Using Rapidminer Build Assignment Help: How to Answer This Question

Understanding this question requires applying core subject principles.

What This Question Is About

This question relates to using rapidminer build and requires a structured academic response.

How to Approach This Question

Break the problem into smaller parts and analyze each logically.

Key Explanation

This topic involves using rapidminer build. A strong answer should include explanation, application, and examples.

Original Question

1. Using RapidMiner, build a logistic regression model, using 70% of the entire data for training and the remaining 30% for testing (using shuffled sampling in RM), to classify patients into those who are likely to have a stroke and those who are not. Use all available variables as predictors except id and the default thresholds by RM to generate the classifications. [40 pts.] (a) Which independent variable(s) are significant at p=0.05? (b) How to interpret the coefficients of the following two variables in this logistic regression: (i) age, and (ii) hypertension? Explain in terms of the effect of a unit change of the independent variable on the odds of the dependent variable and show steps of your derivation to get full credits. For example, how does a unit increase in age affect the odds of stroke? Explain not just whether it affects the odds positively or negatively but also by how much. Similarly, how does hypertension affect the odds of stroke? (c) Evaluate the predictive accuracy of the model using appropriate metrics (e.g., specificity, sensitivity, precision, false positive rate, false negative rate, AUC, etc.) that you learned this week. (Do not just provide the numbers; offer your own analysis of what you think of the model based on those numbers, e.g., what do these numbers represent, are they indicative of good or poor performance, etc.) 2. Assuming that the dataset contains a representative sample of the patients, answer the following questions. [25 pts.] (a) If one randomly selects 100 patients from the dataset, what is the expected number of patients who would potentially have had a stroke? Hint: to obtain the expected number, multiply the probability of a stroke patient from the dataset by 100. (b) Now, if the predictive model that you developed in Q1 selects 100 stroke patients, what is the expected number of patients who would truly have a stroke? (Hint: confusion matrix, multiply the probability that a patient deemed/classified as stroke by the predictive model is truly a stroke patient by 100) (c) Based on your responses to (a) and (b) above, would your predictive model be useful for predicting whether a patient is likely to get a stroke? Explain why. 3. If one uses your model in Q1 to identify the potential stroke patients, how good is your model for this purpose, i.e., out of the patients who truly had a stroke, what percentage of these patients would be correctly classified by the model? Is the performance considered good or bad? Explain. [10 pts.] 4. Suppose one wants to improve the accuracy of identifying the potential stroke patients (i.e., those who would have a stroke) to at least 70% from the number in Q1. [25 pts.] (a) Revise the model in Q1 to achieve this goal. (b) Compare the predictive accuracy of this revised model with that of the model developed in Q1 using appropriate metrics (e.g., specificity, sensitivity, precision, false positive rate, false negative rate, AUC, etc.) that you learned this week. (Again, try to be analytical instead of just providing the numbers, e.g., what do these numbers represent? Does the new model provide better or worse performance based on those numbers compared to the model in Q1?) Q1-3 Image transcription text Attribute gender_Female ever_maried_ Yes work_type_Private work_type_Self-employed work_type_Govt_job work_type_children Res… Show more Image transcription text accuracy: 95.58% true 0 true 1 class precision pred. 0 3281 143 95.82% pred. 1 9 3 25.00% class recall 99.73% 2.05% Question 4 Image transcription text Attribute gender_Female ever_married_Yes work_type_Private work_type_Selfemployed work_type_Govt_job work_type_children Res… Show more

******CLICK ORDER NOW BELOW AND OUR WRITERS WILL WRITE AN ANSWER TO THIS ASSIGNMENT OR ANY OTHER ASSIGNMENT, DISCUSSION, ESSAY, HOMEWORK OR QUESTION YOU MAY HAVE. OUR PAPERS ARE PLAGIARISM FREE*******."