Une approche de prédiction médicale basée sur les données cliniques utilisant des algorithmes d’apprentissage automatique
| dc.contributor.author | Souad BOUTEARI Alla Eddine RICHE | |
| dc.date.accessioned | 2024-09-19T09:23:31Z | |
| dc.date.available | 2024-09-19T09:23:31Z | |
| dc.date.issued | 2021 | |
| dc.description.abstract | In this work, we have designed four models to predict diabetes in order to reduce the risk and the occurrence of complications of this disease on the health of the patient. To design these models, we used four machine learning algorithms, i.e. K nearest neighbors KNN, Decision trees DT, Support Vector Machine SVM, and Logistic Regression LR. The performance of the obtained models was tested according to the accuracy of each model. The highest accuracy rates were obtained in the decision tree model in both the split method (Train / Test Split) and k_fold cross validation splitting model. Keywords: machine Learning, K nearest neighbors, Decision trees, Support vector machine, Logistic Regression, diabetes prediction, medical prediction. | |
| dc.identifier.uri | http://dspace.univ-khenchela.dz:4000/handle/123456789/6640 | |
| dc.language.iso | fr | |
| dc.title | Une approche de prédiction médicale basée sur les données cliniques utilisant des algorithmes d’apprentissage automatique | |
| dc.type | Thesis |