You have developed two classifiers: C1 and C2 which determine the posterior probability of movies being nominated for the Academy Awards. The results of the 2 classifiers are provided in Table 3 Table 3 True Label Not Nominated Nominated Not nominated Nominated Nominated Not nominated Not nominated Not nominated Nominated Nominated 0.15 0.75 0.65 0.70 0.55 0.40 0.20 0.25 0.45 0.15 C2 0.80 0.75 0.30 0.65 0.90 0.35 0.60 0.25 0.85 0.55 6 9 10 a. Plot the ROC curves for both classifiers and determine which classifier is better. (Plot them on the same graph.) Show detailed calculations and provide your explanations b. A movie is speculated to be nominated if the probability p>-6, where is a threshold value. At what value will C1 attain the highest F-measure? Draw the confusion matrix for C1 at that particular value. Compute the precision, recall, and F-measure of C1 at that 0 value. Suppose you are given a weight matrix as shown in Table 4. Use weighted accuracy to determine which classifier is better. Assume -0.6. c. Table 4 Predicted Class Yes es 10 Class You have developed two classifiers: C1 and C2 which determine the posterior probability of movies being nominated for the Academy Awards. The results of the 2 classifiers are provided in Table 3 Table 3 True Label Not Nominated Nominated Not nominated Nominated Nominated Not nominated Not nominated Not nominated Nominated Nominated 0.15 0.75 0.65 0.70 0.55 0.40 0.20 0.25 0.45 0.15 C2 0.80 0.75 0.30 0.65 0.90 0.35 0.60 0.25 0.85 0.55 6 9 10 a. Plot the ROC curves for both classifiers and determine which classifier is better. (Plot them on the same graph.) Show detailed calculations and provide your explanations b. A movie is speculated to be nominated if the probability p>-6, where is a threshold value. At what value will C1 attain the highest F-measure? Draw the confusion matrix for C1 at that particular value. Compute the precision, recall, and F-measure of C1 at that 0 value. Suppose you are given a weight matrix as shown in Table 4. Use weighted accuracy to determine which classifier is better. Assume -0.6. c. Table 4 Predicted Class Yes es 10 Class