CN111126868A - Road traffic accident occurrence risk determination method and system - Google Patents
Road traffic accident occurrence risk determination method and system Download PDFInfo
- Publication number
- CN111126868A CN111126868A CN201911389961.5A CN201911389961A CN111126868A CN 111126868 A CN111126868 A CN 111126868A CN 201911389961 A CN201911389961 A CN 201911389961A CN 111126868 A CN111126868 A CN 111126868A
- Authority
- CN
- China
- Prior art keywords
- traffic accident
- determining
- risk
- classification model
- road
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 206010039203 Road traffic accident Diseases 0.000 title claims abstract description 185
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000013145 classification model Methods 0.000 claims abstract description 67
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 14
- 238000007477 logistic regression Methods 0.000 claims abstract description 9
- 238000003066 decision tree Methods 0.000 claims description 10
- 238000011478 gradient descent method Methods 0.000 claims description 7
- 238000007637 random forest analysis Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 24
- 239000013598 vector Substances 0.000 description 4
- 208000027418 Wounds and injury Diseases 0.000 description 3
- 230000006378 damage Effects 0.000 description 3
- 208000014674 injury Diseases 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000013179 statistical model Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010206 sensitivity analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a method and a system for determining the occurrence risk of a road traffic accident. The method comprises the following steps: acquiring data of traffic accidents in a database; acquiring a plurality of classification models which take all risk factors corresponding to each traffic accident as input and take the probability of causing all traffic accident grades as output; determining the probability of all traffic accident grades according to the data of the traffic accidents by utilizing each classification model; performing weighted regression on the probabilities of all the traffic accident grades output by each classification model by using a logistic regression algorithm, and determining an output result after weighted regression; determining the grade of the occurred traffic accident according to the output result after the weighted regression; and determining the occurrence risk of the road traffic accident according to the level of the occurring traffic accident. The method and the system provided by the invention solve the problem that the occurrence risk of the road traffic accident cannot be effectively determined in the prior art.
Description
Technical Field
The invention relates to the field of road assessment, in particular to a method and a system for determining road traffic accident occurrence risk.
Background
Road traffic accidents can cause personal injury and serious property loss. By collecting traffic accident data and traffic environment data when an accident occurs, such as vehicle conditions, the position of a road section where the accident occurs, the time when the accident occurs, the road design of the road section where the accident occurs, and the like, relevant factors causing the accident influence are analyzed, so that a management department can be facilitated to reduce the occurrence probability of the accident and reduce the loss caused by the accident by improving the conditions of road infrastructure.
In the past decades, a great deal of research and investigation have been conducted on the relationship between accident levels and related risk factors (such as traffic flow, geometric design of roads, the age of drivers and external environmental characteristics, etc.), and the risk of occurrence of traffic accidents can be determined according to the related factors, so that corresponding measures can be taken to avoid occurrence of major traffic accidents.
At present, the method for determining the occurrence risk of the road traffic accident mainly comprises two types. The first type is an analysis method based on a statistical model, the method can analyze the influence of each factor on the severity (grade) by constructing the mathematical and physical relationship between the accident severity and each relevant factor or explanatory variable, and the method has good theoretical interpretability and clear calculation structure. The second category is accident severity classification methods based on machine learning models, such as artificial neural networks, decision trees, support vector machines, and the like.
In summary, early studies aimed at analyzing the relationship between accident levels and different factors were mainly based on statistical methods. However, statistical models have a disadvantage in that most models assume that the factors affect the severity of the accident in a linear manner, and once the assumptions of the statistical models are violated, the inference of the influence of the factors is biased. Moreover, the main disadvantages of the machine learning method are that a black box modeling mechanism is adopted, the explanation of the accident severity and the correlation relationship of related variables is lacked, and each machine learning model brings certain limitations, such as the problems that an artificial neural network is easy to be over-fitted, a decision tree is greatly influenced by sample imbalance, and K-means clustering is sensitive to outliers.
Therefore, according to the method, the occurrence risk of the traffic accident cannot be accurately determined according to the relevant factors in the prior art.
Disclosure of Invention
The invention aims to provide a method and a system for determining the occurrence risk of a road traffic accident, which solve the problem that the occurrence risk of the road traffic accident cannot be effectively determined in the prior art.
