|
AUTHOR(S): Avantika Singh, Komal Saxena
|
TITLE |
 PDF |
ABSTRACT This paper explores the establishment and testing of a healthcare recommendation system for personalized medication suggestions. It employs advanced machine learning methods on various patient profiles, drug prescriptions, and medical conditions using UCI ML repository dataset. A rigorous data collection process is conducted, cleaning and exploratory analysis are performed to elicit insights for model building. Three main models – SVM, XGBoost, Multinomial NB were compared in terms of their performance in medication recommendation tasks where XGBoost performed better than all other models. The research highlights the need for large-scale datasets and more complex algorithms in improving patient care optimization. Additionally, it points out some directions for future work such as feature integration and model refinement to increase adaptability across different clinical settings. |
KEYWORDS Support Vector Machine (SVM), XGBoost, Multinomial Naive Bayes (NB), Exploratory data analysis (EDA) |
|
Cite this paper Avantika Singh, Komal Saxena. (2024) Optimizing Medicine Recommendation Systems: A Comparative Analysis of SVM, XGBoost and Multinomial NB Models. International Journal of Computers, 9, 69-73 |
|
