CN113657488B - User grading method, device, equipment and storage medium based on driving behaviors - Google Patents
User grading method, device, equipment and storage medium based on driving behaviors Download PDFInfo
- Publication number
- CN113657488B CN113657488B CN202110939239.5A CN202110939239A CN113657488B CN 113657488 B CN113657488 B CN 113657488B CN 202110939239 A CN202110939239 A CN 202110939239A CN 113657488 B CN113657488 B CN 113657488B
- Authority
- CN
- China
- Prior art keywords
- value
- user
- driving behavior
- interval
- features
- 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
- 230000006399 behavior Effects 0.000 title claims abstract description 115
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000012216 screening Methods 0.000 claims abstract description 53
- 230000008901 benefit Effects 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 28
- 238000010606 normalization Methods 0.000 claims abstract description 25
- 238000013528 artificial neural network Methods 0.000 claims description 59
- 230000006870 function Effects 0.000 claims description 30
- 238000013507 mapping Methods 0.000 claims description 11
- 230000004913 activation Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000013075 data extraction Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 7
- 238000007726 management method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000004378 air conditioning Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
- G06F18/2113—Selection of the most significant subset of features by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Computing Systems (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Development Economics (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The application relates to an artificial intelligence technology, and discloses a user grading method based on driving behaviors, which comprises the following steps: extracting a historical peak value, a total interval value and an interval extremum of driving behaviors preset by a user during each driving from driving behavior data of the user; acquiring the total amount of the vehicle insurance benefits of the user, and carrying out logarithmic normalization processing on the total amount of the vehicle insurance benefits to obtain normalized benefit parameters; carrying out feature description on the historical peak value, the total value of the interval, the interval extremum and the normalized odds parameter to obtain input features; performing feature screening on the input features to obtain screening features; and calculating the feature level of the screening feature, and taking the feature level as the user level of the user. In addition, the application also relates to a blockchain technology, and driving behavior data can be stored in nodes of the blockchain. The application further provides a user grading device based on driving behaviors, electronic equipment and a storage medium. The application can solve the problem of lower precision of grading the users.
Description
Technical Field
The present application relates to the field of artificial intelligence technologies, and in particular, to a user classification method, apparatus, electronic device, and computer readable storage medium based on driving behavior.
Background
Automobiles are increasingly common transportation means in daily travel of people, a large number of insurance services are developed around the automobiles by insurance companies, and in the insurance services, users are classified according to related data of the users, so that targeted insurance output is realized for the users according to different user grades.
The existing user grading method is mainly based on grading guided by results, namely, grading is carried out on users by judging data such as historical risk number, pay amount and the like of the users.
Disclosure of Invention
The application provides a user grading method and device based on driving behaviors and a computer readable storage medium, and mainly aims to solve the problem that the precision of grading users is low.
In order to achieve the above object, the present application provides a user grading method based on driving behavior, including:
acquiring driving behavior data of a user, and extracting a total value and an extremum of a preset interval of driving behavior and a historical peak value of all intervals when the user drives each time from the driving behavior data;
acquiring the total amount of the vehicle insurance claim of the user, and carrying out logarithmic normalization processing on the total amount of the vehicle insurance claim to obtain normalized claim parameters;
acquiring a pre-trained deep neural network, wherein the deep neural network comprises an input layer, a hidden layer and an output layer;
performing feature description on the historical peak value, the interval total value, the interval extremum and the normalized odds parameter by utilizing an input layer of a pre-trained deep neural network to obtain input features;
utilizing a hidden layer of the deep neural network to perform feature screening on the input features to obtain screening features;
and calculating the feature grade of the screening feature by using the output layer of the deep neural network, and taking the feature grade as the user grade of the user.
Optionally, the extracting, from the driving behavior data, a total value and an extremum of a section of a driving behavior preset when the user drives each time, and a historical peak value of all sections includes:
one driving behavior is selected from the preset driving behaviors one by one;
counting the historical values of the selected driving behaviors in the driving behavior data during each driving, and selecting the maximum value in the historical values as a historical peak value;
calculating the sum of historical values of the selected driving behaviors in the preset time interval during each driving in the driving behavior data to obtain an interval total value;
and counting the maximum value and the minimum value of the selected driving behavior in a preset time interval, and collecting the maximum value and the minimum value as interval extremum.
Optionally, the log normalization processing is performed on the total amount of the vehicle risk claim to obtain normalized claim parameters, including:
carrying out logarithmic operation on the total amount of the vehicle insurance claim payment to obtain a logarithmic value;
judging whether the logarithmic value is larger than a preset threshold value or not;
when the logarithmic value is larger than the preset threshold value, taking a preset first value as a normalized reimbursement parameter of the total vehicle risk reimbursement amount;
and when the logarithmic value is smaller than or equal to the preset threshold value, taking a preset second value as a normalized pay parameter of the total car risk pay amount.
