Opioid Risk Prediction System
Dataset:- https://www.kaggle.com/apryor6/us-opiate-prescriptions
Analysis:- Dataset contains the cases of Opioid Overdose with the prescription and some more related data as Age, gender.
⁌ We can predict the case of Opioid Overdose by this data using some selected as there are too many of them, so we applied PCA for dimensionality reduction.
⁌ Then applied Upsampling of data.
⁌ Then used an ensembled model consisting of logistic regression, random forest, and SVM classifier.
⁌ Used Recall Value for accuracy
Recall Score = 98.943
Auc Score = 71.21
Accuracy = 62.84
