# Conditional Inference trees in R (ctree)

#1

Hello,

I looked up a tutorial on the problem ‘bike sharing demand’ which used the conditional inference trees algorithm from the ‘party’ package on the dataset. I am not able to interpret the result of the algorithm.
How to know which variables are important and which are not?
I get something like this(not putting up whole of the output here)->

Conditional inference tree with 207 terminal nodes

Response: count
Inputs: season, holiday, workingday, weather, temp, atemp, humidity, hour, daypart, sunday
Number of observations: 10886

1. hour == {7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21}; criterion = 1, statistic = 5623.75
1. hour == {8, 16, 17, 18, 19}; criterion = 1, statistic = 1565.212
2. temp <= 19.68; criterion = 1, statistic = 515.184
3. workingday == {1}; criterion = 1, statistic = 185.765
4. hour == {8, 17, 18}; criterion = 1, statistic = 160.553
5. season == {2, 4}; criterion = 1, statistic = 119.435
6. weather == {1, 2}; criterion = 1, statistic = 47.78
7. daypart == {1}; criterion = 0.998, statistic = 16.273
9)* weights = 82
8. daypart == {3}
9. season == {4}; criterion = 0.99, statistic = 10.759
11)* weights = 90
10. season == {2}
12)* weights = 12
11. weather == {3}
12. humidity <= 88; criterion = 0.993, statistic = 11.584
14)* weights = 12
13. humidity > 88
15)* weights = 8
14. season == {1, 3}
15. atemp <= 19.695; criterion = 1, statistic = 33.314
16. daypart == {1}; criterion = 1, statistic = 21.872
18)* weights = 67
17. daypart == {3}
18. temp <= 9.84; criterion = 1, statistic = 21.925
20)* weights = 41
19. temp > 9.84
20. humidity <= 46; criterion = 0.951, statistic = 7.877
22)* weights = 32
21. humidity > 46
23)* weights = 24
22. atemp > 19.695
23. weather == {1}; criterion = 0.961, statistic = 11.058
24. humidity <= 54; criterion = 0.965, statistic = 8.515
26)* weights = 21
25. humidity > 54
27)* weights = 12
26. weather == {2, 3}
28)* weights = 12
27. hour == {16, 19}
28. season == {2, 3, 4}; criterion = 1, statistic = 80.2
29. weather == {1, 2}; criterion = 1, statistic = 26.74
31)* weights = 104
30. weather == {3}
32)* weights = 10
31. season == {1}
32. temp <= 12.3; criterion = 1, statistic = 36.603
34)* weights = 59
33. temp > 12.3
34. weather == {1, 2}; criterion = 0.99, statistic = 13.848
35. atemp <= 20.455; criterion = 0.979, statistic = 9.421
37)* weights = 45
36. atemp > 20.455
38)* weights = 19
37. weather == {3}
39)* weights = 7
38. workingday == {0}
39. hour == {16, 17, 18}; criterion = 1, statistic = 83.651
40. season == {4}; criterion = 1, statistic = 57.151
41. temp <= 12.3; criterion = 1, statistic = 22.09
43)* weights = 11
42. temp > 12.3
43. temp <= 16.4; criterion = 0.982, statistic = 12.293
44. hour == {16, 17}; criterion = 0.994, statistic = 14.739
46)* weights = 33
45. hour == {18}
47)* weights = 17
46. temp > 16.4
48)* weights = 15
47. season == {1, 2}
48. atemp <= 17.425; criterion = 1, statistic = 20.911
50)* weights = 61
49. atemp > 17.425
51)* weights = 43
50. hour == {8, 19}
51. season == {2, 4}; criterion = 1, statistic = 37.591
52. holiday == {1}; criterion = 1, statistic = 27.722
54)* weights = 7
53. holiday == {0}
54. hour == {19}; criterion = 1, statistic = 33.969
56)* weights = 31
55. hour == {8}
57)* weights = 43
56. season == {1, 3}
57. temp <= 16.4; criterion = 0.988, statistic = 10.421
58. hour == {19}; criterion = 0.992, statistic = 11.326
60)* weights = 25
59. hour == {8}
61)* weights = 33
60. temp > 16.4
62)* weights = 12
61. temp > 19.68
62. hour == {17, 18}; criterion = 1, statistic = 218.821
63. workingday == {1}; criterion = 1, statistic = 128.189
64. weather == {1, 2}; criterion = 1, statistic = 40.98
66)* weights = 341
65. weather == {3}
66. humidity <= 77; criterion = 0.964, statistic = 8.426
68)* weights = 18
67. humidity > 77
69)* weights = 20
68. workingday == {0}
69. humidity <= 72; criterion = 1, statistic = 36.114
71)* weights = 140
70. humidity > 72
71. humidity <= 79; criterion = 0.953, statistic = 7.955
73)* weights = 18
72. humidity > 79
74)* weights = 9
73. hour == {8, 16, 19}
74. workingday == {1}; criterion = 1, statistic = 59.767
75. hour == {8}; criterion = 1, statistic = 122.79
76. weather == {1, 2}; criterion = 1, statistic = 28.147
78)* weights = 129
77. weather == {3}
79)* weights = 11
78. hour == {16, 19}
79. hour == {19}; criterion = 1, statistic = 39.096
80. weather == {1, 2}; criterion = 1, statistic = 32.597
81. temp <= 27.06; criterion = 0.99, statistic = 11.689
