CN114997458A - Vehicle insurance claim rate prediction method based on principal component analysis and linear regression - Google Patents
Vehicle insurance claim rate prediction method based on principal component analysis and linear regression Download PDFInfo
- Publication number
- CN114997458A CN114997458A CN202210339332.7A CN202210339332A CN114997458A CN 114997458 A CN114997458 A CN 114997458A CN 202210339332 A CN202210339332 A CN 202210339332A CN 114997458 A CN114997458 A CN 114997458A
- Authority
- CN
- China
- Prior art keywords
- data
- principal component
- component analysis
- vehicle
- linear regression
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000012417 linear regression Methods 0.000 title claims abstract description 24
- 238000000513 principal component analysis Methods 0.000 title claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 8
- 230000001502 supplementing effect Effects 0.000 claims abstract description 8
- 230000001419 dependent effect Effects 0.000 claims abstract description 5
- 230000011218 segmentation Effects 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 5
- 230000001133 acceleration Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000012847 principal component analysis method Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 5
- 238000012795 verification Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012628 principal component regression Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000009466 transformation Effects 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a method for predicting vehicle insurance claim rate based on principal component analysis and linear regression, which comprises the following steps: collecting data of a vehicle network; acquiring insurance data; carrying out data quality processing; then, road information including road grade, speed limit and road name is supplemented according to the longitude and latitude of the reserved data; carrying out journey processing on the data GPS points; supplementing industry information through the longitude and latitude of the end of the journey, and supplementing weather information through the time of the start of the journey and the longitude and latitude of the start time; generating risk factors by the generated Internet of vehicles data; performing principal component analysis on each risk factor, and selecting principal components of the explanation strength; and performing linear fitting by taking the selected main components as independent variables and the calculated odds and payouts as dependent variables, and selecting an optimal prediction model. The method solves the problem of strong correlation among risk factors, and most characteristics of the predicted value can be described by using a small amount of principal components.
Description
Technical Field
The invention relates to a method for extracting and reducing dimensions of risk factors of an internet of vehicles and predicting insurance odds and rates, in particular to a method for predicting the insurance odds and rates of the vehicles based on principal component analysis and linear regression, and belongs to the technical field of the internet of vehicles.
Background
Currently, with the advancement of vehicle sensor technology and the widespread use of satellite signals, the internet of vehicles data contains more and more data information, including vehicle information, driving speed, road conditions, travel records, and the like. The insurance field utilizes the data of the car networking can carry out comparatively accurate prediction to the vehicle, can carry out differentiation pricing to the vehicle of different condition. However, some information has certain correlation, which will have certain influence on the prediction of the subsequent insurance data.
Disclosure of Invention
The invention aims to solve the problem that correlation among risk factors causes multiple collinearity influence on subsequent fitting, dimension reduction is carried out on the risk factors through a principal component analysis method, and linear fitting is carried out on principal components with high feature interpretation strength as independent variables to predict insurance claim payment data, so that the method for predicting the vehicle insurance claim payment rate based on principal component analysis and linear regression is provided.
The invention realizes the aim through the following technical scheme that the method for predicting the vehicle insurance claim rate based on principal component analysis and linear regression comprises the following steps:
the method comprises the following steps: acquiring data of a vehicle network, namely acquiring vehicle data according to embedded vehicle data acquisition equipment, GPS satellite request positioning and other data;
step two: acquiring insurance data, collecting and collecting the insurance application amount and the claim settlement amount of the collected vehicle, and calculating the corresponding odds and payments rate;
step three: and carrying out data quality processing, and filtering out data with poor quality of the GPS points. Then, road information including road grade, speed limit and road name is supplemented according to the longitude and latitude of the reserved data;
step four: step three, performing journey processing on the data GPS points;
step five: supplementing industry information by the longitude and latitude of the end of the four-trip, and supplementing weather information by the time of starting the trip and the longitude and latitude of the starting time;
step six: generating a risk factor according to the Internet of vehicles data generated in the step five;
step seven: performing principal component analysis on each risk factor, and selecting principal components with higher explanation strength (the selected explanation strength is more than 80 percent in total);
step eight: and step seven, taking the principal components selected in the step seven as independent variables, taking the odds calculated in the step two as dependent variables, performing linear fitting, and selecting the optimal prediction model.
As a further embodiment of the invention: in the first step, the collected vehicle data comprises an equipment serial number, equipment system time, satellite positioning time, precision, latitude, speed, GPS direction, vehicle sharp turning times and satellite altitude positioning.
