Description
Upload a comprehensive report (Word document) that contains the following parts:
- Intro to Your Project: CASE STUDY 1 – Health Analytics – Immunization Case Study
- Step 1: Business Problem Understanding
- Step 2: Data Understanding and Collection
- Step 3: Data Preparation and Feature Selection
- Step 4: Modeling Development
- Step 5: Model Evaluation and Interpretation
- Step 6: Model and Results Communication
- Step 7: Model Deployment
- Summary of findings & Conclusion
Note that you must use a provided template.
***I HAVE TO CREATE A 10- 20-minutes Pitch Video: video presentation of the Project. Any pointers, example or actual pitch will be appreciated.
Case Study – Health Analytics – Immunization Case Study
- Question: Which factors best determine whether a patient will receive a flu shot immunization?
- Notes
- The target (y) is FluShotVaccine.
- One advantage is that the model can handle varying input and target with a decision tree. But you can also use Logistic Regression.
- Do the following before creating a model, Detect Missing Data, and if there is any, you need to (replace, impute, or remove) them. The decision of which method to use is up to you — as long as you provide a rational explanation of your decision.
- Detect Outliers – remove these outliers from your data and explain how you accomplished this step.
- Must use more than one model (minimum of two, preferable three)
- Must use the following metrics to evaluate the model:
- Confusion Matrix
- Receiver Operator Characteristics (ROC)
- Lift Chart
- Finally, I assume that you select the best model based on the above metrics and explain your decision in detail.
- Most importantly, interpret your findings and backup your conclusion with screenshots from RapidMiner (Confusion Matrix, Accuracy, Precision, Recall, ROC, and Lift Chart)