Optionally, the performing a logarithmic operation on the total amount of the vehicle risk claim to obtain a logarithmic value includes:
carrying out logarithmic operation on the total sum of the vehicle insurance benefits by using the following logarithmic function to obtain a logarithmic value:
W＝logC
wherein W is the logarithmic value, C is the sum of the vehicle insurance benefits, and log is a logarithmic operator.
Optionally, the characterizing the historical peak, the total interval value, the interval extremum and the normalized pay parameter by using an input layer of a pre-trained deep neural network to obtain input features includes:
performing numerical conversion on the historical peak value, the interval total value, the interval extremum and the normalized odds parameter to obtain a data value corresponding to each data;
and mapping the data values to a pre-constructed space coordinate system by using a preset mapping function, and taking the coordinates of each data value in the space coordinate system as the input characteristics of the data value.
Optionally, the feature screening is performed on the input features by using the hidden layer of the deep neural network to obtain screened features, including:
carrying out batch normalization processing on the input features by utilizing the hidden layer to obtain normalized features;
and randomly discarding the features in the normalized features according to preset weights, and performing full-connection processing for preset times on the randomly discarded normalized features to obtain screening features.
Optionally, the calculating, by using an output layer of the deep neural network, a feature level of the screening feature includes:
respectively calculating relative probability values between the screening features and a plurality of preset feature level labels by using an activation function in an output layer of the deep neural network;
selecting the characteristic grade label with the maximum relative probability value as a target label;
and inquiring from a preset grade table to obtain the characteristic grade corresponding to the target label, and taking the characteristic grade as the characteristic grade of the screening characteristic.
In order to solve the above problems, the present application also provides a user grading apparatus based on driving behavior, the apparatus comprising:
the data extraction module is used for obtaining driving behavior data of a user, and extracting a total value and an interval extremum of a preset driving behavior and a historical peak value of all intervals when the user drives each time from the driving behavior data;
the logarithmic normalization module is used for obtaining the sum of the vehicle insurance benefits of the user, carrying out logarithmic normalization on the sum of the vehicle insurance benefits, and obtaining normalized benefit parameters;
the feature description module is used for acquiring a pre-trained deep neural network, the deep neural network comprises an input layer, a hidden layer and an output layer, and the input layer of the pre-trained deep neural network is used for carrying out feature description on the historical peak value, the interval total value, the interval extremum and the normalized pay parameter to obtain input features;
the feature screening module is used for screening the features of the input features by utilizing the hidden layer of the deep neural network to obtain screening features;
and the grade determining module is used for calculating the characteristic grade of the screening characteristic by utilizing the output layer of the deep neural network, and taking the characteristic grade as the user grade of the user.
In order to solve the above-mentioned problems, the present application also provides an electronic apparatus including:
a memory storing at least one instruction; a kind of electronic device with high-pressure air-conditioning system
And the processor executes the instructions stored in the memory to realize the user grading method based on the driving behavior.
In order to solve the above-mentioned problems, the present application also provides a computer-readable storage medium having stored therein at least one instruction that is executed by a processor in an electronic device to implement the above-mentioned driving behavior-based user grading method.
The embodiment of the application can carry out multidimensional analysis on the driving behavior data of the user such as historical peak value, total interval value, interval extremum and the like, combines the vehicle risk and pay data of the user, realizes data analysis combining behavior and result guidance, and is beneficial to improving the accuracy of grading the user. Therefore, the user grading method, the device, the electronic equipment and the computer readable storage medium based on the driving behavior can solve the problem of lower precision of grading the users.
Drawings
FIG. 1 is a flow chart of a user classification method based on driving behavior according to an embodiment of the present application;
FIG. 2 is a flow chart of a log normalization process according to an embodiment of the present application;
FIG. 3 is a flow chart illustrating the calculation of feature levels according to an embodiment of the present application;
FIG. 4 is a functional block diagram of a user grading device based on driving behavior according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device implementing the driving behavior-based user grading method according to an embodiment of the present application.
The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The embodiment of the application provides a user grading method based on driving behaviors. The execution subject of the user grading method based on driving behavior includes, but is not limited to, at least one of a server, a terminal and the like capable of being configured to execute the method provided by the embodiment of the application. In other words, the driving behavior-based user grading method may be performed by software or hardware installed in a terminal device or a server device, and the software may be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like.
