CN109598931B - Group division and difference analysis method and system based on traffic safety risk - Google Patents
Group division and difference analysis method and system based on traffic safety risk Download PDFInfo
- Publication number
- CN109598931B CN109598931B CN201811463865.6A CN201811463865A CN109598931B CN 109598931 B CN109598931 B CN 109598931B CN 201811463865 A CN201811463865 A CN 201811463865A CN 109598931 B CN109598931 B CN 109598931B
- Authority
- CN
- China
- Prior art keywords
- attribute
- group
- data
- risk
- traffic
- 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.)
- Active
Links
Images
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)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a group division and difference analysis method and system based on traffic safety risks, wherein drivers and motor vehicles are used as objects, the object safety risks are calibrated through an integrated learning algorithm, group division is carried out on the basis, and significant difference indexes are identified through a statistical method; the method is characterized in that features of the traffic participants are mined from the traffic behavior expression of the traffic participants based on an ensemble learning algorithm, the safety risk degree of the traffic participants is calibrated, and a plurality of target groups with different safety levels are divided according to the risk degree in order to reduce the data granularity of analysis, study and judgment application; meanwhile, in order to overcome the problem that the integrated learning algorithm lacks in interpretation degree in the safety risk calibration process, the significant difference indexes among the groups are accurately detected and identified through Fisher exact test, so that the characteristics of each risk level group are accurately described, and data support is provided for active traffic safety control.
Description
Technical Field
The invention relates to a group division and difference analysis method and system based on traffic safety risks.
Background
The traffic accident occurrence probability of the traffic participants is predicted by a machine learning method, a clear safety risk index can be calibrated for each driver and vehicle with traffic violation and accident records, but in the current practical application, the traffic safety management application scene taking an individual as an object is limited.
Under the application condition, the data granularity is reduced, the key safety features are identified from the group perspective, and the method has a more realistic guiding effect on the active safety control. Therefore, a group division and difference analysis method and system based on traffic safety risks are urgently needed at present to achieve the purposes.
Disclosure of Invention
The invention aims to provide a group division and difference analysis method and system based on traffic safety risks, which make up the defects of an ensemble learning algorithm in the description of a risk calibration process through a statistical method, excavate the difference characteristics of groups with different risk levels in accident reasons and accident results, and solve the problem that the traffic safety management application scene taking individuals as objects in the prior art is limited.
The technical solution of the invention is as follows:
a group division and difference analysis method based on traffic safety risks comprises the steps of taking a driver and a motor vehicle as objects, calibrating object safety risks through an integrated learning algorithm, carrying out group division on the basis, and identifying significance difference indexes through a statistical method; comprises the following steps of (a) carrying out,
s1, determining traffic participant objects including drivers and motor vehicles; acquiring traffic violation and traffic accident historical records of the target object according to the target object information, and taking the historical records as sample data;
s2, constructing a risk degree prediction model of the target object based on an ensemble learning algorithm; inputting sample data into a model, and outputting a risk degree index of a target object by the model; the risk degree is the label classification probability of the sample data after model processing;
s3, determining the secondary attribute dimension of the target object according to the sample data acquired in the step S1, and dividing the secondary attribute dimension into an accident cause secondary attribute set and an accident result secondary attribute set; splitting the secondary attributes to three levels, and determining three-level attribute factors corresponding to the secondary attributes;
s4, integrating the processing results of the steps S2 and S3, establishing a group division data table, and determining the sample group attribution;
s5, taking the group as an object, taking the secondary attribute as a statistical dimension, and carrying out third-level attribute data statistics inside the group; integrating the statistical results of all the groups to generate a secondary attribute variable R x C list table, wherein R represents the group quantity, and C represents the number of tertiary attribute factors corresponding to the secondary attributes; using fisher's exact test, assume H0: there are significant differences in the values of attribute variables between different populations, H1: there is no significant difference in attribute variables between different populations; obtaining a fuzzy solution p _ value of a Fisher accurate inspection p value by adopting a Monte Carlo simulation calculation method; and determining a Fisher exact test result according to the p _ value, and taking a variable with a significant difference as a group safety characteristic attribute.
