A comparison of CNN and SVM algorithms for the prediction of growth defects in coffee plants for stable yield and fungal diseases

Coffee plant growth discrepancies can be resolved by using machine learning techniques to predict plant development, reduce yield instability, and minimise fungal infections. use support vector machines (SVM) and convolutional neural networks (CNN) for investigation. Using synthetic datasets with the right sample sizes allows for the dispersion of data. With SPSS software, the CNN and SVM algorithms were contrasted. A total of 40 samples were subjected to the criteria used to determine the fertiliser’s cost (20 samples per group). Group II serves as the control group, while Group I is classified as the experimental group. The G power at 80% and alpha = 0.05, when a significant difference between the samples from the ARIMA and CNN algorithms was found at a significance level of (p < 0.004), is used to calculate the sample size. The present study examines the impact of data dispersion strategies on cloud system availability and redundancy. It evaluates how successfully CNN and SVM algorithms predict stable yields, fungal infections, and growth defects in coffee plants. The SVM method’s effectiveness evaluations, which range from 70.2 to 84.51%, are presented in every cluster. for the intention of forecasting coffee plant growth challenges. With a mean accuracy of 95.805%, the CNN algorithm outperforms the SVM method’s mean accuracy of 79.679%. This suggests that there is more predictive power in the CNN algorithm.