Journal of Chemical and Pharmaceutical Research (ISSN : 0975-7384)

header
Reach Us reach to JOCPR whatsapp-JOCPR +44 1625708989
All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Original Articles: 2014 Vol: 6 Issue: 4

A model of predicting hog prices based on hadoop and SVM

Abstract

In order to improve the result reliability of predicting hog prices, this paper proposes a model of predicting hog prices based on Hadoop and support vector machine (SVM). First, the pig information is mined from the websites about pigs by the distributed computing technology provided by Hadoop. Second, the multiple regression model is constructed by adding arguments from the pig information gradually which can obtain the key indicators of affecting the pig price fluctuation. Then, through rewriting the map and reduce functions in Hadoop, the corresponding data sets of key indicators are extracted. Finally, the model of training and testing hog prices is established by the method of SVM on the basis of key indicators and corresponding data sets. The experimental result shows that the model of predicting hog prices based on Hadoop and SVM is more reliably.