Further, in step S2, the risk prediction model construction process specifically includes data label definition and data set division, model feature variable screening based on an embedding method, data set equalization processing, model training based on cross validation, and model performance evaluation based on an ROC curve, i.e., a receiver operation curve and an area under the curve AUC, to screen out a model with the best fitting effect; the risk degree index output by the model is the label classification probability of the data.
Further, in step S4, the fields of the group partition data table include object information, time, third-level attribute factors, risk degrees, and the group to which the group belongs; the data of the field of the belonging group is determined according to the attribution condition of the risk degree of the object information in the threshold interval of the risk degree of each group.
Further, in step S5, determining a fisher exact test result according to p _ value, and using a variable with a significant difference as a group security feature attribute, specifically, accepting the original assumption H0 if the p _ value of the fuzzy solution of the p value is smaller than a set value; otherwise, the original hypothesis H0 is rejected and the hypothesis H1 is accepted.
The traffic safety risk-based group partitioning and difference analysis system for realizing the traffic safety risk-based group partitioning and difference analysis method comprises a data docking module, a risk degree prediction module, an attribute factor analysis module and a group characteristic identification module,
a data docking module: extracting traffic accident records and traffic violation records from a database;
a risk degree prediction module: accessing historical traffic violation data and traffic accident data of the data docking module as samples for model construction; defining a data label and dividing a sample data set; screening model characteristic variables; carrying out equalization processing on the data set; training the model by adopting a cross validation method, and screening out a model with the best fitting effect according to the ROC curve and the AUC value; completing construction of a risk prediction model, extracting historical traffic violation records of a specified target object from a data docking module according to a user instruction, and outputting a risk prediction value of the target object through model processing; generating a risk table;
an attribute factor analysis module: accessing sample data of the data docking module, and determining a secondary attribute according to an original sample data field; determining a third-level attribute factor corresponding to the second-level attribute according to the specific numerical value of the sample data field, wherein the second-level attribute is discrete data, the third-level attribute factor is a corresponding data value domain, and if the second-level attribute is continuous data, the third-level attribute factor is determined through discretization; generating a second-level attribute table and a third-level attribute table;
a group characteristic identification module: accessing a risk degree table by a risk degree prediction module, and acquiring a secondary attribute table and a tertiary attribute table by an attribute factor analysis module; generating a group division data table according to the setting condition of the risk degree threshold interval; determining a secondary attribute p value by adopting a Fisher precise inspection and Monte Carlo simulation calculation method, and writing the secondary attribute p value into a secondary attribute table; and screening out secondary attributes with the p value smaller than a set value as the difference characteristics of different groups to generate a group characteristic table.
Further, the device also comprises a visualization module which: acquiring a group division data table and a group characteristic table from a group characteristic identification module, and counting each group sample according to three-level attributes corresponding to the difference characteristics to generate a difference characteristic table of each group; and calling a visualization engine, and performing visualization processing and display on the difference characteristics and the third-level attribute sample statistical conditions of each group by adopting a thematic map.
The invention has the beneficial effects that:
the method and the system for group division and difference analysis based on traffic safety risks are characterized in that features of traffic participants are mined from traffic behavior expressions based on an ensemble learning algorithm, safety risk degrees of the traffic participants are calibrated, and a plurality of target groups with different safety levels are divided according to the risk degrees in order to reduce data granularity of analysis research and judgment application; meanwhile, in order to overcome the problem that the integrated learning algorithm lacks in interpretation degree in the safety risk calibration process, the significant difference indexes among the groups are accurately detected and identified through Fisher exact test, so that the characteristics of each risk level group are accurately described, and data support is provided for active traffic safety control.
Secondly, on the basis of traffic safety risk degree prediction of traffic participants such as drivers, vehicles and the like based on integrated learning, groups with different safety levels are divided, an R x C list Fisher accurate inspection method is adopted, and differential attribute factors are identified, in the inspection process, the number of rows and columns of the R x C list Fisher is more than 2, so that the error of accurate solution calculation is obvious, the time consumption is long, therefore, Monte Carlo simulation calculation is adopted, a fuzzy solution of a p value is obtained, and the time cost of the algorithm is effectively saved.
Thirdly, the invention carries out traffic safety risk rating on the traffic participants such as drivers and vehicles, and takes the groups formed by the same-level traffic participants as objects to mine the difference characteristics among the groups, thereby solving the problem that the traffic safety management application scene taking individuals as objects is limited.
