Open AccessSOIL ANALYSIS AND CROP FERTILITY PREDICTION USING MACHINE LEARNING
Author(s) -
Jagdeep Yadav,
Shalu Chopra,
M. Vijayalakshmi
Publication year - 2021
Publication title -
international journal of innovative research in advanced engineering
Language(s) - English
Resource type - Journals
ISSN - 2349-2163
DOI - 10.26562/ijirae.2021.v0803.003
Subject(s) - machine learning , computer science , artificial intelligence , naive bayes classifier , support vector machine , perceptron , random forest , soil fertility , agricultural engineering , crop yield , agriculture , multilayer perceptron , artificial neural network , soil science , environmental science , soil water , agronomy , engineering , biology , ecology
Soil is a critical part of successful agriculture and is the source of the nutrients that we use to grow crops. There are different types of soil and there are different properties of each soil. On these different properties, several types of crops grow. We need to know the properties and characteristics of various soil types to understand which crops sow in certain soil types. Machine Learning allows the user to feed a computer algorithm on an immense amount of data and have the computer analyze, make data-driven recommendations and decisions based to analyze the input data. Machine Learning techniques are used to model this process. Machine Learning has come into the picture with the big data technologies and high-performance computing that create new opportunities for data-intensive science in the multi-disciplinary agri-technology domain. In this paper, we have proposed a model that can find whether the soil is fertile or not, Sowing crop seed on fertile soil, and at last predicting the crop yield on different soil features. According to prediction, it can be suggested and recommended which crops grow more. Various Machine Learning algorithms such as Support Vector Machine (SVM), Random Forest, Naive Bayes, Linear Regression, Multilayer perceptron (MLP), and ANN are used for soil classification and crop yield. Test results show that the proposed ANN method follows a deep learning architecture which means it has several layers for input and output are connected to achieve better accuracy than numerous existing methods.