CN109598931A - Group based on traffic safety risk divides and difference analysis method and system - Google Patents
Group based on traffic safety risk divides and difference analysis method and system Download PDFInfo
- Publication number
- CN109598931A CN109598931A CN201811463865.6A CN201811463865A CN109598931A CN 109598931 A CN109598931 A CN 109598931A CN 201811463865 A CN201811463865 A CN 201811463865A CN 109598931 A CN109598931 A CN 109598931A
- Authority
- CN
- China
- Prior art keywords
- group
- data
- risk
- model
- value
- 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
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- 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
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0133—Traffic data processing for classifying traffic situation
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of group's division and difference analysis method and system based on traffic safety risk, using driver, motor vehicle as object, object security risk is demarcated by Ensemble Learning Algorithms, carries out group's division on this basis, and significant difference index is identified by statistical method；The present invention is based on Ensemble Learning Algorithms to excavate its feature from the performance of the traffic behavior of traffic participant, and demarcates its degree of security risk, and to reduce the data granularity that application is studied and judged in analysis, the target group of several different safety class are divided according to risk；It is spent simultaneously in order to overcome the problems, such as that Ensemble Learning Algorithms lack explanation in the process that security risk is demarcated, the significance difference opposite sex index between identification group is accurately examined by Fisher Fei Sheer, to the feature of each risk class group of accurate description, data supporting is provided for the traffic safety improvement of activeization.
Description
Technical field
The present invention relates to a kind of, and the group based on traffic safety risk divides and difference analysis method and system.
Background technique
The traffic accident probability of happening prediction that traffic participant is carried out with machine learning method, can be existed for each and be handed over
Lead to illegal, accident record driver, vehicle calibration one specific security risk index, but in current practical application,
It is relatively limited as the traffic safety management application scenarios of object using individual.
Under this application conditions, data granularity is reduced, from the critical security feature of group's angle recognition, for master
The safety of dynamicization, which is administered, has more real directive function.For this purpose, being badly in need of a kind of group based on traffic safety risk at present
It divides and difference analysis method and system, Lai Shixian object above.
Summary of the invention
The group that the object of the present invention is to provide a kind of based on traffic safety risk divides and difference analysis method and is
System makes up Ensemble Learning Algorithms defect present in the description of risk calibration process by statistical method, excavates different risks
The group of grade solves existing in the prior art using individual as object in cause of accident, the otherness feature of casualty effect
The relatively limited problem of traffic safety management application scenarios.
The technical solution of the invention is as follows:
A kind of group's division and difference analysis method based on traffic safety risk, using driver, motor vehicle as object,
Object security risk is demarcated by Ensemble Learning Algorithms, carries out group's division on this basis, and identify by statistical method
Significant difference index；Include the following steps,
S1, traffic participant object, including driver, motor vehicle are determined；Its traffic is obtained according to target object information to disobey
Method and traffic accident historical record, as sample data；
S2, the risk prediction model based on Ensemble Learning Algorithms building target object；By sample data input model, mould
The risk index of type output target object；Wherein, risk is labeling probability of the sample data after model treatment；
S3, the secondary attributes dimension of target object is determined according to the sample data that step S1 is obtained, be classified as accident at
Because of secondary attributes set, casualty effect secondary attributes set；Secondary attributes are split to three-level, determine that each secondary attributes are corresponding
Three-level attribute factor；
The processing result of S4, combining step S2, S3 establish group and divide tables of data, determine that sample populations belong to；
S5, the three-level attribute data statistics using group as object, using secondary attributes as statistical dimension, inside progress group；
The statistical result of each group is integrated, generates secondary attributes variable R * C contingency table, wherein the R phenon scale of construction, C characterizes secondary attributes
Corresponding three-level attribute factor number；Using the accurate check system of Fei Sheer, it is assumed that H0: attribute variable's value between different groups is deposited
In significant difference, H1: significant difference is not present in the attribute variable between different groups；It is obtained using Monte Carlo simulation calculation method
Fei Sheer accurately examines the fuzzy solution p_value of p value；The accurate inspection result of Fei Sheer is determined according to p_value, will be present significant
The variable of difference is as group's security feature attribute.
Further, in step S2, the building process of risk prediction model specifically includes data label definition and data
Collection divides, the aspect of model Variable Selection based on embedding inlay technique, data set equilibrium treatment, the model training based on cross validation, base
The optimal mould of fitting effect is filtered out in the assessment of the model performance of ROC curve, that is, receiver operating curve and area under the curve AUC
Type；The risk index of model output is the labeling probability of data.