Drawings
Fig. 1 is a schematic flow chart of a group division and difference analysis method based on traffic safety risk according to an embodiment of the present invention.
Fig. 2 is an explanatory schematic diagram of a group division and difference analysis system based on traffic safety risks according to an embodiment.
Detailed Description
Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
Examples
A group division and difference analysis method based on traffic safety risks comprises the steps of taking a driver and a motor vehicle as objects, calibrating object safety risks through an integrated learning algorithm, carrying out group division on the basis, and identifying significance difference indexes through a statistical method; as shown in fig. 1, the specific steps are as follows:
s1, determining traffic participant objects including drivers and motor vehicles; and acquiring the traffic violation and traffic accident history as sample data according to the target object information.
In the embodiment, the target object information of the driver is a certificate number, and the target object information of the motor vehicle is a combination of a number plate type and a number plate number; the time frame of the history is typically more than 1 year to ensure a sufficient sample size.
S2, constructing a risk degree prediction model of the target object based on an ensemble learning algorithm; and inputting the sample data into the model, and outputting the risk degree index of the target object by the model. And the risk degree is the label classification probability of the sample data after model processing.
The risk prediction model construction process comprises data label definition and data set division, model characteristic variable screening based on an embedding method, data set equalization processing, model training based on cross validation, and model performance evaluation based on an ROC curve (receiver operating curve) and an area under the curve AUC to screen out a model with the best fitting effect; the risk degree index output by the model is the label classification probability of the data.
In the embodiment, a risk prediction model is constructed by combining an improved sampling method and an RF random forest algorithm, the model coverage rate recall is 0.06, and the accuracy is 0.889.
S3, determining the secondary attribute dimension of the target object according to the sample data acquired in the step S1, and dividing the secondary attribute dimension into an accident cause secondary attribute set and an accident result secondary attribute set; and splitting the second-level attributes into three levels, and determining three-level attribute factors corresponding to the second-level attributes.
In the embodiment, a driver is taken as a target object, and elements in the corresponding accident cause secondary attribute set comprise gender, age, nationality, family property, personnel type, driving age, accident identification cause, blood alcohol content, use condition of a safety belt helmet and the like; the vehicle is taken as a target object, and the secondary attributes of the accident cause comprise the type of the vehicle, the traffic mode, the using property of the vehicle, the mileage, the legal condition, the insurance, whether the vehicle is overloaded or not, the state of an illuminating lamp, the load capacity and the like; the secondary attributes of the accident result comprise accident form, accident grade, direct property loss, accident liability and the like.
S4, integrating the processing results of the steps S2 and S3, establishing a group division data table, and determining the sample group attribution; the fields of the group division data table comprise object information, time, third-level attribute factors, risk degrees and groups to which the fields belong; the data of the field of the belonging group is determined according to the attribution condition of the risk degree of the object information in the threshold interval of the risk degree of each group.
In the embodiment, three groups of general, risk and danger are set, the threshold interval of the risk degree of the general group is [0,0.15], the interval of the risk group is (0.15,0.8) and the interval of the danger group is [0.8,1.0 ].
S5, taking the group as an object, taking the secondary attribute as a statistical dimension, and carrying out third-level attribute data statistics inside the group; integrating the statistical results of all the groups to generate a secondary attribute variable R x C list table, wherein R represents the group quantity, and C represents the number of tertiary attribute factors corresponding to the secondary attributes; using Fisher exact test, assume H0: there are significant differences in the values of attribute variables between different populations, H1: there is no significant difference in attribute variables between different populations; considering that three-level attribute factors usually exceed 2 and the number of rows and columns of the list table is generally different, obtaining a fuzzy solution p _ value of a Fisher test p value by adopting a Monte Carlo simulation calculation method; and determining the Fisher accurate test result according to the p _ value, and taking the variable with significant difference as the population security feature attribute.
In the embodiment, a script for testing the difference of attribute variable values among groups is edited by an R language, a fisher.test function in a stats statistical method package is called, a parameter simulate.p.value is set to TRUE, and the Monte Carlo simulation time B is set to 105; and the p value is less than 0.05, the original hypothesis H0 is accepted, and otherwise, the original hypothesis H0 is rejected.
