Agricultural seasonal crops recommendations for classifications and productivity accuracy using random forest compared with KNN Available to Purchase

Crop recommendations for agriculture that take seasonality and productivity into account are the focus of this study, which employs random forests as an alternative to KNN. When it comes to agricultural planning, crop selection is crucial for farmers. Using machine learning approaches, numerous studies investigated crop yield rate, weather prediction, soil classification, and crop classification for agricultural planning. Materials and Methods: With each sample passing through 10 iterations, the experimental study uses 183 crops to issue samples from the agricultural crop recommendations based on productivity and season crops (ACR BPSC) databases. After each operation has completed 10 rounds, the loss values are used to calculate the statistical statistics for comparison. The data is further partitioned into two sets of iterations, one for the train dataset and one for the test dataset, with a ratio of 70:30. In order to analyze and assess the datasets, two sets of forty samples representing different crop seasons are utilized. In order to process and analyze the dataset, two groups were utilized, each with forty samples representing a different agricultural season. A random forest method was used in Group 1, whereas the K nearest neighbors identification based on seasonally recommended crops was used in Group 2. Results: In comparison to a K nearest neighbors algorithm with a significance value of (0.004) where (<0.05), the random forest method achieved a significance level of 86.5010 for accuracy.conclusion: When it comes to evaluating and detecting the season based on crop advice, the random forest algorithm performs better than the K nearest neighbors approach with an accuracy rate of 80.94.