Mining Of Road Accident Data Using K-Mode Clustering And Sampling Algorithm
Author(s):
veera sekhar chamanthula, Mrs.S.Sajida
Keywords:
statistics mining, Sampling, random sampling, pattern.
Abstract
Discovery of association guidelines is a prototypical trouble in statistics mining. The modern algorithms proposed for records mining of association rules make repeated passes over the database to determine the usually occurring object unit (or set of items). For big statistics, the I/O overhead in scanning the facts can be extremely excessive. in this paper we show that random sampling of transactions within the datasets is an powerful technique for locating association guidelines. Sampling can speed up the mining system through greater than an order of importance by way of reducing I/O charges and appreciably shrinking the number of transaction to be considered. moreover, we show that sampling can correctly constitute the records patterns within the dataset with high self assurance. We experimentally evaluate the effectiveness of sampling on specific datasets, and take a look at the connection among the performance, and the accuracy and self assurance of the selected sample.
Article Details
Unique Paper ID: 145565
Publication Volume & Issue: Volume 4, Issue 10
Page(s): 729 - 733
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024