Referring to fig. 1, a flow chart of a user classification method based on driving behavior according to an embodiment of the application is shown. In this embodiment, the driving behavior-based user grading method includes:
s1, acquiring driving behavior data of a user, and extracting a preset total value and an interval extremum of an interval of the driving behavior and a historical peak value of all intervals when the user drives each time from the driving behavior data.
In the embodiment of the application, the driving behavior data refers to related data of various driving behaviors generated by a user in a historical driving process.
For example, data such as a time point at which the user drives each time, a driving time period, a speed, the number of sharp turns, the number of sharp brakes, and the like.
In the embodiment of the application, the kafka message middleware can be utilized to acquire the driving behavior data which can be pre-stored and authorized by the user from a pre-built storage area, wherein the storage area comprises but is not limited to software used by the user, a database corresponding to a system, a blockchain node and a network cache.
In detail, the kafka message middleware is a message middleware for processing active streaming data, and can realize real-time capturing of the data, so that the kafka message middleware is utilized to acquire driving behavior data of a user, timeliness of the acquired driving behavior data can be improved, and further accuracy of grading the user is improved.
Further, a total value and an extremum of a section of the driving behavior preset at each driving of the user and a historical peak value of all sections may be extracted from the driving behavior data.
In detail, the driving behavior includes, but is not limited to, behavior of emergency braking, sharp turns, driving duration, etc. of the user at each driving; the historical peak value, the total interval value and the interval extremum refer to data corresponding to different driving behaviors, for example, the maximum value of the driving speed in the historical driving behaviors of the user, the total number of times of the user turning sharply in a preset time interval, the extremum of the single driving duration of the user in the preset time interval and the like.
In one embodiment of the present application, the extracting, from the driving behavior data, a total value and an extremum of a section of a driving behavior preset when the user drives each time, and a historical peak value of all sections includes:
one driving behavior is selected from the preset driving behaviors one by one;
counting the historical values of the selected driving behaviors in the driving behavior data during each driving, and selecting the maximum value in the historical values as a historical peak value;
calculating the sum of historical values of the selected driving behaviors in the preset time interval during each driving in the driving behavior data to obtain an interval total value;
and counting the maximum value and the minimum value of the selected driving behavior in a preset time interval, and collecting the maximum value and the minimum value as interval extremum.
For example, the selected driving behavior is a driving speed of the user when driving each time, wherein the driving behavior data includes a record of the following driving speeds:
(x 1 ＝90,x 2 ＝30,x 3 ＝50,x 4 ＝40,x 5 ＝75,x 6 ＝60)
wherein ,x1 For the running speed of the user at the 1 st driving, x 2 The driving speed of the user at the 2 nd driving time is the same as the driving speed of the user at the 2 nd driving time.
It is statistically known that the historical peak of the running speed of the user during each driving, i.e. the running speed of all running records of the user is 90 at the most, and the preset time interval is x 4 To x 6 In this case, the maximum value in the predetermined time interval is 75 and the minimum value is 40.
Further, the following summation algorithm may be used to calculate the sum of the historical values of the selected driving behaviors in the preset time interval during each driving in the driving behavior data, so as to obtain an interval total value:
wherein f is the total value of the interval, x i For the i-th value of the selected driving behavior in the preset time interval, N is the number of values of the selected driving behavior in the preset time interval.
S2, obtaining the sum of the vehicle insurance benefits of the user, and carrying out logarithmic normalization processing on the sum of the vehicle insurance benefits to obtain normalized benefit parameters.
In the embodiment of the application, the total amount of the vehicle insurance claim is the total amount of the vehicle insurance claim in the calendar year of the user.
In detail, the step of obtaining the sum of the vehicle insurance claims of the user is consistent with the step of obtaining the driving behavior data of the user in S1, and will not be described herein.
In one practical application scene, the difference of the total amount of the vehicle insurance benefits of different users is larger, so that in order to have a better analysis effect on different users, the total amount of the vehicle insurance benefits can be subjected to logarithmic normalization processing to obtain normalized benefit parameters, and the subsequent efficiency of grading the users is improved.
In the embodiment of the present application, referring to fig. 2, the step of performing logarithmic normalization processing on the total amount of the vehicle risk claims to obtain normalized claim parameters includes:
s21, carrying out logarithmic operation on the total amount of the vehicle insurance claim payment to obtain a logarithmic value;
s22, judging whether the logarithmic value is larger than a preset threshold value or not;
when the logarithmic value is greater than the preset threshold, executing S23, and taking a preset first value as a normalized pay parameter of the total vehicle risk pay;
and when the logarithmic value is smaller than or equal to the preset threshold value, executing S24, and taking the preset second value as a normalized reimbursement parameter of the total automobile insurance reimbursement.
