Five probabilistic classification models are used to create a single ensemble model for a classification task with four categorical labels: A, B, C, and D. A single observation is plugged into the ensemble model, generating the following class probability estimates: Class A Class B Class C Class D Model 1 0.35 0.30 0.20 0.15 Model 2 0.10 0.25 0.15 0.50 Model 3 0.30 0.25 0.25 0.20 Model 4 0.15 0.25 0.25 0.35 Model 5 0.40 0.35 0.10 0.15 Use this table to answer the questions below. 1. Assuming that a soft voting strategy is used to generate the final prediction for the observation, determine the ensemble model's estimate that the observation belongs to Class D. 2. Assuming that a soft voting strategy is used, what is the ensemble model's final prediction for the observation? Assuming that a soft voting strategy is used, what is the ensemble model's final prediction for the observation? Class A Class B Class C Class D