A NOVEL RANK BASED CO-LOCATION PATTERN MINING APPROACH USING MAP-REDUCE

With the increase in spatial data analysis, the co-location patterns and its dependencies are used to discover the complex patterns on spatial databases. Most of the traditional spatial data mining techniques have been implemented based on the assumption that the data is meaningful and clean. It is...

Full description

Saved in:
Bibliographic Details
Published in:Journal of Theoretical and Applied Information Technology 2016-05, Vol.87 (3), p.422-422
Main Authors: Sheshikala, M, Rao, D Rajeswara, Prakash, R Vijaya
Format: Article
Language:English
Subjects:
Algorithms
Cleaning
Data integration
Data mining
Data processing
Information technology
Pattern analysis
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the increase in spatial data analysis, the co-location patterns and its dependencies are used to discover the complex patterns on spatial databases. Most of the traditional spatial data mining techniques have been implemented based on the assumption that the data is meaningful and clean. It is essential to study the data integration issues along with spatial co-locating patterns. Generally, spatial co-location mining algorithms are used to discover the spatial objects and its dependencies among them. As the data size increases, the co-location objects and its patterns are difficult to process on complex spatial objects. In this paper, an optimized spatial co-locating pattern mining framework was developed to discover the highly ranked correlated patterns using the hadoop framework. This MapReduce model was used to minimize computational time and space on complex spatial databases. Finally, the experimental results on the complex spatial data are evaluated using the proposed framework and the traditional hadoop based pattern mining models.
ISSN:1817-3195