Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


ucin1085712249.pdf (527.9 KB)

ETD Abstract Container

Abstract Header

METHODOLOGY FOR CLUSTERING SPATIO-TEMPORAL DATABASES

KAKKAR, SHAGUN
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1085712249

Abstract Details

2004, MS, University of Cincinnati, Engineering : Computer Science.
Data mining aims to discover patterns and extract useful information recorded in databases. Spatial data mining and temporal data mining are two important branches that deal with location data and time series data respectively. Several researchers have studied either spatial data mining or temporal data mining and have proposed algorithms to find clusters. The integration of both spatial and temporal data mining leads to spatio-temporal data mining that deals with the discovery of spatial and temporal relationships. In this thesis, a novel approach is discussed to discover spatio-temporal clusters or patterns of similar characteristics. Regions of similar characteristics in spatio-temporal databases are discovered. The approach considered in this thesis translates each profile into a symbolic sequence and constructs a Generalized Suffix Tree (GST) of all subsequences that are shared by at least two sequences. GST implementation is used for representing multiple sequences and searching patterns in them. The proposed algorithm clusters the profiles which share the same set of subsequences based on temporal hypothesis. To generate more general hypotheses about temporal behavior, the subsequences that define each cluster are generalized. The profiles generated after generalization are further clustered based on a metric ratio. These clusters of temporal subsequences based on spatial hypothesis result in spatio-temporal clustering and help in discovering patterns of similar characteristics. To test and validate the proposed algorithm, different datasets are considered. Details of the implementation and results are provided in the thesis.
Dr. Raj Bhatnagar (Advisor)
88 p.
Computer Science

Recommended Citations

Citations

  • KAKKAR, S. (2004). METHODOLOGY FOR CLUSTERING SPATIO-TEMPORAL DATABASES [Master's thesis, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1085712249

    APA Style (7th edition)

  • KAKKAR, SHAGUN. METHODOLOGY FOR CLUSTERING SPATIO-TEMPORAL DATABASES. 2004. University of Cincinnati, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1085712249.

    MLA Style (8th edition)

  • KAKKAR, SHAGUN. "METHODOLOGY FOR CLUSTERING SPATIO-TEMPORAL DATABASES." Master's thesis, University of Cincinnati, 2004. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1085712249

    Chicago Manual of Style (17th edition)

ucin1085712249
795
© 2004, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.