Construction of Supervised Learning Model for Crop prediction based on Environmental Condition

Abstract

Agriculture is defined as the backbone of our nation. Along with providing food products, it also increases the economic growth of the country. There is a crucial need for technologists and engineers to come up with various technologies and aids to help the farmers to succeed in farming to prevent the death of farmers and urbanization. Urbanization is one of the major threats to human society. If the farmer can identify the property of the soil and nutrients available in the soil even before sowing, it will be extremely helpful for him/her to proceed with the further steps i.e., to pick a perfect crop that can produce a maximum yield. If that happens and a perfect cycle repeats every year, it will also increase the number of nutrients in the soil. This proj ect includes the analysis of the property of the soil by measuring certain environmental parameters such as nitrogen, potassium, and the phosphorous content in the soil and also environmental parameters such as temperature, humidity, ph. level and the rainfall amount. This data acquired further undergoes some pre-processing techniques like cleaning the data and transformation of the data to the desired format. The cleaned data is then split into two divisions. One is for training and the other one is for testing the software. The data used for training is then used analyzed using various machine learning algorithms such as Linear Discriminant Analysis, Decision Tree algorithm, and the Random Forest algorithm. Then a graph is generated for each of the algorithm based on certain parameters. These parameters include precision comparison, recall comparison, and the Fl score comparison for various fruits like apple, banana, grapes, mango, muskmelon, orange, papaya, etc., and other crops varieties such as chickpea, coffee, kidney beans, lentils, moth beans, maize, etc. Once all the parameters are analyzed using the graph, the best crop that is suitable for the soil will be suggested to the farmer. Using this data, he/she can sow a suitable crop and increase the average yield of the soil.