Illustratively, the vehicle insurance claim total can be logarithmically calculated using the following logarithmic function, resulting in a logarithmic value:
W＝logC
wherein W is the logarithmic value, C is the sum of the vehicle insurance benefits, and log is a logarithmic operator.
In detail, the normalized pay parameter of the total amount of the vehicle insurance pay can be generated according to the judging structure by judging the magnitude relation between the logarithmic value and the preset threshold.
For example, when the preset threshold is 0.5 and the logarithmic value is greater than 0.5, it may be indicated that the total amount of vehicle insurance benefits for the user over the years is too high, and the user is a dangerous user, so that the preset first value (e.g. value 1) may be used as the normalized benefit parameter of the total amount of vehicle insurance benefits for the user; when the number is less than or equal to 0.5, it may be indicated that the total amount of vehicle insurance benefits for the user over the years is too high, and the user is a dangerous user, so that a preset first value (e.g., a value of 0) may be used as a normalized benefit parameter of the total amount of vehicle insurance benefits for the user.
In the embodiment of the application, the total amount of the vehicle insurance claim is normalized, so that the numerical diversity of the total amount of the vehicle insurance claim data can be reduced, and the efficiency and the accuracy of grading the users are improved.
S3, acquiring a pre-trained deep neural network, wherein the deep neural network comprises an input layer, a hidden layer and an output layer.
In the embodiment of the application, the pre-trained deep neural network can be utilized to analyze various data obtained through calculation and statistics.
In detail, the pre-trained deep neural network may include the following network hierarchy:
input layer: the method comprises the steps of carrying out feature description on input data so as to facilitate the network hierarchy behind an input layer to process the described input features;
hidden layer: the method comprises the steps of performing operations such as dimension reduction, mapping, selective discarding and the like on input features transmitted by an input layer to realize feature screening, wherein a hidden layer can comprise a plurality of network layers (such as a batch normalization layer, a discarding layer, a full connection layer and the like);
output layer: and the data processing module is used for carrying out final classified output on the data obtained by the hidden layer processing.
Specifically, the deep neural network may be a pre-trained artificial intelligent network model with a data feature analysis function, and the deep neural network includes, but is not limited to, a CNN network, an RNN network, and an RCNN network.
And S4, carrying out feature description on the historical peak value, the interval total value, the interval extremum and the normalized odds parameter by utilizing an input layer of a pre-trained deep neural network to obtain input features.
In the embodiment of the present application, the feature description is performed on the historical peak value, the total interval value, the interval extremum and the normalized odds parameter by using an input layer of a pre-trained deep neural network, so as to obtain an input feature, including:
performing numerical conversion on the historical peak value, the interval total value, the interval extremum and the normalized odds parameter to obtain a data value corresponding to each data;
and mapping the data values to a pre-constructed space coordinate system by using a preset mapping function, and taking the coordinates of each data value in the space coordinate system as the input characteristics of the data value.
In detail, the historical peak value, the total interval value, the interval extremum and the normalized odds parameter may be encoded according to a preset encoding mode, so as to convert the data into a digital form, and obtain a data value corresponding to each data. Wherein, the coding mode includes but is not limited to: ASCII encoding, ANSI encoding, GBK encoding.
Specifically, since the data obtained after the numerical conversion is only data in the form of a numeric string and has no sortability, the data numerical value can be mapped to a space coordinate system constructed in advance by using a preset mapping function to convert the data numerical value into a vector form, so that the sortability of the data numerical value is improved, wherein the mapping function comprises, but is not limited to, a gaussian function and a map function.
For example, the numerical data is mapped into the two-dimensional coordinate system through a preset mapping function and expressed in the form of two-dimensional coordinates, or the numerical data is mapped into the three-dimensional coordinate system through a preset mapping function and expressed in the form of three-dimensional coordinates.
And S5, utilizing the hidden layer of the deep neural network to perform feature screening on the input features to obtain screening features.
In one practical application scenario of the present application, since the input features generated by the input layer are obtained by processing various data (the historical peak value, the total interval value, the interval extremum and the normalized odds parameter), the input features may include some redundant features or error features, that is, some features in the input features cannot reflect relevant data of the user, if the input features are directly analyzed to classify the user, a large amount of computing resources are occupied, resulting in low analysis efficiency, and the error features included in the input features affect the accuracy of the classification result of the final user.
Therefore, the embodiment of the application can utilize the hidden layer of the deep neural network to perform feature screening on the input features so as to remove redundant features and error features in the input features, reduce the data volume of the input features and improve the accuracy of user classification by utilizing the input features.
