Comparative Accuracy Of Prediction Classification Using Supervised Machine Learning

Le résumé fourni par la source

Machine learning, a field of Artificial Intelligence, gives systems the ability to learn from data and experience, allowing them to make predictions or decisions without being explicitly programmed. Supervised learning is a common approach within machine learning where algorithms are trained using data that has already been labeled. The main purpose of classification is to recognize patterns or relationships in data so that the model can provide accurate predictions for new data that has never been seen before. From several models formed from the supervised learning classification model, researchers will compare datasets from Indian employees taken from public datasets to obtain the most accurate level of accuracy from several models formed. The results of testing machine learning methods showed an accuracy rate for neural networks of 61.36%, logistic regression of 65.12%, SVC of 61.36%, gradient boosting classifier of 79.59%, extra trees classifier of 72.36%, bagging classifier of 74.38%, Boost classifier of 75.40%, gaussian NB of 64.25% MLP classifier of 49.06%, XGB classifier of 74.67%, LGBM Classifier classifier of 77.86%, K nearest neighbor classifier of 69.32%, Decison Tree Classifier is 70.48%, Random Forest Classifier is 72.94%. From 14 supervised machine learning models, it can be concluded that the ensemble learning algorithm has the highest accuracy rate of 79.59%.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.