Order from us for quality, customized work in due time of your choice.
You are a data and business-intelligence analyst working for a network of hospitals in rural districts. The hospital’s chief of behavioral health services (CBHS) would like to use existing research data to create a model to predict or classify newer patients as at-risk or not at-risk for clinical depression. Such predictions will enable them to provide early mental health interventions to at-risk individuals. The CBHS has tasked your team with creating the predictive model and then testing the model on five new patients at the hospital.
You applied logistic regression to build a predictive model for the data set and presented your report to the CBHS. After reviewing the report, the CBHS is now interested in exploring additional predictive methodologies that could be used to analyze the historical depression data set, and potentially create more accurate predictions to improve the hospital’s ability to detect patients who are at-risk for depression.
In Module Two, you applied logistic regression techniques to build a predictive model in Excel. In this assignment, you will apply advanced predictive models such as a decision tree and a random forest model to the same historical clinical depression data set using Rattle, a package for R Studio, within the VDI. You will then use the predictive model to classify five new patients at the hospital network as at-risk or not at-risk. You will then compare your results from the different models and share your analysis with the CBHS.
Directions
Create a report with your analysis about advanced predictive analysis models, such as the decision tree and random forest models. Include relevant screenshots from Rattle.
Specifically, you must address the following rubric criteria:
Build a Decision Tree: Build a decision tree prediction model for the given historical data set. Include relevant screenshots.
Identify the independent and dependent variables in the given data set.
Apply the decision tree algorithm on the given data set to produce a binary classification model.
Generate a visualization of the decision tree.
Interpret the results of the algorithm.
What does the output tell you about the risk factors for depression?
What business rules did the model generate, and what do they tell you about the data set?
Apply the Decision Tree: Apply the decision tree model to make predictions about new data. Include relevant screenshots.
Apply this model to the new set of patient data you have been given.
Make predictions about the risk of depression in each of the new patients.
Build a Random Forest: Build a random forest prediction model for the given historical data set. Include relevant screenshots.
Apply the random forest algorithm on the given data set to produce a binary classification model.
Interpret the results of the algorithm.
What does the output tell you about the risk factors for depression?
Apply the Random Forest: Apply the random forest model you built to make predictions about new data. Include relevant screenshots.
Apply this model to the new set of patient data you have been given.
Make predictions about the risk of depression in each of the new patients.
Compare Predictive Models : Compare and contrast the random forest model, the decision tree model, and logistic regression.
Discuss the differences (if any) in the final predictions for the five new patients using each of the models.
What do you think led to these differences?
Compare the accuracy of your prediction models.
Discuss the out-of-bag (OOB) error rate of the advanced predictive models and how this might affect your predictions.
Which model do you think is more accurate? Why?
Share your thoughts about the pros and cons of each model.
For which type of situations or business requirements would you choose each model?
Order from us for quality, customized work in due time of your choice.