Testing Employee Attrition Modelling

Employee Attrition modelling
Author
Affiliation
Course
Instructor
Due Date
Testing Your Model
What results did you expect?
After reading the verification dataset, the model should use the features available in the
dataset to predict. The expected result is a binary output with a "yes" or "no" prediction based on the model's assessment of the employees' profile likelihood to churn. We expect the model to produce a true positive rate of 95 with a false-positive rate of 22 percent.
What results did you not expect?
The results not expected were an error. Furthermore, we do not expect the model to predict only one categorical value, and hence, the prediction should have both a "yes" and "No" mixture. The model could be biased or guessing if it can only predict one categorical value.
How do the results for your model change when using the verification data set, as opposed to the training data?
The model was cabale of making prediction ; however, the accuracy of the model could not be verified. To very the accuracy, we need the actual data on the “Attrition” in order to make comparison with...