PREDICTION OF CROPS CULTIVATION CLASSIFYING SOIL THROUGH ARTIFICIAL INTELLIGENCE
No Thumbnail Available
Date
2022-11-30
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The field of agriculture is one of the significant sources of people surviving in
Bangladesh. Soil is one of the prominent components in agriculture for cultivating crops.
There are different types of soil, and there are various features of each soil. Based on the
soil types, several kinds of crops are grown. However, the agricultural land is decreasing
day by day while the number of people is increased. On the other hand, farmers do not
know what crops can be cultivated in their land. There are many lands in our country
where the people think that cultivation is not possible at all. Due to these reasons, many
lands are uncultivated in our country every year. Considering these issues, this research
work aims to predict the crops cultivation for a particular soil so that the people can select
and cultivates crops undoubtedly and correctly in their land. Consequently, all the
agricultural land will be cultivated and we will be able to provide our own food. The soil
types are determined first based on its distinct characteristics for a particular zone then
the crops cultivation is ascertained according to the soil types. Artificial intelligence is an
emerging and challenging research field in agriculture. Data is collected from different
zone of Noakhali district. Then preprocessed data is splitted into training and testing.
Several artificial intelligence techniques such as Support Vector Machine, Decision Tree,
Multilayer Perceptron Neural Network, Random Forest, Logistic Regression and Naïve
Bayes are used for prediction. For soil classification, Random Forest algorithm is selected
on the basis of accuracy, precision, recall, f1-score and ROC. The accuracy of Random
forest is 95.15% which is higher than others algorithm. And for crops prediction, Support
vector machine is selected. The accuracy of support vector machine is 91.17% which is
higher than other algorithms. Thus, this research work will become an effective and
intelligent tool to enhance cultivation in our country and can play a significant role in
economy development of Bangladesh.