PAFHWKM: An Enhanced Parallel Approach to Forecast Time Series Data Using Holt-Winters and K-Means Algorithm

Main Article Content

B. Arputhamary, Dr. L Arockiam

Abstract

Big Data is a recent research style which brings up challenges in decision making process. The size of the dataset turn intotremendously big, the process of extracting valuablefacts by analyzing these data also has become tedious. To solve this problem of information extraction with Big Data, parallel programming models can be used. Parallel Programming model achieves information extraction by partitioning the huge data into smaller chunks. MapReduce is one of the parallel programming models which works well with Hadoop Distributed File System(HDFS) that can be used to partition the data in a more efficient and effective way. In MapReduce, once the data is partitioned based on the pair, it is ready for data analytics. Time Series data play an important role in Big Data Analytics where Time Series analysis can be performed with many machine learning algorithms as well as traditional algorithmic concepts such as regression, exponential smoothing, moving average, classification, clustering and model-based recommendation. For Big Data, these algorithms can be used with MapReduce programming model on Hadoop clusters by translating their data analytics logic to the MapReduce job which is to be run over Hadoop clusters. But Time Series data are sequential in nature so that the partitioning of Time Series data must be carefully done to retain its prediction accuracy.In this paper, a novel parallel approach to forecast Time Series data with Holt-Winters model (PAFHW) is proposed and the proposed approach PAFHW is enhanced by combining K-means clusteringfor forecasting the Time Series data in distributed environment.

Article Details

How to Cite
, B. A. D. L. A. (2017). PAFHWKM: An Enhanced Parallel Approach to Forecast Time Series Data Using Holt-Winters and K-Means Algorithm. International Journal on Future Revolution in Computer Science &Amp; Communication Engineering, 3(8), 65–73. Retrieved from http://www.ijfrcsce.org/index.php/ijfrcsce/article/view/181
Section
Articles