CN110599353A

CN110599353A - Vehicle insurance and claims rate prediction method, device, equipment and medium

Info

Publication number: CN110599353A
Application number: CN201810608430.XA
Authority: CN
Inventors: 葛婷婷; 甘勋; 崔勇
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Baidu Online Network Technology Beijing Co Ltd; Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2018-06-13
Filing date: 2018-06-13
Publication date: 2019-12-20

Abstract

The embodiment of the invention discloses a method, a device, equipment and a medium for predicting vehicle insurance and odds ratio, wherein the method comprises the following steps: acquiring navigation track data of a target vehicle, wherein the navigation track data is obtained based on navigation software installed on a mobile terminal; extracting driving behavior characteristic data corresponding to the target vehicle according to the navigation track data; based on the driving behavior characteristic data, calling a pre-trained risk probability prediction model and an odds rate prediction model to obtain a risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model, and extracting driving behavior characteristic data from navigation track data recorded in navigation software for predicting vehicle risk and odds rate, so that the accuracy of predicting the risk condition and the odds rate of the vehicle is improved, and the map data assets are realized.

Description

Vehicle insurance and claims rate prediction method, device, equipment and medium

Technical Field

The embodiment of the invention relates to the technical field of data processing, in particular to a method, a device, equipment and a medium for predicting vehicle insurance and claims rate.

Background

As the vehicle holding capacity increases year by year, the vehicle insurance traffic of each insurance company also increases. Vehicle insurance surveys show that only 17.5% of vehicle owners in a vehicle insurance applicant frequently have road safety accidents, while 82.5% of vehicle owners driving safely buy bills for 17.5% of the vehicle owners by assuming a premium. Therefore, how to accurately evaluate the insurance risk of the insured vehicle to formulate the vehicle insurance services such as reasonable vehicle insurance acceptance, pricing and service items is a technical problem to be solved urgently.

In the prior art, when calculating the vehicle insurance premium, the vehicle insurance situation and the claim data need to be predicted to calculate a reasonable vehicle insurance premium, for example, if the probability of predicting the vehicle insurance is higher and the claim amount is higher, the calculated vehicle insurance premium should be higher.

The vehicle insurance situation and the claim data can be predicted according to the attribute data (such as the brand, the model, the purchasing age and the like) of the vehicle and the historical claim data. The method has the defects that on one hand, the attributes of domestic vehicle brands, models, vehicle purchasing years and the like are similar, on the other hand, the vehicle owner does not report the vehicle at risk, the vehicle owner reports the vehicle at a malicious rate and the like, and the historical data has certain sporadic nature, so that the risk condition and the claim data of the vehicle in the next year cannot be accurately predicted. Vehicle insurance and claims data may also be predicted based on person attribute data such as insurance applicant's age, marital status, driving age, family members, number of vehicles in possession, educational programs, occupation, and residence. The disadvantage is that these data are not diverse enough and accurate predictions of the individual's venture behavior and claims data cannot be made.

Disclosure of Invention

The embodiment of the invention provides a method, a device, equipment and a medium for predicting vehicle insurance leaving and odds ratio, which aim to improve the accuracy of predicting the insurance leaving situation and the odds ratio of a vehicle.

In a first aspect, an embodiment of the present invention provides a vehicle insurance and odds ratio prediction method, including:

acquiring navigation track data of a target vehicle, wherein the navigation track data is obtained based on navigation software installed on a mobile terminal;

extracting driving behavior characteristic data corresponding to the target vehicle according to the navigation track data;

and calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data to obtain a risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model.

In a second aspect, an embodiment of the present invention further provides a vehicle insurance and dividend rate prediction apparatus, where the apparatus includes:

the track data acquisition module is used for acquiring navigation track data of a target vehicle, wherein the navigation track data is obtained based on navigation software installed on the mobile terminal;

the driving behavior acquisition module is used for extracting driving behavior characteristic data corresponding to the target vehicle according to the navigation track data;

and the result prediction module is used for calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle output by the odds rate prediction model.

In a third aspect, an embodiment of the present invention further provides a computer device, where the computer device includes:

one or more processors;

storage means for storing one or more programs;

when executed by the one or more processors, cause the one or more processors to implement a vehicle venture and odds prediction method as provided by any embodiment of the invention.

In a fourth aspect, embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements a vehicle risk and odds prediction method according to any of the embodiments of the present invention.

The embodiment of the invention collects the navigation track data of a target vehicle based on navigation software installed on a mobile terminal, and extracts the driving behavior characteristic data corresponding to the target vehicle according to the navigation track data; based on the driving behavior characteristic data, calling a pre-trained risk probability prediction model and an odds rate prediction model to obtain a risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model, and extracting driving behavior characteristic data from navigation track data recorded in navigation software for predicting vehicle risk and odds rate, so that the accuracy of predicting the risk condition and the odds rate of the vehicle is improved, and the map data assets are realized.

Drawings

FIG. 1 is a flow chart of a vehicle insurance and dividend rate prediction method according to an embodiment of the present invention;

FIG. 2 is a flowchart of a vehicle insurance and dividend rate prediction method according to a second embodiment of the present invention;

FIG. 3 is a flowchart of a vehicle insurance and dividend rate prediction method according to a third embodiment of the present invention;

FIG. 4 is a flowchart of a vehicle insurance and dividend rate prediction method according to a fourth embodiment of the present invention;

FIG. 5 is a flowchart of a vehicle insurance and dividend rate prediction method according to a fifth embodiment of the present invention;

FIG. 6 is a flowchart of a vehicle insurance and dividend rate prediction method according to a sixth embodiment of the present invention;

fig. 7 is a flowchart of a vehicle insurance and dividend rate prediction method according to a seventh embodiment of the present invention;

fig. 8 is a flowchart of a vehicle insurance and dividend rate prediction method according to an eighth embodiment of the present invention;

FIG. 9 is a flowchart of a vehicle insurance and dividend rate prediction method according to a ninth embodiment of the present invention;

fig. 10 is a flowchart of a vehicle insurance and dividend rate prediction method according to a tenth embodiment of the present invention;

fig. 11 is a schematic structural diagram of a vehicle risk occurrence and odds ratio prediction apparatus according to an eleventh embodiment of the present invention;

fig. 12 is a schematic structural diagram of a computer device in a twelfth embodiment of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.

Example one

Fig. 1 is a flowchart of a vehicle insurance and dividend rate prediction method according to an embodiment of the present invention, which can be applied to a situation when insurance and dividend rate prediction is performed on a vehicle. The method may be performed by a vehicle insurance and odds prediction apparatus, which may be implemented in software and/or hardware, for example, which may be configured in a computer device. As shown in fig. 1, the method specifically includes:

s110, acquiring navigation track data of the target vehicle, wherein the navigation track data is obtained based on navigation software installed on the mobile terminal.

