Prediction of diabetes for Big data using Classification comparison - Shraddha Shukla
Two-class boosted decision tree classification algorithm is used to predict the chances of diabetes in a patient with around 95% accuracy.
This model uses big data from two different sources. The first is from azure cloud while the second data sourse is an SQL database. Both the data are joined for further processing. The label for prediction is the chance for diabetes, The prediction of diabetes is quite accurate by use of Two-class boosted decision tree classification algorithm when compared to Two-class logistic classification algorithm. The accuracy for Two-class boosted decision tree classification algorithm is around 95% while the recall i.e. the fraction of positive cases correctly classified is around 92%.
