SHRINKING THE UNCERTAINTY IN ONLINE SALES PREDICTION WITH TIME SERIES ANALYSIS
Abstract
vioft2nntf2t|tblJournal|Abstract_paper|0xf4ff1afd17000000aa49030001000800
In any production environment, processing is centered on the manufacture of products. It is important to get adequate volumes of orders for those products. However, merely getting orders is not enough for the long-term sustainability of multinationals. They need to know the demand for their products well in advance in order to compete and win in a highly competitive market. To assess the demand of a product we need to track its order behavior and predict the future response of customers depending on the present dataset as well as historical dataset. In this paper we propose a systematic, time-series based scheme to perform this task using the Hadoop framework and Holt-Winter prediction function in the R environment to show the sales forecast for forthcoming years.

Authors
Rashmi Ranjan Dhal, B.V.A.N.S.S. Prabhakar Rao
VIT University, Chennai, India

Keywords
Time Series Analysis, Sales Prediction, Hadoop Distributed File System, Holt-Winters Function, Seasonal and Level Variant
Yearly Full Views
JanuaryFebruaryMarchAprilMayJuneJulyAugustSeptemberOctoberNovemberDecember
001000000100
Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 5 , Issue: 1 , Pages: 869-874 )
Date of Publication :
October 2014
Page Views :
682
Full Text Views :
5

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.