Further, in step S4, group divide tables of data field include object information, the time, three-level attribute factor,
Risk, affiliated group；Wherein affiliated group's field data is according to the risk of object information in each group risk degree threshold zone
Between ownership situation determine.
Further, in step S5, the accurate inspection result of Fei Sheer is determined according to p_value, significant difference will be present
Variable then receives null hypothesis H0 specifically, the fuzzy solution p_value of p value is less than setting value as group's security feature attribute；
Otherwise refuse null hypothesis H0, receive to assume H1.
It is a kind of to realize that the group described in any of the above embodiments based on traffic safety risk divides and difference analysis method
Group based on traffic safety risk divides and difference analysis system, including data are to connection module, risk prediction module, category
Sex factor analysis module and population characteristic identification module,
Data are to connection module: traffic accident record, traffic law violation recording are extracted from database；
Risk prediction module: historical traffic unlawful data, traffic accident data of the access data to connection module, as mould
The sample of type building；Data label is defined, sample data set is divided；Screening model characteristic variable；It carries out at the equilibrium of data set
Reason；Using cross validation method training pattern, according to ROC curve and AUC value filters out the optimal model of fitting effect；It completes
The building of risk prediction model, and the history that specified target object is extracted in connection module is handed over from data according to user instructions
Lead to illegal record, the risk predicted value of the target object is exported by model treatment；Generate risk table；
Attribute factor analysis module: access data determine the sample data of connection module according to raw sample data field
Secondary attributes；The corresponding three-level attribute factor of secondary attributes is determined according to the specific value of sample data field, wherein second level category
Property is discrete data, then three-level attribute factor is corresponding data value field, and secondary attributes then pass through if continuity data
Sliding-model control determines three-level attribute factor；Generate secondary attributes table, three-level attribute list；
Population characteristic identification module: risk table is accessed by risk prediction module, is obtained by attribute factor analysis module
Secondary attributes table, three-level attribute list；Group, which is generated, according to risk threshold interval facilities divides tables of data；Using Fei Sheer
Accurate inspection and Monte Carlo simulation calculating method method, determine secondary attributes p value, secondary attributes table are written；P value is filtered out to be less than
The secondary attributes of setting value generate population characteristic table as the otherness feature of different groups.
Further, further include visualization model, visualization model: obtain group from population characteristic identification module and divide
Each population sample is counted according to the corresponding three-level attribute of otherness feature, generates each group by tables of data, population characteristic table
Otherness mark sheet；Visualization engine is called, is united using otherness feature and three-level attribute sample of the thematic map to each group
Situation is counted to carry out visualization processing and show.
The beneficial effects of the present invention are:
One, group division and difference analysis method and system of this kind based on traffic safety risk, is based on integrated study
Algorithm excavates its feature from the performance of the traffic behavior of traffic participant, and demarcates its degree of security risk, grinds to reduce analysis
The data granularity for sentencing application, the target group of several different safety class are divided according to risk；It is integrated in order to overcome simultaneously
Learning algorithm lacks the problem of explanation is spent in the process that security risk is demarcated, and identification group is accurately examined by Fisher Fei Sheer
Between significance difference opposite sex index administer and mention for the traffic safety of activeization thus the feature of each risk class group of accurate description
For data supporting.
Two, on the traffic safety risk fundamentals of forecasting of the traffic participants such as driver, vehicle based on integrated study,
The group for dividing different safety grades, using the accurate method of inspection of R*C contingency table Fisher, recognition differential opposite sex attribute factor,
In checkout procedure, because R*C contingency table ranks number is all larger than 2, the error for causing its calculating accurately to solve is significant, and time-consuming for calculating,
Thus present invention employs Monte Carlo simulation calculating, the fuzzy solution of p value is obtained, is effectively saved the time cost of the algorithm.
Three, the present invention carries out traffic safety risk rating to traffic participants such as driver, vehicles, and with ad eundem traffic
Patcicipant's gruop at group be object, excavate group between otherness feature, solve the traffic safety pipe using individual as object
Manage the relatively limited problem of application scenarios.
Detailed description of the invention
Fig. 1 is that the embodiment of the present invention is divided and the signal of the process of difference analysis method based on the group of traffic safety risk
Figure.
