Prediction of Acidity Level of Avomango (Gadung Klonal 21) Using Local Polynomial Estimator
- DOI
- 10.2991/assehr.k.201010.015How to use a DOI?
- Keywords
- Titratable acidity, acidity level, avomango, local polynomial estimator
- Abstract
The acidity level is one of the parameters that determine fruit maturity. Mature avomango has a low acidity level, indicated by its titratable acidity value. Non-destructive analysis using a near infrared (NIR) spectroscopy tool produces wavelength data that is used as a predictor variable with a titratable acidity value as a response variable. This study aims to predict the acidity level (titratable acidity) of avomangoes using nonparametric regression based on local linear estimators and compare them with the prediction using the parametric regression approach. This study uses 100 mango data as samples, which are divided into 80 into sample data for the training process and 20 out sample data for the prediction process. The results showed that the second degree local polynomial estimator gives the best results. From the 5 iterations, the 4th iteration data gives the mean square error value of prediction using local polynomial estimator and global polynomial regression method is 0.0651and 0.0733, respectively. The mean absolute percentage error of prediction using local polynomial estimators and global polynomial regression is 39.6411 and 43.3676, respectively. It means that the nonparametric regression based on local polynomial estimator provides better predictive results compared to the parametric regression approach.
- Copyright
- © 2020, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - M Ulya AU - N Chamidah PY - 2020 DA - 2020/10/11 TI - Prediction of Acidity Level of Avomango (Gadung Klonal 21) Using Local Polynomial Estimator BT - Proceedings of the 2nd International Seminar on Science and Technology (ISSTEC 2019) PB - Atlantis Press SP - 92 EP - 100 SN - 2352-5398 UR - https://doi.org/10.2991/assehr.k.201010.015 DO - 10.2991/assehr.k.201010.015 ID - Ulya2020 ER -