College of Computer Science and Engineering Department of Computer Science and Artificial Intelligence CCAI-312: Pattern Recognition 14 points Question 2: CLO 1.2 (K2) 17 Copacent 15 thered the linear pension, Baye, PCA, KA Aparat / 3 points) Choose the correct answer . (1+1) Assume you are building ader to detect sentiment analysis from the working with has 95 positive tweets and the restare negative. Now, after building your model, you need to evaluate it Which of the following metrics best suits your problem. Justify your answer fexplain who chose this werd. Mean Absolute Error (MAE) it. Accuracy F1 score Justifications b. 1) Assume you have trained a model, and it underfits your data. Cheskall solutions below that can help for an underfitting problem. 1. Use basis functions to fit non-linear data ll. Increase regularization parameter Il Decrease regularization parameter