Abstract

Soil temperature prediction is an important task since soil temperature plays an important role in agriculture and land use. Although some progress has been made in this area, the existing methods provide a regression or nominal classification task. However, ordinal classification is yet to be explored. To bridge the gap, this paper proposes a novel approach: Soil Temperature Ordinal Classification (STOC), which considers the relationships between the class labels during soil temperature level prediction. To demonstrate the effectiveness of the proposed approach, the STOC method using five different traditional machine learning methods (Decision Tree, Naive Bayes, K-Nearest Neighbors, Support Vector Machines, and Random Forest) was applied on daily values of meteorological and soil data obtained from 16 stations in three states (Utah, Alabama, and New Mexico) of United States at five soil depths (2, 4, 8, 20, and 40 inches) between the years of 2011 and 2020. The experiments show that the proposed STOC approach is an efficient method for soil temperature level (very low, low, medium, high, and very high) prediction. The applied STOC models (STOC.DT, STOC.NB, STOC.KNN, STOC.SVM, and STOC.RF) showed average accuracy rates of 90.95%, 77.09%, 90.84%, 89.94%, and 90.91% on the experimental datasets, respectively. It was observed from the experimental results that the STOC.DT method achieved the best soil temperature level prediction among the others.

Highlights

  • Soil temperature greatly influences plant growth and development, soil water and salt transport, soil carbon balance, and microbial activity and chemical reactions inside the soil (Onwuka & Mang 2018)

  • The novelty and main contributions of this work are as follows. (i) It is the first attempt to apply ordinal classification for soil temperature level prediction. (ii) This paper proposes a novel method, called Soil Temperature Ordinal Classification (STOC), which takes into account the relationships between the class labels during soil temperature level prediction. (iii) This study is original in that it compares alternative base learners in conjunction with the proposed method, including decision tree (DT), Naive Bayes (NB), k-nearest neighbors (KNN), support vector machines (SVM), and random forest (RF). (iv) This is the first study using ordinal classification to predict soil temperature levels at five different soil depths

  • When the STOC approach was tested in combination with different classification algorithms (DT, NB, KNN, SVM, and RF), the SSOC.DT method achieved higher accuracy (90.95%) than the rest on average

Read more

Summary

A Novel Machine Learning Approach

Soil Temperature Ordinal Classification (STOC) Cansel KUCUKa, Derya BIRANTb, Pelin YILDIRIM TASERc aDokuz Eylul University, Graduate School of Natural and Applied Sciences, Izmir, Turkey bDokuz Eylul University, Department of Computer Engineering, Izmir, Turkey cIzmir Bakircay University, Department of Computer Engineering, Izmir, Turkey This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi:10.15832/ankutbd.866045

Introduction
Material and Methods 7
Results
Discussion
Conclusions
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call