns. An example of calendar schema is (year, month, day), which provides a framework for calendar-based temporal patterns of the form -38352 , where each is either an integer or the symbol . For example, is such a pattern, which corresponds to the time intervals consisting of all the 16th days of all months in year 2000. This paper also evaluates the data generator through a series of experiments. The synthetic data generator is intended to provide support for data mining community in evaluating various aspects (especially the temporal aspects and the scalability) of data mining algorithms.