Fig. 2 group based on traffic safety risk that is embodiment, which divides, illustrates schematic diagram with difference analysis system.
Specific embodiment
The preferred embodiment that the invention will now be described in detail with reference to the accompanying drawings.
Embodiment
A kind of group's division and difference analysis method based on traffic safety risk, using driver, motor vehicle as object,
Object security risk is demarcated by Ensemble Learning Algorithms, carries out group's division on this basis, and identify by statistical method
Significant difference index；Such as Fig. 1, specific steps are as follows:
S1, traffic participant object, including driver, motor vehicle are determined；Its traffic is obtained according to target object information to disobey
Method and traffic accident historical record, as sample data.
In embodiment, the target object information of driver is passport NO., and the target object information of motor vehicle is number plate class
Type is combined with brand number；The time range of historical record is usually more than 1 year, to guarantee enough sample sizes.
S2, the risk prediction model based on Ensemble Learning Algorithms building target object；By sample data input model, mould
The risk index of type output target object.Wherein, risk is labeling probability of the sample data after model treatment.
Wherein, risk prediction model building process includes that data label definition is divided with data set, based on embedding inlay technique
Aspect of model Variable Selection, data set equilibrium treatment, the model training based on cross validation, based on ROC curve (recipient's operation
Curve) with the assessment of the model performance of area under the curve AUC filter out the optimal model of fitting effect；The risk of model output
Index is the labeling probability of data.
In embodiment, the method building risk combined using the improved methods of sampling with RF random forests algorithm is pre-
Model is surveyed, model test coverage recall is 0.06, accuracy 0.889.
S3, the secondary attributes dimension of target object is determined according to the sample data that step S1 is obtained, be classified as accident at
Because of secondary attributes set, casualty effect secondary attributes set；Secondary attributes are split to three-level, determine that each secondary attributes are corresponding
Three-level attribute factor.
In embodiment, to drive artificial target object, corresponding accident origin cause of formation secondary attributes set interior element is inclusive
Not, age, nationality, registered permanent residence property, personnel's type, driving age, accident assert that reason, blood alcohol content, the safety belt helmet use
Situation etc.；Using vehicle as target object, accident origin cause of formation secondary attributes include type of vehicle, mode of transportation, vehicle character of use, inner
Whether number of passes legal status, insurance, overloads, lamp state, capacity value etc.；Casualty effect secondary attributes include Crash characteristics,
Incident classification, direct property loss, accident responsibility etc..
The processing result of S4, combining step S2, S3 establish group and divide tables of data, determine that sample populations belong to；Group draws
The field of divided data table includes object information, time, three-level attribute factor, risk, affiliated group；Wherein affiliated group's field
Data are determined according to the risk of object information in the ownership situation of each group risk degree threshold interval.
In embodiment, be provided with general, risk, dangerous three types of populations, the risk threshold interval of general groups be [0,
0.15], risk group section is (0.15,0.8), and risk groups section is [0.8,1.0].
S5, the three-level attribute data statistics using group as object, using secondary attributes as statistical dimension, inside progress group；
The statistical result of each group is integrated, generates secondary attributes variable R * C contingency table, wherein the R phenon scale of construction, C characterizes secondary attributes
Corresponding three-level attribute factor number；Using the accurate check system of Fisher Fei Sheer, it is assumed that H0: the attribute between different groups becomes
There are significant differences for magnitude, and H1: significant difference is not present in the attribute variable between different groups；It is usually super in view of three-level attribute factor
2 are crossed, and the ranks number of contingency table is generally different, the mould that Fisher examines p value is obtained using Monte Carlo simulation calculation method
Paste solution p_value；The accurate inspection result of Fisher is determined according to p_value value, and the variable of significant difference will be present as group
Security feature attribute.
In embodiment, the script of test of difference is worth by attribute variable between R language editor group, calls stats statistics
Fisher.test function in method packet, parameter simulate.p.value are set as TRUE, and Monte Carlo simulation number B is set
It is set to 105；P value receives null hypothesis H0 less than 0.05, otherwise refuses null hypothesis H0.