In the embodiment of the present application, the feature screening of the input features by using the hidden layer of the deep neural network, to obtain screening features, includes:
carrying out batch normalization processing on the input features by utilizing the hidden layer to obtain normalized features;
and randomly discarding the features in the normalized features according to preset weights, and performing full-connection processing for preset times on the randomly discarded normalized features to obtain screening features.
In detail, since the input features generated by the input layer are input features corresponding to each of a plurality of data (the historical peak value, the total value of the interval, the extremum of the interval and the normalized pay parameter), the data scale in the input features is not uniform due to different data types, which is not beneficial to the accuracy of grading the user by using the data.
Therefore, batch normalization processing can be performed on the input features, and unification of data scales in the input features is achieved.
Further, after the batch normalization process, the obtained normalized features are still numerous and may include partial redundant features, and if the normalized features are subjected to redundancy elimination, detailed analysis needs to be performed on each normalized feature, so that the screening efficiency of the input features is low.
According to the embodiment of the application, the features in the normalized features can be randomly discarded according to the preset weight, so that the data volume in the normalized features is reduced, each normalized feature is prevented from being analyzed in detail, and the analysis efficiency is improved.
For example, when the normalized features include 10 features, 3 features may be randomly selected from the normalized features to be discarded when the preset weight is 0.3.
In the embodiment of the application, the normalization feature after random discarding can be subjected to full connection processing for a preset number of times by using the preset activation function in the hidden layer, wherein the preset number of times can be two, and the activation function can be a relu activation function.
In detail, the complexity of the network structure of the deep neural network can be improved by performing the full connection processing for a plurality of times, so that the accuracy of grading the users is improved.
And S6, calculating the characteristic grade of the screening characteristic by using the output layer of the deep neural network, and taking the characteristic grade as the user grade of the user.
In the embodiment of the application, the filtering characteristics can be calculated by utilizing the output layer of the deep neural network so as to obtain the characteristic grade corresponding to the filtering characteristics, and the users are graded according to the characteristic grade.
In an embodiment of the present application, referring to fig. 3, the calculating, by using an output layer of the deep neural network, a feature level of the screening feature includes:
s31, respectively calculating relative probability values between the screening features and a plurality of preset feature level labels by using an activation function in an output layer of the deep neural network;
s32, selecting the characteristic grade label with the maximum relative probability value as a target label;
s33, inquiring from a preset level table to obtain the feature level corresponding to the target label, and taking the feature level as the feature level of the screening feature.
In detail, the activation function may be a sigmoid activation function, and the selection feature may be calculated by using the activation function to obtain a relative probability value between the selection feature and a preset feature tag.
The relative probability value is used to identify the probability that the screening feature belongs to the preset label, for example, the relative probability value between the screening feature and the feature level label a is 80, that is, the probability value indicating that the screening feature belongs to the feature level label a is eighty percent.
In the embodiment of the application, the relative probability value between the screening feature and a plurality of preset feature grade labels can be calculated respectively, and the feature grade label with the maximum relative probability value is selected from the plurality of preset feature grade labels as the target label.
The embodiment of the application can carry out multidimensional analysis on the driving behavior data of the user such as historical peak value, total interval value, interval extremum and the like, combines the vehicle risk and pay data of the user, realizes data analysis combining behavior and result guidance, and is beneficial to improving the accuracy of grading the user. Therefore, the user grading method based on driving behaviors can solve the problem of lower precision of grading the users.
Fig. 4 is a functional block diagram of a user grading device based on driving behavior according to an embodiment of the present application.
The user grading apparatus 100 based on driving behavior according to the present application may be installed in an electronic device. Depending on the functions implemented, the driving behavior based user grading device 100 may include a data extraction module 101, a log normalization module 102, a feature description module 103, a feature screening module 104, and a grade determination module 105. The module of the application, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the present embodiment, the functions concerning the respective modules/units are as follows:
the data extraction module 101 is configured to obtain driving behavior data of a user, and extract a total value and an interval extremum of an interval of a driving behavior preset when the user drives each time and a historical peak value of all intervals from the driving behavior data;
the logarithmic normalization module 102 is configured to obtain a total amount of vehicle insurance benefits of the user, and perform logarithmic normalization processing on the total amount of vehicle insurance benefits to obtain normalized benefit parameters;
the feature description module 103 is configured to obtain a pre-trained deep neural network, where the deep neural network includes an input layer, a hidden layer, and an output layer, and perform feature description on the historical peak value, the total interval value, the interval extremum, and the normalized odds parameter by using the input layer of the pre-trained deep neural network to obtain an input feature;
the feature screening module 104 is configured to perform feature screening on the input features by using a hidden layer of the deep neural network to obtain screening features;
the level determining module 105 is configured to calculate a feature level of the filtering feature by using an output layer of the deep neural network, and take the feature level as a user level of the user.
