Tutorial 1 (Decision Trees) 1) Consider the following dataset: High True No Outlook Temperature Humidity Windy Play 1 Sunny Hot False No 2 Sunny Hot High True No 3 Overcast Hot High False Yes 4 Rainy Mild Normal False Yes 5 Rainy Cool Normal False Yes 6 Rainy Cool High 7 Overcast Cool Normal True Yes & Sunny Mild High False No 9 Sunny Cool Normal False Yes 10 Rainy Mild Normal False 11 Sunny Mild Normal True Yes 12 | Overcast Mild High False Yes 13 | Overcast Hot Normal False Yes 14 Rainy Mild High Truc No Yes a. Construct a decision tree that can help in deciding whether to play or not based on various weather conditions. What is the accuracy taking into consideration all 14 records? b. Construct another decision tree? What is the accuracy taking into consideration all 14 records? . Some variables can be excluded as they do not affect the play-decision. What are they? d. Use the decision tree you constructed in (a) to decide in the following situations. Outlook Temperature Humidity Windy Play 15 Sunny Hot High False 16 Sunny Hot Normal True "Irisercast Hot High False 462 Which decision tree is better in terms of accuracy? In terms of complexity? In terms of tree size? 2) Consider the following dataset: Tear Astigmatism Production Rate Recommended Lens yes yes none hard nonc hard yes Spectacle Age Prescriptio n 1 young myope 2 young myope 3 young hypermetrope 4 young hypermetrope 5 pre-presbyopic myope 6 pre-presbyopic myope 7 pre-presbyopic hypermetrope 8 pre-presbyopic hypermetrope 9 presbyopic myope 10 presbyopic myope 11 presbyopic hypermetrope 12 presbyopic hypermetrope yes yes yes yes yes none hard reduced normal reduced normal reduced normal reduced normal reduced normal reduced normal none none yes yes none hard yes none yes none a. Construct a decision tree that can help in determining the recommended type of contact lenses. What is the accuracy taking into consideration all 12 records? b. Construct another decision tree? What is the accuracy taking into consideration all 12 records? c. Some variables can be excluded as they do not affect the play-decision. What are they? d. Which decision tree is better in terms of accuracy? In terms of complexity? In terms of tree size? 3) Consider the following dataset: ITIS Age Astigmatism Tear production rate Young Young Young Young Young Spectacle prescription Myope Myope Myope Myope Hypermetrop No NO Yes Yes No Reduced Normal Reduced Normal Reduced lenses type None Soft None Hard None Young Hypermetrop NO Normal Soft Young Hypermetrop Yes Reduced None Young Hypermetrop Yes Normal hard Myope No Reduced None Pre- presbyopic Pre- presbyopic Myope NO Normal Soft Pre- Myope Yes Reduced None presbyopic Pre Myope Yes Normal Hard Hypermetrop No Reduced None Hypermetrop No Normal Soft Yes Reduced None presbyopic Pre- presbyopic Pre- presbyopic Pre- presbyopic Pre- presbyopic Presbyopic Presbyopic Presbyopic Presbyopic Presbyopic Yes Normal None Hypermetrop e Hypermetrop e Myope Myope Myope Myope Hypermetrop e Hypermetrop No No Yes Yes No Reduced Normal Reduced Normal Reduced None Soft None Hard None Presbyopic No Normal Soft Presbyopic Hypermetrop Yes Reduced None Presbyopic Hypermetrop Yes Normal None ..Construct a decision tree that can help in determining the recommended type of contact lenses. What is the accuracy taking into consideration all 24 records? Construct another decision tree? What is the accuracy taking into consideration all 12 records? Some variables can be excluded as they do not affect the play-decision. What are they? Which decision tree is better in terms of accuracy? In terms of complexity? In terms of tree size? c