82. humidity <= 77; criterion = 0.983, statistic = 9.885
84)* weights = 72
83. humidity > 77
85)* weights = 13
84. temp > 27.06
86)* weights = 82
85. weather == {3}
87)* weights = 13
86. hour == {16}
87. humidity <= 73; criterion = 1, statistic = 18.148
88. season == {2, 3, 4}; criterion = 0.99, statistic = 16.231
90)* weights = 166
89. season == {1}
91)* weights = 16
90. humidity > 73
92)* weights = 16
91. workingday == {0}
92. hour == {16, 19}; criterion = 1, statistic = 123.327
93. humidity <= 54; criterion = 1, statistic = 43.586
94. hour == {16}; criterion = 1, statistic = 21.514
96)* weights = 61
95. hour == {19}
96. season == {3}; criterion = 0.964, statistic = 13.489
98)* weights = 13
97. season == {1, 2, 4}
98. atemp <= 30.305; criterion = 0.979, statistic = 9.423
100)* weights = 9
99. atemp > 30.305
101)* weights = 14
100. humidity > 54
101. humidity <= 73; criterion = 0.998, statistic = 14.037
103)* weights = 41
102. humidity > 73
103. atemp <= 26.515; criterion = 0.96, statistic = 8.237
105)* weights = 7
104. atemp > 26.515
106)* weights = 22
105. hour == {8}
106. sunday == {0}; criterion = 0.981, statistic = 9.624
108)* weights = 33
107. sunday == {1}
109)* weights = 27
108. hour == {7, 9, 10, 11, 12, 13, 14, 15, 20, 21}
109. atemp <= 18.18; criterion = 1, statistic = 879.89
110. season == {3, 4}; criterion = 1, statistic = 298.428
111. temp <= 14.76; criterion = 1, statistic = 44.97
112. daypart == {1, 2}; criterion = 1, statistic = 43.705
113. atemp <= 15.15; criterion = 0.988, statistic = 12.076
114. workingday == {1}; criterion = 0.999, statistic = 14.469
115. daypart == {1}; criterion = 1, statistic = 59.582
116. weather == {2}; criterion = 0.97, statistic = 11.568
118)* weights = 12
117. weather == {1, 3}
119)* weights = 35
118. daypart == {2}
120)* weights = 50
119. workingday == {0}
120. daypart == {2}; criterion = 1, statistic = 51.128
122)* weights = 28
121. daypart == {1}
122. hour == {9}; criterion = 1, statistic = 23.964
124)* weights = 14
123. hour == {7}
125)* weights = 20
124. atemp > 15.15
126)* weights = 158
125. daypart == {3}
126. workingday == {1}; criterion = 1, statistic = 27.091
127. weather == {1}; criterion = 0.999, statistic = 18.54
128. hour == {20}; criterion = 0.998, statistic = 14.1
130)* weights = 22
129. hour == {21}
131)* weights = 22
130. weather == {2, 3}
132)* weights = 18
131. workingday == {0}
133)* weights = 38
132. temp > 14.76
134)* weights = 10
133. season == {1, 2}
134. temp <= 9.84; criterion = 1, statistic = 64.16
135. daypart == {1}; criterion = 1, statistic = 56.008
136. workingday == {1}; criterion = 1, statistic = 56.483
137. atemp <= 9.09; criterion = 0.978, statistic = 9.397
139)* weights = 39
138. atemp > 9.09
140)* weights = 52
139. workingday == {0}
140. hour == {9}; criterion = 1, statistic = 21.192
142)* weights = 20
141. hour == {7}
143)* weights = 24
142. daypart == {2, 3}
143. workingday == {0}; criterion = 1, statistic = 39.152
144. daypart == {2}; criterion = 1, statistic = 22.123
146)* weights = 57
145. daypart == {3}
147)* weights = 21
146. workingday == {1}
147. temp <= 9.02; criterion = 0.981, statistic = 17.808
148. hour == {20}; criterion = 0.975, statistic = 22.057
150)* weights = 21
149. hour == {10, 11, 12, 13, 14, 15, 21}
151)* weights = 130
150. temp > 9.02
152)* weights = 37
151. temp > 9.84
152. weather == {1, 2}; criterion = 1, statistic = 37.741
153. daypart == {1}; criterion = 1, statistic = 38.516
154. workingday == {1}; criterion = 1, statistic = 41.504
156)* weights = 51
155. workingday == {0}
156. hour == {9}; criterion = 0.977, statistic = 9.308
158)* weights = 12
157. hour == {7}
159)* weights = 14
158. daypart == {2, 3}
159. workingday == {0}; criterion = 1, statistic = 46.192
160. daypart == {2}; criterion = 1, statistic = 48.461
162)* weights = 87
161. daypart == {3}
163)* weights = 28
162. workingday == {1}
163. hour == {12, 20}; criterion = 0.972, statistic = 21.724
165)* weights = 62
164. hour == {10, 11, 13, 14, 15, 21}
166)* weights = 168
165. weather == {3}
167)* weights = 36

Thanks.

#2

try plot this ctree and you will find it. for reference see

#3

Use varimp or varimpAUC function and pass you fitted object to it.
check this:

varimp(object, mincriterion = 0, conditional = FALSE,
threshold = 0.2, nperm = 1, OOB = TRUE, pre1.0_0 = conditional)
varimpAUC(object, mincriterion = 0, conditional = FALSE,
threshold = 0.2, nperm = 1, OOB = TRUE, pre1.0_0 = conditional)