Data mining for ranking sorghum seed lots

ABSTRACT The ranking of seed lots is a fundamental process for all companies in the industry. This work aims to demonstrate data mining methods sorghum during processing through analysis quality control data. Germination and cold tests were performed verify physiological lots. Seed samples from each lot evaluated two moments: post-cleaning finished product (ready marketing). results after pre-processing totaled 188 rows with six attributes, encompassing 150 accepted marketing, 6 rejected, 32 intermediate classifiers used J48, Random Forest, Classification Via Regression, Naive Bayes, Multilayer Perceptron, IBk. Resample filter was adjustment k-fold technique training, ten folds. metrics Accuracy, Precision, Recall, F-measure, ROC Area accuracy algorithms. obtained determine best machine-learning algorithm. IBk J48 presented highest data; results. essential solving imbalance problem. Sorghum can be classified great precision artificial intelligence machine learning technique.