Being able to predict machine failures before they happen can save millions of dollars for manufacturing companies. Manufacturers want to be able to perform preventive maintenance or repairs in advance to minimize machine downtime and often install electronic sensors to monitor the machines and their surrounding environment. However, the more sophisticated the machine, the more difficult it is to diagnose and predict the failure rate. Data mining has been used to analyze environmental factors to predict whether or not complex machines such as nanotechnology equipment will fail from one production period to another. The accompanying data set contains 480 observations and three environmental variables: level of humidity in the room where the equipment is located (Humid, in percentage); overall temperature in the room (Temp, in Fahrenheit); and a target variable that indicates whether or not the equipment broke down during the next production period (Breakdown = 1 if breakdown, 0 otherwise). 

a. Perform KNN analysis on the data set. What is the optimal value of k? 

b. What is the misclassification rate for the optimal k? 

c. Report the accuracy, specificity, sensitivity, and precision rates for the test data set (for Analytic Solver) or validation data set (for R). 

d. What is the area under the ROC curve (or the AUC value)? 

e. Based on your answers in part c and d, is KNN an effective way to classify potential customers? 

f. Change the cutoff value to 0.3. Report the accuracy, specificity, sensitivity, and precision rates for the test data set (for Analytic Solver) or validation data set (for R).

Transcribed Image Text:

Humid 9.36 10.37 ⠀ 15.02 Temp 79.99 61.89 ⠀ 38.02 Breakdown 0 0 ⠀ 1