Professional Documents
Culture Documents
تحلیل شدت مصدومیت ناشی از تصادفها در راههای دوخطه برونشهری PDF
تحلیل شدت مصدومیت ناشی از تصادفها در راههای دوخطه برونشهری PDF
E-mail: shariat@iust.ac.ir
-1388/09/21 :1389/03/12:
.
.
.
.
) (- .
.
.
.
: ) (CART
.1
. 3
24000
%70
] .[1388
3.
50 60
1 ] .[1386
169648 7241
. 90
153
1389
.2
10
5 .
) .(Al-Ghamdi, 2002
7 8
9.
] .2003
1389
154
[Sohn (et. al.),
.3
] . 2001 12
30353
14 .
].[Delen (et. al.), 2006
-1-3
13
)(
12604
15
) (
( .
.
16
155
1389
120km/h 120km/h
.2
21 :
)( 1
) ( j) N j (m
)p( j, m
= ), p( j, m
,
)p(m
Nj
= )p( j | m
J
)p(m) = p( j, m
j =1
)( 2
)Gini(m) = 1 p 2 ( j | m
j =1
:
*
J ( j )
22 j
.1
) 2-5 ( Nj(m)
j Nj m
17 18
j p(j|m)
) N1
j m ) Gini(m
N2( 19 .
Gini(m)
) Gini(m
20 .
2 .
1389
156
) 23
N1 (1
)(
24 -
25
- :
T
J
] .[Giudici, 2003
) p (t t
T .
)
( .
. 26
27
28
- 29 .
) (
.
.3
. 3
2-3
)( .
) (
) xj (...
nm
))Gini( S ( x j , m
m =1 N
= ) VIM ( x j
)(4
) S ( x j , m xj m
)) Gini ( S ( x j , m m
nm xj M m
N
157
1389
N .
169648
xj
7241
. nm
N
.
.
xj
. 13 1
1 .
) (
11 )
(
.4
1 ) (
. ]Stewart,
) (1387-1385 .
[1996
30 31
) (21579 km
%70
%30 .
114
k 32 .
) (
] .[Stewart, 1996
) (... )
%30
(... )
(...
) (... ) (
.5
1-5
) (
) (
( .
) (
2 ( . %27/9
1389
158
( )
) .((j)=0.5 3
)) ((j
) 2-1 -
. 4
12 (
. 1-2-
2-1-
((1)= (11138+1791)/(156719+11138+1791)=0/08
2-1 -
)(
) (
] . 2007
2-5
. 3
%60
. 8
)) (j 1
159
1389
.1
) (
.1 : .2 .3
) (
.1 :.2 .3
.1.2
) (
) (
.1 .2 .3
.1 .2 .3 .4
.5 .6 .7 .8
.9 .10 .11 .13
.14 .15 .16 .17
.18 .19 .20 .21 4
.22 .23 .24 .25 .26
.1 .2
.3 .4 .5
.6 .7 .8 .9
.10 .11
.1.2 .3 .4 .5 .6
.1.2 .3
.1.2 .3 .4 .5 .6 .7
.1.2 .3 .4 .5 .6 .7
.1 .2 .3 ) (.4 ) (.5 .6
.1.2 .3
) (
) (
.1.2 .3 .4 .5 .6 .7 .8 .9
.10.11 .12 .13 .14 .15
.2
109656
)83463 (76/11%
47063
)2324 (29/93%
7765
)35736 (75/93%
3373
)941 (27/90%
1297
)665 (51/27%
494
)244 (49/39%
118718
)86452(72/82%
50930
)36921 (72/49%
)43/79% (3171
)49/10% (3556
)7/10% (514
1.1
1.2
2.1
2.2
-4
1.1
1.2
2.1
2.2
-4
.4 : :
1389
160
.3
1-2
1-1
2-2
2-1
56/80%
55/00%
58/12%
71/10%
57/55%
73/36%
40/70%
47/40%
58/33%
88/05%
59/88%
67/33%
57/62%
63/17%
69/51%
70/00%
56/91%
55/32%
59/06%
70/86%
57/59%
72/59%
44/43%
50/35%
3-5
) (Al-Ghamdi, 2002
) (
.
4 -2
1 2-1
-4
-4
4-5
-4
33
1 -2
5 .
161
1389
-4 .
1.2
1-
2
11.1-1
2.2
2-2
2.1
2-
1
0/65
0/94
0/96
0/4
0/07
0/01
0/01
0/13
0/07
0/01
0/01
0/09
0/07
0/01
0/01
0/08
0/07
0/01
0/01
0/04
0/07
0/01
0/01
0/04
0/01
0/01
0/04
0/01
0/01
0/04
0/04
0/04
0/04
-4 .
1.21
2
1.1
1-
1
2.2
2-2
2.12
-1
0/61
0/59
0/53
0/31
0/11
0/2
0/23
0/26
0/07
0/06
0/06
0/14
0/07
0/05
0/06
0/14
0/07
0/04
0/04
0/14
0/07
0/03
0/03
0/02
0/02
0/02
0/02
0/02
. 6
12 11
. 17 18
. 4
) (5111718 18
4 3 . 3
( .
2 5 6
. 5
( 14 5 3 1
23 19 18 16 15 25 %68
1389
162
( )
3 (.
4 9
5 3
15 16
1 -2
.5 1.2
163
1389
5. Functional form
)6. Classification and Regression Tree (CART
7. Independent variable
8. Target variable
9. Classification and Prediction
10. Logit models
11. Ordered probit
12. Artificial Neural Networks
13. Decision tree
14. Variable importance
15. Nominal
16. Leaf or terminal node
17. Splitter
18. Purity
19. Child node
20. Parent node
21. Gini index
22. Prior probability
23. Root node
24. Misclassification cost
25. Goodness of fit
26. Maximal tree
27. Overfitting
28. Pruning
29. Cost-complexity
30. Train
31. Test
32. K-fold cross validation
33. Rules
.6
.
" "if-then .
.
.
.
.
.
.
.9
- )(1386
"
( .
"
3.221-213 .
.7
)
(
. 8
1. Two-lane Two-way rural roads
2. Injury severity
3. Data mining
4. Regression type generalized linear models
1389
164
- Kweon, Y. and Kockelman, K. (2003) "Driver
attitudes and choices: Seatbelt use, speed limits,
alcohol consumption and crash histories". 82nd.
Annual Meeting of Transportation Research Board,
Washington D. C.
- Bedard, M. and
Guyatt, G. (2002) "The
independent contribution of driver, crash and vehicle
characteristics to driver fatalities", Accident Analysis
and Prevention, 34(6), p. 717.
1389
165
ABSTRACT
In this study, factors influencing driver and occupant injury severity in two lane , two way roads
of Iran are identified. Using statistical models is one of the most common methods that were
employed to analyze crash severity. In this study, in addition to examining such models and
expressing their weak points, CART method is introduced. Classification and regression trees
(CART), which is one of the most common methods of data mining, was employed to analyze
the traffic crash data over a three-year period (2006-2008). Expressing models in decision tree
format with explicit rules is one of their most useful outputs. In the analysis procedure, the
problem of three-class prediction was decomposed into a set of binary prediction models and
then eight models were analyzed. This resulted higher overall accuracy of predicting the model,
besides the prediction accuracy of the fatality class, which was nearly 0%, and in most of the
previous studies, increased significantly. Results indicated that improper overtaking and ignoring
seat belts are the most important factors affecting the severity of injuries.
Keywords: Data mining, Classification and Regression Trees (CART), injury severity, rural
pathaaaaaa roads
200