Updating the Built Prelarge Fast Updated Sequential Pattern Trees with Sequence Modification

Jerry Chun-Wei Lin, Wensheng Gan, Tzung-Pei Hong and Jingliang Zhang
Additional contact information
Jerry Chun-Wei Lin: Innovative Information Industry Research Center (IIIRC) & School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen Graduate School, Shenzhen, China
Wensheng Gan: Innovative Information Industry Research Center (IIIRC) & School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen Graduate School, Shenzhen, China
Tzung-Pei Hong: Department of Computer Science and Information Engineering, National University of Kaohsiung, Kaohsiung, Taiwan & Department of Computer Science and Engineering, National SunYat-sen University, Kaohsiung, Taiwan
Jingliang Zhang: Shandong Sport University, Jinan, China

International Journal of Data Warehousing and Mining (IJDWM), 2015, vol. 11, issue 1, 1-22

Abstract: Mining useful information or knowledge from a very large database to aid managers or decision makers to make appropriate decisions is a critical issue in recent years. Sequential patterns can be used to discover the purchased behaviors of customers or the usage behaviors of users from Web log data. Most approaches process a static database to discover sequential patterns in a batch way. In real-world applications, transactions or sequences in databases are frequently changed. In the past, a fast updated sequential pattern (FUSP)-tree was proposed to handle dynamic databases whether for sequence insertion, deletion or modification based on FUP concepts. Original database is required to be re-scanned if it is necessary to maintain the small sequences which was not kept in the FUSP tree. In this paper, the prelarge concept was adopted to maintain and update the built prelarge FUSP tree for sequence modification. A prelarge FUSP tree is modified from FUSP tree for preserving not only the frequent 1-sequences but also the prelarge 1-sequences in the tree structure. The PRELARGE-FUSP-TREE-MOD maintenance algorithm is proposed to reduce the rescans of the original database due to the pruning properties of prelarge concept. When the number of modified sequences is smaller than the safety bound of the prelarge concept, better results can be obtained by the proposed PRELARGE-FUSP-TREE-MOD maintenance algorithm for sequence modification in dynamic databases.

Date: 2015
References: Add references at CitEc
Citations:

Downloads: (external link)
http://services.igi-global.com/resolvedoi/resolve. ... 018/ijdwm.2015010101 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:igg:jdwm00:v:11:y:2015:i:1:p:1-22

Access Statistics for this article

International Journal of Data Warehousing and Mining (IJDWM) is currently edited by Eric Pardede

More articles in International Journal of Data Warehousing and Mining (IJDWM) from IGI Global
Bibliographic data for series maintained by Journal Editor ().