In order to achieve the purpose, the invention provides the following scheme:
a method for determining the occurrence risk of a road traffic accident comprises the following steps:
acquiring data of traffic accidents in a database; the data of the traffic accident comprises risk factors of the traffic accident and corresponding traffic accident grade; the risk factors include: the length of a ramp of the expressway, the annual average daily traffic volume of a main road, the gradient of a road, weather, the age and driving age of a driver, whether the driver drinks wine or not and the number of people on board the vehicle;
acquiring a plurality of classification models which take all risk factors corresponding to each traffic accident as input and take the probability of causing all traffic accident grades as output; the classification model is a random forest model, a gradient lifting decision tree and a self-adaptive lifting algorithm;
determining the probability of all traffic accident grades caused by all risk factors corresponding to each traffic accident by utilizing each classification model according to the data of the traffic accidents;
performing weighted regression on the probabilities of all the traffic accident grades output by each classification model by using a logistic regression algorithm, and determining an output result after weighted regression;
determining the grade of the occurred traffic accident according to the output result after the weighted regression;
and determining the occurrence risk of the road traffic accident according to the level of the occurring traffic accident.
Optionally, the obtaining a plurality of classification models which take all risk factors as input and take the probability of causing all traffic accident levels as output further includes:
determining a weight for each of the risk factors according to each of the classification models;
carrying out weighted average on the multiple weights of each risk factor to determine the final weight of each risk factor;
sequencing the final weights of all the risk factors and determining key risk factors; the key risk factor is the risk factor with the highest final weight.
Optionally, the determining the level of the traffic accident according to the output result after the weighted regression further includes:
constructing a cross entropy loss function according to the level of the occurred traffic accident and the level of the traffic accident in the database; the cross entropy loss function is an error function between the level of the occurring traffic accident and the level of the traffic accident in the database;
minimizing the cross entropy loss function by adopting a gradient descent method to obtain the weight of each classification model;
and optimizing the output result of the corresponding classification model by adopting the weight.
Optionally, the determining the level of the traffic accident according to the output result after the weighted regression specifically includes:
and determining the grade of the occurred traffic accident by adopting a Sigmoid function according to the output result after the weighted regression.
A road traffic accident occurrence risk determination system, comprising:
the data acquisition module is used for acquiring the data of the traffic accidents in the database; the data of the traffic accident comprises risk factors of the traffic accident and corresponding traffic accident grade; the risk factors include: the length of a ramp of the expressway, the annual average daily traffic volume of a main road, the gradient of a road, weather, the age and driving age of a driver, whether the driver drinks wine or not and the number of people on board the vehicle;
the classification model acquisition module is used for acquiring a plurality of classification models which take all risk factors corresponding to each traffic accident as input and take the probability of causing all traffic accident grades as output; the classification model is a random forest model, a gradient lifting decision tree and a self-adaptive lifting algorithm;
the output result module is used for determining the probability of all traffic accident grades caused by all risk factors corresponding to each traffic accident by utilizing each classification model according to the data of the traffic accidents;
the weighted regression determining module is used for performing weighted regression on the probabilities of all the traffic accident grades output by each classification model by using a logistic regression algorithm and determining an output result after the weighted regression;
the occurring traffic accident grade determining module is used for determining the occurring traffic accident grade according to the output result after the weighted regression;
and the road traffic accident occurrence risk determining module is used for determining the road traffic accident occurrence risk according to the level of the occurring traffic accident.
Optionally, the method further includes:
a risk factor weight determining module for determining the weight of each risk factor according to each classification model;
a risk factor final weight determining module, configured to perform weighted average on a plurality of weights of each risk factor, and determine a final weight of each risk factor;
the key risk factor determining module is used for sequencing the final weights of all the risk factors and determining the key risk factors; the key risk factor is the risk factor with the highest final weight.
Optionally, the method further includes:
the cross entropy loss function construction module is used for constructing a cross entropy loss function according to the level of the generated traffic accident and the level of the traffic accident in the database; the cross entropy loss function is an error function between the level of the occurring traffic accident and the level of the traffic accident in the database;
the weight determining module of the classification model is used for minimizing the cross entropy loss function by adopting a gradient descent method to obtain the weight of each classification model;
and the optimization module is used for optimizing the output result of the corresponding classification model by adopting the weight.
