Integrating machine learning models for linear and exponential regression to predict wheat area, productivity and population

Muhammad Islam; Farrukh Shehzad; Muhammad Omar; Abdul Qayyum; Rabia Siddiqui

doi:10.17582/journal.sja/2022/38.3.894.901

No Detail Found!

Integrating machine learning models for linear and exponential regression to predict wheat area, productivity and population

Author(s)

Muhammad Islam The Islamia University of Bahawalpur, Punjab, Pakistan; Crop Reporting Service, Government of the Punjab, Pakistan.

Farrukh Shehzad Department of Statistics, The Islamia University of Bahawalpur, Punjab, Pakistan;

Muhammad Omar Department of Computer Science, The Islamia University of Bahawalpur, Pakistan

Abdul Qayyum Director of Agriculture, Crop Reporting Service, Government of the Punjab, Pakistan

Rabia Siddiqui Crop Reporting Service, Government of the Punjab, Pakistan.

Abstract

The true prediction of food crop with exact estimation of population growth can play a crucial role for evolving efective agricultural policy decisions for food concerns. This study integrated the eficacies of supervised machine learning (ML) algorithms using compound growth exponential regression models (CGREM) with linear regression models (LRM) to predict the wheat area, yield and population explosion. hTe historical data are collected from 1950 to 2020 and ML models are deployed using the 80% train and 20% test datasets. Various combinations of train test split have been applied to check the precisions of data partitions subset on the deployed models. ML model predict that the wheat area will rise up to 15%, 51.7%, wheat yield will grow up to 28%, 109.7% and population will rise up to 34.2%, 140.6%, respectively for the year 2030 and 2050. Population will upturn about 88.9% and 30.9% more than from wheat area and wheat yield up to 2050 and it might explode the food critiques in the region. It is concluded that wheat productivity must be raise to ensure the food demands. The results of this study demonstrated that the CGREM found to be superior comparing with benchmark LRM.

Publication Details

Page(s) 894-901

DOI 10.17582/journal.sja/2022/38.3.894.901

Published Journal: Sarhad Journal of Agriculture, Volume: 38, Issue: 3, Year: 2022

Request the full text from PASTIC

Keywords

Prediction, machine learning, Population explosion, Wheat area and productivity, Exponential regression