DEVELOPING A SMART INTEGRATED MODEL BASED ON MACHINE LEARNING FOR THE EFFECTIVE DETECTION AND DIAGNOSIS OF A SPECTRUM OF DISEASES

, and ABSTRACT Disease diagnosis is crucial in the medical field, and timely and accurate diagnosis is necessary for efficient treatment and management. AI methods, including Naive Bayes, have shown guarantee in disease detection and analysis. A machine learning-based, Naive Bayesian network-based, multi-disease prediction system is presented in this study. The proposed method aims to provide accurate disease predictions for several diseases immediately. We also talk about the work's social relevance, focusing on the potential impact of accurate disease prediction on improving patient outcomes and lowering healthcare costs, in addition to describing the methods used, which included feature selection, pre-processing, dataset selection, and the Naive Bayesian network algorithm. To assess the presence of the proposed model, we executed tests using an openly accessible disease dataset. The outcomes exhibited that the proposed model accomplished high precision of 91.2% and outflanked other best-in-class models for multi-disease prediction; Random Forest, which got 85.7%, and Decision Tree, which got 81.3%, are two examples. In conclusion, the proposed system demonstrates how well Naive Bayesian networks can predict multiple diseases and potentially enhance medical disease diagnosis and treatment.