一、使用SAS/EM界面生成CHAID決策樹

CHAID (Chi-Square Automatic Interaction

Detector)演算法為利用卡方分析(Chi-Square

Test)預測二個變數是否需要合併，如能夠產生最大的類別差異的預測變數，將成為節點的分隔變數。透過計算節點中類別的 P值

(P-Value)，以P值大小來決定決策樹是否繼續生長，所以不需像C4.5或CART要再做決策樹修剪的動作。CHAID

與CART、C4.5

之差異在於，CHAID只限於處理類別變數，如連續變數必須採用區段的方式，轉換成類別變數。另一差異部分在於修剪的部分，CART、C4.5

是先過度套用資料訓練，之後再修剪。但CHAID 則是在過度套用之前即停止支點蔓生擴大。

二、通過R和SAS代碼實現

樣例數據來自 R 包，主要是想比較 R 和 SAS

生成的決策樹是否相同。運行的結果確實不同，估計是運行參數的不同導致，仍有待研究。

1) Build CHAID tree using

R

# Train

data:

library(partykit)

library("CHAID")

data("BreastCancer", package = "mlbench")

# Build model:

ctrl

-1, alpha4 = 0.05,

minsplit=2, minbucket = 5, minprob = 0.01, stump = FALSE, maxheight

= 6)

b_chaid

Cl.thickness + Cell.size + Cell.shape + Marg.adhesion +

Epith.c.size + Bare.nuclei + Bl.cromatin + Normal.nucleoli +

Mitoses,

data = BreastCancer, na.action = na.pass, control =

ctrl)

plot(b_chaid)

2) Build CHAID tree using SAS/EM

SAS/EM Chaid Tree:

SAS Code:

proc iml;

submit /R;

#setInternet2(TRUE)

#install.packages("CHAID", repos="http://R-Forge.R-project.org")

# Train data:

library(partykit)

library("CHAID")

data("BreastCancer", package = "mlbench")

# Build model:

b_chaid

Cell.size + Cell.shape + Marg.adhesion +

Epith.c.size + Bare.nuclei + Bl.cromatin + Normal.nucleoli +

Mitoses,

data = BreastCancer)

png("D:/sbjgay/Chaid_r_plot.png")

plot(b_chaid)

dev.off()

endsubmit;

call ImportDataSetFromR("work.BreastCancer",

"BreastCancer");

run;quit;

filename rulecode "c:tempem_chaid_rules.sas";

*------------------------------------------------------------*;

* Tree: Run ARBOR procedure;

*------------------------------------------------------------*;

proc arbor data=work.BreastCancer

Leafsize=1

Splitsize=2

Mincatsize = 5

Maxbranch=10

Maxdepth=6

Criterion=PROBCHISQ

alpha = 0.05

Padjust= CHAIDAFTER

DEPTH

MAXRULES=5

MAXSURRS=0

Missing=USEINSEARCH

Exhaustive=0

event='malignant'

;

input Cl_thickness Cell_size

Cell_shape Marg_adhesion

Epith_c_size Bare_nuclei

Bl_cromatin Normal_nucleoli Mitoses / level=nominal;

target Class / level=NOMINAL

Criterion=PROBCHISQ;

Performance DISK

NodeSize=20000;

Assess NoValidata

measure=MISC;

SUBTREE LARGEST;

MAKEMACRO

NLEAVES=nleaves;

save

MODEL=Tree_EMTREE

SEQUENCE=Tree_OUTSEQ

IMPORTANCE=Tree_OUTIMPORT

NODESTAT=Tree_OUTNODES

SUMMARY=Tree_OUTSUMMARY

STATSBYNODE=Tree_OUTSTATS

Topology=Tree_OUTTOPOLOGY

Path = Tree_OUTPATH

Rules=Tree_OUTRules

;

code file=rulecode;

run;

quit;

3) Build CHAID tree using

TreeDisc.sas?in SAS 9.3

NOTE: Treedisc.sas does not work in 9.4.

%inc 'c:tempchaidxmacro.sas';

%inc 'c:tempchaidtreedisc.sas';

data set2;

set breastcancer;

run;

%treedisc(data=set2, depvar=class, freq=, ordinal=,

nominal=Cl_thickness Cell_size Cell_shape Marg_adhesion

Epith_c_size Bare_nuclei Bl_cromatin Normal_nucleoli Mitoses,

alpha=0.05,

outtree=trd,

options=noformat,

trace=long);

%treedisc(intree=trd, draw=graphics);

NOTE: Trees built by the above 3 methods are

different.?Not sure why this

happened.

Reference:

總結

以上是生活随笔為你收集整理的代码chaid_[转载]经典决策树之SAS实现--CHAID的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。