In one embodiment, the Fisher accurate inspection is carried out on the secondary attributes of the accident result by taking personnel as a target object, and the generated R x C list of the accident grade, the accident form, the accident responsibility and the direct property loss is as follows:
TABLE 1 Accident rating R x C series table
TABLE 2 Accident form R x C series connection table
TABLE 3 Accident responsibility R C List
TABLE 4 direct property loss R x C series table
Calculating Fisher exact test p values by Monte Carlo simulation, and the results are respectively:
| secondary attributes of accident outcomes | Grade of accident | Form of accident | Accident responsibility | Direct property loss |
| p_value | 1.0 | 0.58790 | 0.03533 | 0.01469 |
The accident liability and direct property loss of different groups are obviously different, and the two variables are used as the group security feature attribute variables.
According to the group division and difference analysis method based on traffic safety risks, traffic safety risk division is carried out on traffic participants such as drivers and vehicles, the groups formed by the same-level traffic participants are used as objects, difference characteristics among the groups are mined, and the problem that traffic safety management application scenes using individuals as the objects are limited is solved.
A traffic participant group division and feature study and judgment system is shown in figure 2 and comprises a data docking module, a risk degree prediction module, an attribute factor analysis module, a group feature identification module and a visualization module.
And the data docking module is used for extracting the traffic accident record and the traffic violation record from the database.
The risk degree prediction module is accessed into the historical traffic violation data and the traffic accident data of the data docking module and used as a sample for model construction; defining a data label and dividing a sample data set; screening model characteristic variables; carrying out equalization processing on the data set; training the model by adopting a cross validation method, and screening out a model with the best fitting effect according to the ROC curve and the AUC value; the module completes the construction of a risk prediction model, extracts the historical traffic violation record of the specified target object from the data docking module according to the user instruction, and outputs the risk prediction value of the target object through model processing; and generating a risk table.
The attribute factor analysis module is accessed to the sample data of the data docking module and determines a secondary attribute according to an original sample data field; determining a third-level attribute factor corresponding to the second-level attribute according to the specific numerical value of the sample data field, wherein the second-level attribute is discrete data, the third-level attribute factor is a corresponding data value domain, and if the second-level attribute is continuous data, the third-level attribute factor is determined through discretization; and generating a secondary attribute table and a tertiary attribute table.
The group characteristic identification module is accessed into the risk table by the risk prediction module, and the attribute factor analysis module acquires a secondary attribute table and a tertiary attribute table; generating a group division data table according to the setting condition of the risk degree threshold interval; determining a secondary attribute p value by adopting a Fisher precise inspection and Monte Carlo simulation calculation method, and writing the secondary attribute p value into a secondary attribute table; and screening out secondary attributes with the p value smaller than a set value as the difference characteristics of different groups to generate a group characteristic table. Among them, the set value is preferably 0.05.
The visualization module is used for acquiring a group division data table and a group characteristic table from the group characteristic identification module, counting samples of each group according to three-level attributes corresponding to the difference characteristics, and generating a difference characteristic table of each group; and calling a visualization engine, and performing visualization processing and display on the difference characteristics and the three-level attribute sample statistical conditions of each group by adopting a thematic map, wherein the thematic map comprises the forms of comparing type graphs and comparing type graphs such as word clouds, a histogram, a pie chart, a ring chart, a number chart and the like.
According to the group division and difference analysis method and system based on traffic safety risks, features of traffic participants are mined from traffic behavior performance of the traffic participants based on an ensemble learning algorithm, safety risk degrees of the traffic participants are calibrated, and a plurality of target groups with different safety levels are divided according to the risk degrees in order to reduce data granularity of analysis research and judgment application; meanwhile, in order to overcome the problem that the integrated learning algorithm lacks in interpretation degree in the safety risk calibration process, the significant difference indexes among the groups are accurately detected and identified through Fisher exact test, so that the characteristics of each risk level group are accurately described, and data support is provided for active traffic safety control.
On the basis of traffic safety risk degree prediction of traffic participants such as drivers and vehicles based on ensemble learning, groups with different safety levels are divided, an R, C and column link table Fisher accurate inspection method is adopted, differential attribute factors are identified, in the inspection process, the number of rows and columns of the R, C and column link table is greater than 2, so that the error of accurate solution calculation is obvious, the time consumption is long, Monte Carlo simulation calculation is adopted in the embodiment method and the system, the fuzzy solution of the p value is obtained, and the time cost of the algorithm is effectively saved.
