Tree Methods

Decision Tree, Random Forest를 dataitgirls3 학생들에게 가르치기 위해 만든 수업자료입니다.

Transcript

1. Tree methods dataitgirls3 Instructor Sunmi Yoon
2. Decision Tree
3. Sex <= 0.5 gini = 0.473 samples = 891 value = [549, 342] class = Survived Fare <= 26.269 gini = 0.306 samples = 577 value = [468, 109] class = Survived True Fare <= 48.2 gini = 0.383 samples = 314 value = [81, 233] class = Dead False gini = 0.226 samples = 415 value = [361, 54] class = Survived gini = 0.448 samples = 162 value = [107, 55] class = Survived gini = 0.447 samples = 225 value = [76, 149] class = Dead gini = 0.106 samples = 89 value = [5, 84] class = Dead
4. Sex <= 0.5 gini = 0.473 samples = 891 value = [549, 342] class = Survived Fare <= 26.269 gini = 0.306 samples = 577 value = [468, 109] class = Survived True Fare <= 48.2 gini = 0.383 samples = 314 value = [81, 233] class = Dead False gini = 0.226 samples = 415 value = [361, 54] class = Survived gini = 0.448 samples = 162 value = [107, 55] class = Survived gini = 0.447 samples = 225 value = [76, 149] class = Dead gini = 0.106 samples = 89 value = [5, 84] class = Dead Root Node (ࡸܻ) Intermediate Node (о૑) Terminal Node, Leaf (੖)
5. Sex <= 0.5 gini = 0.473 samples = 891 value = [549, 342] class = Survived Fare <= 26.269 gini = 0.306 samples = 577 value = [468, 109] class = Survived True Fare <= 48.2 gini = 0.383 samples = 314 value = [81, 233] class = Dead False gini = 0.226 samples = 415 value = [361, 54] class = Survived gini = 0.448 samples = 162 value = [107, 55] class = Survived gini = 0.447 samples = 225 value = [76, 149] class = Dead gini = 0.106 samples = 89 value = [5, 84] class = Dead അ ਤ஖ী ؘ੉ఠо ݻ ѐ ਤ஖೧ ੓ח૑ Ӓ ؘ੉ఠٜ੉ যڃ ۄ߰ਸ о஖Ҋ ੓ח૑
6. Sex <= 0.5 gini = 0.473 samples = 891 value = [549, 342] class = Survived Fare <= 26.269 gini = 0.306 samples = 577 value = [468, 109] class = Survived True Fare <= 48.2 gini = 0.383 samples = 314 value = [81, 233] class = Dead False gini = 0.226 samples = 415 value = [361, 54] class = Survived gini = 0.448 samples = 162 value = [107, 55] class = Survived gini = 0.447 samples = 225 value = [76, 149] class = Dead gini = 0.106 samples = 89 value = [5, 84] class = Dead যڃ ӝળਵ۽ о૑஖ӝܳ ೮ח૑ (gini ژח entropy)
7. Sex <= 0.5 gini = 0.473 samples = 891 value = [549, 342] class = Survived Fare <= 26.269 gini = 0.306 samples = 577 value = [468, 109] class = Survived True Fare <= 48.2 gini = 0.383 samples = 314 value = [81, 233] class = Dead False gini = 0.226 samples = 415 value = [361, 54] class = Survived gini = 0.448 samples = 162 value = [107, 55] class = Survived gini = 0.447 samples = 225 value = [76, 149] class = Dead gini = 0.106 samples = 89 value = [5, 84] class = Dead Terminal Nodeী ب଱ೠ ؘ੉ఠٜਸ যڌѱ ࠙ܨೡ Ѫੋ૑
8. sklearn Code
9. Impurity
10. Impurity ੄ࢎѾ੿աޖח Impurity (ࠛࣽب, ࠛഛपࢿ)੉ ծই૑ח ߑߨਵ۽ ೟ण೤פ׮. ࣽبо ૐоೞח Ѫਸ فҊ Information gain੉ۄҊ ೞӝب ೤פ׮. য়ט਷ ੄ࢎѾ੿աޖ੄ ࠛࣽب ஏ੿ ߑߨ ઺, Gini Indexܳ ҕࠗ೤פ׮.
11. Sex <= 0.5 gini = 0.473 samples = 891 value = [549, 342] class = Survived Fare <= 26.269 gini = 0.306 samples = 577 value = [468, 109] class = Survived True Fare <= 48.2 gini = 0.383 samples = 314 value = [81, 233] class = Dead False gini = 0.226 samples = 415 value = [361, 54] class = Survived gini = 0.448 samples = 162 value = [107, 55] class = Survived gini = 0.447 samples = 225 value = [76, 149] class = Dead gini = 0.106 samples = 89 value = [5, 84] class = Dead G = d ∑ i=1 Ri ( 1 − m ∑ k=1 p2 ik) Step 1. gini = 0.473 ਸ ૒੽ ҅࢑೧ ঳যࠁࣁਃ Step 2. gini = 0.226 ਸ ૒੽ ҅࢑೧ ঳যࠁࣁਃ
12. https://imgur.com/n3MVwHW
13. Random Forest
14. ৈ۞ ౟ܻٜਸ ‘׮ܰѱ’ ݅ٚ׮. https://www.researchgate.net/figure/Architecture-of-the-random-forest-model_fig1_301638643
15. https://community.alteryx.com/t5/Alteryx-Designer-Knowledge-Base/Seeing-the-Forest-for-the-Trees-An-Introduction-to-Random-Forest/ta-p/158062 bagging = bootstrap aggregating
16. Bagging ߓӦ(bagging)਷ bootstrap aggregating੄ ড੗۽, ࠗ౟झ౟ە(bootstrap)ਸ ా೧ ઑӘঀ ׮ܲ ള۲ ؘ੉ఠী ؀೧ ള۲ػ ӝୡ ࠙ܨӝ(base learner)ٜਸ Ѿ೤(aggregating)दఃח ߑߨ੉׮. ࠗ౟झ౟ە੉ۆ, ઱য૓ ള۲ ؘ੉ఠীࢲ ઺ࠂਸ ೲਊೞৈ ਗ ؘ੉ఠࣇҗ э਷ ௼ӝ੄ ؘ੉ఠࣇਸ ݅٘ח җ੿ਸ ݈ೠ׮. ߓӦਸ ా೧ ےؒ ನۨझ౟ܳ ള۲दఃח җ੿਷ ׮਺җ э੉ ࣁ ױ҅۽ ૓೯ػ׮. 1. ࠗ౟झ౟ە ߑߨਸ ా೧ Nѐ੄ ള۲ ؘ੉ఠࣇਸ ࢤࢿೠ׮. 2. Nѐ੄ ӝୡ ࠙ܨӝ(౟ܻ)ٜਸ ള۲दఅ׮. 3. ӝୡ ࠙ܨӝ(౟ܻ)ٜਸ ೞա੄ ࠙ܨӝ(ےؒ ನۨझ౟)۽ Ѿ೤ೠ׮(ಣӐ ژח җ߈ࣻై಴ ߑध ੉ਊ). Wikipedia ےؒನۨझ౟ > ߓӦਸ ੉ਊೠ ನۨझ౟ ҳࢿ
17. sklearn Code
18. ՘