TY  - JOUR
T1  - Asynchronous Pattern Mining for Mixed Sequences Using Ant Colony Optimization
AU - Parimala, M. AU - Sathiyabama, S. 
JO  - Research Journal of Applied Sciences
VL  - 7
IS  - 7
SP  - 354
EP  - 358
PY  - 2012
DA  - 2001/08/19
SN  - 1815-932x
DO  - rjasci.2012.354.358
UR  - https://makhillpublications.co/view-article.php?doi=rjasci.2012.354.358
KW  - Pattern mining
KW  -asynchronous pattern mining
KW  -long sequence
KW  -closed sequence
KW  -ant colony optimizations
AB  - In data mining, sequential pattern mining is a fundamental 
  and essential field since of its broad possibility of applications spanning 
  from forecasting the user shopping patterns and scientific discoveries. In many 
  applications, the most vital part of mining problem is mining periodic patterns 
  in time series databases. It can be visualized as a tool for forecasting and 
  identifying the potential activities of time-series data. Previous studies have 
  measured hash based asynchronous periodic patterns where uneven occurrences 
  of datasets are sequenced. Still, asynchronous periodic pattern mining has expected 
  less attention and in effective. Most researches are explained about an occurrence 
  of long and closed sequence in an asynchronous pattern by a form of numerical 
  derivatives. To improve the asynchronous pattern mining more effective in this 
  study, researchers are going to present an Ant Colony Optimization technique 
  to identify the pattern matching in a mixed sequences (long and closed sequences) 
  raised in the asynchronous pattern mining. The Ant Colony Optimization algorithm 
  (ACO) is an optimized technique for resolving computational problems which can 
  be condensed to discovering good paths and identified the best pattern from 
  the given database. Using ACO, the optimal pattern matching is made efficiently 
  for mixed sequences (closed or long set of sequences) in the given database. 
  Experimental evaluation will conduct with time series dense data sets in arff 
  format in Weka tool and show the performance improvement of asynchronous pattern 
  mining for mining mixed sequences in the given database in terms of running 
  time, pattern matching factor, optimality rate and scalability.
ER  -