Disease prediction using naive bayes, random forest, decision tree, KNN algorithms

Abstract

In contemporary society, encountering individuals afflicted with various diseases is a common occurrence, emphasizing the critical need for accurate disease prediction as an integral facet of effective treatment. This paper focuses on leveraging classification algorithms such as Naive Bayes, Random Forest, Decision Tree, and KNN to predict diseases based on patient symptoms. This system enables users to input symptoms and, through meticulous analysis, accurately forecast the disease the patient may be suffering from. The prediction model extends to specific diseases like heart disease and diabetes, providing the outcome of the presence or absence of a particular ailment. The potential impact of such a predictive system on the future of medical treatment is substantial. Upon disease prediction, the system not only identifies the ailment but also recommends the appropriate type of doctor for consultation. This paper reviews recent advancements in utilizing machine learning for disease prediction and emphasizes the creation of an interactive interface as the front-end for user-friendly symptom input. By leveraging machine learning algorithms, this system extracts valuable insights from medical databases, aiding in early disease prediction, patient care, and community services. A comprehensive analysis was conducted using a dataset comprising 4920 patient records with 41 diseases. This integrated machine learning-based disease prediction system represents a significant step forward in leveraging advanced technologies for enhancing healthcare outcomes.