As a further embodiment of the invention: and in the second step, corresponding insurance application and claim settlement data are obtained according to the equipment serial number of the data collected by the Internet of vehicles, and corresponding claim rate data are calculated.
As a further embodiment of the invention: in the third step, when data is screened, data with poor data quality is filtered (for example, the GPS point reports once in 15 seconds, if the time interval between two data is within 15 seconds, the poor data is filtered, and the recording interval is guaranteed to be 15 seconds or more).
As a further embodiment of the invention: in the fourth step, the trip segmentation takes the vehicle record interval time as a judgment standard, and the specific segmentation rule depends on the stability of the time interval.
As a further embodiment of the invention: and in the fifth step, the weather interface is accessed by the travel starting time and the longitude and latitude to return the weather data. And returning the industry information by accessing the map interface according to the longitude and latitude after the journey is finished.
As a further embodiment of the invention: in the sixth step, the extracted risk factors comprise the number of times of sudden acceleration of hundreds of kilometers, the proportion of active days, the turning degree of a travel route, the proportion of traveled mileage in severe weather, the coefficient of variation of speed per hour, the number of times of sudden deceleration of hundreds of kilometers, the proportion of fatigue driving travel times in active days, the number of times of sudden change of hundreds of kilometers, the median of maximum overspeed degree of travel, the number of times of overspeed of hundreds of kilometers, the complexity of road conditions and the like.
As a further embodiment of the invention: in the seventh step, the principal component analysis method comprises
1) The method firstly tests the correlation between risk factors;
2) each risk factor is standardized before principal component analysis, and the elimination amount is just right;
3) the explaining strength of each main component to the vehicle characteristics can be visually observed, and the main component PC1, PC2.
As a further embodiment of the invention: in the step eight, main components of different linear regression methods (linear regression, laser regression and edge regression) are adopted in the step seven as independent variables, the odds paid by the corresponding vehicle is adopted as dependent variables to carry out model training, and an optimal model and corresponding optimal parameters are selected by comparing error items in a cross validation mode.
The invention has the beneficial effects that: 1) the method for extracting the risk factors and predicting the corresponding insurance claim rate based on principal component regression solves the problem that the risk factors have strong correlation, and most characteristics of predicted values can be described by using a small amount of principal components; 2) the selected principal components are subjected to a stepwise linear regression prediction model, a ridge regression prediction model, a last regression prediction model and an elastic net regression prediction model. And selecting an optimal regression model by comparing the relative square error of the models and the root-mean-square error.
Drawings
FIG. 1 is a schematic overall flow diagram of the present invention;
FIG. 2 is a schematic diagram of the principal component analysis process and principal component selection process for risk factors according to the present invention;
FIG. 3 is a schematic flow chart of the training selection of the regression model 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.
Example one
As shown in fig. 1 to 3: a method for predicting vehicle insurance claim rate based on principal component analysis and linear regression comprises the following steps:
and step S10, collecting vehicle information.
In the embodiment, the vehicle data is acquired according to the embedded vehicle data acquisition equipment, the positioning request of the GPS satellite and other data, and the acquired vehicle data comprises an equipment serial number, equipment system time, satellite positioning time, precision, latitude, speed, GPS direction, vehicle sharp turning times, satellite altitude positioning and the like.
Step S20, acquiring insurance data.
And acquiring corresponding insurance application and claim settlement data according to the equipment serial number of the data acquired by the Internet of vehicles, and calculating corresponding odds and rate data.
And step S30, speed is compensated.
And data quality processing of the vehicle networking data is carried out, and data with poor quality of the GPS points are filtered. And then road information including road grade, speed limit, road name and the like is supplemented according to the latitude and longitude of the reserved data.
Step S40, a stroking process.
And stroke segmentation is carried out according to the definition and the processed data of the Internet of vehicles, the segmentation rule depends on the stability of the data, the data can be used as a section of stroke for stroke segmentation within 15 minutes or more if the data is stable, and the data can be used as a section of stroke for stroke segmentation within 10 minutes or more if the data is unstable and stable.
And step S50, supplementing weather and industry information data.
And (4) accessing a weather interface to return weather data by using the journey data after the journey segmentation according to the journey starting time and the longitude and latitude. And returning the industry information by accessing the map interface according to the longitude and latitude after the journey is finished. Weather and industry information of the completion journey are supplemented.
