<![CDATA[Rice seeds are one of the main elements in agricultural businesses. The choice of type of rice seed planted can influence the quality of the harvest obtained. The large number of varieties of rice seeds with similar shapes makes identifying the type of rice seed an activity that is not easy and requires experts to do. One fairly fast way to identify rice seed varieties is to use machine learning technology. This research will implement machine learning classification algorithms, namely KNN, Naïve Bayes, and Random Forest. Identification of rice seed varieties is carried out based on the morphological features of the seeds. The dataset used is in the form of seed morphological feature values, namely aspect ratio, solidity, circumference, area, area, roundness, circularity and equivalent diameter. Research stages starting from preprocessing, feature extraction, and experimental parameter values were carried out to find the model with the best performance. Feature selection can increase the testing accuracy on KNN and Random Forest models. The test results obtained an accuracy of 78.3% with KNN, 61.7% using Naïve Bayes, and 90% using Random Forest.]]>
Copyrights © 2024