In detail, each module in the driving behavior-based user grading device 100 in the embodiment of the present application adopts the same technical means as the driving behavior-based user grading method described in fig. 1 to 3, and can produce the same technical effects, which are not described herein.
Fig. 5 is a schematic structural diagram of an electronic device for implementing a user grading method based on driving behavior according to an embodiment of the present application.
The electronic device 1 may comprise a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may further comprise a computer program stored in the memory 11 and executable on the processor 10, such as a user-grading program based on driving behaviour.
The processor 10 may be formed by an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be formed by a plurality of integrated circuits packaged with the same function or different functions, including one or more central processing units (Central Processing unit, CPU), a microprocessor, a digital processing chip, a graphics processor, a combination of various control chips, and so on. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various parts of the entire electronic device using various interfaces and lines, executes various functions of the electronic device and processes data by running or executing programs or modules stored in the memory 11 (e.g., executing a user-grading program based on driving behavior, etc.), and calls data stored in the memory 11.
The memory 11 includes at least one type of readable storage medium including flash memory, a removable hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. The memory 11 may in other embodiments also be an external storage device of the electronic device, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device. The memory 11 may be used not only for storing application software installed in an electronic device and various types of data, such as codes of user-classified programs based on driving behavior, etc., but also for temporarily storing data that has been output or is to be output.
The communication bus 12 may be a peripheral component interconnect standard (peripheral component interconnect, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.
The communication interface 13 is used for communication between the electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), or alternatively a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.
Fig. 5 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 5 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.
For example, although not shown, the electronic device may further include a power source (such as a battery) for supplying power to the respective components, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device may further include various sensors, bluetooth modules, wi-Fi modules, etc., which are not described herein.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The user-grading program based on driving behavior stored in the memory 11 in the electronic device 1 is a combination of instructions which, when run in the processor 10, can realize:
acquiring driving behavior data of a user, and extracting a total value and an extremum of a preset interval of driving behavior and a historical peak value of all intervals when the user drives each time from the driving behavior data;
acquiring the total amount of the vehicle insurance claim of the user, and carrying out logarithmic normalization processing on the total amount of the vehicle insurance claim to obtain normalized claim parameters;
acquiring a pre-trained deep neural network, wherein the deep neural network comprises an input layer, a hidden layer and an output layer;
performing feature description on the historical peak value, the interval total value, the interval extremum and the normalized odds parameter by utilizing an input layer of a pre-trained deep neural network to obtain input features;
utilizing a hidden layer of the deep neural network to perform feature screening on the input features to obtain screening features;
and calculating the feature grade of the screening feature by using the output layer of the deep neural network, and taking the feature grade as the user grade of the user.
In particular, the specific implementation method of the above instructions by the processor 10 may refer to the description of the relevant steps in the corresponding embodiment of the drawings, which is not repeated herein.
Further, the modules/units integrated in the electronic device 1 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. The computer readable storage medium may be volatile or nonvolatile. For example, the computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
The present application also provides a computer readable storage medium storing a computer program which, when executed by a processor of an electronic device, can implement:
acquiring driving behavior data of a user, and extracting a total value and an extremum of a preset interval of driving behavior and a historical peak value of all intervals when the user drives each time from the driving behavior data;
acquiring the total amount of the vehicle insurance claim of the user, and carrying out logarithmic normalization processing on the total amount of the vehicle insurance claim to obtain normalized claim parameters;
acquiring a pre-trained deep neural network, wherein the deep neural network comprises an input layer, a hidden layer and an output layer;
performing feature description on the historical peak value, the interval total value, the interval extremum and the normalized odds parameter by utilizing an input layer of a pre-trained deep neural network to obtain input features;
utilizing a hidden layer of the deep neural network to perform feature screening on the input features to obtain screening features;
and calculating the feature grade of the screening feature by using the output layer of the deep neural network, and taking the feature grade as the user grade of the user.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the application 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 signs in the claims shall not be construed as limiting the claim concerned.
The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The Blockchain (Blockchain), which is essentially a decentralised database, is a string of data blocks that are generated by cryptographic means in association, each data block containing a batch of information of network transactions for verifying the validity of the information (anti-counterfeiting) and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the system claims can also be implemented by means of software or hardware by means of one unit or means. The terms first, second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present application and not for limiting the same, and although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present application without departing from the spirit and scope of the technical solution of the present application.
