Tutorial 1 (Decision Trees) Age Spectacle Prescriptio Tear Astigmatism Production Rate Recommended Lens n yes yes yes yes yes Bone hard bone hard none hard yes 1 young myope 2 young myone 3 young hypermetrope 4 young hypermetrope Spre-presbyopic myope 6 pre-presbyopic myope 7 pre-presbyopic hypermetrope 8 pre-presbyopic hypermetrope 9 presbyopic myope 10 presbyopic myope 11 presbyopic hypermetrope 12 presbyopie hypermetrope reduced normal roduced normal reduced normal Teduced normal reduced normal reduced normal yes yes none none none hard yes yes yes yes hond bone 1. Construct a decision tree using J48 algorithm, and setting Test Option to training set in order to help the to predict the lens. 2. Visualize the decision tree. 3. How many leaves does the tree contain? 4. What is the size of the tree? 5. Select the attribute to be ignored and not to be used in the classification? 6. How accurately the classifier was able to predict the true class of instances under the chosen test module"setting Test Option": 7. Construct another decision tree using J48 algorithm, and setting Test Option to 66% Percentage split in order to help the to predict the lens 8. Visualize the decision tree. 9. How many instances were used to perform Percentage split test? 10. How accurately the classifier was able to predict the true class of instances under the chosen test module" Percentage split"? 233 yos 229 Age Gender Chest Pain Type Cholestrol Exercise Induced Angina 63 Male atypangina DO 67 Female non anginal 286 no 67 Females 263 yes 38 Male 222 63 Female panna DO 40 Male atypangina 223 yes 62 Females ansinal 286 DO 38 | Male nan anginal 223 DO 63 Male typ angina 212 yes 67 Female Pangina 67 Female asympt 229 yes 38 Female atypangina 233 246 yes 1. Construct a decision tree using J48 algorithm, and setting Test Option to training set in order to help the to predict whether the person have Exercise_Induced_Angina 2. Visualize the decision tree. 3. How many leaves does the tree contain? 4. What is the size of the tree? 5. Select the attribute to be ignored and not to be used in the classification? 6. How accurately the classifier was able to predict the true class of instances under the chosen test module"setting Test Option": 7. Construct another decision tree using J48 algorithm, and setting Test Option to 66% Percentage split in order to help the to predict whether the person will have Exercise_Induced_Angina 8. Visualize the decision tree. 9. How many instances were used to perform Percentage split test? 10. How accurately the classifier was able to predict the true class of instances under the chosen test module" Percentage split