Diabetic Risk Prediction Using Data Mining Techniques

Description

Data mining techniques are very useful in various real-world applications for decision making, prediction, and relationship building. therefore a number of engineering and scientific applications are designed using data mining and machine learning techniques. in this work, a data mining model has been presented which involves five supervised learning models namely Support Vector Machine (SVM), Artificial Neural Network (ANN), C4.5 Decision Tree, Bays Classifier, and CART decision tree. these algorithms are taking training from two different datasets obtained from Kaggle and UCI for Diabetic Risk Prediction. In Order to implement the weka-based library has used. the performance of the classifiers for Diabetic Risk Prediction is measured and stored in MySql database. The cross-validation carried out for testing of the models. based on experiments the comparative results are also described in terms of precision, recall, and F1-Score. additionally, the time requirements and memory usage of the algorithms are also computed. the comparative study is visualized using a line graph.

project includes:

1. A java based simulation
2.  database script
3. datasets

References 

[1] https://www.kaggle.com/uciml/pima-indians-diabetes-database

[2]https://archive.ics.uci.edu/ml/datasets/Diabetes