Step S60, a risk factor is calculated.
Generating a risk factor from the data generated in step S50: the extracted risk factors comprise the number of sudden acceleration times of hundred kilometers, the proportion of active days, the turning degree of a travel route, the proportion of traveled mileage in severe weather, the coefficient of variation of speed per hour, the number of sudden deceleration times of hundred kilometers, the proportion of fatigue driving travel times of active days, the number of sudden change times of hundred kilometers, the median of maximum overspeed degree of travel, the number of overspeed times of hundred kilometers, the complexity of road conditions and the like.
In step S70, the generated risk factors are subjected to principal component analysis.
And performing principal component analysis according to the generated risk factors, and selecting principal components with principal component accumulation and characteristic accumulated interpretation strength of more than 80% as independent variable parameters of the next step. Fig. two depicts the detailed steps in this embodiment.
S7001: the data set is read. The data set survived and warehoused in S60 is read from the database.
S7002: and (6) standardizing. And (4) normalizing the data set according to different risk factor data to eliminate the dimension.
Direct principal component determination without dimensional elimination can produce unreasonable results.
Setting: x ═ x 1 ,x 2 .....x m ) T Is a random variable of dimension m, x i Is the ith random variable, i ═ 1, 2.. m, order
Wherein, E (x) i ),var(x i ) Are respectively random variables x i Mean and variance of whenIs x i Normalized random variables.
S7003: and calculating a correlation coefficient matrix. The data set processed in step S7002 is subjected to covariance matrix calculation of correlation coefficients.
Let m-dimensional random variable x be (x) 1 ,x 2 ,.....,x m ) T Carry out n independent observations, x 1 ,x 2 ,.....x n Represents an observation sample, wherein x j =(x 1j ,x 2j ,....,x mj ) T Denotes the jth observed sample, x ij The ith variable, j 1, 2.. n, representing the jth observed sample. The observation data are represented by a matrix X.
The covariance matrix S of the samples is
S=[S ij ] m×n
Wherein, the first and the second end of the pipe are connected with each other,is the mean of the samples of the ith variable,is the sample mean of the jth variable.
The sample correlation matrix R is
S7004: and solving the eigenvalue and the eigenvector. And solving the characteristic value and the characteristic vector of the data set processed in the step S7003.
And solving k eigenvalues of the R and corresponding k unit eigenvectors.
Solving the eigen equation of R
|R-λI|=0
M characteristic values of R are obtained
λ 1 ≥λ 2 ...≥λ m
S7005: and extracting characteristic values in the result, and selecting principal components participating in the regression prediction model. The processing data of S7004 is subjected to extraction of feature values, that is, the variance of the principal component, calculation of the contribution ratio of the variance after conversion to the standard deviation, and cumulative contribution ratio. The principal components PC1 to PCn are selected so that the cumulative contribution rate reaches more than 80%. The extracted principal component is selected as an argument component of the prediction model in step S80.
Calculating unit characteristic vector corresponding to first K unit characteristic values
a i =(a 1i ,a 2i ,.....,a mi ) T ,i=1,2....,k
Solving K sample principal components
Linear transformation is carried out by taking unit eigenvector corresponding to the k eigenvalues as coefficients to obtain k sample principal components
Calculating k principal components y i With the original variable x i Correlation coefficient of (p) ((x)) i ,y i ) And k sample principal components versus original variable x i Is given by i 。
In step S80, the regression model is compared to select the best prediction model.
The principal component and insurance data odds data itinerary data set generated in S70 are combined, and a principal component-based regression model can be trained and verified, so that an optimal regression model is selected. And obtaining a prediction model for predicting the insurance claim rate based on the principal component. Fig. three depicts the detailed steps in this embodiment.
S8001: the data set is read. And reading the selected principal component from the S70 data set, and storing the generated payout rate into the data set by insurance in the step S20.
S8002: and (5) splitting a variable. The data set is partitioned into principal component factors and insurance claim rates. The principal component factors serve as independent variables of the model, and the insurance claim rate serves as a target variable of the model.
S8003: and splitting the data set. The data set is partitioned into a training set, a validation set, and a test set, where the test set fraction may be set to 10%.
S800401: and (4) training a linear regression model. And performing model learning on the training set and the verification set according to an algorithm of linear regression.
S800402: and (5) verifying a linear regression model. And S800401, the trained model enters verification set verification. And calculating the model error MSE.
S800411: lasso regression, edge regression, and elastic net regression models.
Regularization is the embodiment of a structure risk (loss function + regularization term) minimum strategy, and a regularization term is added to the empirical risk.
Regularization generally has an optimization objective of the form
Where λ ≧ 0 is used to balance the regularization term and the empirical risk factor.
If it is the basic linear regression,
given data set
The cost function is:
if L is used 2 Norm regularization is then ridge regression
A cost function of
If L is used 1 Norm regularization, which is lasso regression
A cost function of
If L is used 1 Regularization term L 2 The regular term join is then elastic net
A cost function of
Model training can be performed by adjusting the lambda value and the value of alpha (alpha is more than or equal to 0 and less than or equal to 1) by using a grid search method of formula 1-1
The latter term of the formula is elastic-net regularization, lasso regression when α is 0, edge regression when α is 1, and elastic-net regression when α < 1. Model learning is performed on the training set and validation set using a cross-validation method according to equation 1-1.
S800412: verification of the edge regression, laser regression, elastic-net regression test set. And calculating MSE corresponding to different regressions on the verification set, and selecting an optimal regression model (MSE is minimum).
Step S8005: and comparing error terms of different regression models to select a final prediction model. The error term of the model selected in S800402 is compared with the error term of the model selected in S800412. And selecting a model with smaller error terms as a final prediction insurance claim rate prediction model.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (9)
1. A method for predicting vehicle insurance claim rate based on principal component analysis and linear regression, the method comprising the steps of:
the method comprises the following steps: acquiring data of a vehicle network, namely acquiring vehicle data according to embedded vehicle data acquisition equipment, GPS satellite request positioning and other data;
step two: acquiring insurance data, collecting and collecting the insurance application amount and the claim settlement amount of the collected vehicle, and calculating the corresponding odds and payments rate;
step three: performing data quality processing, filtering data with poor quality of GPS points, and supplementing road information including road grade, speed limit and road name according to the longitude and latitude of the reserved data;
step four: step three, performing journey processing on the data GPS points;
step five: supplementing industry information by the longitude and latitude of the end of the four-trip, and supplementing weather information by the time of starting the trip and the longitude and latitude of the starting time;
step six: generating a risk factor according to the Internet of vehicles data generated in the fifth step;
step seven: performing principal component analysis on each risk factor, and selecting principal components of interpretation dynamics, wherein the total selected interpretation dynamics is more than 80%;
step eight: and step seven, taking the selected principal components as independent variables, taking the calculated odds and payouts in step two as dependent variables, performing linear fitting, and selecting the optimal prediction model.
2. The method for vehicle insurance claim rate prediction based on principal component analysis and linear regression of claim 1, wherein: in the first step, the acquired vehicle data comprises an equipment serial number, equipment system time, satellite positioning time, precision, latitude, speed, GPS direction, vehicle sharp turning times and satellite altitude positioning.
3. The method for vehicle insurance claim rate prediction based on principal component analysis and linear regression of claim 1, wherein: in the second step, corresponding insurance application and claim settlement data are obtained according to the equipment serial number of the data collected by the Internet of vehicles, and corresponding claim rate data are calculated.
4. The method for vehicle insurance claim rate prediction based on principal component analysis and linear regression of claim 1, wherein: in the third step, when data is screened, data with poor data quality is filtered, if the time interval between two certain data is within 15 seconds, the bad data is filtered, and the recording interval is ensured to be 15 seconds or more.
5. The method for vehicle insurance claim rate prediction based on principal component analysis and linear regression of claim 1, wherein: in the fourth step, the trip segmentation takes the vehicle record interval time as a judgment standard, and the specific segmentation rule depends on the stability of the time interval.
6. The method for vehicle insurance claim rate prediction based on principal component analysis and linear regression of claim 1, wherein: and in the fifth step, the weather interface is accessed by the travel starting time and the longitude and latitude to return weather data, and the map interface is accessed by the longitude and latitude to return industry information after the travel is finished.
7. The principal component analysis and linear regression-based method for predicting vehicle insurance claim rate according to claim 1, wherein: in the sixth step, the extracted risk factors comprise the number of times of sudden acceleration of hundreds of kilometers, the proportion of active days, the turning degree of a travel route, the proportion of traveled mileage in severe weather, the coefficient of variation of speed per hour, the number of times of sudden deceleration of hundreds of kilometers, the proportion of fatigue driving travel times in active days, the number of times of sudden change of hundreds of kilometers, the median of maximum overspeed degree of travel, the number of times of overspeed of hundreds of kilometers and the complexity of road conditions.
8. The method for vehicle insurance claim rate prediction based on principal component analysis and linear regression of claim 1, wherein: in the seventh step, the principal component analysis method comprises
1) The method firstly tests the correlation between risk factors;
2) each risk factor is standardized before principal component analysis, and dimensions are eliminated;
3) and visually observing the interpretation strength of each main component on the vehicle characteristics, and selecting the main components PC1, PC2.
9. The principal component analysis and linear regression-based method for predicting vehicle insurance claim rate according to claim 1, wherein: and in the step eight, selecting the principal components of the method in different linear regression modes in the step seven as independent variables, carrying out model training by taking the odds and the odds of the corresponding vehicle as dependent variables, and selecting the optimal model and the corresponding optimal parameters by comparing error items in a grid search and cross validation mode.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210339332.7A CN114997458A (en) | 2022-04-01 | 2022-04-01 | Vehicle insurance claim rate prediction method based on principal component analysis and linear regression |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210339332.7A CN114997458A (en) | 2022-04-01 | 2022-04-01 | Vehicle insurance claim rate prediction method based on principal component analysis and linear regression |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114997458A true CN114997458A (en) | 2022-09-02 |
Family
ID=83023398
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210339332.7A Pending CN114997458A (en) | 2022-04-01 | 2022-04-01 | Vehicle insurance claim rate prediction method based on principal component analysis and linear regression |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114997458A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116091254A (en) * | 2023-04-11 | 2023-05-09 | 天津所托瑞安汽车科技有限公司 | Commercial vehicle risk analysis method |
-
2022
- 2022-04-01 CN CN202210339332.7A patent/CN114997458A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116091254A (en) * | 2023-04-11 | 2023-05-09 | 天津所托瑞安汽车科技有限公司 | Commercial vehicle risk analysis method |
CN116091254B (en) * | 2023-04-11 | 2023-08-01 | 天津所托瑞安汽车科技有限公司 | Commercial vehicle risk analysis method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112446591B (en) | Zero sample evaluation method for student comprehensive ability evaluation | |
CN109050535B (en) | Rapid terrain condition identification method based on vehicle attitude | |
CN109670970B (en) | Driving behavior scoring method and device and computer readable storage medium | |
CN111428960B (en) | Intelligent vehicle driving automatic evaluation method fusing multi-source vehicle-mounted sensor information | |
CN110705774A (en) | Vehicle energy consumption analysis prediction method and system | |
CN105975757B (en) | A kind of anxious deceleration Activity recognition method based on vehicle operation data | |
CN110399364B (en) | Data fusion method based on multiple road detector data | |
CN110619482A (en) | Driving behavior scoring method based on logistic regression and single-level analysis weighting method | |
CN113688558A (en) | Automobile driving condition construction method and system based on large database samples | |
CN114997458A (en) | Vehicle insurance claim rate prediction method based on principal component analysis and linear regression | |
CN110910531A (en) | Rapid pavement friction coefficient detection method based on vehicle-mounted OBD information | |
CN112784173B (en) | Recommendation system scoring prediction method based on self-attention confrontation neural network | |
CN117436653A (en) | Prediction model construction method and prediction method for travel demands of network about vehicles | |
CN117078077A (en) | Ecological vulnerability evaluation method for expressway road domain | |
CN115774942A (en) | Driving style identification model modeling and statistical method based on Internet of vehicles real vehicle data and SVM | |
CN105869412B (en) | A kind of anxious acceleration Activity recognition method based on vehicle operation data | |
CN114493093A (en) | Evaluation method and device for automatic driving prediction module | |
CN113252057A (en) | Method and system for identifying driving tendency based on high altitude navigation data | |
CN113870565B (en) | Method and system for estimating unknown flow of urban road network based on transfer learning model | |
Nurcahya et al. | Forecasting fuel consumption based-on OBD II data | |
US20220340148A1 (en) | Method for estimating an accident risk of an autonomous vehicle | |
CN117985026A (en) | Driving style analysis method and device and vehicle | |
Xia et al. | Study of comprehensive evaluation for L2 automated vehicles on field test | |
CN116665342B (en) | New energy automobile driving behavior analysis method, system and equipment | |
CN116884225B (en) | Method and device for evaluating influence of proportion of automatic driving vehicles on traffic efficiency |
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 |