Optionally, the traffic accident grade determining module specifically includes:
and the occurring traffic accident grade determining unit is used for determining the occurring traffic accident grade by adopting a Sigmoid function according to the output result after the weighted regression.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
according to the method and the system for determining the road traffic accident occurrence risk, the grade of the occurring traffic accident is determined by performing weighted regression on the output structures of the plurality of classification models, the characteristics of the plurality of classification models are considered in the grade of the occurring traffic accident, the problem of limitation of a single classification model is avoided, the accuracy of determining the grade of the occurring traffic accident is improved, and the road traffic accident occurrence risk can be effectively determined.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a method for determining the risk of occurrence of a road traffic accident according to the present invention;
fig. 2 is a schematic structural diagram of a system for determining the occurrence risk of a road traffic accident according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a method and a system for determining the occurrence risk of a road traffic accident, which solve the problem that the occurrence risk of the road traffic accident cannot be effectively determined in the prior art.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Fig. 1 is a schematic flow chart of a method for determining a road traffic accident occurrence risk according to the present invention, and as shown in fig. 1, the method for determining a road traffic accident occurrence risk according to the present invention includes:
and S101, acquiring the data of the traffic accidents in the database. The data for the traffic accident includes risk factors for the traffic accident and a corresponding traffic accident rating. The risk factors include: the length of the ramp of the expressway, the annual average daily traffic volume of the main road, the gradient of the road, the weather, the age and the driving age of a driver, whether the driver drinks wine or not and the number of people on board the vehicle.
S102, acquiring a plurality of classification models which take all risk factors corresponding to each traffic accident as input and take the probability of causing all traffic accident grades as output. The classification model is a random forest model, a gradient lifting decision tree and a self-adaptive lifting algorithm.
S103, determining the probability of all traffic accident grades caused by all risk factors corresponding to each traffic accident by using each classification model according to the data of the traffic accidents;
the plurality of classification models may be considered to be highly complex non-linear feature converters, with data for a traffic accident being fed into each classification model separately, each classification model outputting a vector, each value in the vector representing a probability of occurrence of a level of each traffic accident. The three vectors output by the multiple classification models represent the input features from the perspective of three different methods. These three vectors are stitched together as a high dimensional representation of the original features.
And S104, performing weighted regression on the probabilities of all the traffic accident grades output by each classification model by using a logistic regression algorithm, and determining an output result after weighted regression.
To prevent overfitting, a logistic regression algorithm with L2 regularization was used.
And S105, determining the grade of the occurred traffic accident according to the output result after the weighted regression.
And determining the grade of the occurred traffic accident by adopting a Sigmoid function according to the output result after the weighted regression.
And mapping the output result after weighted regression to be between 0 and 1 by adopting a Sigmoid function.
And determining the grade of the traffic accident with the highest probability.
And S106, determining the occurrence risk of the road traffic accident according to the level of the occurring traffic accident.
In order to take measures to prevent the occurrence risk of the traffic accident, the method for determining the occurrence risk of the road traffic accident provided by the invention further comprises the following steps:
1) determining a weight for each of the risk factors based on each of the classification models.
2) And carrying out weighted average on the multiple weights of each risk factor, and determining the final weight of each risk factor.
3) Sequencing the final weights of all the risk factors and determining key risk factors; the key risk factor is the risk factor with the highest final weight.
Each classification model is an integrated model based on a decision tree, and the decision tree can determine the importance of the input factors by calculating a normalized value of the amount of information entropy reduction. Extracting the importance of each risk factor from each classification model as weight, calculating the average value of the weight of the risk factors in each classification model, arranging the weight from low to high according to the average weight, and determining the influence level of the risk factors, wherein the influence degree is larger when the value is higher. The risk factors belonging to the highest level are the key risk factors. And performing sensitivity analysis on the obtained key risk factors to quantify the contribution degree of the key risk factors to the injury severity. The specific process is that each time, a key risk factor is disturbed, other risk factors are kept unchanged, and the change degree of classification accuracy on the verification data set is observed. The perturbation of a risk factor ranges from 1 unit increase to 10 units increase, each unit referring to one tenth of the average of all samples of the risk factor. During the disturbance process of the risk factor, the greater the influence on the classification of the traffic accident grade, the more sensitive the risk factor is, namely the greater the influence on the severity of the accident. In the traffic management and control process, the risk factor is improved due to the gravity, so that the probability of traffic accidents is reduced, and the injury degree of the traffic accidents is reduced.
