Effective Data Mining Techniques Performance Analysis to Predict Anemia Disease Using Orange Tools
##plugins.themes.academic_pro.article.main##
Abstract
The most well-known type of dietary inadequacy is anemia disease. It is more common in undernourished people and affects humans equally. There has been a worldwide awareness of the use of anemia supplements for people due to this anaemia deficiency. Diagnosing the condition at an earlier stage of life is preferable to prevent further harm and create an appropriate treatment. In this study, anemia is taken into account for early disease prediction and diagnosis by analyzing the data and evaluating the training performance of classifiers in terms of accuracy, precision, sensitivity or recall, specificity, F-measure, Matthews correlation coefficient, and training duration. The volume of data in healthcare organizations is greater. An efficient way to extract knowledge from this kind of data is needed. Data mining is used to discover knowledge from large amounts of data in databases. A classification technique, which is a data mining technique, is used to classify the stages of anemia. The data was collected from 397 households of visitors to the Medical Alabideen Lab in Wadi Etabah, Libya. The research is carried out using Orange software. An experimental study will be conducted with the Anemia data set to determine the best prediction utilizing multiple data mining techniques. As a result, the performance of eight classification techniques is examined, and their training performance is compared using a confusion matrix. It has been determined that the Ensemble classifier outperforms the other techniques in terms of training performance precision.