The RMS Titanic crashed into an iceberg in the North Atlantic on 1912, killing 1500 out...

Transcribed Image Text:

The RMS Titanic crashed into an iceberg in the North Atlantic on 1912, killing 1500 out of 2224 passengers and crew onboard. In the following table, we have sample data from a census about the Titanic ship. The data profiles passengers who survived the RMS Titanic or not. We have as attributes, or independent variables, 'gender', 'class' and 'age range'. The target variable for prediction is 'survived?'. Gender (G) Class (C) Age range (A) | Survived? (S) 20s 30s 20s 30s 30s M F F F M 3rd 1st 3rd 3rd 3rd No Yes No Yes No In this problem, build by hand two decision tree classification models from this training data, to predict whether a passenger survives the Titanic or not. The decision trees should have all leafs pure (i.e. all data points in the leaf belong to the same class). If needed, you can break ties in any way you want it, but write in your answer how you did it. For full credits, show how each split in the decision tree was decided, in addition to the final tree. (a) (4 pts) Build a decision tree using Information Gain as the criterion to choose the feature for each split. (b) (4 pts) Build a decision tree using Gini Gain as the criterion to choose the feature for each split. A (c) (2 pts) In the Oscar winning movie Titanic (1997), would the fictional character Rose, who is in her 20's and in 1st class, survive her doomed lover Jack as in the movie, according to each of G your models? The RMS Titanic crashed into an iceberg in the North Atlantic on 1912, killing 1500 out of 2224 passengers and crew onboard. In the following table, we have sample data from a census about the Titanic ship. The data profiles passengers who survived the RMS Titanic or not. We have as attributes, or independent variables, 'gender', 'class' and 'age range'. The target variable for prediction is 'survived?'. Gender (G) Class (C) Age range (A) | Survived? (S) 20s 30s 20s 30s 30s M F F F M 3rd 1st 3rd 3rd 3rd No Yes No Yes No In this problem, build by hand two decision tree classification models from this training data, to predict whether a passenger survives the Titanic or not. The decision trees should have all leafs pure (i.e. all data points in the leaf belong to the same class). If needed, you can break ties in any way you want it, but write in your answer how you did it. For full credits, show how each split in the decision tree was decided, in addition to the final tree. (a) (4 pts) Build a decision tree using Information Gain as the criterion to choose the feature for each split. (b) (4 pts) Build a decision tree using Gini Gain as the criterion to choose the feature for each split. A (c) (2 pts) In the Oscar winning movie Titanic (1997), would the fictional character Rose, who is in her 20's and in 1st class, survive her doomed lover Jack as in the movie, according to each of G your models? The RMS Titanic crashed into an iceberg in the North Atlantic on 1912, killing 1500 out of 2224 passengers and crew onboard. In the following table, we have sample data from a census about the Titanic ship. The data profiles passengers who survived the RMS Titanic or not. We have as attributes, or independent variables, 'gender', 'class' and 'age range'. The target variable for prediction is 'survived?'. Gender (G) Class (C) Age range (A) | Survived? (S) 20s 30s 20s 30s 30s M F F F M 3rd 1st 3rd 3rd 3rd No Yes No Yes No In this problem, build by hand two decision tree classification models from this training data, to predict whether a passenger survives the Titanic or not. The decision trees should have all leafs pure (i.e. all data points in the leaf belong to the same class). If needed, you can break ties in any way you want it, but write in your answer how you did it. For full credits, show how each split in the decision tree was decided, in addition to the final tree. (a) (4 pts) Build a decision tree using Information Gain as the criterion to choose the feature for each split. (b) (4 pts) Build a decision tree using Gini Gain as the criterion to choose the feature for each split. A (c) (2 pts) In the Oscar winning movie Titanic (1997), would the fictional character Rose, who is in her 20's and in 1st class, survive her doomed lover Jack as in the movie, according to each of G your models? The RMS Titanic crashed into an iceberg in the North Atlantic on 1912, killing 1500 out of 2224 passengers and crew onboard. In the following table, we have sample data from a census about the Titanic ship. The data profiles passengers who survived the RMS Titanic or not. We have as attributes, or independent variables, 'gender', 'class' and 'age range'. The target variable for prediction is 'survived?'. Gender (G) Class (C) Age range (A) | Survived? (S) 20s 30s 20s 30s 30s M F F F M 3rd 1st 3rd 3rd 3rd No Yes No Yes No In this problem, build by hand two decision tree classification models from this training data, to predict whether a passenger survives the Titanic or not. The decision trees should have all leafs pure (i.e. all data points in the leaf belong to the same class). If needed, you can break ties in any way you want it, but write in your answer how you did it. For full credits, show how each split in the decision tree was decided, in addition to the final tree. (a) (4 pts) Build a decision tree using Information Gain as the criterion to choose the feature for each split. (b) (4 pts) Build a decision tree using Gini Gain as the criterion to choose the feature for each split. A (c) (2 pts) In the Oscar winning movie Titanic (1997), would the fictional character Rose, who is in her 20's and in 1st class, survive her doomed lover Jack as in the movie, according to each of G your models?