In order to improve the accuracy of determining the grade of the occurred traffic accident, the method for determining the risk of the road traffic accident further comprises the following steps:
1) constructing a cross entropy loss function according to the level of the occurred traffic accident and the level of the traffic accident in the database; the cross entropy loss function is an error function between the level of the occurring traffic accident and the level of the traffic accident in the database.
2) And minimizing the cross entropy loss function by adopting a gradient descent method to obtain the weight of each classification model.
The gradient descent method iterates by continually calculating gradients and updating the weights of each of the classification models until convergence is reached.
3) And optimizing the output result of the corresponding classification model by adopting the weight.
The method for determining the road traffic accident occurrence risk provided by the invention uses a double-layer Stacking framework for analysis. And performing logistic regression on output results of the plurality of classification models to obtain a second layer of the Stacking framework.
Corresponding to the method for determining the risk of occurrence of the road traffic accident provided by the present invention, the present invention also provides a system for determining the risk of occurrence of the road traffic accident, as shown in fig. 2, the method for determining the risk of occurrence of the road traffic accident provided by the present invention comprises: the system comprises a data acquisition module 201, a classification model acquisition module 202, an output result module 203, a weighted regression determination module 204, an occurred traffic accident grade determination module 205 and a road traffic accident occurrence risk determination module 206.
The data acquisition module 201 is used for acquiring data of traffic accidents in the database; the data of the traffic accident comprises risk factors of the traffic accident and corresponding traffic accident grade; the risk factors include: the length of the ramp of the expressway, the annual average daily traffic volume of the main road, the gradient of the road, the weather, the age and the driving age of a driver, whether the driver drinks wine or not and the number of people on board the vehicle.
The classification model acquisition module 202 is configured to acquire a plurality of classification models that take all risk factors corresponding to each traffic accident as input and take the probability of causing all levels of the traffic accidents as output; the classification model is a random forest model, a gradient lifting decision tree and a self-adaptive lifting algorithm.
The output result module 203 is configured to determine, according to the data of the traffic accident, probabilities of all traffic accident grades caused by all risk factors corresponding to each traffic accident by using each classification model.
The weighted regression determining module 204 is configured to perform weighted regression on the probabilities of all the traffic accident classes output by each of the classification models by using a logistic regression algorithm, and determine an output result after the weighted regression.
The occurring traffic accident grade determining module 205 is configured to determine the occurring traffic accident grade according to the output result after the weighted regression.
The road traffic accident occurrence risk determining module 206 is configured to determine the road traffic accident occurrence risk according to the level of the occurring traffic accident.
The invention provides a road traffic accident occurrence risk determining system, which further comprises: the system comprises a risk factor weight determining module, a risk factor final weight determining module and a key risk factor determining module.
And the risk factor weight determining module is used for determining the weight of each risk factor according to each classification model.
The risk factor final weight determining module is used for carrying out weighted average on a plurality of weights of each risk factor and determining the final weight of each risk factor.
The key risk factor determination module is used for sequencing the final weights of all the risk factors and determining the key risk factors; the key risk factor is the risk factor with the highest final weight.
The invention provides a road traffic accident occurrence risk determining system, which further comprises: the device comprises a cross entropy loss function building module, a weight determining module of a classification model and an optimizing module.
The cross entropy loss function construction module is used for constructing a cross entropy loss function according to the level of the generated traffic accident and the level of the traffic accident in the database; the cross entropy loss function is an error function between the level of the occurring traffic accident and the level of the traffic accident in the database.
And the weight determining module of the classification model is used for minimizing the cross entropy loss function by adopting a gradient descent method to obtain the weight of each classification model.
And the optimization module is used for optimizing the output result of the corresponding classification model by adopting the weight.
The occurring traffic accident level determining module 205 specifically includes: and an occurring traffic accident grade determining unit.