Claims (8)
1. A method of user classification based on driving behavior, the method comprising:
acquiring driving behavior data of a user, and extracting a total value and an extremum of a preset interval of driving behavior and a historical peak value of all intervals when the user drives each time from the driving behavior data;
acquiring the total amount of the vehicle insurance claim of the user, and carrying out logarithmic normalization processing on the total amount of the vehicle insurance claim to obtain normalized claim parameters;
acquiring a pre-trained deep neural network, wherein the deep neural network comprises an input layer, a hidden layer and an output layer;
performing feature description on the historical peak value, the interval total value, the interval extremum and the normalized odds parameter by utilizing an input layer of a pre-trained deep neural network to obtain input features;
utilizing a hidden layer of the deep neural network to perform feature screening on the input features to obtain screening features;
calculating the feature grade of the screening feature by using the output layer of the deep neural network, and taking the feature grade as the user grade of the user;
the extracting, from the driving behavior data, a total value and an extremum of a section of a preset driving behavior of the user during each driving, and a historical peak value of all sections, includes: one driving behavior is selected from the preset driving behaviors one by one; counting the historical values of the selected driving behaviors in the driving behavior data during each driving, and selecting the maximum value in the historical values as a historical peak value; calculating the sum of historical values of the selected driving behaviors in the preset time interval during each driving in the driving behavior data to obtain an interval total value; counting the maximum value and the minimum value of the selected driving behavior in a preset time interval, and collecting the maximum value and the minimum value as interval extremum;
and carrying out logarithmic normalization processing on the total amount of the vehicle insurance claims to obtain normalized claim parameters, wherein the method comprises the following steps: carrying out logarithmic operation on the total amount of the vehicle insurance claim payment to obtain a logarithmic value; judging whether the logarithmic value is larger than a preset threshold value or not; when the logarithmic value is larger than the preset threshold value, taking a preset first value as a normalized reimbursement parameter of the total vehicle risk reimbursement amount; and when the logarithmic value is smaller than or equal to the preset threshold value, taking a preset second value as a normalized pay parameter of the total car risk pay amount.
2. The driving behavior-based user grading method according to claim 1, wherein the logarithm operation of the total amount of the vehicle risk claim is performed to obtain a logarithm value, and the method comprises:
carrying out logarithmic operation on the total sum of the vehicle insurance benefits by using the following logarithmic function to obtain a logarithmic value:
wherein ,for the logarithmic value, +_>Pay a sum for said vehicle insurance claim,/->Is a logarithmic operator.
3. The driving behavior based user grading method according to claim 1, wherein the characterizing the historical peak, the total interval value, the interval extremum and the normalized odds parameter with the input layer of the pre-trained deep neural network to obtain an input feature comprises:
performing numerical conversion on the historical peak value, the interval total value, the interval extremum and the normalized odds parameter to obtain a data value corresponding to each data;
and mapping the data values to a pre-constructed space coordinate system by using a preset mapping function, and taking the coordinates of each data value in the space coordinate system as the input characteristics of the data value.
4. The driving behavior-based user grading method according to claim 1, wherein the feature screening the input features by using the hidden layer of the deep neural network to obtain screened features comprises:
carrying out batch normalization processing on the input features by utilizing the hidden layer to obtain normalized features;
and randomly discarding the features in the normalized features according to preset weights, and performing full-connection processing for preset times on the randomly discarded normalized features to obtain screening features.
5. A driving behavior based user grading method according to any of claims 1-4, wherein the calculating feature grades of the screening features using the output layer of the deep neural network comprises:
respectively calculating relative probability values between the screening features and a plurality of preset feature level labels by using an activation function in an output layer of the deep neural network;
selecting the characteristic grade label with the maximum relative probability value as a target label;
and inquiring from a preset grade table to obtain the characteristic grade corresponding to the target label, and taking the characteristic grade as the characteristic grade of the screening characteristic.
6. A driving behavior based user grading apparatus for implementing a driving behavior based user grading method according to any of claims 1-5, characterized in that the apparatus comprises:
the data extraction module is used for obtaining driving behavior data of a user, and extracting a total value and an interval extremum of a preset driving behavior and a historical peak value of all intervals when the user drives each time from the driving behavior data;
the logarithmic normalization module is used for obtaining the sum of the vehicle insurance benefits of the user, carrying out logarithmic normalization on the sum of the vehicle insurance benefits, and obtaining normalized benefit parameters;
the feature description module is used for acquiring a pre-trained deep neural network, the deep neural network comprises an input layer, a hidden layer and an output layer, and the input layer of the pre-trained deep neural network is used for carrying out feature description on the historical peak value, the interval total value, the interval extremum and the normalized pay parameter to obtain input features;
the feature screening module is used for screening the features of the input features by utilizing the hidden layer of the deep neural network to obtain screening features;
and the grade determining module is used for calculating the characteristic grade of the screening characteristic by utilizing the output layer of the deep neural network, and taking the characteristic grade as the user grade of the user.
