A.
By creating a scatter plot of Experience vs. Training, it was possible to visually examine the
relationship between these predictors and task completion. The scatter plot distinguished
between administrators who completed the task and those who did not. By analyzing the plot,
one could identify patterns or trends that indicate the potential usefulness of each predictor for
classifying task completion.
B.
A recommendation I can make to improve the model is to use cross validation to ensure the
model performs well on unseen data and is not overfitting.
C.
When I looked at the data I noticed the data change between each model. The metrics
completed tasks, prediction, confidence, experience, and training. For the completed task there
were more “No” than “Yes”. But for the other model there were more “No” than “Yes” but a
decrease. The confidence rate average of “Yes” for the first model was slightly higher compared
to the second model and the confidence rate average of “No” was slightly lower compared to the
second model average. The average for Experience for the first model was lightly lower than the
second model average. Then the average for Training was for the first model was lightly higher
than the second model average.