And the occurring traffic accident grade determining unit is used for determining the occurring traffic accident grade by adopting a Sigmoid function according to the output result after the weighted regression.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (8)
1. A method for determining the occurrence risk of a road traffic accident is characterized by comprising the following steps:
acquiring data of traffic accidents in a database; the data of the traffic accident comprises risk factors of the traffic accident and corresponding traffic accident grade; the risk factors include: the length of a ramp of the expressway, the annual average daily traffic volume of a main road, the gradient of a road, weather, the age and driving age of a driver, whether the driver drinks wine or not and the number of people on board the vehicle;
acquiring a plurality of classification models which take all risk factors corresponding to each traffic accident as input and take the probability of causing all traffic accident grades as output; the classification model is a random forest model, a gradient lifting decision tree and a self-adaptive lifting algorithm;
determining the probability of all traffic accident grades caused by all risk factors corresponding to each traffic accident by utilizing each classification model according to the data of the traffic accidents;
performing weighted regression on the probabilities of all the traffic accident grades output by each classification model by using a logistic regression algorithm, and determining an output result after weighted regression;
determining the grade of the occurred traffic accident according to the output result after the weighted regression;
and determining the occurrence risk of the road traffic accident according to the level of the occurring traffic accident.
2. The method according to claim 1, wherein the obtaining of the plurality of classification models with all risk factors as input and all probability of causing the traffic accident as output further comprises:
determining a weight for each of the risk factors according to each of the classification models;
carrying out weighted average on the multiple weights of each risk factor to determine the final weight of each risk factor;
sequencing the final weights of all the risk factors and determining key risk factors; the key risk factor is the risk factor with the highest final weight.
3. The method for determining the risk of occurrence of a road traffic accident according to claim 1, wherein the determining the level of the occurring traffic accident according to the output result after the weighted regression further comprises:
constructing a cross entropy loss function according to the level of the occurred traffic accident and the level of the traffic accident in the database; the cross entropy loss function is an error function between the level of the occurring traffic accident and the level of the traffic accident in the database;
minimizing the cross entropy loss function by adopting a gradient descent method to obtain the weight of each classification model;
and optimizing the output result of the corresponding classification model by adopting the weight.
4. The method for determining the occurrence risk of the road traffic accident according to claim 1, wherein the determining the level of the occurring traffic accident according to the output result after the weighted regression specifically comprises:
and determining the grade of the occurred traffic accident by adopting a Sigmoid function according to the output result after the weighted regression.
5. A system for determining the risk of occurrence of a road traffic accident, comprising:
the data acquisition module is used for acquiring the data of the traffic accidents in the database; the data of the traffic accident comprises risk factors of the traffic accident and corresponding traffic accident grade; the risk factors include: the length of a ramp of the expressway, the annual average daily traffic volume of a main road, the gradient of a road, weather, the age and driving age of a driver, whether the driver drinks wine or not and the number of people on board the vehicle;
the classification model acquisition module is used for acquiring a plurality of classification models which take all risk factors corresponding to each traffic accident as input and take the probability of causing all traffic accident grades as output; the classification model is a random forest model, a gradient lifting decision tree and a self-adaptive lifting algorithm;
the output result module is used for determining the probability of all traffic accident grades caused by all risk factors corresponding to each traffic accident by utilizing each classification model according to the data of the traffic accidents;
the weighted regression determining module is used for performing weighted regression on the probabilities of all the traffic accident grades output by each classification model by using a logistic regression algorithm and determining an output result after the weighted regression;
the occurring traffic accident grade determining module is used for determining the occurring traffic accident grade according to the output result after the weighted regression;
and the road traffic accident occurrence risk determining module is used for determining the road traffic accident occurrence risk according to the level of the occurring traffic accident.
6. The system of claim 5, further comprising:
a risk factor weight determining module for determining the weight of each risk factor according to each classification model;
a risk factor final weight determining module, configured to perform weighted average on a plurality of weights of each risk factor, and determine a final weight of each risk factor;
the key risk factor determining module is used for sequencing the final weights of all the risk factors and determining the key risk factors; the key risk factor is the risk factor with the highest final weight.