7. An electronic device, the electronic device comprising:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the driving behavior based user grading method according to any of claims 1-5.
8. A computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the driving behavior based user grading method according to any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110939239.5A CN113657488B (en) | 2021-08-16 | 2021-08-16 | User grading method, device, equipment and storage medium based on driving behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110939239.5A CN113657488B (en) | 2021-08-16 | 2021-08-16 | User grading method, device, equipment and storage medium based on driving behaviors |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113657488A CN113657488A (en) | 2021-11-16 |
| CN113657488B true CN113657488B (en) | 2023-08-25 |
Family
ID=78491265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110939239.5A Active CN113657488B (en) | 2021-08-16 | 2021-08-16 | User grading method, device, equipment and storage medium based on driving behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113657488B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109767346A (en) * | 2018-12-17 | 2019-05-17 | 中国平安财产保险股份有限公司 | CAR SERVICE processing method, device, computer equipment and storage medium |
| CN111861077A (en) * | 2019-08-22 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | Method and system for determining driving habits of users and pushing service information |
| CN113178071A (en) * | 2021-04-22 | 2021-07-27 | 深圳壹账通智能科技有限公司 | Driving risk level identification method and device, electronic equipment and readable storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11080725B2 (en) * | 2019-04-17 | 2021-08-03 | Capital One Services, Llc | Behavioral data analytics platform |
| US11704554B2 (en) * | 2019-05-06 | 2023-07-18 | Baidu Usa Llc | Automated training data extraction method for dynamic models for autonomous driving vehicles |
-
2021
- 2021-08-16 CN CN202110939239.5A patent/CN113657488B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109767346A (en) * | 2018-12-17 | 2019-05-17 | 中国平安财产保险股份有限公司 | CAR SERVICE processing method, device, computer equipment and storage medium |
| CN111861077A (en) * | 2019-08-22 | 2020-10-30 | 北京嘀嘀无限科技发展有限公司 | Method and system for determining driving habits of users and pushing service information |
| CN113178071A (en) * | 2021-04-22 | 2021-07-27 | 深圳壹账通智能科技有限公司 | Driving risk level identification method and device, electronic equipment and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113657488A (en) | 2021-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113688923B (en) | Order abnormity intelligent detection method and device, electronic equipment and storage medium | |
| CN114399212A (en) | Ecological environment quality evaluation method and device, electronic equipment and storage medium | |
| CN115170869A (en) | Repeated vehicle damage claim identification method, device, equipment and storage medium | |
| CN113918718A (en) | Vehicle insurance user classification method, device, equipment and medium based on artificial intelligence | |
| CN116662839A (en) | Associated big data cluster analysis method and device based on multidimensional intelligent acquisition | |
| CN114187489B (en) | Method and device for detecting abnormal driving risk of vehicle, electronic equipment and storage medium | |
| CN113628043B (en) | Complaint validity judging method, device, equipment and medium based on data classification | |
| CN117155771B (en) | Equipment cluster fault tracing method and device based on industrial Internet of things | |
| CN114066664A (en) | Risk level assessment method, device, equipment and medium based on behavior portrait | |
| CN114418360A (en) | Smart city operation sign big data analysis method and device | |
| CN113657488B (en) | User grading method, device, equipment and storage medium based on driving behaviors | |
| CN116823437A (en) | Access method, device, equipment and medium based on configured wind control strategy | |
| CN116468547A (en) | Credit card resource allocation method and system based on data mining | |
| CN116401606A (en) | Fraud identification method, device, equipment and medium | |
| CN113706019B (en) | Service capability analysis method, device, equipment and medium based on multidimensional data | |
| CN113626605B (en) | Information classification method, device, electronic equipment and readable storage medium | |
| CN113191805B (en) | Vehicle owner replacement evaluation method, system, electronic equipment and storage medium | |
| CN111860661B (en) | Data analysis method and device based on user behaviors, electronic equipment and medium | |
| CN113657546A (en) | Information classification method and device, electronic equipment and readable storage medium | |
| CN113706207A (en) | Order transaction rate analysis method, device, equipment and medium based on semantic analysis | |
| CN113920590A (en) | Living body detection method, living body detection device, living body detection equipment and readable storage medium | |
| CN113486238A (en) | Information pushing method, device and equipment based on user portrait and storage medium | |
| CN111161042A (en) | Personal risk assessment method, device, terminal and storage medium | |
| CN113793037B (en) | Service distribution method, device, equipment and storage medium based on data analysis | |
| CN113312409B (en) | Task monitoring method and device, electronic equipment and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |