PowerPoint 프레젠테이션 - Kangwon

PowerPoint 프레젠테이션 - Kangwon

2014 ([email protected]) : (Classification) Classification) ) ( : ) : , (Classification) ) . : . : (Classification) , ) 2 Decision) Trees 3

Example 1: Iris Datasets: 150 Iris Sepal.Len) gth: Sepal.Width: Petal.Len) gth: Petal.Width: Species: setosa versicolor virgin) ica Sepal.Leng Sepal.Wid Petal.Leng Petal.Wid Specie th th th th s 1 5.1 3.5 1.4

0.2 setosa 2 4.9 3 1.4 0.2 setosa 3 4.7 3.2 1.3 0.2 setosa 4 4.6 3.1 1.5 0.2 setosa 5 5 3.6 1.4 0.2 setosa 6

5.4 3.9 1.7 0.4 setosa 7 4.6 3.4 1.4 0.3 setosa 8 5 3.4 1.5 0.2 setosa 9 4.4 2.9 1.4 0.2 setosa 10 4.9 3.1

1.5 0.1 setosa 11 5.4 3.7 1.5 0.2 setosa 12 4.8 3.4 1.6 0.2 setosa 13 4.8 3 1.4 0.1 setosa 14 4.3 3 1.1 0.1 setosa

15 5.8 4 1.2 0.2 setosa # Iris ? 4 Ex.1: Decision) Tree http://cran) .r-project.org/web/packages/party/in) dex.html http://cran) .r-project.org/web/packages/zoo/in) dex.html http://cran) .r-project.org/web/packages/san) dwich/in) dex.html http://cran) .r-project.org/web/packages/strucchan) ge/in) dex.html http://cran) .r-project.org/web/packages/modeltools/in) dex.html http://cran) .r-project.org/web/packages/coin) /in) dex.html http://cran) .r-project.org/web/packages/mvtn) orm/in) dex.html R

5 Ex.1: R iris 6 Ex.1: sample(Classification) ) 70% 30% replace=TRUE: prob=c(Classification) 0.7, 0.3): 7 Ex.1: Decision) Tree ctree(Classification) ) : Decision) Tree 8

Ex.1: Decision) Tree Decision) Tree 9 Ex.1: Decision) Tree Decision) Tree 10 Ex.1: Decision) Tree predict(Classification) ) : 11 The k-Nearest Neighbor Algorithm kn) n) (Classification) : )

k Euclidean) distan) ce ? A 3 104

20.5 B 2 100 18.7 C 1 81 19.2

D 101 10 115.3 E 99 5 117.4 F

98 2 118.9 ? 18 90 (Classification) ) k=4 3 , 1 ? 12 Example 2: Foren) sic Glass Datasets: 6 214

Win) F: float glass win) dow # RI 1 3.01 13.64 4.49 Win) NF: n) on) -float glass win) dow Na Mg Al Si K

Ca Ba Fe type 1.1 71.78 0.06 8.75 0 0 Win) F 2 -0.39 13.89 3.6 1.36 72.73 0.48 7.83 0

0 Win) F 3 -1.82 13.53 3.55 1.54 72.99 0.39 7.78 0 0 Win) F Con) : con) tain) er (Classification) bottles) 4 -0.34 13.21 3.69 1.29 72.61 0.57 8.22 0 0 Win) F Tabl: Tableware

5 -0.58 13.27 3.62 1.24 73.08 0.55 8.07 0 0 Win) F Head: vehicle headlamp 6 -2.04 12.79 3.61 1.62 72.97 0.64 8.07 0 0.26 Win) F 7 -0.57 3.6 1.14 73.09 0.58 8.17

0 0 Win) F 8 -0.44 13.15 3.61 1.05 73.24 0.57 8.24 0 0 Win) F 0 0 Win) F Veh: vehicle win) dow

RI: (Classification) refractive in) dex) 9 13.3 1.18 14.04 3.58 1.37 72.08 0.56 8.3 Percen) tages of Na, Mg, Al, Si, K, Ca, Ba, an) d Fe type: ? 13 Ex.2: , http://cran) .r-project.org/web/packages/textir/in) dex.html http://cran) .r-project.org/web/packages/distrom/in) dex.html http://cran) .r-project.org/web/packages/gamlr/in) dex.html , library

R 14 Ex. 2: str(Classification) ) : 10 (Classification) RI, Percen) tages of elemen) ts, type) 214 15 Ex. 2: Box plots (Classification) 1/2) type box plot par(Classification) ) : par(mfrow=c(3,3), mai=c(.3,.6,.1,.1)) plot(RI ~ type, data=fgl, col=c(grey(.2),2:6)) plot(Al ~ type, data=fgl, col=c(grey(.2),2:6)) plot(Na ~ type, data=fgl, col=c(grey(.2),2:6)) plot(Mg ~ type, data=fgl, col=c(grey(.2),2:6)) plot(Ba ~ type, data=fgl, col=c(grey(.2),2:6)) plot(Si ~ type, data=fgl, col=c(grey(.2),2:6))

plot(K ~ type, data=fgl, col=c(grey(.2),2:6)) plot(Ca ~ type, data=fgl, col=c(grey(.2),2:6)) plot(Fe ~ type, data=fgl, col=c(grey(.2),2:6)) 16 Ex. 2: Box plots (Classification) 2/2) 17 Ex. 2: RIxAl (Classification) 200 ) (Classification) 14 ) n) t : sample(Classification) x, size, ) : x n) t 18 Ex. 2: RIxAl kNN

19 Ex. 2: RIxAl kNN plot plot(Classification) ) open) symbol poin) ts(Classification) ) solid symbol 20 Ex. 2: RIxAl kNN kNN type

1NN 78.6 , 5NN 71.4 21 #4 RI Al RI Percen) tages of Na, Mg, Al, Si, K, Ca, Ba, an) d Fe 214 (Classification) ) [email protected]) gwon) .ac.kr : [ ][ ]HW#4 , 20% 22

Recently Viewed Presentations

  • 5.2 Calorimetry and Enthalpy

    5.2 Calorimetry and Enthalpy

    Connecting Enthalpy Change and Molar Enthalpy Change. The amount of enthalpy change depends on the quantity of matter that undergoes the change . For example, it takes more energy to convert 1 L of liquid water to water vapor than...
  • Neuroanatomy and Neuroimaging: and how to apply that to ...

    Neuroanatomy and Neuroimaging: and how to apply that to ...

    Neuroanatomyand Neuroimaging: Application to Alzheimer's disease diagnosis. Megha Vasavada. Good morning. Today I am going to teach you basic neuroanatomy, a little bit about neuroimaging, and how you can take the anatomy of the brain and neuroimaging to study a...
  • The City of Pompeii

    The City of Pompeii

    The city of Pompeii was a major resort city during the times of Ancient Rome. However, in 79 AD, disaster struck the city when it was buried under 20 feet of ash and debris from the eruption of the nearby...
  • Kuby's Immunology \chapter 12 - University of Windsor

    Kuby's Immunology \chapter 12 - University of Windsor

    Hugh B. Fackrell T cell Maturation, Activation & Differentation Assigned Reading Content Outline Performance Ojectives Key terms Key Concepts Short Answer Questions Assigned Reading Chapter: 12 pp 285-310 Janis Kuby's Immunology 3rd Ed Content Outline T Cell Maturation T helper...
  • PHPH 101 Lecture 4 - Rick Baumgartner

    PHPH 101 Lecture 4 - Rick Baumgartner

    PHPH 101. Class 4: "Evidence-based Public Health I" ... Major champion of sanitary movement. Strongly argued for government action and public support for building sewers, cleaning refuse from streets, draining swamps. ... Personal cleanliness of residents (and whether they smell)
  • Welcome to the Start Small and Retire Early

    Welcome to the Start Small and Retire Early

    Time Warp Weekly Option Strategy (4-Variations) ($497) "Risk Free Projection" Weekly Option Strategy ($197) How to Own a Stock at a Cost Basis of $0 ($97) 12-Month Subscription to PDS Trader ($497) All for Only $497(You Won't See This Offer...
  • 03/01/2020 Dot Density Maps Gary Christopherson, Monica Stephens,

    03/01/2020 Dot Density Maps Gary Christopherson, Monica Stephens,

    Dot Density Map. Dot Density Map. Represents the distribution of a certain attribute through an arrangement of dots in enumeration areas. Density is not a calculated value. Each dot represents a specified amount of the attribute. Dots do not represent...
  • In Search Of Excellence In Project Management

    In Search Of Excellence In Project Management

    Times New Roman Monotype Sorts Book Antiqua Arial Zurich BT diamondc ENGR 281 Engineering Scholars Program Project Management Overview K. F. Reinschmidt J. L. Frank / Marathon Ashland Petroleum LLC Chair in Engineering Project Management [email protected] Slide 2 Slide 3...