|
| Citations | Objectives | Parameters | Methodology | Shortcomings |
|
| G Sai Pravallika et al., 2020 [31] | Crop selection | Temperature, moisture, humidity, and pH | No ML algorithm is applied. Simply compare the sensor data with the static data store in data base by SQL query to predict desirable crop. | Important parameters of soil fertility (NPK) are missing. An important factor rainfall is missing. |
| Bakthavatchalam et al., 2022 [21] | Crop prediction | Temperature, humidity, NPK, pH, and rainfall | Multilayer perceptron, JRip, Decision table | The performance metric results in Python are more reliable than in WEKA. |
| Waikar et al., 2020 [32] | Crop prediction | NPK, pH, and EC | SVM, Naïve Bayes, Artificial Neural Network, AdaBoost; Bagged Tree (Ensemble technique) | Results show less accuracy as compared to SCS. Important parameters, temperature, humidity, and rainfall, are missing |
| Jain and Ramesh, 2020 [33] | Crop selection | pH, NPK, and drainage capacity | RNN for weather prediction and Random Forest algorithm for crop selection | Random Forest is slow and inefficient for real-time predictions. Important parameters are missing. |
| A. Chlingaryan et al., 2020 [34] | Crop selection and fertilizer recommendation | Temperature, humidity, and water level | Linear Regression, Decision tree, K-Nearest Neighbors, and XGBoost | Limited parameters are considered for crop selection. |
|
