Driving style scoring method and device, electronic equipment and storage medium

Abstract

The invention discloses a driving style scoring method and device, electronic equipment and a storage medium, and is applied to the field of big data. The method comprises the following steps: firstly, acquiring historical driving data, then preprocessing the historical driving data to remove invalid data to obtain first data, and performing feature extraction on the first data to obtain an initial data set for distinguishing a positive sample and a negative sample; and finally, training a driving style scoring model by taking the initial data set as input data, and scoring the driving style of the user by using the trained driving style scoring model. According to the method, the risk degree of the driving data is rapidly measured, and then the driving style score of the driver is efficiently and accurately determined.

Description

Technical Field

[0001] This application relates to the field of big data technology, and particularly to a method, device, electronic device, and storage medium for driving style scoring. Background Art

[0002] Currently, traffic safety issues have become increasingly prominent and have become a hot topic of concern to the whole society. Existing research shows that accidents related to driver behavior factors account for more than 90% of all accidents. Therefore, objectively evaluating the driving style of drivers and carrying out active intervention based on this have important practical significance for improving road traffic safety.

[0003] In the related art, when evaluating the driving style of a driver, it is impossible to quickly measure the risk degree of driving data and efficiently and accurately determine the driving style score of the driver. Summary of the Invention

[0004] In view of this, embodiments of this application provide a method, device, electronic device, and storage medium for driving style scoring, aiming to quickly measure the risk degree of driving data and then efficiently and accurately determine the driving style score of the driver.

[0005] In a first aspect, embodiments of this application provide a method for driving style scoring, the method including:

[0006] Obtain historical driving data, where the historical driving data is historical data generated by a user driving a vehicle;

[0007] Preprocess the historical driving data to remove invalid data to obtain first data, where the invalid data is used to represent data irrelevant to driving style evaluation;

[0008] Extract features from the first data to obtain an initial data set that distinguishes positive samples and negative samples;

[0009] Use the initial data set as input data to train a driving style scoring model, and use the trained driving style scoring model to score the driving style of the user.

[0010] Optionally, the extracting features from the first data to obtain an initial data set that distinguishes positive samples and negative samples includes:

[0011] Obtain the first data;

[0012] Based on the first data, calculate and determine the evaluation weights of different first data features, where the first data features are used to represent the data types in the first data, and the evaluation weights are used to represent the importance degree of the first data features for driving style evaluation;

[0013] Based on the result calculated by the evaluation weight of the first data feature, screen and extract the first data feature in the first data to obtain an initial data set for distinguishing positive samples and negative samples.

[0014] Optionally, the calculating the evaluation weights of different first data features based on the first data includes:

[0015] Set a sample discrimination rule, which is used to distinguish the first data into positive sample data and negative sample data, and the sample discrimination rule has an associated relationship with the first data feature;

[0016] Determine the sample discrimination rule corresponding to the first data feature in the first data;

[0017] Use the sample discrimination rule to distinguish the positive sample data and negative sample data in the first data feature;

[0018] Calculate the evaluation weights of different first data features based on the positive sample data and the negative sample data.

[0019] Optionally, the calculating the evaluation weights of different first data features based on the positive sample data and the negative sample data includes:

[0020] Perform binning processing on the positive sample data and the negative sample data according to the first data feature;

[0021] The training of the driving style scoring model with the initial data set as the input data includes:

[0022] Calculate the evaluation weights of different first data features based on the binned positive sample data and negative sample data.

[0023] Optionally, the preprocessing of the historical driving data to remove invalid data to obtain the first data includes:

[0024] Determine the corresponding category of the data in the historical driving data;

[0025] Set the target value range of the corresponding data based on the category;

[0026] Screen the data in the historical driving data according to the target value range, and remove the invalid data to obtain the first data, where the invalid data is the data whose value does not conform to the target value range.

[0027] Optionally, the method further includes:

[0028] Divide the initial data set into a first data set and a second data set, use the first data set as the training data set of the driving style scoring model, and use the second data set as the test data set of the driving style scoring model;

[0029] Training the driving style scoring model with the initial data set as input data, and scoring the driving style of the user by using the trained driving style scoring model, includes:

[0030] Use the first data set as input data to train the driving style scoring model, and use the trained driving style scoring model to score the driving style of the user in the second data set.

[0031] Optionally, the driving style scoring model is a linear regression prediction model, and scoring the driving style of the user by using the trained driving style scoring model, includes:

[0032] Obtain the driving data of the vehicle;

[0033] Input the driving data into the trained linear regression prediction model to obtain a prediction result;

[0034] Convert the prediction result to obtain the score of the driving style of the user.

[0035] In a second aspect, an embodiment of the present application provides a device for scoring driving style, the device includes: an acquisition module, a processing module, an extraction module, and a scoring module;

[0036] The acquisition module is configured to acquire historical driving data, and the historical driving data is historical data generated by a user driving a vehicle;

[0037] The processing module is configured to preprocess the historical driving data to remove invalid data to obtain first data, and the invalid data is used to represent data irrelevant to driving style evaluation;

[0038] The extraction module is configured to perform feature extraction on the first data to obtain an initial data set;

[0039] The scoring module is configured to use the initial data set as input data to train a driving style scoring model, and use the trained driving style scoring model to score the driving style of the user.

[0040] In a third aspect, the present application provides an electronic device, the device includes: a processor, a memory, and a system bus;

[0041] The processor and the memory are connected through the system bus;

[0042] The memory is used to store one or more programs, and the one or more programs include instructions, which when executed by the processor cause the processor to execute the method described in the first aspect.

[0043] In a fourth aspect, an embodiment of the present application provides a computer storage medium, in which code is stored, and when the code is run, the device running the code implements the method described in any one of the foregoing first aspects.

[0044] The present application provides a method and system for driving style scoring. When executing the method, first obtain historical driving data, then preprocess the historical driving data to remove invalid data to obtain first data, and extract features from the first data to obtain an initial data set for distinguishing positive samples and negative samples. Finally, use the initial data set as input data to train a driving style scoring model, and use the trained driving style scoring model to score the driving style of the user. In this way, by preprocessing the historical driving data to remove invalid data, it can be ensured that the training data of the driving style evaluation model are all valid data, which guarantees the training effect of the driving style evaluation model. Since the redundant interference of invalid data is avoided, the training speed of the model can be improved. By distinguishing the data into positive samples and negative samples to obtain positive sample data and negative sample data, the risk degree of the driving data can be quickly measured, and then the driving style score of the driver can be determined efficiently and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] To more clearly illustrate the technical solutions in the embodiments or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0046] Figure 1 It is a flowchart of a method for driving style scoring provided by an embodiment of the present application;

[0047] Figure 2 It is a flowchart of a method for obtaining an initial data set provided by an embodiment of the present application;

[0048] Figure 3 It is a schematic structural diagram of a device for driving style scoring provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0050] In the research of related technologies, it is found that when evaluating the driving style of a driver, only the risk degree of different driving data can be measured by the user-defined method. This method makes the scoring process very complicated, and different users have different understandings of the risk degree of different driving data, so it is impossible to uniformly and efficiently evaluate the driving style.

[0051] Based on this, the present application proposes a method, device, electronic device and storage medium for driving style scoring. It can distinguish positive samples and negative samples from driving data, quickly measure the risk degree of driving data, and then efficiently and accurately determine the driving style score of the driver.

[0052] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0053] Figure 1 For the flowchart of a method for driving style scoring provided in an embodiment of the present application, see Figure 1 As shown, a method for driving style scoring provided in an embodiment of the present application includes:

[0054] S11: Obtain historical driving data.

[0055] The above-mentioned historical driving data is the historical data generated by the user driving a vehicle, and can also be understood as the driving history data included in the vehicle driving cycle. Specific historical driving data may include, but is not limited to, the following types: vin number, time, vehicle speed, acceleration, mileage, angular velocity, steering wheel angle, longitude, latitude, accelerator pedal travel, brake pedal travel, airbag status, temperature, assisted driving function activation status, etc.

[0056] The historical driving data in the embodiments of the present application can be understood as raw data, which needs to be processed to obtain the input data for training the driving style scoring model.

[0057] S12: Preprocess the historical driving data to remove invalid data and obtain the first data.

[0058] After obtaining the historical driving data, it needs to be preprocessed. The purpose of preprocessing is to extract the valid data in the historical driving data, that is, the data related to the driving style evaluation. In one possible implementation, the method for preprocessing the historical driving data can be: first, determine the corresponding categories of the data in the historical driving data, and then set the target value range of the corresponding data based on the categories. Finally, screen the data in the historical driving data according to the target value range, remove the invalid data, and obtain the first data. The invalid data is the data whose value does not match the target value range.

[0059] Specifically, as can be seen from the above description, the historical driving data can include various different categories of data, and data screening is required during the preprocessing of the historical driving data. The specific screening process can be to set the same or different target value ranges for different data categories in advance, and then determine the valid data in the data of this category according to the target value range, identify the data that does not belong to the target value range as invalid data, and remove the invalid data to obtain the first data. The above invalid data can be used to represent the data unrelated to the driving style evaluation.

[0060] For example, preprocessing the historical driving data to remove invalid data can include the following methods: ① Filter out abnormal vehicle speed amounts by checking the driving duration, longitude and latitude, and driving mileage in the driving itinerary. ② Filter out abnormal angular velocity and steering wheel angle amounts by longitude and latitude, and filter out abnormal acceleration amounts by the accelerator pedal and brake pedal strokes. ③ Filter out non-driver driving itineraries by the auxiliary driving function activation status.

[0061] By preprocessing the historical driving data, screening the valid data and removing the invalid data. In this way, it is convenient to obtain the data related to the driving style evaluation, which is beneficial to improving the efficiency of training the driving style scoring model. At the same time, the training accuracy of the driving style scoring model can be improved because the interference of irrelevant data is avoided.

[0062] S13: Extract features from the first data to obtain an initial data set for distinguishing positive samples and negative samples.

[0063] As mentioned above, "extract features from the first data to obtain an initial data set for distinguishing positive samples and negative samples". Figure 2 This is a flowchart of a method for obtaining an initial data set provided by an embodiment of the present application. As Figure 2 shown, the method can specifically be:

[0064] S131: Obtain the first data.

[0065] S132: Calculate and determine the evaluation weights of different first data features, where the first data features are used to represent the data types in the first data, and the evaluation weights are used to represent the importance of the first data features for driving style evaluation.

[0066] The first data may include different data types, and the importance of different data types for evaluating the user's driving style is different. Therefore, it is necessary to determine the proportion of different data types in evaluating the driving style. According to the different proportions, the data situations of all types are comprehensively considered to finally determine the driving style score of the user.

[0067] The evaluation weights for different data need to be obtained through calculation. The specific calculation method can be:

[0068]

[0069] In the formula, P ratio represents the proportion of positive samples, N ratio represents the proportion of negative samples, and WOE (weight of evidence) represents the evidence weight of the feature.

[0070]

[0071] In the formula, WOE i represents the WOE value of the i-th bin of a certain feature variable. The IV value is in the range of [0, 1], which can represent the prediction ability of the variable. The larger the value, the better the prediction ability of the feature. Different intervals of the IV value can be divided to represent different prediction abilities. Feature variables with an IV value less than the threshold are excluded. The magnitude of the IV value reflects the contribution ability to prediction. According to the magnitude of the IV value, features with small contributions to prediction can be excluded to avoid adverse effects on the scoring results. The above-mentioned threshold can be set by those skilled in the art according to the actual situation and application scenario, and is not set here. For example, it can be 0.03.

[0072] In a possible implementation manner, when determining the evaluation weights of different first data features, it is necessary to distinguish positive samples and negative samples from the first data. Specifically: First, set a sample discrimination rule, which is used to distinguish the first data into positive sample data and negative sample data, and the sample discrimination rule has an association relationship with the first data feature. Then, determine the sample discrimination rule corresponding to the first data feature in the first data, and use the sample discrimination rule to distinguish the positive sample data and negative sample data in the first data feature. Finally, calculate the evaluation weights of different first data features based on the positive sample data and the negative sample data.

[0073] The above-mentioned sample discrimination rules need to be set in combination with the different data types in the first data feature. For example, positive and negative sample data can be distinguished according to whether the airbag on the vehicle has been activated. Record the vehicle with the activated airbag as a negative sample, and the vehicle with the unactivated airbag as a positive sample. The acceleration data can determine positive and negative samples according to the value of the acceleration itself. A positive acceleration value is recorded as a positive sample, and a negative acceleration value is recorded as a negative sample. The first data feature can include features of different data types. For example, it can include features such as acceleration standard deviation feature and speed standard deviation feature.

[0074] Different data features need to be determined through calculations. For example: ① Calculate the speed mean feature and speed standard deviation feature of each vehicle through speed data; ② Calculate the acceleration standard deviation feature and deceleration standard deviation feature of each vehicle by distinguishing the positive and negative signs of the acceleration data; ③ Calculate the positive and negative mean features of the steering angle and the steering angle standard deviation feature of each vehicle through angular velocity and steering wheel angle; ④ Divide the driving speed range of each vin number into three intervals: low, medium, and high speed, and calculate the proportion of the three intervals to form three proportion features: low, medium, and high speed, etc.

[0075] Regarding the aforementioned "calculating the evaluation weights of different first data features based on the positive sample data and the negative sample data", in one possible implementation, before the calculation, the positive sample data and the negative sample data can be binned according to the first data feature. Binning means binning and discretizing the first data feature. To ensure the balance of positive and negative samples in different bins, the equal-frequency binning method is used. Binning and discretization can present data information more concisely and reflect data features more clearly. The reason for using equal-frequency binning is that the WOE value of each bin is calculated based on the ratio of positive and negative samples in each bin. If the number of a certain type of sample is too small in multiple bins, it means that the binning rule is unreasonable. Using equal-frequency binning can avoid such situations.

[0076] By dividing the positive and negative samples as mentioned above, it is possible to more conveniently determine the risk levels of different types of data. The more negative samples, the higher the driving risk can be considered. The more positive samples, the lower the driving risk can be considered, and the corresponding score in the final driving style score is also higher. This division method makes the evaluation of driving style more understandable, convenient, and clear, and can improve the efficiency of driving style evaluation.

[0077] S133: According to the result calculated by the evaluation weight of the first data feature, screen and extract the first data feature in the first data to obtain an initial data set for distinguishing positive samples and negative samples.

[0078] The specific method for calculating the weight (IV) has been described in the previous description and will not be elaborated here.

[0079] A method for obtaining an initial data set provided by an embodiment of the present application can determine the evaluation weights occupied by different data types when evaluating driving styles. According to the obtained weights, the data types that are more critical for evaluation can be obtained, thereby making the evaluation of the user's driving style more targeted and the results more reliable.

[0080] S14: Use the initial data set as input data to train a driving style scoring model, and use the trained driving style scoring model to score the user's driving style.

[0081] Specifically, the above-mentioned initial data set can be divided into a first data set and a second data set. The first data set is used as the training data set of the driving style scoring model, and the second data set is used as the test data set of the driving style scoring model. The initial data set can be divided in a certain proportion, such as dividing it in a ratio of 8:2, taking 80% of the data as the first data set and 20% of the data as the second data set. The specific division method can be determined by those skilled in the art according to the actual situation and application scenarios, and is not limited herein.

[0082] In a possible implementation manner, the first data set can be used as input data to train a driving style scoring model, and the trained driving style scoring model is used to score the driving style of the user in the second data set.

[0083] Input the data of the first data set into the driving style scoring model for model training. After the driving style scoring model is trained, the second data set can be input into the model, and the model scores the driving style of the corresponding user in the second data. The data volume of the first data set is generally larger than that of the second data set because the larger the data volume, the better the training effect of the driving style scoring model. Having more sample data can increase the reference for model calibration, making the training efficiency of the driving style scoring model higher and the accuracy stronger.

[0084] As mentioned in S14, "use the trained driving style scoring model to score the user's driving style". In a possible implementation manner, the driving style scoring model can be a linear regression prediction model. The linear regression prediction model is a classic statistical model, and its application scenario is to predict a continuous numerical variable (dependent variable) based on known variables (independent variables). In the embodiments of the present application, the driving style of the user can be scored and predicted through the input data set.

[0085] The specific implementation method may include: first, obtaining the driving data of the vehicle, then inputting the driving data into the trained linear regression prediction model to obtain a prediction result. Finally, converting the prediction result to obtain the score of the user's driving style.

[0086] When it is necessary to score the driving style of a user (i.e., the driver), it is first necessary to know the driving style of the driver. In the embodiments of the present application, the driving style of the driver is determined by obtaining the driving data of the vehicle driven by the driver. By processing and converting the driving data, the score of the driver's driving style can be finally obtained. Therefore, it is first necessary to obtain the driving data of the vehicle, then input the driving data into the linear regression prediction model, use the model to obtain the prediction result related to the driving style, and convert the prediction result into the score of the driving style, so as to realize the scoring of the driver's driving style.

[0087] The specific method for converting the prediction result into a score can be:

[0088]

[0089]

[0090] A = P 0 -B × ln(θ 0 )

[0091]

[0092] score = A + B × ln(odds)

[0093] In the formula, p represents the probability of the prediction result of the regression model, β n is the feature weight, x 1 , x 2 ,... x n is the WOE value, n is the number of retained feature variables, odds represents the odds ratio, P 0 and PDO are two constant values for adjusting score, controlling the range of the final output score. The higher the score, the safer the driving style of the driver, and vice versa, the more aggressive.

[0094] The above-mentioned "first" and "second" are only used to distinguish two data sets, and do not represent the meaning of sequence, priority or importance, etc.

[0095] In this embodiment, a method for driving style scoring is proposed. The method first obtains historical driving data, then preprocesses the historical driving data to remove invalid data to obtain first data, and extracts features from the first data to obtain an initial data set for distinguishing positive samples and negative samples. Finally, the initial data set is used as input data for training a driving style scoring model, and the trained driving style scoring model is used to score the driving style of the user. In this way, by preprocessing the historical driving data to remove invalid data, the training data of the driving style evaluation model can be all valid data, which ensures the training effect of the driving style evaluation model. Since the redundant interference of invalid data is avoided, the training speed of the model can be improved. By distinguishing the data into positive samples and negative samples, positive sample data and negative sample data can be obtained, and the risk degree of the driving data can be quickly measured, and then the driving style score of the driver can be determined efficiently and accurately.

[0096] Figure 3 The following is a schematic structural diagram of a device for driving style scoring provided by an embodiment of the present application. As Figure 3 shown, a device for driving style scoring specifically includes: an acquisition module 100, a processing module 200, an extraction module 300, and a scoring module 400;

[0097] The acquisition module 100 is configured to acquire historical driving data, where the historical driving data is historical data generated by a user driving a vehicle;

[0098] The processing module 200 is configured to preprocess the historical driving data to remove invalid data to obtain first data, where the invalid data is used to represent data irrelevant to driving style evaluation;

[0099] The extraction module 300 is configured to extract features from the first data to obtain an initial data set;

[0100] The scoring module 400 is configured to use the initial data set as input data for training a driving style scoring model, and use the trained driving style scoring model to score the driving style of the user. In a possible implementation manner, the extraction module 300 is specifically configured to:

[0101] Obtain the first data;

[0102] Based on the first data, calculate and determine the evaluation weights of different first data features, where the first data features are used to represent the data types in the first data, and the evaluation weights are used to represent the importance of the first data features for driving style evaluation;

[0103] Based on the result calculated according to the evaluation weight of the first data feature, screen and extract the first data feature in the first data to obtain an initial data set for distinguishing positive samples and negative samples.

[0104] In a possible implementation manner, the extraction module 300 is specifically configured to:

[0105] Set a sample discrimination rule, which is used to distinguish the first data into positive sample data and negative sample data, and the sample discrimination rule has an associated relationship with the first data feature;

[0106] Determine the sample discrimination rule corresponding to the first data feature in the first data;

[0107] Use the sample discrimination rule to distinguish positive sample data and negative sample data in the first data feature;

[0108] Calculate the evaluation weights of different first data features based on the positive sample data and the negative sample data.

[0109] In a possible implementation manner, the extraction module 300 is specifically configured to:

[0110] Perform binning processing on the positive sample data and the negative sample data according to the first data feature;

[0111] Using the initial data set as input data for training a driving style scoring model includes:

[0112] Calculate the evaluation weights of different first data features based on the binned positive sample data and negative sample data.

[0113] In a possible implementation manner, the processing module 200 is specifically configured to:

[0114] Determine the corresponding category of the data in the historical driving data;

[0115] Set a target value range for the corresponding data based on the category;

[0116] Screen the data in the historical driving data according to the target value range, and remove invalid data to obtain first data, where the invalid data is data whose value does not conform to the target value range.

[0117] In a possible implementation manner, the device is specifically configured to:

[0118] Divide the initial data set into a first data set and a second data set, use the first data set as the training data set of the driving style scoring model, and use the second data set as the test data set of the driving style scoring model;

[0119] The scoring module 400 is specifically configured to:

[0120] Use the first data set as input data to train a driving style scoring model, and use the trained driving style scoring model to score the driving style of the users in the second data set.

[0121] In a possible implementation, the driving style scoring model is a linear regression prediction model. The scoring module 400 is specifically configured to:

[0122] Obtain the driving data of the vehicle;

[0123] Input the driving data into the trained linear regression prediction model to obtain a prediction result;

[0124] Convert the prediction result to obtain the score of the driving style of the user.

[0125] In this embodiment, a device for scoring driving style is proposed. The device includes an acquisition module, a processing module, an extraction module, and a scoring module. The acquisition module is used to acquire historical driving data, and the historical driving data is the historical data generated by a user driving a vehicle; the processing module is used to preprocess the historical driving data to remove invalid data to obtain first data, and the invalid data is used to represent data irrelevant to driving style evaluation; the extraction module is used to extract features from the first data to obtain an initial data set; the scoring module is used to use the initial data set as input data to train a driving style scoring model, and use the trained driving style scoring model to score the driving style of the user. In this way, by first using the acquisition module to acquire historical driving data, then using the processing module to preprocess the historical driving data to remove invalid data to obtain first data, and using the extraction module to extract features from the first data to obtain an initial data set for distinguishing positive samples and negative samples. Finally, use the scoring module to use the initial data set as input data to train a driving style scoring model, and use the trained driving style scoring model to score the driving style of the user. In this way, by preprocessing the historical driving data to remove invalid data, it can be ensured that the training data of the driving style evaluation model are all valid data, which guarantees the training effect of the driving style evaluation model, and can improve the model training speed because the redundant interference of invalid data is avoided. By distinguishing the data into positive samples and negative samples to obtain positive sample data and negative sample data, the risk degree of the driving data can be quickly measured, and then the driving style score of the driver can be determined efficiently and accurately.

[0126] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of devices and methods according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0127] Embodiments of the present application also provide corresponding devices and computer-readable storage media for implementing the solutions provided by the embodiments of the present application.

[0128] Among them, the device includes a memory and a processor. The memory is used to store instructions or code, and the processor is used to execute the instructions or code so that the device performs a method for driving style scoring according to any embodiment of the present application.

[0129] In practical applications, the computer-readable storage medium may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0130] A computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program code is carried. Such a propagated data signal may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0131] The program code contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0132] The computer program code for performing the operations of the present invention may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, execute as a stand-alone software package, partly on the user's computer and partly on a remote computer, or execute entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0133] It should also be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including", or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such a process, method, article, or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0134] As described above, it is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed by the present application should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for driving style scoring, characterized in that, the method includes: Obtain historical driving data, where the historical driving data is historical data generated by a user driving a vehicle; Preprocess the historical driving data to remove invalid data to obtain first data, where the invalid data is used to represent data unrelated to driving style evaluation; Extract features from the first data to obtain an initial data set for distinguishing positive samples and negative samples; Use the initial data set as input data to train a driving style scoring model, and use the trained driving style scoring model to score the driving style of the user.

2. The method according to claim 1, characterized in that, The extracting features from the first data to obtain an initial data set for distinguishing positive samples and negative samples includes: Obtain the first data; Based on the first data, calculate and determine the evaluation weights of different first data features, where the first data features are used to represent the data types in the first data, and the evaluation weights are used to represent the importance of the first data features for driving style evaluation; According to the calculation results of the evaluation weights of the first data features, screen and extract the first data features in the first data to obtain an initial data set for distinguishing positive samples and negative samples.

3. The method according to claim 2, characterized in that, The calculating and determining the evaluation weights of different first data features based on the first data includes: Set a sample discrimination rule, where the sample discrimination rule is used to distinguish the first data into positive sample data and negative sample data, and the sample discrimination rule has an association relationship with the first data features; Determine the sample discrimination rule corresponding to the first data features in the first data; Use the sample discrimination rule to distinguish the positive sample data and negative sample data in the first data features; Based on the positive sample data and the negative sample data, calculate the evaluation weights of different first data features.

4. The method according to claim 3, characterized in that, The calculating the evaluation weights of different first data features based on the positive sample data and the negative sample data includes: Perform binning processing on the positive sample data and the negative sample data according to the first data features; The using the initial data set as input data to train a driving style scoring model includes: Based on the binned positive sample data and negative sample data, calculate the evaluation weights of different first data features.

5. The method according to claim 1, characterized in that, The preprocessing the historical driving data to remove invalid data to obtain first data includes: Determine the corresponding categories of the data in the historical driving data; Based on the categories, set the target value ranges of the corresponding data; Filter the data in the historical driving data according to the target value ranges to remove invalid data to obtain first data, where the invalid data is data whose values do not conform to the target value ranges.

6. The method according to claim 1, characterized in that, The method further includes: Divide the initial data set into a first data set and a second data set, use the first data set as the training data set of the driving style scoring model, and use the second data set as the test data set of the driving style scoring model; Training the driving style scoring model with the initial data set as input data, and scoring the driving style of the user by using the trained driving style scoring model, includes: Train the driving style scoring model with the first data set as input data, and score the driving style of the users in the second data set by using the trained driving style scoring model.

7. The method according to claim 1, wherein, The driving style scoring model is a linear regression prediction model. Scoring the driving style of the user by using the trained driving style scoring model includes: Obtain the driving data of the vehicle; Input the driving data into the trained linear regression prediction model to obtain a prediction result; Convert the prediction result to obtain the score of the driving style of the user.

8. An apparatus for driving style scoring, wherein, The apparatus includes: an acquisition module, a processing module, an extraction module and a scoring module; The acquisition module is configured to acquire historical driving data, where the historical driving data is historical data generated by a user driving a vehicle; The processing module is configured to preprocess the historical driving data to remove invalid data to obtain first data, where the invalid data is used to represent data irrelevant to driving style evaluation; The extraction module is configured to extract features from the first data to obtain an initial data set; The scoring module is configured to train the driving style scoring model with the initial data set as input data, and score the driving style of the user by using the trained driving style scoring model.

9. An electronic device, wherein, The device includes: a processor, a memory, and a system bus; The processor and the memory are connected through the system bus; The memory is used to store one or more programs, where the one or more programs include instructions, and when the instructions are executed by the processor, the processor executes the driving style scoring method according to any one of claims 1-7.

10. A computer-readable storage medium, wherein, An implementation program for implementing the driving style scoring method is stored on the computer-readable storage medium, and when the implementation program for implementing the driving style scoring method is executed by a processor, the steps of the method according to any one of claims 1-7 are implemented.

Images