2. (a) CORE-TEMP STABLE- TEMP GENDER DECISION JID 1 2 3 4 5 6 high low high high low low TRUE male Igen TRUE female gen FALSE female licu FALSE male FALSE female icu TRUE male icu licu The training dataset above contains data about whether post-operative patients were sent to an intensive care unit (icu) or a general ward (gen) for recovery. You are required to develop a 1R model that can be used to predict the target DECISION for future patients. When answering each part below, you should provide detailed comments explaining your calculations. i. Calculate the entropy of the dataset with respect to the target variable DECISION. [2] ii. Calculate the information gain when the dataset is partitioned using the CORE- TEMP feature. [4] iii. Calculate the information gain when the dataset is partitioned using the STABLE-TEMP feature. [4] iv. Calculate the information gain when the dataset is partitioned using the GENDER feature. [4] v. Based on your answers to parts ii, iii and iv above, draw a diagram showing the completed IR model. [2] vi. In your opinion, would a decision tree be more appropriate than a 1R model for this dataset? [2] 2. (a) CORE-TEMP STABLE- TEMP GENDER DECISION JID 1 2 3 4 5 6 high low high high low low TRUE male Igen TRUE female gen FALSE female licu FALSE male FALSE female icu TRUE male icu licu The training dataset above contains data about whether post-operative patients were sent to an intensive care unit (icu) or a general ward (gen) for recovery. You are required to develop a 1R model that can be used to predict the target DECISION for future patients. When answering each part below, you should provide detailed comments explaining your calculations. i. Calculate the entropy of the dataset with respect to the target variable DECISION. [2] ii. Calculate the information gain when the dataset is partitioned using the CORE- TEMP feature. [4] iii. Calculate the information gain when the dataset is partitioned using the STABLE-TEMP feature. [4] iv. Calculate the information gain when the dataset is partitioned using the GENDER feature. [4] v. Based on your answers to parts ii, iii and iv above, draw a diagram showing the completed IR model. [2] vi. In your opinion, would a decision tree be more appropriate than a 1R model for this dataset? [2]