Prediction of soil cation exchange capacity using enhanced machine learning approaches in the southern region of the Caspian Sea

Abstract

Cation exchange capacity (CEC) has a key role in soil studies such as agriculture, energy balance, characteristics of the soil for food, maintaining water in the soil as well as soil pollution management. Its measurement is difficult and time-consuming. So, its prediction using artificial intelligent (AI) models with soil readily available properties can be the proper solution. In this study, the physical and chemical properties of the soil, such as pH, EC, organic carbon, clay content, sands, and total nitrogen used as input data for the AI models. The adaptive-network-based fuzzy inference system (ANFIS), ANFIS model coupled by differential evolution (ANFIS-DE), and ANFIS model coupled by particle swarm optimization (ANFIS-PSO) are used for the prediction of the CEC. Then the ability of those methods in the prediction of the CEC. Results showed higher efficiency of the coupled models (ANFIS-DE and ANFIS-PSO) compared to the ordinary ANFIS model.

Keywords

• Artificial intelligence model
• Cation exchange capacity
• Differential evolution algorithm
• Multidisciplinary research
• Multiple soil classes
• Particle swarm optimization