Claims (5)
1. A group division and difference analysis method based on traffic safety risks is characterized by comprising the following steps: the method comprises the steps of taking a driver and a motor vehicle as objects, calibrating object safety risks through an integrated learning algorithm, carrying out group division on the basis, and identifying significance difference indexes through a statistical method; comprises the following steps of (a) carrying out,
s1, determining traffic participant objects including drivers and motor vehicles; acquiring traffic violation and traffic accident historical records of the target object according to the target object information, and taking the historical records as sample data;
s2, constructing a risk degree prediction model of the target object based on an ensemble learning algorithm; inputting sample data into a model, and outputting a risk degree index of a target object by the model; the risk degree is the label classification probability of the sample data after model processing; in step S2, the risk prediction model construction process specifically includes data label definition and data set division, model feature variable screening based on an embedding method, data set equalization processing, model training based on cross validation, and model performance evaluation based on an ROC curve, i.e., a receiver operation curve and an area under the curve AUC, to screen out a model with the best fitting effect; the risk degree index output by the model is the label classification probability of the data;
s3, determining the secondary attribute dimension of the target object according to the sample data acquired in the step S1, and dividing the secondary attribute dimension into an accident cause secondary attribute set and an accident result secondary attribute set; splitting the secondary attributes to three levels, and determining three-level attribute factors corresponding to the secondary attributes;
s4, integrating the processing results of the steps S2 and S3, establishing a group division data table, and determining the sample group attribution;
s5, taking the group as an object, taking the secondary attribute as a statistical dimension, and carrying out third-level attribute data statistics inside the group; integrating the statistical results of all the groups to generate a secondary attribute variable R x C list table, wherein R represents the group quantity, and C represents the number of tertiary attribute factors corresponding to the secondary attributes; using fisher's exact test, assume H0: there are significant differences in the values of attribute variables between different populations, H1: there is no significant difference in attribute variables between different populations; obtaining a fuzzy solution p _ value of a Fisher accurate inspection p value by adopting a Monte Carlo simulation calculation method; and determining a Fisher exact test result according to the p _ value, and taking a variable with a significant difference as a group safety characteristic attribute.
2. The traffic safety risk-based population partitioning and variability analysis method of claim 1, wherein: in step S4, the fields of the group division data table include object information, time, third-level attribute factors, risk degrees, and the group to which the group belongs; the data of the field of the belonging group is determined according to the attribution condition of the risk degree of the object information in the threshold interval of the risk degree of each group.
3. The traffic safety risk-based population partitioning and differential analysis method according to claim 1 or 2, wherein: in step S5, determining a fisher exact test result according to p _ value, and using a variable with a significant difference as a group security feature attribute, specifically, accepting an original assumption H0 if a fuzzy solution p _ value of a p value is smaller than a set value; otherwise, the original hypothesis H0 is rejected and the hypothesis H1 is accepted.
4. A traffic safety risk-based population partitioning and dissimilarity analysis system that implements the traffic safety risk-based population partitioning and dissimilarity analysis method according to any one of claims 1 to 3, characterized in that: comprises a data docking module, a risk degree prediction module, an attribute factor analysis module and a group characteristic identification module,
a data docking module: extracting traffic accident records and traffic violation records from a database;
a risk degree prediction module: accessing historical traffic violation data and traffic accident data of the data docking module as samples for model construction; defining a data label and dividing a sample data set; screening model characteristic variables; carrying out equalization processing on the data set; training the model by adopting a cross validation method, and screening out a model with the best fitting effect according to the ROC curve and the AUC value; completing construction of a risk prediction model, extracting historical traffic violation records of a specified target object from a data docking module according to a user instruction, and outputting a risk prediction value of the target object through model processing; generating a risk table;
an attribute factor analysis module: accessing sample data of the data docking module, and determining a secondary attribute according to an original sample data field; determining a third-level attribute factor corresponding to the second-level attribute according to the specific numerical value of the sample data field, wherein the second-level attribute is discrete data, the third-level attribute factor is a corresponding data value domain, and if the second-level attribute is continuous data, the third-level attribute factor is determined through discretization; generating a second-level attribute table and a third-level attribute table;
a group characteristic identification module: accessing a risk degree table by a risk degree prediction module, and acquiring a secondary attribute table and a tertiary attribute table by an attribute factor analysis module; generating a group division data table according to the setting condition of the risk degree threshold interval; determining a secondary attribute p value by adopting a Fisher precise inspection and Monte Carlo simulation calculation method, and writing the secondary attribute p value into a secondary attribute table; and screening out secondary attributes with the p value smaller than a set value as the difference characteristics of different groups to generate a group characteristic table.
