Diabetes is a disease that occurs when your blood glucose, also called blood sugar, is too high. Victims of this disease are increasing day by day.
In this respository, a diabetes prediction application is created having a web based frontend and a machine learning backend. Here different classification models are used to predict if a patient is diabetic or not by taking essential patient details as input such as number of pregnancies, Plasma glucose concentration, Diastolic blood pressure, Triceps skin fold thickness, 2-Hour serum insulin, Body mass index, Diabetes pedigree function, and Age to predict if the patient is diabetic or normal.
The dataset is obtained from kaggle: https://www.kaggle.com/uciml/pima-indians-diabetes-database
Below are the screenshots:
Below are the various classification models applied to the dataset are compared using accuracy_score r2_score:
| Regression Models | Accuracy Score | R2 Score |
|---|---|---|
| K Nearest Neighbor | 75.32 | -0.1636 |
| Naive Bayes | 79.22 | 0.0200 |
| Random Forest Classification | 75.32 | -0.1636 |
| Linear Support Vector Classification | 70.12 | -0.3780 |
| Support Vector Classification | 81.81 | 0.1425 |
accuracy_score is the percentage of the success of a model to predcit the independent attribute and r2_score is a statistical measure that represents the goodness of fit of a regression model. The ideal value for r2_score is 1, its range is from -1 to 1. Some other methods to determine the success of a classification model are mean_squared_error, mean_absolute_error, confusion_matrix, calssification_report.
Out of all the Classification models above, Support Vector Classification has the highest accuracy of 81.81%, this model is deployed using flask inorder to provide a web based interactive interface for users.