Title: A machine learning based crop recommendation system and user-friendly android application for cultivation

Authors: Kaniz Fatima Tonni; Mahfuzulhoq Chowdhury

Addresses: Computer Science and Engineering Department, Chittagong University of Engineering and Technology, Chittagong 4349, Bangladesh ' Computer Science and Engineering Department, Chittagong University of Engineering and Technology, Chittagong 4349, Bangladesh

Abstract: Bangladesh is essentially an agricultural nation, and its economy is heavily dependent on it. A farmer could plant a crop if he knew which one would yield more. The existing literature works fail to provide a user-friendly mobile application for cultivation as well as machine learning-based crop recommendation by taking different factors into account. This paper creates a mobile application that enables farmers to forecast viable crops based on climate factors like humidity, rainfall, and temperature as well as soil characteristics. The suggested model is used to forecast agricultural production using crop records of diverse crops with various properties of soil and climate parameters. The suggested model offers farmers a comprehensive list of recommendations to help them choose crops that are best for them based on particular considerations like production costs and fertiliser recommendations. The user's feedback shows satisfactory remarks in terms of its usefulness.

Keywords: crop recommendation; machine learning; prediction; evaluation; cultivation; Bangladesh; Android application.

DOI: 10.1504/IJSMARTTL.2023.129648

International Journal of Smart Technology and Learning, 2023 Vol.3 No.2, pp.168 - 186

Received: 03 Oct 2022
Received in revised form: 21 Jan 2023
Accepted: 25 Jan 2023
Published online: 17 Mar 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article