5. The traffic safety risk-based population partitioning and variability analysis system of claim 4, wherein: still include visual module, visual module: acquiring a group division data table and a group characteristic table from a group characteristic identification module, and counting each group sample according to three-level attributes corresponding to the difference characteristics to generate a difference characteristic table of each group; and calling a visualization engine, and performing visualization processing and display on the difference characteristics and the third-level attribute sample statistical conditions of each group by adopting a thematic map.
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 CN109598931A (en) | 2019-04-09 |
| CN109598931B true 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) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN110570655B (en) * | 2019-09-19 | 2021-03-05 | 安徽百诚慧通科技有限公司 | Vehicle feature evaluation method based on hierarchical clustering and decision tree |
| 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 |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8909582B2 (en) * | 2013-02-04 | 2014-12-09 | Nec Corporation | Hierarchical latent variable model estimation device, hierarchical latent variable model estimation method, and recording medium |
| EP2778007B1 (en) * | 2013-03-12 | 2022-09-07 | INRIA - Institut National de Recherche en Informatique et en Automatique | Method and system to assess abnormal driving behaviour of vehicles travelling on road |
| 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 |
| CN107886723B (en) * | 2017-11-13 | 2021-07-20 | 深圳大学 | Traffic travel survey data processing method |
| CN108573601B (en) * | 2018-03-26 | 2021-05-11 | 同济大学 | Traffic safety risk field construction method based on WIM data |
| CN108596409B (en) * | 2018-07-16 | 2021-07-20 | 江苏智通交通科技有限公司 | Method for improving accident risk prediction precision of traffic hazard personnel |
| 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
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020108219A1 (en) | 2020-06-04 |
| CN109598931A (en) | 2019-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109598931B (en) | Group division and difference analysis method and system based on traffic safety risk | |
| CN113688042B (en) | Determination method and device of test scene, electronic equipment and readable storage medium | |
| EP3716165A1 (en) | Esg criteria-based enterprise evaluation device and operation method thereof | |
| CN107203774A (en) | The method and device that the belonging kinds of data are predicted | |
| CN109635010B (en) | User characteristic and characteristic factor extraction and query method and system | |
| CN108833139B (en) | OSSEC alarm data aggregation method based on category attribute division | |
| CN104636449A (en) | Distributed type big data system risk recognition method based on LSA-GCC | |
| CN112232944B (en) | Method and device for creating scoring card and electronic equipment | |
| CN112686388A (en) | Data set partitioning method and system under federated learning scene | |
| CN114879613A (en) | Industrial control system information security attack risk assessment method and system | |
| CN111259167B (en) | User request risk identification method and device | |
| TWI677830B (en) | Method and device for detecting key variables in a model | |
| CN116414815A (en) | Data quality detection method, device, computer equipment and storage medium | |
| CN115080638A (en) | Microscopic simulation multi-source data fusion analysis method, electronic equipment and storage medium | |
| CN115203167A (en) | Data detection method and device, computer equipment and storage medium | |
| JP6943242B2 (en) | Analytical instruments, analytical methods, and programs | |
| CN106815320B (en) | Investigation big data visual modeling method and system based on expanded three-dimensional histogram | |
| CN110059749B (en) | Method and device for screening important features and electronic equipment | |
| CN113673609B (en) | Questionnaire data analysis method based on linear hidden variables | |
| CN112488236B (en) | Integrated unsupervised student behavior clustering method | |
| KR102110350B1 (en) | Domain classifying device and method for non-standardized databases | |
| Thompson | Data mining methods and the rise of big data | |
| Keskin et al. | Cohort fertility heterogeneity during the fertility decline period in Turkey | |
| Jacintho et al. | Brazilian presidential elections: Analysing voting patterns in time and space using a simple data science pipeline | |
| CN112733897B (en) | Method and apparatus for determining abnormality cause of multi-dimensional sample 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 |
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. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |