Improved accuracy in rice quality prediction using logistic regression in comparison with linear regression

The major purpose of this project is to improve the accuracy of rice quality in real-time images using Logistic Regression instead of Linear Regression. Research Methods and Equipment: The suggested approach uses 273 rice samples for training and 83 samples for testing out of a total of 356 samples. A sample size of N=10, a G-power of 80%, and a significance level of 0.05 are utilised in both the logistic and linear regression analyses. Experiments with the rice dataset are carried out using the Jupyter software application tool. Findings: The accuracy of Logistic Regression for Rice Quality Prediction is higher than that of Linear Regression. With a proposed Logistic Regression accuracy of 92.12% and a Linear Regression accuracy of 88.10%, there is a statistically significant difference between the two approaches. With a p-value of 0.001 (p