Prediction of Different Types of Wine Using Nonlinear and Probabilistic Classifiers
In the past few years, machine-learning techniques have garnered much attention across disciplines. Most of these techniques are capable of producing highly accurate results that compel a majority of scientists to implement the approach in cases of predictive analytics. Few works related to wine data have been undertaken using different classifiers, and thus far, no studies have compared the performance metrics of the different classifiers with different feature sets for the prediction of quality among types of wine. In this chapter, an intelligent approach is proposed by considering a recursive feature elimination (RFE) algorithm for feature selection, as well as nonlinear and probabilistic classifiers. Performance metrics including accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) are compared by implementing different classifiers with original feature sets (OFS) as well as reduced feature sets (RFS). The results show accuracy ranging from 97.61 to 99.69% among the different feature sets. This analysis will aid wine experts in differentiating various wines according to their features.
KeywordsMachine learning Feature selection Classifiers Performance metrics Prediction
- 2.Preedy, V., M.L.R. Mendez. 2016. Wine applications with electronic noses. In Electronic noses and tongues in food science, 137–151. Cambridge, MA, USA: Academic Press.Google Scholar
- 6.Beltran, N.H., M.A. Duarte-MErmound, V.A.S. Vicencio, S.A. Salah, and M.A. Bustos. 2008. Chilean wine classification using volatile organic compounds data obtained with a fast GC analyzer. IEEE Transactions on Instrumentation and Measurement 57: 2421–2436.Google Scholar
- 7.Chen, B., C. Rhodes, A. Crawford, and L. Hambuchen. 2014. Wineinformatics: Applying data mining on wine sensory reviews processed by the computational wine wheel. In IEEE international conference on data mining workshop, 142–149, Dec. 2014.Google Scholar
- 8.Forina, M., R. Leardi, C. Armanino, and S. Lanteri. 1998. PARVUS an extendible package for data exploration, classification and correla.Google Scholar
- 10.https://en.wikipedia.org/wiki/Recursive_partitioning, Retrieved 19 August 2017.
- 11.Vijayarani, S., and M. Divya. 2011. An efficient algorithm for generating classification rules. International Journal of Computer Science and Technology 2 (4).Google Scholar
- 12.http://machinelearningmastery.com/non-linear-classification-in-r-with-decision-trees/, Retrieved 19 August 2017.
- 14.Sun, X. 2002. Pitch accent prediction using ensemble machine learning. In Seventh international conference on spoken language processing.Google Scholar
- 17.Is See5/C5.0 Better Than C4.5?. 2009. (Online). http://www.rulequest.com/see5-comparison.html.
- 19.http://www.dtreg.com Software for Predictive Modelling and Forecasting.