Disease Prediction System Using Machine Learning

Abstract

The healthcare system requires reliable and timely disease prediction, which is a basic requirement forits preventive healthcare services. The online Disease Prediction System applies supervised machine learning to align the user’s symptoms with their respective disease likelihoods based on the design and implementation of this paper. The system performs critical operations that include data preprocessing and feature selection and the application of three classification algorithms that include Random Forest and Support Vector Machine and Logistic Regression and Voting Classifier that functions as an ensemble learning model. A carefully selected symptom-disease dataset with 132 symptom features and 41 disease classes was applied in the development of the system. The Voting Classifier outperforms the other models with a 98.41

Introduction

Early disease detection is crucial for reducing healthcare costs, complications, and mortality, especially in rural areas with limited access to medical services. To address this, the study proposes a web-based Disease Prediction System that uses machine learning to provide preliminary diagnoses based on patient symptoms.

The system leverages machine learning algorithms to analyze large datasets and identify patterns between symptoms and diseases. It uses an ensemble approach (Voting Classifier) combining Random Forest, Support Vector Machine (SVM), and Logistic Regression to improve prediction accuracy and reduce errors. The system is implemented using a Flask-based web application where users input symptoms and receive predicted diseases along with precautionary advice.

The dataset consists of 132 binary symptom features and 41 disease classes. Data preprocessing, feature engineering, and techniques like normalization, correlation analysis, and feature importance are applied to improve model performance. The methodology includes training models with cross-validation and evaluating them using accuracy, precision, recall, and F1-score.

Results show that the Voting Classifier achieves the highest accuracy (98.41%), outperforming individual models. The system enables real-time predictions, efficient processing, and user-friendly interaction.

The literature review highlights that machine learning significantly improves disease prediction compared to traditional methods, but existing systems often lack real-time functionality, multi-disease prediction, and user-centric design. The proposed system addresses these gaps by integrating multiple models, advanced preprocessing, and a web interface.

However, the system has limitations such as dependence on data quality, lack of clinical inputs like lab reports, and potential bias. It is intended as a decision-support tool, not a replacement for medical professionals, and emphasizes data privacy and ethical considerations.

Conclusion

In this research paper, a reliable and functional Disease Prediction System was proposed using ensemble machine learningtechniques.TheVotingClassifierattainedanaccuracy of 98.41 The project shows how machine learning algorithms are used to predict diseases based on the assessment of patients’ symptoms. The project attains a high level of accuracy using its implementation of an ensemble Voting Classifier algo- rithm, which also improves the reliability of the system. The web-based system implementation allows users to access the system’s functionality in an easy-to-use manner. The system serves as a decision-making tool that assists users in recogniz- ing medical problems when they still need to visit a doctor. Futureworkincludes: • Expandingthedatasetwithreal-worldclinicalrecords and diverse demographics. • Incorporatinglaboratorytestinputsandtemporalhealth records to improve prediction reliability. • Exploring deep learning approaches for feature learning from raw clinical text. • Integratingsecureclouddeploymentandmobileapp front-ends for wider accessibility.

References

[1] T.M.Mitchell,MachineLearning.McGraw-Hill,1997. [2] L.Breiman,“RandomForests,”MachineLearning,vol.45,no.1,pp.5–32, 2001. [3] C.CortesandV.Vapnik,“Support-VectorNetworks,”MachineLearning,vol. 20, pp. 273–297, 1995. [4] L.Rokach,“Ensemble-basedClassifiers,”ArtificialIntelligenceReview,vol. 33, pp. 1–39, 2009. [5] S.PatelandA.Shah,“Web-basedSymptomCheckers:Stateofthe Art,”InternationalJournalofHealthcareInformatics,2021. [6] F. Pedregosa et al., “Scikit-learn: Machine learning in Python,” Journalof Machine Learning Research, vol. 12, pp. 2825–2830, 2011. [7] A. Esteva et al., “Dermatologist-level classification of skin cancer withdeep neural networks,”Nature, vol. 542, pp. 115–118, 2017. [8] A. Rajkomar, J. Dean, and I. Kohane, “Machine learning in medicine,”New England Journal of Medicine, vol. 380, no. 14, pp. 1347–1358,2019. [9] I. Kononenko, “Machine learning for medical diagnosis: History, stateof the art and perspective,” Artificial Intelligence in Medicine, vol. 23,no. 1, pp. 89–109, 2001. [10] B. Shickel et al., “Deep EHR: A survey of recent advances in deeplearning techniques for electronic health record analysis,” IEEE Journalof Biomedical and Health Informatics, vol. 22, no. 5, pp. 1589–1604,2018. [11] J. Han, M. Kamber, and J. Pei, Data Mining: Concepts and Techniques,3rd ed., Morgan Kaufmann, 2011. [12] K. Kourou et al., “Machine learning applications in cancer prognosisand prediction,” Computational and Structural Biotechnology Journal,vol. 13, pp. 8–17, 2015.

Copyright

Copyright © 2026 Dr. Ganapathi Rao Gajula, Pravalika Bandi, Maheshwari Jonnada, Sai Deekshitha Golla. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.