Predicting Soil pH by Using Nearest Fields

TL;DR

This paper presents a novel approach for predicting soil pH in precision agriculture using nearest neighbor fields and regression techniques, reducing the need for costly soil testing.

Contribution

It introduces a spatial data mining method combining nearest neighbor queries with regression models for soil pH prediction, demonstrating high accuracy with real-world data.

Findings

LR, SVR, and GBRT achieved R^2 of about 0.718

MAE values around 0.29

The approach is effective and promising for precision agriculture

Abstract

In precision agriculture (PA), soil sampling and testing operation is prior to planting any new crop. It is an expensive operation since there are many soil characteristics to take into account. This paper gives an overview of soil characteristics and their relationships with crop yield and soil profiling. We propose an approach for predicting soil pH based on nearest neighbour fields. It implements spatial radius queries and various regression techniques in data mining. We use soil dataset containing about 4,000 fields profiles to evaluate them and analyse their robustness. A comparative study indicates that LR, SVR, and GBRT techniques achieved high accuracy, with the R_2 values of about 0.718 and MAE values of 0.29. The experimental results showed that the proposed approach is very promising and can contribute significantly to PA.