In the embodiment, the navigation software installed on the mobile terminal is used as a data source for predicting the vehicle insurance and the payout rate of the target vehicle. With the development of navigation software technology, many navigation software can plan various paths for a user according to current road condition information when planning the paths for the user according to the initial address of the user, such as paths with low traffic jam risk, paths with less traffic lights, and the like. Therefore, most users rely on route planning on navigation software when driving to travel. That is, most users install navigation software on the mobile terminal and use the navigation software when traveling, that is, most of traveling information of the vehicle is recorded in the navigation software. In the process of navigating by using navigation software, a user needs to use a Global Positioning System (GPS), an accelerometer and other sensors to collect Positioning information of the user, the data is inherent data serving the navigation software, and the user uses the navigation software to be inherent behavior of the user without additionally downloading applications. Therefore, the navigation track data in the navigation software installed on the mobile terminal is used as a data source when the vehicle insurance and the pay rate of the target vehicle are predicted, a user does not need to install specific equipment on the vehicle, or install special software on the smart phone to collect vehicle driving data of the vehicle, the distribution problem is solved, the problem that the data volume is insufficient due to insufficient distribution of the specific equipment or the special software is avoided, and the map data assets are changed.

Optionally, the mobile terminal may be a smart phone, a tablet computer, or a vehicle-mounted computer. The navigation software installed on the mobile terminal may be at least one of navigation software installed on a user smart phone, navigation software installed on a user tablet computer, and navigation software installed on a target vehicle-mounted computer. When the vehicle insurance and the claim rate of the target vehicle need to be predicted, at least one of navigation track data in navigation software installed on a user smart phone, a tablet personal computer and a vehicle-mounted computer can be acquired as the navigation track data of the target vehicle. In consideration of the fact that a user may be located on another vehicle when using a smart phone or a tablet computer for navigation, the navigation track information recorded by the navigation software in the smart phone or the tablet computer of the user may include track information other than the navigation track information of the target vehicle. Optionally, the navigation trajectory data in the navigation software installed on the vehicle-mounted computer, that is, the navigation trajectory data in the vehicle-mounted navigation system, is used as the navigation trajectory data of the target vehicle. The navigation track data in the vehicle-mounted navigation system is generally the navigation track data of the target vehicle, and the accuracy of the data can be ensured by using the navigation track data in the vehicle-mounted navigation system as the navigation track data of the target vehicle. In the present embodiment, the navigation trajectory data includes the travel time, the travel route, the travel speed, and the like of the vehicle corresponding to each navigation behavior of the target vehicle.

Optionally, an expiration date or a time range may be set when the navigation track data is collected from the navigation software, and the navigation track data in a preset range, such as the navigation track data in the range from the expiration date to the current date, is obtained according to the expiration date or the time range. For example, if the preset time range is one year, navigation track data recorded in the navigation software is acquired within one year based on the current date when the navigation track data is acquired. Optionally, different time ranges may be set according to different requirements. For example, if the accuracy of the data is higher, a longer time range may be set; a shorter time range may be set in order to save computation time.

And S120, extracting driving behavior characteristic data corresponding to the target vehicle according to the navigation track data.

In the embodiment, the historical insurance policy data of the vehicle is not taken as the standard for judging the premium of the vehicle, but the driving behavior characteristic data corresponding to the target vehicle is taken as the standard for judging the premium of the vehicle, and the premium of the vehicle is determined based on the daily driving characteristics of the target vehicle, so that the calculation of the premium of the vehicle is more accurate and reasonable.

Optionally, the driving behavior feature data includes: at least one of sharp acceleration data, sharp deceleration data, sharp turn data, overspeed data, mileage data, and speed per hour data.

Optionally, each data included in the navigation track data may be extracted from the navigation track data by means of keyword extraction, such as speed per hour data, mileage data, overspeed data, real-time vehicle traveling speed, position, and the like of the current vehicle navigation, and then data such as rapid acceleration data, rapid deceleration data, or rapid turning data of the vehicle may be calculated according to a preset algorithm, and a speed per hour standard deviation of the vehicle may also be calculated according to the speed per hour data of the vehicle, where the speed per hour standard deviation represents a change condition of the vehicle speed, and is compared with a speed limit of a road on which the vehicle travels, and the speed of the vehicle is determined according to a comparison result.

It should be noted that the navigation track data recorded by different navigation software is different. For example, some navigation software directly records the rapid acceleration data, the rapid deceleration data, the speed data and the like of the vehicle in a single navigation behavior, but some navigation software does not record the data. Therefore, the driving behavior characteristic data of the vehicle can be directly extracted from the navigation track data, and can also be calculated according to the data extracted from the vehicle track data.

S130, calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data to obtain a risk probability prediction result output by the risk probability prediction model and an odds rate prediction result output by the odds rate prediction model.

In this embodiment, the driving behavior feature data is used as an input parameter of the risk probability prediction model and/or the odds ratio prediction model, and is respectively input into the risk probability prediction model and the odds ratio prediction model trained in advance, so as to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds ratio prediction result corresponding to the target vehicle output by the odds ratio prediction model.

Optionally, the risk probability prediction result and the odds ratio prediction result are represented in a score form, and the larger the score is, the larger the corresponding risk probability and the corresponding odds ratio are. In this embodiment, a score range for outputting the score may be preset, and the prediction result within the score range is output, where the larger the score of the prediction result is, the larger the corresponding risk probability and odds ratio are. For example, the output score range is set to (0, 20), and if the prediction result of the probability of occurrence output by the target vehicle a is 10, the prediction result of the odds ratio is 11, the prediction result of the probability of occurrence output by the target vehicle B is 8, and the prediction result of the odds ratio is 7, the probability of occurrence of the target vehicle a is higher than the probability of occurrence of the target vehicle B, and the odds ratio of the target vehicle a is higher than the odds ratio of the target vehicle B. Accordingly, when determining the premiums of the target vehicle a and the target vehicle B, the premiums of the target vehicle a are also higher than those of the target vehicle B.

Optionally, the risk probability prediction result and the odds ratio prediction result are expressed in a grade form, and the higher the grade is, the higher the corresponding risk probability and the corresponding odds ratio are. In the present embodiment, the level of the output level may be set in advance. For example, the output levels are set to three levels, i.e., low, medium, and high, and if the risk probability prediction result and the odds prediction result output by the target vehicle a are both high and the risk probability prediction result and the odds prediction result output by the target vehicle B are both medium, the risk probability and the odds of the target vehicle a are both higher than those of the target vehicle B, and accordingly, when the insurance premiums of the target vehicle a and the target vehicle B are determined, the insurance premiums of the target vehicle a are also higher than those of the target vehicle B.

Preferably, the risk probability prediction result and the odds prediction result are expressed in the form of scores. The prediction result of the risk probability and the prediction result of the odds and the payoff rate are expressed in a score form, so that the prediction result is more accurate, and the actuarial pricing of the vehicle insurance is better assisted.

Example two

Fig. 2 is a flowchart of a vehicle insurance and dividend rate prediction method according to a second embodiment of the present invention, which is further optimized based on the above embodiments. As shown in fig. 2, the method includes:

s210, collecting navigation track data of the target vehicle, wherein the navigation track data is obtained based on navigation software installed on the mobile terminal.

And S220, extracting driving behavior characteristic data corresponding to the target vehicle according to the navigation track data.

And S230, collecting at least one of map data, weather data and user reported data.

Generally, driving behaviors (such as rapid acceleration, rapid deceleration, rapid turning, overspeed and other behaviors) of vehicle owners can be completely different under different environments, for example, under the conditions of road condition congestion and dangerous terrain, the driving behavior of the same vehicle owner is much worse on a large probability than that of the same vehicle owner on a smooth and gentle road. The accident is more likely to happen on congested road sections with severe terrain. Therefore, the ambient environment factors when the owner drives are one of the criteria for judging the premium of the vehicle. Optionally, the environmental factors include road grade, road shape, speed limit grade, weather conditions, driving time interval, surrounding environment (such as whether there is school, whether animals often appear or disappear, etc.), and the like.

Optionally, the map data may reflect the road shape, speed limit level, and other conditions of the user during driving, for example, when the driving road segment is displayed in the map data as a mountain road segment, the map data indicates that the road segment terrain is severe during driving; the weather data can reflect the weather condition of the user during driving, for example, the weather data shows that the environment is severe during driving when rainstorm exists; the user reported data can reflect the surrounding environment of the user when driving, for example, when the user reported data shows that animals often appear and disappear on the road section, or when the user reported the driving road section and has temporary management measures, the user reported the surrounding environment is not good when driving. Based on the method, at least one of map data, weather data and user reported data is collected so as to extract environmental characteristic data when the user drives.

Optionally, the map data includes: road network data and road condition data; the road network data comprises at least one of road information, geographical position information, traffic identification information and regional information; the road condition data includes: at least one of congestion data and average speed of a road on which the navigation is performed; the data reported by the user comprises: at least one of road congestion data, road temporary management measure data, and traffic event data.

And S240, extracting the environmental characteristic data corresponding to the navigation track data according to at least one of the map data, the weather data and the user reported data.

Optionally, the environment characteristic data includes: at least one of weather characteristics, road condition characteristics, time characteristics and surrounding environment characteristics. The road characteristics may include characteristics such as road grade, road form, speed limit grade, and the like.

Optionally, a certain parameter range may be set for each feature, and the size of the parameter indicates the degree of influence of the feature on the driving behavior. For example, it may be set that the larger the parameter, the greater the degree of influence on the driving behavior. Taking the weather characteristic as an example, the parameter range of the weather characteristic may be set to (0, 5) in advance. If the weather is good (such as cloudy, sunny, etc.), the weather characteristic is set to 0, if the weather is light rain, the weather characteristic is set to 1, and if the weather is heavy rain, the weather characteristic is set to 4.

And S250, calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data and the environment characteristic data to obtain a risk probability prediction result output by the risk probability prediction model and an odds rate prediction result output by the odds rate prediction model.

In this embodiment, the driving behavior characteristic data and the environmental characteristic data are used as input parameters of the risk probability prediction model and/or the odds ratio prediction model, and are respectively input into the risk probability prediction model and the odds ratio prediction model trained in advance, so as to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds ratio prediction result corresponding to the target vehicle output by the odds ratio prediction model.

According to the technical scheme of the embodiment of the invention, on the basis of the embodiment, the environmental characteristic data and the driving behavior characteristic data are jointly used as the prediction parameters of the vehicle risk probability and the loss rate, the environmental characteristic data corresponding to the navigation track data are extracted according to at least one of the map data, the weather data and the user reported data by collecting at least one of the map data, the weather data and the user reported data, and the vehicle risk probability and the loss rate are predicted on the basis of the environmental characteristic data and the driving behavior characteristic data, so that the prediction results of the vehicle risk probability and the loss rate are more accurate, and the vehicle insurance acceptance and the pricing are more reasonable.

EXAMPLE III

Fig. 3 is a flowchart of a vehicle insurance and dividend rate prediction method according to a third embodiment of the present invention, which is further optimized based on the above embodiments. As shown in fig. 3, the method includes:

s310, collecting navigation track data of the target vehicle, wherein the navigation track data is obtained based on navigation software installed on the mobile terminal.

And S320, extracting driving behavior characteristic data corresponding to the target vehicle according to the navigation track data.

And S330, acquiring vehicle use characteristic data of the target vehicle.

Generally, the emergency behavior of vehicles for different purposes can vary greatly. For example, the probability of occurrence will be different for long-distance vehicles and for commuters on and off duty. The accident-prone vehicle is reflected in the danger, and the accident is easier to happen to the long-distance vehicle compared with the vehicles for riding on and off duty. Therefore, the vehicle usage when the owner is driving is one of the criteria for judging the premium of the vehicle.

Optionally, the vehicle use characteristic data is mined by a machine learning method based on the trajectory data, the navigation destination and the navigation time in the navigation trajectory data. For example, a long-distance vehicle has a long travel time, a corresponding navigation time is also long, and the distance between the navigation destination and the navigation start position is also long. Optionally, the vehicle use characteristic data extraction model may be trained according to the navigation track data of a plurality of vehicles, the vehicle use and a preset machine learning algorithm. The machine learning algorithm is a linear regression algorithm, a neural network algorithm, a classification algorithm or a decision tree algorithm. When the vehicle use characteristic data of the target vehicle needs to be acquired, the track data, the navigation destination and the navigation time of the target vehicle are used as input, a trained vehicle use characteristic extraction model is called, and vehicle use characteristics corresponding to the navigation track data output by the vehicle use characteristic extraction model are acquired. The vehicle use characteristics are extracted by using a machine learning algorithm, so that the vehicle use characteristics corresponding to the navigation track data are more accurate, and further the prediction results of the risk and the odds ratio of the vehicle are more accurate.

And S340, calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data and the vehicle use characteristic data to obtain a risk probability prediction result output by the risk probability prediction model and an odds rate prediction result output by the odds rate prediction model.

In this embodiment, the driving behavior characteristic data and the vehicle usage characteristic data are used as input parameters of the risk probability prediction model and/or the odds ratio prediction model, and are respectively input into the risk probability prediction model and the odds ratio prediction model trained in advance, so as to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds ratio prediction result corresponding to the target vehicle output by the odds ratio prediction model.

According to the technical scheme of the embodiment of the invention, on the basis of the embodiment, the vehicle use characteristic data and the driving behavior characteristic data are jointly used as the prediction parameters of the vehicle risk probability and the risk loss rate, and the vehicle risk probability and the risk loss rate are predicted on the basis of the vehicle use characteristic data and the driving behavior characteristic data by acquiring the vehicle use characteristic data of the target vehicle, so that the prediction results of the vehicle risk probability and the risk loss rate are more accurate, and the vehicle insurance acceptance and pricing are more reasonable.

Example four

Fig. 4 is a flowchart of a vehicle insurance and dividend rate prediction method according to a fourth embodiment of the present invention, which is further optimized based on the foregoing embodiments. As shown in fig. 4, the method includes:

s410, acquiring navigation track data of the target vehicle, wherein the navigation track data is obtained based on navigation software installed on the mobile terminal.

And S420, extracting driving behavior characteristic data corresponding to the target vehicle according to the navigation track data.

And S430, acquiring driving behavior distinguishing characteristic data of the target vehicle and the surrounding vehicles according to the navigation track data of the target vehicle and the navigation track data of the surrounding vehicles.

In general, different drivers may have different driving behaviors under the same condition. In the present embodiment, the driving behavior of the target vehicle and the surrounding vehicle in the same situation is distinguished as one of the criteria for judging the premium level of the target vehicle.

Optionally, the navigation software stores navigation track data of all users using the navigation software for navigation. And extracting navigation track data of surrounding vehicles on the same driving road at the same driving time point as the target vehicle from the navigation software, and comparing the navigation track data of the target vehicle with the navigation track data of the surrounding vehicles to determine the driving behavior distinguishing characteristic data of the target vehicle and the surrounding vehicles.

In the present embodiment, the driving behavior distinction feature data from the nearby vehicle includes: driving behavior distinguishing features and speed distinguishing features on the same road and at the same point in time. Wherein the driving behavior distinguishing characteristics comprise distinguishing characteristics of sharp acceleration, sharp deceleration, sharp turning and/or overspeed. The speed distinguishing characteristic is whether the target vehicle is driven at a speed significantly lower or higher than the surrounding vehicles. Alternatively, the driving behavior distinguishing characteristic data of the target vehicle may be represented by a parameter. The larger the parameter value is, the larger the difference between the driving behaviors of the target vehicle and the surrounding vehicle is, which means that the current driving behavior of the driver of the target vehicle is significantly different from that of the surrounding vehicle, and the risk probability of the target vehicle is likely to be higher than that of the surrounding vehicle.

And S440, calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data and the driving behavior distinguishing characteristic data to obtain a risk probability prediction result output by the risk probability prediction model and an odds rate prediction result output by the odds rate prediction model.

In this embodiment, the driving behavior feature data and the driving behavior distinguishing feature data of the surrounding vehicle are used as input parameters of the risk probability prediction model and/or the odds prediction model, and are input into the risk probability prediction model and the odds prediction model which are trained in advance, respectively, so as to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds prediction result corresponding to the target vehicle output by the odds prediction model.

According to the technical scheme of the embodiment of the invention, on the basis of the embodiment, the driving behavior distinguishing characteristic data and the driving behavior characteristic data are jointly used as the prediction parameters of the vehicle risk probability and the risk loss rate, the driving behavior distinguishing characteristic data of the target vehicle and the surrounding vehicles are obtained according to the navigation track data of the target vehicle and the navigation track data of the surrounding vehicles, and the vehicle risk probability and the risk loss rate are predicted on the basis of the driving behavior distinguishing characteristic data and the driving behavior characteristic data, so that the prediction results of the vehicle risk probability and the risk loss rate are more accurate, and further the vehicle insurance acceptance and pricing are more reasonable.

EXAMPLE five

Fig. 5 is a flowchart of a vehicle insurance and dividend rate prediction method according to a fifth embodiment of the present invention, which is further optimized based on the foregoing embodiments. As shown in fig. 5, the method includes:

and S510, acquiring navigation track data of the target vehicle, wherein the navigation track data is obtained based on navigation software installed on the mobile terminal.

And S520, extracting driving behavior characteristic data corresponding to the target vehicle according to the navigation track data.

S530, acquiring other characteristic data, wherein the other characteristic data comprises at least one of violation data, user attribute data and frequent place classification data corresponding to the target vehicle.

In this embodiment, the other characteristic data includes at least one of violation data, user attribute data, and frequent location classification data corresponding to the target vehicle. Optionally, the violation data of the target vehicle can reflect the daily driving habits of the driver of the target vehicle. For example, if the number of violations of the target vehicle is large, the probability of a traffic accident occurring in the vehicle may be large. The user attribute can embody the reaction sensitivity or driving experience of the user. The frequent classification can embody the degree of safety of the driving environment when the target vehicle is driven daily.

Wherein the frequented classification is obtained based on navigation trajectory data of the target vehicle; the frequented classifications include: home, corporate, etc. The user attribute data is obtained based on the network retrieval behavior and the access log of the corresponding user; the user attribute data includes: sex, age, etc.

Optionally, whether the target vehicle violates regulations may be determined by comparing the driving behavior data of the target vehicle with pre-stored standard driving behavior data. For example, if the current road speed limit is 80KM/h and the traveling speed of the target vehicle is 90KM/h, it is determined that the target vehicle is overspeed, i.e. the target vehicle is violated. And the violation data can be obtained based on the violation record of the target vehicle in the traffic management bureau. The attribute of the user can be judged by combining the personal information of the user and the navigation track data, or the attribute of the user can be determined according to the network retrieval behavior and the access log of the corresponding user. For example, if the user routinely searches for places such as beauty salons and shopping malls, it can be determined that the user is a woman. The frequented region of the target vehicle can be determined according to the navigation track data of the user, and the frequented region classification data of the target vehicle can be determined according to the frequented region classification.

And S540, calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data and the other characteristic data to obtain a risk probability prediction result output by the risk probability prediction model and an odds rate prediction result output by the odds rate prediction model.

In this embodiment, the other feature data and the vehicle usage feature data are used as input parameters of the risk probability prediction model and/or the odds prediction model, and are respectively input into the risk probability prediction model and the odds prediction model trained in advance, so as to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds prediction result corresponding to the target vehicle output by the odds prediction model.

According to the technical scheme of the embodiment of the invention, on the basis of the embodiment, other characteristic data and driving behavior characteristic data are jointly used as prediction parameters of the vehicle risk probability and the claim rate, at least one of violation data, user attribute data and frequent classification data corresponding to the target vehicle is obtained as other characteristic data of the target vehicle, and the vehicle risk probability and the claim rate are predicted on the basis of the other characteristic data and the driving behavior characteristic data, so that the prediction results of the vehicle risk probability and the claim rate are more accurate, and further the vehicle insurance acceptance and pricing are more reasonable.

It should be noted that the vehicle insurance and reimbursement rate prediction method provided by any embodiment of the present invention can be randomly combined for use, and is not limited herein. For example, the driving behavior feature data, the environmental feature data, and the vehicle use feature data are collectively used as prediction parameters of vehicle risk and loss rate, or the driving behavior feature data, the vehicle use feature data, the driving behavior distinguishing feature data, and other feature data are collectively used as prediction parameters of vehicle risk and loss rate.

EXAMPLE six

Fig. 6 is a flowchart of a vehicle insurance and dividend rate prediction method according to a sixth embodiment of the present invention, which is further optimized based on the above embodiments. As shown in fig. 6, the method includes:

s610, obtaining historical navigation track data of a plurality of vehicles and insurance data and odds rate data corresponding to the historical navigation track data.

In this embodiment, the navigation software stores navigation track data of all users using the navigation software for navigation, and optionally, the navigation software installed on the mobile terminal may be used as a training data source for vehicle insurance and pay rate of the target vehicle, or historical navigation track data of a plurality of vehicles may be directly obtained from a background database of the navigation software.

Optionally, an expiration date or a time range may be set when the navigation track data is collected from the navigation software, and the navigation track data within a preset range is obtained according to the expiration date or the time range. Alternatively, different expiration dates or time ranges may be set according to different requirements.

And S620, extracting historical driving behavior characteristic data of the corresponding vehicle according to the historical navigation track data.

In the embodiment, the historical driving behavior characteristic data corresponding to the vehicle is used as the input parameters for training the risk probability prediction model and the odds ratio prediction model. Optionally, the historical driving behavior feature data includes: at least one of historical hard acceleration data, historical hard deceleration data, historical hard turn data, historical overspeed data, historical mileage data, and historical speed-per-hour data.

Optionally, each data included in the historical navigation track data may be extracted from the historical navigation track data by means of keyword extraction, and a historical speed-per-hour standard deviation of the vehicle may be calculated according to the historical speed-per-hour data of the vehicle, where the historical speed-per-hour standard deviation represents a change condition of the vehicle speed, and is compared with the speed limit of the road on which the vehicle is traveling, and the speed of the vehicle is determined according to the comparison result.

The manner of obtaining the historical navigation tracks of the multiple vehicles and extracting the historical driving behavior features from the historical navigation tracks is similar to the manner corresponding to the target vehicle in the above embodiments, and for more detailed contents, reference may be made to the above embodiments, which are not described herein again.

S630, based on the historical driving behavior characteristic data, the insurance data and the dividend rate data, training by adopting a machine learning algorithm to obtain an insurance probability prediction model and an dividend rate prediction model.

Optionally, the historical driving behavior feature data, the risk data and the odds ratio data are used as training parameters of the risk probability prediction model and the odds ratio prediction model, and the pre-established machine learning model is trained to obtain the trained risk probability prediction model and the trained odds ratio prediction model.

Optionally, the historical driving behavior feature data and the risk data are used as training parameters of the risk probability prediction model, and the trained risk probability prediction model is obtained through training. And taking the historical driving behavior characteristic data and the odds ratio data as training parameters of the risk probability prediction model, and training to obtain a trained odds ratio prediction model. And respectively selecting a proper machine learning algorithm according to the characteristics of the risk probability prediction and the odds ratio prediction to establish a proper machine learning model. That is, the machine learning algorithms used to train the risk probability prediction model and the odds prediction model may be the same or different.

Optionally, the machine learning algorithm is a linear regression algorithm, a neural network algorithm, a classification algorithm, or a decision tree algorithm. The corresponding machine learning model is a linear regression model, a neural network model, a classification model or a decision tree model.

And S640, acquiring navigation track data of the target vehicle, wherein the navigation track data is obtained based on navigation software installed on the mobile terminal.

And S650, extracting the driving behavior characteristic data corresponding to the target vehicle according to the navigation track data.

And S660, calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data to obtain a risk probability prediction result output by the risk probability prediction model and an odds rate prediction result output by the odds rate prediction model.

It should be noted that the training methods of the risk probability prediction model and the odds prediction model provided by the embodiments of the present invention may be performed separately. That is, the training of the risk probability prediction model and the odds prediction model can be completed by using the operation steps in S610-S630 provided in the embodiment of the present invention alone, and the prediction of the risk probability and the odds of the target vehicle in the subsequent steps S640-S660 is not performed.

According to the technical scheme of the embodiment of the invention, the operation of training the risk probability prediction model and the odds ratio prediction model is added on the basis of the embodiment, and historical navigation track data of a plurality of vehicles and risk data and odds ratio data corresponding to the historical navigation track data are obtained; extracting historical driving behavior characteristic data of a corresponding vehicle according to historical navigation track data; based on historical driving behavior characteristic data, insurance data and odds ratio data, a machine learning algorithm is adopted for training to obtain an insurance probability prediction model and an odds ratio prediction model, and the insurance probability prediction model and the odds ratio prediction model are trained according to the historical driving behavior characteristics of a plurality of vehicles, so that the insurance probability prediction model and the odds ratio prediction model are more accurate and reasonable, prediction results obtained through the insurance probability prediction model and the odds ratio prediction model are more accurate, and map data asset achievement is achieved.

EXAMPLE seven

Fig. 7 is a flowchart of a vehicle risk occurrence and odds ratio prediction method according to a seventh embodiment of the present invention, which further optimizes the training method of the risk occurrence probability prediction model and the odds ratio prediction model based on a sixth embodiment of the present invention. As shown in fig. 7, the method includes:

s710, obtaining historical navigation track data of a plurality of vehicles and corresponding risk data and odds rate data of the historical navigation track data.

And S720, extracting historical driving behavior characteristic data of the corresponding vehicle according to the historical navigation track data.

And S730, acquiring at least one of historical map data, historical weather data and historical user reported data corresponding to the historical navigation track data.

And S740, extracting historical environmental characteristic data corresponding to the historical navigation track data according to at least one of the historical map data, the historical weather data and the historical user reported data.

In this embodiment, the ambient environment factors when the vehicle owner drives are used as one of the training parameters for training the risk probability prediction model and the odds ratio prediction model.

Optionally, the historical map data includes: historical road network data and historical road condition data; the historical road network data comprises at least one of historical road information, historical geographical position information, historical traffic identification information and historical region information; the historical road condition data comprises: at least one of congestion data and historical average speed of a road where historical navigation is located; the historical user reported data comprises: at least one of historical road congestion data, historical road temporary management action data, and historical traffic event data.

Optionally, the historical environmental characteristic data includes: at least one of historical weather characteristics, historical road conditions characteristics, historical time characteristics and historical surrounding environment characteristics. The historical road characteristics can include historical road grades, historical road forms, historical speed limit grades and other characteristics.

The manner of obtaining the historical environmental characteristic data of the multiple vehicles is similar to that of the target vehicle in the above embodiments, and for more details, reference may be made to the above embodiments, which are not described herein again.

And S750, training by adopting a machine learning algorithm to obtain an insurance probability prediction model and an odds rate prediction model based on the historical driving behavior characteristic data, the historical environment characteristic data, the insurance data and the odds rate data.

It should be noted that, by using the risk probability prediction model and the loss rate prediction model trained in the seventh embodiment of the present invention to predict vehicle risks and loss rates, the vehicle risk probability prediction method and the vehicle loss probability prediction method provided in any of the first to fifth embodiments of the present invention can be implemented. Preferably, when the vehicle insurance probability prediction model and the claim rate prediction model trained in the seventh embodiment of the present invention are used to predict vehicle insurance and claim rate, the method for vehicle insurance and claim rate provided in the second embodiment of the present invention is used to predict the insurance and claim rate of the target vehicle, so that the prediction result is more accurate.

According to the technical scheme of the embodiment of the invention, the environmental characteristic data is used as the training parameters of the risk probability prediction model and the odds ratio prediction model on the basis of the embodiment, so that the risk probability prediction model and the odds ratio prediction model are more accurate and reasonable.

Example eight

Fig. 8 is a flowchart of a vehicle risk occurrence and odds ratio prediction method according to an eighth embodiment of the present invention, which further optimizes the training method of the risk occurrence probability prediction model and the odds ratio prediction model based on the sixth embodiment. As shown in fig. 8, the method includes:

s810, obtaining historical navigation track data of a plurality of vehicles and insurance data and odds rate data corresponding to the historical navigation track data.

And S820, extracting historical driving behavior characteristic data of the corresponding vehicle according to the historical navigation track data.

And S830, obtaining historical vehicle use characteristic data corresponding to the historical navigation track data.

In the present embodiment, the historical vehicle use characteristic data is used as one of the training parameters for training the risk probability prediction model and the odds ratio prediction model. Optionally, the historical vehicle use characteristic data is mined by a machine learning method based on historical track data, historical navigation destinations and historical navigation time in the historical navigation track data.

The manner of obtaining the historical vehicle usage characteristic data is similar to that of the target vehicle in the above embodiment, and for more details, reference may be made to the above embodiment, which is not described herein again.

And S840, training by adopting a machine learning algorithm based on the historical driving behavior characteristic data, the historical vehicle use characteristic data, the insurance data and the odds rate data to obtain an insurance probability prediction model and an odds rate prediction model.

It should be noted that, by using the risk probability prediction model and the loss rate prediction model trained in the eighth embodiment of the present invention to predict vehicle risks and loss rates, the vehicle risk probability prediction method and the vehicle loss probability prediction method provided in any of the first to fifth embodiments of the present invention can be implemented. Preferably, when the vehicle insurance probability prediction model and the claim rate prediction model trained in the eighth embodiment of the present invention are used to predict vehicle insurance and claim rate, the vehicle insurance and claim rate prediction method provided in the third embodiment of the present invention is used to predict the insurance and claim rate of the target vehicle, so that the prediction result is more accurate.

According to the technical scheme of the embodiment of the invention, on the basis of the embodiment, the historical vehicle use characteristic data is used as the training parameters of the risk probability prediction model and the odds ratio prediction model, so that the risk probability prediction model and the odds ratio prediction model are more accurate and reasonable.

Example nine

Fig. 9 is a flowchart of a vehicle risk occurrence and odds ratio prediction method according to a ninth embodiment of the present invention, and the present embodiment further optimizes the training method of the risk occurrence probability prediction model and the odds ratio prediction model based on a sixth embodiment. As shown in fig. 9, the method includes:

s910, historical navigation track data of a plurality of vehicles and insurance data and odds rate data corresponding to the historical navigation track data are obtained.

And S920, extracting historical driving behavior characteristic data of the corresponding vehicle according to the historical navigation track data.

And S930, acquiring historical driving behavior distinguishing characteristic data of the vehicle and the surrounding vehicles corresponding to the historical navigation track data according to the historical navigation track data.

In the present embodiment, the historical driving behavior distinguishing feature data is used as one of the training parameters for training the risk probability prediction model and the odds ratio prediction model. Optionally, the characteristic data for distinguishing the historical driving behavior of the nearby vehicle includes: a historical driving behavior distinguishing feature and a historical speed distinguishing feature on the same road and at the same point in time. Wherein the historical driving behavior distinguishing characteristics comprise distinguishing characteristics of sharp acceleration, sharp deceleration, sharp turning and/or overspeed. The historical speed distinguishing feature is whether the target vehicle is driving at a speed significantly lower or higher than the surrounding vehicles.

The manner of extracting the historical driving behavior distinguishing characteristic data of the corresponding vehicle is similar to that of the target vehicle in the above embodiment, and for more detailed contents, reference may be made to the above embodiment, which is not described herein again.

And S940, training by adopting a machine learning algorithm based on the driving behavior characteristic data, the historical driving behavior distinguishing characteristic data, the insurance data and the odds rate data to obtain an insurance probability prediction model and an odds rate prediction model.

It should be noted that, by using the risk probability prediction model and the odds prediction model trained in the ninth embodiment of the present invention to predict vehicle risks and odds, the vehicle risk probability prediction method and the vehicle odds prediction method provided in any of the first to fifth embodiments of the present invention can be implemented. Preferably, when the vehicle insurance probability prediction model and the claim rate prediction model trained in the ninth embodiment of the present invention are used to predict vehicle insurance and claim rate, the vehicle insurance and claim rate prediction method provided in the fourth embodiment of the present invention is used to predict the insurance and claim rate of the target vehicle, so that the prediction result is more accurate.

Example ten

Fig. 10 is a flowchart of a vehicle risk occurrence and odds ratio prediction method according to a tenth embodiment of the present invention, which further optimizes the training method of the risk occurrence probability prediction model and the odds ratio prediction model based on the sixth embodiment. As shown in fig. 10, the method includes:

s1010, historical navigation track data of a plurality of vehicles and insurance data and odds rate data corresponding to the historical navigation track data are obtained.

And S1020, extracting historical driving behavior characteristic data of the corresponding vehicle according to the historical navigation track data.

S1030, acquiring other historical characteristic data, wherein the other historical characteristic data comprises at least one of historical violation data, historical user attribute data and historical frequent place classification data of the vehicle corresponding to the historical navigation track data.

In the present embodiment, other historical feature data is used as one of the training parameters for training the risk probability prediction model and the odds ratio prediction model. Optionally, the historical frequent classification is obtained based on historical navigation trajectory data of the target vehicle; the history frequented classifications include: home, corporate, etc. The historical user attribute data is obtained based on the network retrieval behavior and the access log of the corresponding user; the historical user attribute data includes: sex, age, etc.

The manner of acquiring other historical characteristic data is similar to that of the target vehicle in the above embodiment, and for more details, reference may be made to the above embodiment, which is not described herein again.

And S1040, training by adopting a machine learning algorithm based on the historical driving behavior characteristic data, other historical characteristic data, the insurance data and the odds rate data to obtain an insurance probability prediction model and an odds rate prediction model.

It should be noted that, by using the risk probability prediction model and the loss rate prediction model trained in the tenth embodiment of the present invention to predict vehicle risks and loss rates, the vehicle risk probability prediction method and the vehicle loss probability prediction method provided in any of the first to fifth embodiments of the present invention can be implemented. Preferably, when the vehicle insurance probability prediction model and the claim rate prediction model trained in the tenth embodiment of the present invention are used to predict vehicle insurance and claim rate, the vehicle insurance and claim rate prediction method provided in the fifth embodiment of the present invention is used to predict insurance and claim rate of the target vehicle, so that the prediction result is more accurate.

It should be noted that the vehicle risk occurrence and odds ratio training method provided by any embodiment of the present invention can be randomly combined for use, and is not limited herein. Correspondingly, the vehicle insurance and the claim rate are predicted by using the prediction parameters corresponding to the training parameters of the trained insurance probability prediction model and the trained claim rate prediction model, so that the vehicle insurance and the claim rate are more accurately predicted.

Optionally, the historical driving behavior feature data, the historical environment feature data and the historical vehicle usage feature data are used as training parameters of the risk probability prediction model and the odds rate prediction model together, and the risk probability prediction model and the odds rate prediction model are obtained by adopting machine learning algorithm training in combination with the risk data and the odds rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the environmental characteristic data and the vehicle use characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the environmental characteristic data and the vehicle use characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and a risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model are obtained.

Optionally, the historical driving behavior feature data, the historical environment feature data and the historical driving behavior distinguishing feature data are used as training parameters of an insurance probability prediction model and an insurance rate prediction model together, and the insurance probability prediction model and the insurance rate prediction model are obtained by adopting machine learning algorithm training in combination with the insurance data and the insurance rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the environmental characteristic data and the driving behavior distinguishing characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the environmental characteristic data and the driving behavior distinguishing characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and the risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and the odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model are obtained.

Optionally, the historical driving behavior feature data, the historical environment feature data and the historical other feature data are used as training parameters of the risk probability prediction model and the odds rate prediction model together, and the risk probability prediction model and the odds rate prediction model are obtained by training through a machine learning algorithm in combination with the risk data and the odds rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the environmental characteristic data and other characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the environmental characteristic data and other characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and a risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model are obtained.

Optionally, the historical driving behavior feature data, the historical vehicle usage feature data and the historical driving behavior distinguishing feature data are used as training parameters of an insurance probability prediction model and an insurance rate prediction model together, and the insurance probability prediction model and the insurance rate prediction model are obtained by adopting machine learning algorithm training in combination with the insurance data and the insurance rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the vehicle usage characteristic data and the driving behavior distinguishing characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the vehicle usage characteristic data and the driving behavior distinguishing characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and the risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and the odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model are obtained.

Optionally, the historical driving behavior feature data, the historical vehicle usage feature data and the historical other feature data are used as training parameters of the risk probability prediction model and the odds rate prediction model together, and the risk probability prediction model and the odds rate prediction model are obtained by adopting machine learning algorithm training in combination with the risk data and the odds rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the vehicle usage characteristic data and other characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the vehicle usage characteristic data and other characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and a risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model are obtained.

Optionally, the historical driving behavior feature data, the historical driving behavior distinguishing feature data and the historical other feature data are used as training parameters of the risk probability prediction model and the odds rate prediction model together, and the risk probability prediction model and the odds rate prediction model are obtained by adopting machine learning algorithm training in combination with the risk data and the odds rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the driving behavior distinguishing characteristic data and other characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the driving behavior distinguishing characteristic data and other characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and the risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and the odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model are obtained.

Optionally, the historical driving behavior feature data, the historical environment feature data, the historical vehicle usage feature data and the historical driving behavior distinguishing feature data are used as training parameters of an insurance probability prediction model and an insurance rate prediction model together, and the insurance probability prediction model and the insurance rate prediction model are obtained by training through a machine learning algorithm in combination with the insurance data and the insurance rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the environmental characteristic data, the vehicle usage characteristic data and the driving behavior distinguishing characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the environmental characteristic data, the vehicle usage characteristic data and the driving behavior distinguishing characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and the risk probability prediction result corresponding to the target vehicle and the odds rate prediction result corresponding to the target vehicle and output by the risk probability prediction model are obtained.

Optionally, the historical driving behavior feature data, the historical environment feature data, the historical vehicle usage feature data and the historical other feature data are used as training parameters of an insurance probability prediction model and an insurance rate prediction model together, and the insurance probability prediction model and the insurance rate prediction model are obtained by training through a machine learning algorithm in combination with the insurance data and the insurance rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the environmental characteristic data, the vehicle usage characteristic data and other characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the environmental characteristic data, the vehicle usage characteristic data and other characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and the risk probability prediction result corresponding to the target vehicle and the odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model are obtained.

Optionally, the historical driving behavior feature data, the historical vehicle usage feature data, the historical driving behavior distinguishing feature data and the historical other feature data are used as training parameters of an insurance probability prediction model and an insurance rate prediction model together, and the insurance probability prediction model and the insurance rate prediction model are obtained by training through a machine learning algorithm in combination with the insurance data and the insurance rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the vehicle usage characteristic data, the driving behavior distinguishing characteristic data and other characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the vehicle usage characteristic data, the driving behavior distinguishing characteristic data and other characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and a risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model are obtained.

Optionally, the historical driving behavior feature data, the historical environment feature data, the historical vehicle usage feature data, the historical driving behavior distinguishing feature data and the historical other feature data are used as training parameters of the risk probability prediction model and the odds rate prediction model together, and the risk probability prediction model and the odds rate prediction model are obtained by training through a machine learning algorithm in combination with the risk data and the odds rate data corresponding to the historical navigation trajectory data. Correspondingly, when the risk probability prediction model and the odds rate prediction model are used for predicting the risk and the odds rate of the target vehicle, the driving behavior characteristic data, the environmental characteristic data, the vehicle usage characteristic data, the driving behavior distinguishing characteristic data and other characteristic data of the target vehicle are jointly used as prediction parameters for predicting the risk and the odds rate of the target vehicle, namely the driving behavior characteristic data, the environmental characteristic data, the vehicle usage characteristic data, the driving behavior distinguishing characteristic data and other characteristic data of the target vehicle are used as input parameters of the risk probability prediction model and the odds rate prediction model, and the risk probability prediction result corresponding to the target vehicle and the odds rate prediction result corresponding to the target vehicle and output by the risk probability prediction model are obtained.

EXAMPLE eleven

Fig. 11 is a schematic structural diagram of a vehicle risk and odds prediction apparatus according to an eleventh embodiment of the present invention. The vehicle insurance and dividend rate prediction device can be implemented in software and/or hardware, for example, the vehicle insurance and dividend rate prediction device can be configured in a computer device, as shown in fig. 11, and the device includes: trajectory data acquisition module 1110, driving behavior acquisition module 1120, and outcome prediction module 1130. Wherein:

a track data acquisition module 1110, configured to acquire navigation track data of a target vehicle, where the navigation track data is obtained based on navigation software installed on a mobile terminal;

a driving behavior extraction module 1120, configured to extract driving behavior feature data corresponding to the vehicle according to the navigation track data;

the result prediction module 1130 is configured to call a pre-established risk probability prediction model and an odds rate prediction model based on the driving behavior feature data, so as to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle output by the odds rate prediction model.

The embodiment of the invention collects navigation track data of a target vehicle through a track data acquisition module based on navigation software installed on a mobile terminal, and a driving behavior acquisition module extracts driving behavior characteristic data corresponding to the target vehicle according to the navigation track data; the result prediction module calls a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior feature data to obtain a risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle and output by the odds rate prediction model, and extracts the driving behavior feature data from navigation track data recorded in navigation software for predicting vehicle risk and odds rate, so that the accuracy of predicting the risk condition and the odds rate of the vehicle is improved, and the achievement of map data assets is realized.

On the basis of the above scheme, the apparatus further comprises:

the environment characteristic acquisition module is used for acquiring at least one of map data, weather data and user reported data;

the environment feature extraction module is used for extracting environment feature data corresponding to the navigation track data according to at least one of the map data, the weather data and the user reported data;

accordingly, the result prediction module 1130 is specifically configured to:

and calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data and the environment characteristic data to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle output by the odds rate prediction model.

On the basis of the above scheme, the apparatus further comprises:

the vehicle use acquisition module is used for acquiring vehicle use characteristic data of the target vehicle;

accordingly, the result prediction module 1130 is specifically configured to:

and calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data and the vehicle use characteristic data to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle output by the odds rate prediction model.

On the basis of the above scheme, the apparatus further comprises:

the distinguishing characteristic acquisition module is used for acquiring the driving behavior distinguishing characteristic data of the target vehicle and the surrounding vehicles according to the navigation track data of the target vehicle and the navigation track data of the surrounding vehicles;

accordingly, the result prediction module 1130 is specifically configured to:

and calling a pre-trained risk probability prediction model and an indemnity rate prediction model based on the driving behavior characteristic data and the driving behavior distinguishing characteristic data of the surrounding vehicles to obtain a risk probability prediction result corresponding to the target vehicle and output by the risk probability prediction model and an indemnity rate prediction result corresponding to the target vehicle and output by the indemnity rate prediction model.

On the basis of the above scheme, the apparatus further comprises:

the other characteristic acquisition module is used for acquiring other characteristic data, wherein the other characteristic data comprises at least one of violation data, user attribute data and frequent place classification data corresponding to the target vehicle;

accordingly, the result prediction module 1130 is specifically configured to:

and calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior characteristic data and the other characteristic data to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle output by the odds rate prediction model.

On the basis of the above scheme, the apparatus further comprises:

the training data acquisition module is used for acquiring historical navigation track data of a plurality of vehicles and insurance data and odds rate data corresponding to the historical navigation track data before acquiring navigation track data of a target vehicle;

the historical driving extraction module is used for extracting the historical driving behavior characteristic data of the corresponding vehicle according to the historical navigation track data;

and the model training module is used for training by adopting a machine learning algorithm to obtain an insurance probability prediction model and an odds rate prediction model based on the historical driving behavior characteristic data, the insurance data and the odds rate data.

On the basis of the above scheme, the apparatus further comprises:

the historical environment acquisition module is used for acquiring at least one of historical map data, historical weather data and historical user reported data corresponding to the historical navigation track data;

the historical environment extraction module is used for extracting historical environment characteristic data corresponding to the historical navigation track data according to at least one of the historical map data, the historical weather data and the historical user reported data;

correspondingly, the model training module is specifically configured to:

and training by adopting a machine learning algorithm to obtain an insurance probability prediction model and an odds rate prediction model based on the historical driving behavior characteristic data, the historical environment characteristic data, the insurance data and the odds rate data.

On the basis of the above scheme, the apparatus further comprises:

the historical use acquisition module is used for acquiring historical vehicle use characteristic data corresponding to the historical navigation track data;

correspondingly, the model training module is specifically configured to:

and training by adopting a machine learning algorithm to obtain an insurance probability prediction model and an odds rate prediction model based on the historical driving behavior characteristic data, the historical vehicle use characteristic data, the insurance data and the odds rate data.

On the basis of the above scheme, the apparatus further comprises:

the historical distinguishing acquisition module is used for acquiring historical driving behavior distinguishing characteristic data of the vehicle and the surrounding vehicles corresponding to the historical navigation track data according to the historical navigation track data;

correspondingly, the model training module is specifically configured to:

and training by adopting a machine learning algorithm to obtain an insurance probability prediction model and an insurance rate prediction model based on the driving behavior characteristic data, the historical driving behavior distinguishing characteristic data, the insurance data and the insurance rate data.

On the basis of the above scheme, the apparatus further comprises:

the historical other acquisition module is used for acquiring other historical characteristic data, and the other historical characteristic data comprises at least one of historical violation data, historical user attribute data and historical frequent place classification data of the vehicle corresponding to the historical navigation track data;

correspondingly, the model training module is specifically configured to:

and training by adopting a machine learning algorithm to obtain an insurance probability prediction model and an odds rate prediction model based on the historical driving behavior characteristic data, the other historical characteristic data, the insurance data and the odds rate data.

The vehicle insurance and claim rate prediction device provided by the embodiment of the invention can execute the vehicle insurance and claim rate prediction method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects for executing the vehicle insurance and claim rate prediction method.

Example twelve

Fig. 12 is a schematic structural diagram of a computer device in a twelfth embodiment of the present invention. FIG. 12 illustrates a block diagram of an exemplary computer device 1212 suitable for use in implementing embodiments of the invention. Computer device 1212 shown in fig. 12 is only an example, and should not bring any limitations to the functionality or scope of use of embodiments of the present invention.

As shown in fig. 12, computer device 1212 takes the form of a general purpose computing device. Components of computer device 1212 may include, but are not limited to: one or more processors 1216, a system memory 1228, and a bus 1218 that couples the various system components (including the system memory 1228 and the processors 1216).

Bus 1218 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor 1216 or local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

Computer device 1212 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by computer device 1212 and includes both volatile and nonvolatile media, removable and non-removable media.

The system memory 1228 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)1230 and/or cache memory 1232. Computer device 1212 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, the storage device 1234 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 12, and commonly referred to as a "hard disk drive"). Although not shown in FIG. 12, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 1218 by one or more data media interfaces. Memory 1228 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.

Program/utility 1240 having a set (at least one) of program modules 1242, such program modules 1242 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, may be stored in, for example, memory 1228, each of which examples or some combination may include an implementation of a network environment. Program modules 1242 generally perform the functions and/or methodologies of embodiments of the invention as described herein.

Computer device 1212 may also communicate with one or more external devices 1214 (e.g., keyboard, pointing device, display 1224, etc.), and may also communicate with one or more devices that enable a user to interact with computer device 1212, and/or any devices (e.g., network card, modem, etc.) that enable computer device 1212 to communicate with one or more other computing devices. Such communication may occur via an input/output (I/O) interface 1222. Also, computer device 1212 can communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the internet) through network adapter 1220. As shown, network adapter 1220 communicates with the other modules of computer device 1212 through bus 1218. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with computer device 1212, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.

The processor 1216 executes various functional applications and data processing by executing programs stored in the system memory 1228, for example, implementing a vehicle risk and odds prediction method provided by an embodiment of the present invention, the method including:

Of course, those skilled in the art will appreciate that the processor may also implement the technical solutions of the vehicle risk and odds prediction methods provided by any embodiments of the present invention.

EXAMPLE thirteen

Thirteenth embodiment of the present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a vehicle risk and odds prediction method according to an embodiment of the present invention, the method including:

Of course, the computer program stored on the computer-readable storage medium provided by the embodiments of the present invention is not limited to the method operations described above, and may also perform related operations in the vehicle risk occurrence and odds prediction method provided by any embodiments of the present invention.

Computer storage media for embodiments of the invention may employ 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. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied 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.

Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or 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 the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A vehicle insurance and odds ratio prediction method, comprising:

2. The method of claim 1, further comprising:

collecting at least one of map data, weather data and user reported data;

extracting environment characteristic data corresponding to the navigation track data according to at least one of the map data, the weather data and the user reported data;

correspondingly, the step of calling a pre-trained risk probability prediction model and an odds rate prediction model based on the driving behavior feature data to obtain a risk probability prediction result corresponding to the target vehicle output by the risk probability prediction model and an odds rate prediction result corresponding to the target vehicle output by the odds rate prediction model includes:

3. The method of claim 1, further comprising:

acquiring vehicle use characteristic data of the target vehicle;

4. The method of claim 1, further comprising:

acquiring driving behavior distinguishing characteristic data of the target vehicle and the surrounding vehicles according to the navigation track data of the target vehicle and the navigation track data of the surrounding vehicles;

5. The method of claim 1, further comprising:

acquiring other characteristic data, wherein the other characteristic data comprises at least one of violation data, user attribute data and frequent place classification data corresponding to the target vehicle;

6. The method of claim 1, further comprising, prior to acquiring the navigation trajectory data of the target vehicle:

acquiring historical navigation track data of a plurality of vehicles, and insurance data and odds rate data corresponding to the historical navigation track data;

extracting historical driving behavior characteristic data of a corresponding vehicle according to the historical navigation track data;

and training by adopting a machine learning algorithm to obtain an insurance probability prediction model and an odds rate prediction model based on the historical driving behavior characteristic data, the insurance data and the odds rate data.

7. The method of claim 6, further comprising:

acquiring at least one of historical map data, historical weather data and historical user reported data corresponding to the historical navigation track data;

extracting historical environment characteristic data corresponding to the historical navigation track data according to at least one of the historical map data, the historical weather data and the historical user reported data;

correspondingly, the obtaining of the risk probability prediction model and the odds rate prediction model by adopting machine learning algorithm training based on the historical driving behavior feature data, the risk data and the odds rate data comprises:

8. The method of claim 6, further comprising:

acquiring historical vehicle use characteristic data corresponding to the historical navigation track data;

9. The method of claim 6, further comprising:

acquiring historical driving behavior distinguishing characteristic data of a vehicle and a surrounding vehicle corresponding to the historical navigation track data according to the historical navigation track data;

10. The method of claim 6, further comprising:

acquiring other historical characteristic data, wherein the other historical characteristic data comprises at least one of historical violation data, historical user attribute data and historical frequent place classification data of a vehicle corresponding to the historical navigation track data;

11. A vehicle insurance and odds prediction apparatus, comprising:

12. A computer device, the device comprising:

one or more processors;

storage means for storing one or more programs;

when executed by the one or more processors, cause the one or more processors to implement the vehicle venture and odds prediction method of any one of claims 1-10.

13. A computer-readable storage medium on which a computer program is stored, the program, when executed by a processor, implementing a vehicle risk and odds prediction method according to any one of claims 1-10.