This paper uses machine learning techniques to predict human diseases based on a Kaggle dataset. Key steps in the workflow include data preprocessing and dimensionality reduction in which principal component analysis was employed to reduce the feature dimensions, simplifying the dataset while retaining important variance. A convolutional neural network was employed to extract high-level characteristics from the data. Three machine learning models were utilized: Classification and Regression Trees, Support Vector Machine, and Logistic Regression. The goal was to determine which model provided the best predictive performance. The SVM classifier showed superior performance with a training accuracy of 99.63%, testing accuracy of 97.62%, and an AUC value of 99.85%, indicating a highly effective model for disease classification. Based on the success of the SVM classifier, the model was further developed into a mobile application capable of predicting diseases from user-provided data.