e-ISSN 2231-8542
ISSN 1511-3701

Home / Regular Issue / JTAS Vol. 28 (2) Apr. 2020 / JST-1786-2019


Comparison of Sparse and Robust Regression Techniques in Efficient Model Selection for Moisture Ratio Removal of Seaweed using Solar Drier

Anam Javaid, Mohd. Tahir Ismail and Majid Khan Majahar Ali

Pertanika Journal of Tropical Agricultural Science, Volume 28, Issue 2, April 2020

Keywords: All possible models, LASSO, model selection, robust, seaweed, selection criteria

Published on: 15 April 2020

Solar drier is considered to be an important product used in the internet of things (IoT). It is used to dry different kinds of products used in agriculture or aquaculture. There are many factors that have different effects on the drying of items in the solar drier. The current study focused on the removal of the moisture ratio in the drying process for seaweed using solar drier. For this purpose, a dataset containing 1924 observations was used to study the effect of six different independent variables on the dependent variable. Moisture ratio removal (%) was considered to be dependent variable with ambient temperature, chamber temperature, collector temperature, chamber relative humidity, ambient relative humidity and solar radiation as independent variables. All possible models were used in the analysis till fifth order interaction terms. Hybrid model of LASSO with bisquare M was proposed for efficient selection of the model. The procedure based on four phases was used for efficient model selection and a comparison was made with other existing sparse and robust regression techniques. The result indicates that the proposed technique is better than other existing techniques in terms of mean squared error (MSE) and mean absolute percentage error (MAPE).

ISSN 1511-3701

e-ISSN 2231-8542

Article ID


Download Full Article PDF

Share this article

Recent Articles