In one embodiment, using personnel as target object, Fisher is carried out to casualty effect secondary attributes and is accurately examined,
Incident classification, Crash characteristics, accident responsibility, the R*C contingency table of direct property loss of generation are as follows:
1. incident classification R*C contingency table of table
2. Crash characteristics R*C contingency table of table
3. accident responsibility R*C contingency table of table
4. direct property loss R*C contingency table of table
Fisher Fei Sheer is calculated by Monte Carlo simulation and accurately examines p value, is as a result respectively as follows:
| Casualty effect secondary attributes | Incident classification | Crash characteristics | Accident responsibility | Direct property loss |
| p_value | 1.0 | 0.58790 | 0.03533 | 0.01469 |
Accident responsibility, the direct property loss significant difference of different groups, using the two variables as group's security feature
Attribute variable.
This kind is divided based on the group of traffic safety risk and difference analysis method, joins to traffic such as driver, vehicles
Traffic safety risk rating is carried out with person, and using the group that ad eundem traffic participant forms as object, excavates the difference between group
Anisotropic feature solves the problems, such as relatively limited using individual as the traffic safety management application scenarios of object.
A kind of traffic participant group, which divides, studies and judges system, such as Fig. 2 with feature, including data are pre- to connection module, risk
Survey module, attribute factor analysis module, population characteristic identification module, visualization model.
Data extract traffic accident record, traffic law violation recording to connection module from database.
Risk prediction module accesses historical traffic unlawful data, traffic accident data of the data to connection module, as mould
The sample of type building；Data label is defined, sample data set is divided；Screening model characteristic variable；It carries out at the equilibrium of data set
Reason；Using cross validation method training pattern, according to ROC curve and AUC value filters out the optimal model of fitting effect；The mould
Block completes the building of risk prediction model, and according to user instructions from data to extracting specified target object in connection module
The illegal record of historical traffic exports the risk predicted value of the target object by model treatment；Generate risk table.
Attribute factor analysis module, access data determine the sample data of connection module according to raw sample data field
Secondary attributes；The corresponding three-level attribute factor of secondary attributes is determined according to the specific value of sample data field, wherein second level category
Property is discrete data, then three-level attribute factor is corresponding data value field, and secondary attributes then pass through if continuity data
Sliding-model control determines three-level attribute factor；Generate secondary attributes table, three-level attribute list.
Population characteristic identification module accesses risk table by risk prediction module, is obtained by attribute factor analysis module
Secondary attributes table, three-level attribute list；Group, which is generated, according to risk threshold interval facilities divides tables of data；Using Fei Sheer
Accurate inspection and Monte Carlo simulation calculating method method, determine secondary attributes p value, secondary attributes table are written；P value is filtered out to be less than
The secondary attributes of setting value generate population characteristic table as the otherness feature of different groups.Wherein, setting value is preferably
0.05。
Visualization model obtains group from population characteristic identification module and divides tables of data, population characteristic table, according to difference
The property corresponding three-level attribute of feature counts each population sample, generates the otherness mark sheet of each group；Call visualization
Engine carries out visualization processing and exhibition using otherness feature and three-level attribute sample statistics situation of the thematic map to each group
Show, thematic map includes the accountings classes such as word cloud, histogram, pie chart, doughnut, number figure, compares class graphic form.
This kind is divided based on the group of traffic safety risk and difference analysis method and system, is based on Ensemble Learning Algorithms
Its feature is excavated from the performance of the traffic behavior of traffic participant, and demarcates its degree of security risk, is answered to reduce to analyze to study and judge
Data granularity divides the target group of several different safety class according to risk；While in order to overcome integrated study
Algorithm lacks the problem of explanation is spent in the process that security risk is demarcated, and is accurately examined between identifying group by Fisher Fei Sheer
Significance difference opposite sex index, so that the feature of each risk class group of accurate description, provides number for the traffic safety improvement of activeization
According to support.
On the traffic safety risk fundamentals of forecasting of the traffic participants such as driver, vehicle based on integrated study, draw
The group of point different safety grades, using the accurate method of inspection of R*C contingency table Fisher, recognition differential opposite sex attribute factor,
In checkout procedure, because R*C contingency table ranks number is all larger than 2, the error for causing its calculating accurately to solve is significant, and time-consuming for calculating, is
This embodiment method and system using Monte Carlo simulation calculate, obtain the fuzzy solution of p value, be effectively saved the algorithm when
Between cost.