7. The system of claim 5, further comprising:
the cross entropy loss function construction module is used for constructing a cross entropy loss function according to the level of the generated traffic accident and the level of the traffic accident in the database; the cross entropy loss function is an error function between the level of the occurring traffic accident and the level of the traffic accident in the database;
the weight determining module of the classification model is used for minimizing the cross entropy loss function by adopting a gradient descent method to obtain the weight of each classification model;
and the optimization module is used for optimizing the output result of the corresponding classification model by adopting the weight.
8. The system for determining the risk of occurrence of a road traffic accident according to claim 5, wherein the occurring traffic accident grade determination module specifically comprises:
and the occurring traffic accident grade determining unit is used for determining the occurring traffic accident grade by adopting a Sigmoid function according to the output result after the weighted regression.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911389961.5A CN111126868B (en) | 2019-12-30 | 2019-12-30 | Road traffic accident occurrence risk determination method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911389961.5A CN111126868B (en) | 2019-12-30 | 2019-12-30 | Road traffic accident occurrence risk determination method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111126868A true CN111126868A (en) | 2020-05-08 |
| CN111126868B CN111126868B (en) | 2023-07-04 |
Family
ID=70504654
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911389961.5A Active CN111126868B (en) | 2019-12-30 | 2019-12-30 | Road traffic accident occurrence risk determination method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111126868B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111784873A (en) * | 2020-07-01 | 2020-10-16 | 上海城市交通设计院有限公司 | Machine learning principle-based expressway safety vehicle-mounted intelligent system and working process |
| CN112561175A (en) * | 2020-12-18 | 2021-03-26 | 深圳赛安特技术服务有限公司 | Traffic accident influence factor prediction method, device, equipment and storage medium |
| CN113946943A (en) * | 2021-09-26 | 2022-01-18 | 清华大学 | Human-vehicle-road micro traffic system modeling and risk identification method and device |
| CN115565373A (en) * | 2022-09-22 | 2023-01-03 | 中南大学 | Real-time risk prediction method, device, equipment and medium for highway tunnel accident |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708244A (en) * | 2012-05-08 | 2012-10-03 | 清华大学 | Automatic conceptual graph layout method based on measurement of importance degree |
| CN105045975A (en) * | 2015-06-30 | 2015-11-11 | 北京师范大学 | Bayesian network model based risk evaluation method for road transportation accident |
| US20160047666A1 (en) * | 2014-08-15 | 2016-02-18 | Gil Emanuel Fuchs | Determination and Display of Driving Risk |
| CN106897826A (en) * | 2017-02-23 | 2017-06-27 | 吉林大学 | A kind of street accidents risks appraisal procedure and system |
| CN107273340A (en) * | 2017-06-01 | 2017-10-20 | 南京邮电大学 | A kind of road traffic accident factor-analysis approach based on Logistic models |
| CN109191828A (en) * | 2018-07-16 | 2019-01-11 | 江苏智通交通科技有限公司 | Traffic participant accident risk prediction method based on integrated study |
| CN109558969A (en) * | 2018-11-07 | 2019-04-02 | 南京邮电大学 | A kind of VANETs car accident risk forecast model based on AdaBoost-SO |
| CN109871439A (en) * | 2019-02-18 | 2019-06-11 | 华南理工大学 | A kind of Ask-Answer Community problem method for routing based on deep learning |
-
2019
- 2019-12-30 CN CN201911389961.5A patent/CN111126868B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708244A (en) * | 2012-05-08 | 2012-10-03 | 清华大学 | Automatic conceptual graph layout method based on measurement of importance degree |
| US20160047666A1 (en) * | 2014-08-15 | 2016-02-18 | Gil Emanuel Fuchs | Determination and Display of Driving Risk |
| CN105045975A (en) * | 2015-06-30 | 2015-11-11 | 北京师范大学 | Bayesian network model based risk evaluation method for road transportation accident |
| CN106897826A (en) * | 2017-02-23 | 2017-06-27 | 吉林大学 | A kind of street accidents risks appraisal procedure and system |
| CN107273340A (en) * | 2017-06-01 | 2017-10-20 | 南京邮电大学 | A kind of road traffic accident factor-analysis approach based on Logistic models |
| CN109191828A (en) * | 2018-07-16 | 2019-01-11 | 江苏智通交通科技有限公司 | Traffic participant accident risk prediction method based on integrated study |
| CN109558969A (en) * | 2018-11-07 | 2019-04-02 | 南京邮电大学 | A kind of VANETs car accident risk forecast model based on AdaBoost-SO |
| CN109871439A (en) * | 2019-02-18 | 2019-06-11 | 华南理工大学 | A kind of Ask-Answer Community problem method for routing based on deep learning |
Non-Patent Citations (3)
| Title |
|---|
| ROSE YU等: "Deep Learning:A Generic Approach for Extreme Condition Traffic Forecasting", 《PROCEEDINGS OF THE 2017 SIAM INTERNATIONAL CONFERENCE ON DATA MINING》 * |
| 徐萌等: "基于集成学习的航空发动机故障诊断方法", pages 2 - 5 * |
| 石雪怀等: "基于组合模型的交通事故严重程度预测方法", pages 1 - 5 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111784873A (en) * | 2020-07-01 | 2020-10-16 | 上海城市交通设计院有限公司 | Machine learning principle-based expressway safety vehicle-mounted intelligent system and working process |
| CN112561175A (en) * | 2020-12-18 | 2021-03-26 | 深圳赛安特技术服务有限公司 | Traffic accident influence factor prediction method, device, equipment and storage medium |
| CN113946943A (en) * | 2021-09-26 | 2022-01-18 | 清华大学 | Human-vehicle-road micro traffic system modeling and risk identification method and device |
| CN113946943B (en) * | 2021-09-26 | 2022-06-21 | 清华大学 | Human-vehicle-road micro traffic system modeling and risk identification method and device |
| CN115565373A (en) * | 2022-09-22 | 2023-01-03 | 中南大学 | Real-time risk prediction method, device, equipment and medium for highway tunnel accident |
| CN115565373B (en) * | 2022-09-22 | 2024-04-05 | 中南大学 | Expressway tunnel accident real-time risk prediction method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111126868B (en) | 2023-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111126868B (en) | Road traffic accident occurrence risk determination method and system | |
| CN106095963B (en) | Vehicle driving behavior analysis big data public service platform under internet + era | |
| CN110287180B (en) | Wind control modeling method based on deep learning | |
| CN110288137B (en) | Road network level road infrastructure structure comprehensive maintenance decision method and system | |
| CN112149898B (en) | Training of failure rate prediction model, failure rate prediction method and related device | |
| CN112735097A (en) | Regional landslide early warning method and system | |
| CN104636449A (en) | Distributed type big data system risk recognition method based on LSA-GCC | |
| CN112966871A (en) | Traffic jam prediction method and system based on convolution long-short term memory neural network | |
| CN110222733B (en) | High-precision multi-order neural network classification method and system | |
| CN110555565A (en) | Decision tree model-based expressway exit ramp accident severity prediction method | |
| CN111915022B (en) | Gaussian process method and device for rapidly identifying stability coefficient of sliding karst dangerous rock | |
| CN114331122A (en) | Key person risk level assessment method and related equipment | |
| CN115796606A (en) | Quantitative evaluation method and device for highway operation safety index and server | |
| CN116933946A (en) | Rail transit OD passenger flow prediction method and system based on passenger flow destination structure | |
| CN112149922A (en) | Method for predicting severity of accident in exit and entrance area of down-link of highway tunnel | |
| CN116432810A (en) | Traffic flow prediction model determination method, device, apparatus and readable storage medium | |
| CN111260490A (en) | Rapid claims settlement method and system based on tree model for car insurance | |
| Lei et al. | Condition level deteriorations modeling of RC beam bridges with U-Net convolutional neural networks | |
| CN114418748A (en) | Vehicle credit evaluation method, device, equipment and storage medium | |
| CN114154622A (en) | Algorithm model for traffic operation system flow data acquisition missing completion | |
| CN113837266A (en) | Software defect prediction method based on feature extraction and Stacking ensemble learning | |
| CN113222208A (en) | Ada-XGboost-based traffic accident prediction system | |
| CN113642162A (en) | Simulation comprehensive analysis method for urban road traffic emergency plan | |
| CN113487223A (en) | Risk assessment method and risk assessment system based on information fusion | |
| CN110717623B (en) | Photovoltaic power generation power prediction method, device and equipment integrating multiple weather conditions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |