1) Build a Naive Gaussian model to predict GRADUATED based on available data.

2) Evaluate the accuracy of your model.

Function call 1:

predict_graduation(df_train, df_test)

# optional: process data to attempt to increase accuracy

return predicted outcomes for df_test

Input:

df_train = a data frame like the assignment3 ones...

df_test = a data frame of unknown size but otherwise identical to DATA_ASSIGNMENT_3_n.csv except it does not have the GRADUATED

column.

Output:

predicted outcomes is a sequence of values (in this case predictions for GRADUATED)

Function call 2:

evaluate_model(predicted, observed)

#

#determine True Positive (predicted_values = observed = True)

#determine True Negative (predicted_values = observed = False)

#determine False Positive (predicted_values = True, observed = False)

#determine False Negative (predicted_values = False, observed = True)

#determine specificity (True Negative / (True Negative + False Positive))

#determine sensitivity (True Positive / (True Positive + False Negative))

return TP, TN, FP, FN, specificity, sensitivity

input: predicted = sequence and observed = sequence

dataset:

SMARTS,MOTIVATION,HOME_LIFE,GRADUATED

2,2,2,True

2,3,2,True

2,2,2,True

3,1,2,True

2,2,2,True

2,2,2,True

2,3,3,False

2,2,2,True

2,2,2,False

2,2,2,True

2,2,2,True

2,2,2,True

1,3,1,True

2,2,2,True

2,2,2,True

2,2,2,True

1,2,2,False

2,2,1,False

2,3,2,True

2,2,2,False

3,2,2,True

2,1,3,True

2,2,3,True

2,2,1,False

2,2,2,False

3,2,3,True

3,2,2,True

2,2,2,True

2,2,3,False

2,2,2,True

3,2,2,True

2,3,2,True

2,3,2,True

2,2,3,False

2,1,2,True

3,2,2,True

2,3,2,False

2,2,2,False

3,2,2,True

3,2,2,True

2,2,2,True

2,2,2,True

3,1,2,True

2,2,1,True

3,2,3,True

2,1,2,False

2,2,2,True

2,2,2,False

2,2,2,False

2,2,2,True

2,2,1,False

2,2,2,True

3,2,2,False

2,2,2,True

2,2,1,True

2,1,2,False

2,3,1,True

2,2,2,True

2,3,2,True

2,2,2,True

1,2,2,False

2,2,2,False

2,3,3,True

2,2,2,False

2,2,2,False

2,2,3,True

2,3,2,True

1,1,3,False