Claims (6)
1. a kind of group based on traffic safety risk divides and difference analysis method, it is characterised in that: with driver, motor-driven
Vehicle is object, demarcates object security risk by Ensemble Learning Algorithms, carries out group's division on this basis, and pass through statistics
Method identifies significant difference index；Include the following steps,
S1, traffic participant object, including driver, motor vehicle are determined；According to target object information obtain its traffic offence with
Traffic accident historical record, as sample data；
S2, the risk prediction model based on Ensemble Learning Algorithms building target object；By sample data input model, model is defeated
The risk index of target object out；Wherein, risk is labeling probability of the sample data after model treatment；
S3, the secondary attributes dimension that target object is determined according to the sample data that step S1 is obtained, are classified as the accident origin cause of formation two
Grade attribute set, casualty effect secondary attributes set；Secondary attributes are split to three-level, determine the corresponding three-level of each secondary attributes
Attribute factor；
The processing result of S4, combining step S2, S3 establish group and divide tables of data, determine that sample populations belong to；
S5, the three-level attribute data statistics using group as object, using secondary attributes as statistical dimension, inside progress group；It is integrated
The statistical result of each group generates secondary attributes variable R * C contingency table, wherein the R phenon scale of construction, and it is corresponding that C characterizes secondary attributes
Three-level attribute factor number；Using the accurate check system of Fei Sheer, it is assumed that H0: attribute variable's value between different groups exists aobvious
Write difference, H1: significant difference is not present in the attribute variable between different groups；House is taken using the acquisition of Monte Carlo simulation calculation method
You are the accurate fuzzy solution p_value for examining p value；The accurate inspection result of Fei Sheer is determined according to p_value, and significant difference will be present
Variable as group's security feature attribute.
2. the group based on traffic safety risk divides and difference analysis method as described in claim 1, it is characterised in that:
In step S2, the building process of risk prediction model specifically includes data label definition and divides with data set, is based on embedding inlay technique
Aspect of model Variable Selection, data set equilibrium treatment, the model training based on cross validation, be based on ROC curve, that is, recipient
The assessment of the model performance of operating curve and area under the curve AUC filters out the optimal model of fitting effect；The wind of model output
Dangerous degree index is the labeling probability of data.
3. the group based on traffic safety risk divides and difference analysis method as described in claim 1, it is characterised in that:
In step S4, the field that group divides tables of data includes object information, time, three-level attribute factor, risk, affiliated group；
Wherein affiliated group's field data is determined according to the risk of object information in the ownership situation of each group risk degree threshold interval.
4. the group as described in any one of claims 1-3 based on traffic safety risk divides and difference analysis method,
It is characterized in that: in step S5, the accurate inspection result of Fei Sheer being determined according to p_value, the variable conduct of significant difference will be present
Group's security feature attribute then receives null hypothesis H0 specifically, the fuzzy solution p_value of p value is less than setting value；Otherwise refuse
Null hypothesis H0 receives to assume H1.
5. a kind of realize that the described in any item groups based on traffic safety risk of claim 1-4 divide and difference analysis side
The group based on traffic safety risk of method divides and difference analysis system, it is characterised in that: including data to connection module, wind
Dangerous degree prediction module, attribute factor analysis module and population characteristic identification module,
Data are to connection module: traffic accident record, traffic law violation recording are extracted from database；
Risk prediction module: historical traffic unlawful data, traffic accident data of the access data to connection module, as model structure
The sample built；Data label is defined, sample data set is divided；Screening model characteristic variable；Carry out the equilibrium treatment of data set；It adopts
With cross validation method training pattern, according to ROC curve and AUC value filters out the optimal model of fitting effect；Complete risk
The building of prediction model is spent, and the historical traffic for extracting specified target object in connection module is disobeyed from data according to user instructions
Method record, the risk predicted value of the target object is exported by model treatment；Generate risk table；
Attribute factor analysis module: access data determine second level according to raw sample data field to the sample data of connection module
Attribute；The corresponding three-level attribute factor of secondary attributes is determined according to the specific value of sample data field, and wherein secondary attributes are
Discrete data, then three-level attribute factor is corresponding data value field, and secondary attributes are if continuity data, then by discrete
Change processing, determines three-level attribute factor；Generate secondary attributes table, three-level attribute list；
Population characteristic identification module: accessing risk table by risk prediction module, obtains second level by attribute factor analysis module
Attribute list, three-level attribute list；Group, which is generated, according to risk threshold interval facilities divides tables of data；It is accurate using Fei Sheer
Inspection and Monte Carlo simulation calculating method method, determine secondary attributes p value, and secondary attributes table is written；It filters out p value and is less than setting
The secondary attributes of value generate population characteristic table as the otherness feature of different groups.
