From symptom to solution : leveraging machine learning for improving healthcare prognosis
Από το σύμπτωμα στη λύση : αξιοποιώντας τη μηχανική μάθηση για τη βελτίωση της πρόγνωσης της υγείας
View/
Keywords
Machine learning ; Deep learning ; Medical science ; Disease diagnosis ; Healthcare ; Classification models ; Random Forest Classifier ; Decision Tree Classifier ; Logistic regression ; Gradient Boosting Classifier ; Data preprocessing ; Hyper parameterAbstract
This study evaluates and compares the performance of five machine learning algorithms for
medical diagnosis using a Kaggle dataset. The algorithms include Gradient Boosting, Decision
Tree, Random Forest, Logistic Regression, and Multinomial Naive Bayes classifiers. The dataset
was preprocessed and partitioned using cross-validation for robust model evaluation. The
evaluation metrics used were accuracy, precision, recall, and F1-score. All algorithms, except
the Multinomial Naive Bayes Classifier, exhibited commendable performance across all metrics.
The findings underline the efficacy of machine learning algorithms in medical diagnosis and
present a foundation for future explorations in this domain.
Postgraduate Studies Programme
ΠληροφορικήDepartment
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα ΠληροφορικήςNumber of pages
47Language
EnglishCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα