Frequent and Sequential Pattern Mining with Period of Interest Awareness
Date Issued
2009
Date
2009
Author(s)
Huang, Jen-Wei
Abstract
In this dissertation, we addressed the frequent and sequential pattern mining problem with period of interest awareness. It is noted that users are usually more interested in recent data than old ones. Taking the period of interest into consideration, we are able to derive most interesting frequent patterns in time domain in a transaction database or sequential patterns in a sequence database.e first explored the general model of mining associations in a temporal database, where exhibition periods of items are allowed to be different from one to another. To address this issue, we proposed an efficient algorithm Twain, standing for TWo end AssocIation miNer to give more precise frequent exhibition periods of frequent temporal itemsets. Twain not only generates frequent patterns with more precise frequent exhibition periods, but also discovers more interesting frequent patterns.e also proposed a general model of sequential pattern mining with a progressive database while the data in the database may be static, inserted or deleted. In addition, we presented a progressive algorithm Pisa, standing for Progressive mIning of Sequential pAtterns, to progressively discover sequential patterns in a defined period of interest. Pisa utilizes a progressive sequential tree to efficiently maintain the latest data sequences, discover the complete set of up-to-date sequential patterns, and delete obsolete data and patterns accordingly.n addition, we examined the intrinsic scalability problem of mining progressive sequential patterns. When the number of sequences grows and the POI becomes larger, the time and space used to conduct progressive sequential patterns increases dramatically. Due to the limited computing power and working space, single processors usually struggle to scale up. Therefore, we designed a distributed algorithm DPSP, standing for Distributed Progressive Sequential Pattern mining algorithm, to deal with large amounts of data. At each timestamp, DPSP is able to delete obsolete itemsets, update current candidate sequential patterns and report up-to-date frequent sequential patterns within the current POI.
Subjects
frequent pattern
sequential pattern
period of interest
data mining
distributed algorithm
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-98-F91921025-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):d05412057fee5d10bb8c3fbaa08dd7a3