6. the group based on traffic safety risk divides and difference analysis system as claimed in claim 5, it is characterised in that:
Further include visualization model, visualization model: obtains group from population characteristic identification module and divide tables of data, population characteristic
Each population sample is counted according to the corresponding three-level attribute of otherness feature, generates the otherness mark sheet of each group by table；
Visualization engine is called, is visualized using otherness feature and three-level attribute sample statistics situation of the thematic map to each group
Processing and displaying.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811463865.6A CN109598931B (en) | 2018-11-30 | 2018-11-30 | Group division and difference analysis method and system based on traffic safety risk |
| PCT/CN2019/114373 WO2020108219A1 (en) | 2018-11-30 | 2019-10-30 | Traffic safety risk based group division and difference analysis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811463865.6A CN109598931B (en) | 2018-11-30 | 2018-11-30 | Group division and difference analysis method and system based on traffic safety risk |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109598931A true CN109598931A (en) | 2019-04-09 |
| CN109598931B CN109598931B (en) | 2021-06-11 |
Family
ID=65959467
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811463865.6A Active CN109598931B (en) | 2018-11-30 | 2018-11-30 | Group division and difference analysis method and system based on traffic safety risk |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109598931B (en) |
| WO (1) | WO2020108219A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110458425A (en) * | 2019-07-25 | 2019-11-15 | 腾讯科技（深圳）有限公司 | Risk analysis method, device, readable medium and the electronic equipment of risk subject |
| CN110555277A (en) * | 2019-09-09 | 2019-12-10 | 山东科技大学 | method for evaluating sand burst risk of mining water under loose aquifer |
| CN110570655A (en) * | 2019-09-19 | 2019-12-13 | 安徽百诚慧通科技有限公司 | Vehicle feature evaluation method based on hierarchical clustering and decision tree |
| WO2020108219A1 (en) * | 2018-11-30 | 2020-06-04 | 江苏智通交通科技有限公司 | Traffic safety risk based group division and difference analysis method and system |
| CN117274762A (en) * | 2023-11-20 | 2023-12-22 | 西南交通大学 | Real-time track extraction method based on vision under subway tunnel low-illumination scene |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2778007A1 (en) * | 2013-03-12 | 2014-09-17 | Inria Institut National de Recherche en Informatique et en Automatique | Method and system to assess abnormal driving behaviour of vehicles travelling on road |
| CN104969216A (en) * | 2013-02-04 | 2015-10-07 | 日本电气株式会社 | Hierarchical latent variable model estimation device |
| CN107886723A (en) * | 2017-11-13 | 2018-04-06 | 深圳大学 | A kind of traffic trip survey data processing method |
| CN108573601A (en) * | 2018-03-26 | 2018-09-25 | 同济大学 | A kind of traffic safety risk field construction method based on WIM data |
| CN108596409A (en) * | 2018-07-16 | 2018-09-28 | 江苏智通交通科技有限公司 | The method for promoting traffic hazard personnel's accident risk prediction precision |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160292599A1 (en) * | 2015-04-06 | 2016-10-06 | Fmr Llc | Analyzing and remediating operational risks in production computing systems |
| CN106022787A (en) * | 2016-04-25 | 2016-10-12 | 王琳 | People-vehicle multifactorial assessment method and system based on big data |
| CN106126960A (en) * | 2016-07-25 | 2016-11-16 | 东软集团股份有限公司 | Driving safety appraisal procedure and device |
| CN109598931B (en) * | 2018-11-30 | 2021-06-11 | 江苏智通交通科技有限公司 | Group division and difference analysis method and system based on traffic safety risk |
-
2018
- 2018-11-30 CN CN201811463865.6A patent/CN109598931B/en active Active
-
2019
- 2019-10-30 WO PCT/CN2019/114373 patent/WO2020108219A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104969216A (en) * | 2013-02-04 | 2015-10-07 | 日本电气株式会社 | Hierarchical latent variable model estimation device |
| EP2778007A1 (en) * | 2013-03-12 | 2014-09-17 | Inria Institut National de Recherche en Informatique et en Automatique | Method and system to assess abnormal driving behaviour of vehicles travelling on road |
| CN107886723A (en) * | 2017-11-13 | 2018-04-06 | 深圳大学 | A kind of traffic trip survey data processing method |
| CN108573601A (en) * | 2018-03-26 | 2018-09-25 | 同济大学 | A kind of traffic safety risk field construction method based on WIM data |
| CN108596409A (en) * | 2018-07-16 | 2018-09-28 | 江苏智通交通科技有限公司 | The method for promoting traffic hazard personnel's accident risk prediction precision |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020108219A1 (en) * | 2018-11-30 | 2020-06-04 | 江苏智通交通科技有限公司 | Traffic safety risk based group division and difference analysis method and system |
| CN110458425A (en) * | 2019-07-25 | 2019-11-15 | 腾讯科技（深圳）有限公司 | Risk analysis method, device, readable medium and the electronic equipment of risk subject |
| CN110555277A (en) * | 2019-09-09 | 2019-12-10 | 山东科技大学 | method for evaluating sand burst risk of mining water under loose aquifer |
| CN110570655A (en) * | 2019-09-19 | 2019-12-13 | 安徽百诚慧通科技有限公司 | Vehicle feature evaluation method based on hierarchical clustering and decision tree |
| CN117274762A (en) * | 2023-11-20 | 2023-12-22 | 西南交通大学 | Real-time track extraction method based on vision under subway tunnel low-illumination scene |
| CN117274762B (en) * | 2023-11-20 | 2024-02-06 | 西南交通大学 | Real-time track extraction method based on vision under subway tunnel low-illumination scene |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020108219A1 (en) | 2020-06-04 |
| CN109598931B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109598931A (en) | Group based on traffic safety risk divides and difference analysis method and system | |
| CN105303197B (en) | A kind of vehicle follow the bus safety automation appraisal procedure based on machine learning | |
| CN111462488B (en) | Intersection safety risk assessment method based on deep convolutional neural network and intersection behavior characteristic model | |
| Chen et al. | A graphical modeling method for individual driving behavior and its application in driving safety analysis using GPS data | |
| CN110188807A (en) | Tunnel pedestrian target detection method based on cascade super-resolution network and improvement Faster R-CNN | |
| CN105787472B (en) | A kind of anomaly detection method based on the study of space-time laplacian eigenmaps | |
| CN105513370B (en) | The traffic zone division methods excavated based on sparse license plate identification data | |
| CN110929939B (en) | Landslide hazard susceptibility spatial prediction method based on clustering-information coupling model | |
| CN109635852B (en) | User portrait construction and clustering method based on multi-dimensional attributes | |
| CN108922168B (en) | A kind of mid-scale view Frequent Accidents road sentences method for distinguishing | |
| CN108847022B (en) | Abnormal value detection method of microwave traffic data acquisition equipment | |
| CN105160330A (en) | Vehicle logo recognition method and vehicle logo recognition system | |
| CN111598069B (en) | Highway vehicle lane change area analysis method based on deep learning | |
| CN112966941A (en) | Accident black spot identification method and system based on traffic accident big data | |
| CN110562261A (en) | Method for detecting risk level of driver based on Markov model | |
| CN115080638A (en) | Microscopic simulation multi-source data fusion analysis method, electronic equipment and storage medium | |
| CN112149922A (en) | Method for predicting severity of accident in exit and entrance area of down-link of highway tunnel | |
| CN113420059A (en) | Method and device for actively treating citizen hot line problem | |
| CN106815320B (en) | Investigation big data visual modeling method and system based on expanded three-dimensional histogram | |
| CN109101568A (en) | Traffic high-risk personnel recognition methods based on XgBoost algorithm | |
| Engstrom et al. | Evaluating the Relationship between Contextual Features Derived from Very High Spatial Resolution Imagery and Urban Attributes: A Case Study in Sri Lanka | |
| CN109849926B (en) | Method and system for distinguishing whether taxi is handed to others for driving | |
| Marciniuk et al. | Machine learning applied to acoustic-based road traffic monitoring | |
| CN113192340B (en) | Method, device, equipment and storage medium for identifying highway construction vehicles | |
| CN115497306A (en) | Speed interval weight calculation method based on GIS data |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 211106 19 Su Yuan Avenue, Jiangning economic and Technological Development Zone, Nanjing, Jiangsu Applicant after: JIANGSU ZHITONG TRAFFIC TECHNOLOGY Co.,Ltd. Address before: 210006, Qinhuai District, Jiangsu, Nanjing should be 388 days street, Chenguang 1865 Technology Creative Industry Park E10 building on the third floor Applicant before: JIANGSU ZHITONG TRAFFIC TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |