A vehicle abnormal hoarding identification method and device, a vehicle, and a storage medium

Abstract

The application provides a vehicle abnormal hoarding identification method and device, a vehicle and a storage medium, wherein the method comprises the following steps: obtaining sales data and vehicle condition data of the vehicle, extracting sales feature information according to the sales data, extracting vehicle condition feature information according to the vehicle condition data, inputting the sales feature information and the vehicle condition feature information into a vehicle detection model for detection, judging whether the vehicle is an abnormal hoarding vehicle according to a detection result, and outputting information of the abnormal hoarding vehicle. The beneficial effects of the application include: based on the vehicle Internet of Things data, the feature information of the abnormal hoarding vehicle is identified, the vehicle detection model is trained according to the feature information, the abnormal hoarding condition of the vehicle in the market is identified by using the vehicle detection model, the actual market demand of the vehicle is effectively predicted, the inventory is effectively adjusted, the vehicle production is targeted, the market demand is better adapted and satisfied, and the interests of the manufacturer are protected.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle body control, in particular to a vehicle abnormal accumulation identification method and device, a vehicle and a storage medium. BACKGROUND

[0002] With the continuous production and sale of various brands, various functions and various sizes of vehicles by vehicle manufacturers, there are often situations where the inventory of a certain brand or a certain model of vehicle increases, while the inventory of a certain brand or a certain model of vehicle is insufficient. Therefore, for vehicle manufacturers, how to effectively control the abnormal accumulation of vehicle dealers and produce vehicles in a targeted manner is a problem that needs to be solved at present. SUMMARY

[0003] In view of the above problem that dealers order vehicles by claiming that customers buy vehicles and accumulate vehicles, and then flow into the market in the form of used cars, disrupting the market sales strategy of the manufacturer, the present application is proposed to provide a vehicle abnormal accumulation identification method, device, vehicle and storage medium to overcome the above problems or at least partially solve the above problems.

[0004] To solve the above problems, on the one hand, the present application discloses a vehicle abnormal accumulation identification method, comprising:

[0005] obtaining sales data and vehicle condition data of a vehicle;

[0006] extracting sales feature information of the vehicle according to the sales data, and extracting vehicle condition feature information of the vehicle according to the vehicle condition data;

[0007] inputting the sales feature information and the vehicle condition feature information into a preset vehicle detection model for detection to obtain a detection result, wherein the vehicle detection model is used to detect abnormal accumulation of vehicles;

[0008] judging whether the vehicle is an abnormal accumulation vehicle according to the detection result;

[0009] outputting information of the abnormal accumulation vehicle.

[0010] Further, the process of training the vehicle detection model comprises:

[0011] obtaining a first training set for model training, the first training set comprising sales data and vehicle condition data of a plurality of vehicles, the first training set being obtained by aggregating the sales feature information and the vehicle condition feature information;

[0012] selecting a vehicle identified as abnormal hoarding from sales data and vehicle condition data of the vehicle, the vehicle identified as abnormal hoarding being obtained through on-site investigation by a marketing personnel;

[0013] After data labeling is performed on the vehicle identified as abnormal hoarding, sales feature information and vehicle condition feature information of a vehicle in the first training set that is not labeled are classified to obtain all vehicle sales feature information and vehicle condition feature information under abnormal hoarding classification;

[0014] The all vehicle sales feature information and vehicle condition feature information under abnormal hoarding classification are taken as a second training set;

[0015] The vehicle detection model is trained by using the second training set, and the vehicle detection model is tested by using the cumulative mileage of the vehicle, the use time interval of the vehicle, and the vehicle sale time in the first training set to obtain a trained preset vehicle detection model.

[0016] Further, the first training set includes a vehicle identification code, a vehicle cumulative mileage, a use time interval, and a vehicle sale time, and the classification of the sales feature information and the vehicle condition feature information of the vehicle in the first training set that is not labeled includes:

[0017] According to the sales feature information of the vehicle in the first training set that is not labeled, a first set of vehicle identification codes with a sales time greater than a preset time interval is obtained;

[0018] According to the sales feature information of the vehicle, a second set of vehicle identification codes with a vehicle cumulative mileage less than a preset mileage is obtained;

[0019] From the vehicle condition feature information corresponding to the first set and the second set, vehicle operation data corresponding to the vehicle identification code is obtained, and the operation data includes a vehicle cumulative mileage of the vehicle, a use time interval, a data collection time, and a latitude and longitude in a vehicle operation process;

[0020] According to the vehicle cumulative mileage, the use time interval, the data collection time, and the latitude and longitude, the vehicle condition feature information is classified.

[0021] Further, the sales feature information of the vehicle is extracted according to the sales data, and the sales feature information of the vehicle includes:

[0022] According to the sales data of the vehicle, a sales time of the vehicle is obtained;

[0023] A time difference between the sales time and a current time is calculated;

[0024] The time difference is recorded as the sold-out time of the vehicle, and the sold-out time of the vehicle is taken as the sales feature information of the vehicle.

[0025] Further, the vehicle condition feature information of the vehicle is extracted according to the vehicle condition data, comprising:

[0026] The vehicle condition data is obtained, and the vehicle condition data comprises cumulative mileage of the vehicle, mean value of use time interval, minimum value of use time interval and maximum value of use time interval;

[0027] After data cleaning of the vehicle condition data, the vehicle condition data is classified according to the vehicle identification code;

[0028] The vehicle condition feature information of the vehicle is extracted according to the classified vehicle condition data.

[0029] Further, the vehicle condition data is classified according to the vehicle identification code, comprising:

[0030] The vehicle condition data comprises vehicle cumulative mileage data, and after classification of the vehicle cumulative mileage data according to the vehicle identification code, the vehicle cumulative mileage data is arranged in ascending order of data collection time;

[0031] The first mileage difference value of adjacent two vehicle cumulative mileage data arranged in ascending order is calculated respectively;

[0032] If the first mileage difference value is less than a preset mileage, the first mileage difference value is reserved;

[0033] If the first mileage difference value is greater than or equal to the preset mileage, the first mileage difference value is recorded as zero;

[0034] The minimum value of the vehicle cumulative mileage arranged in ascending order is taken as the starting point, and the vehicle cumulative mileage data is rearranged.

[0035] Further, the vehicle condition data is classified according to the vehicle identification code, comprising:

[0036] The vehicle condition data is arranged in ascending order of vehicle condition data collection time;

[0037] The second time difference value of adjacent two vehicle condition data is calculated respectively;

[0038] If the second time difference value is less than or equal to a preset time interval, the corresponding journey of the vehicle condition data is divided into the same journey;

[0039] If the second time difference value is greater than the preset time interval, the corresponding journey of the vehicle condition data is divided into different journeys;

[0040] The start time of each vehicle trip is calculated respectively.

[0041] Further, the vehicle condition data is classified according to the vehicle identification code, comprising:

[0042] The vehicle condition data is arranged in ascending order according to the start time of the vehicle trip;

[0043] The third time difference value between two adjacent vehicle condition data is calculated respectively;

[0044] The use time interval of the vehicle is calculated according to the third time difference value, including the average value of the use time interval, the minimum value of the use time interval, the maximum value of the use time interval, and the variance of the use time interval.

[0045] Further, the information of the abnormal stock vehicle is output, comprising:

[0046] If the vehicle is determined to be an abnormal stock vehicle, the information of the city and the dealer to which the vehicle belongs is obtained, and the identification result is visually displayed.

[0047] In another aspect, the present application also provides a vehicle abnormal stock identification device, comprising:

[0048] A vehicle data acquisition module is configured to acquire sales data and vehicle condition data of a vehicle.

[0049] A vehicle feature information extraction module is configured to extract sales feature information of the vehicle according to the sales data, and extract vehicle condition feature information of the vehicle according to the vehicle condition data.

[0050] A vehicle abnormal stock detection module is configured to input the sales feature information and the vehicle condition feature information into a preset vehicle detection model for detection to obtain a detection result, wherein the vehicle detection model is used to detect the abnormal stock of the vehicle.

[0051] A vehicle abnormal stock judgment module is configured to judge whether the vehicle is an abnormal stock vehicle according to the detection result.

[0052] A detection data output module is configured to output information of the abnormal stock vehicle.

[0053] Further, the vehicle abnormal stock detection module comprises:

[0054] The first training set acquisition submodule is configured to acquire a first training set of the vehicle detection model, the first training set comprising sales data and vehicle condition data of a plurality of vehicles, the first training set being obtained by aggregating the sales feature information and the vehicle condition feature information, and the first training set comprising a vehicle identification code, a vehicle cumulative mileage, a vehicle use time interval, and a vehicle sale time;

[0055] The abnormal hoarding vehicle identification submodule is configured to select, from the sales data and the vehicle condition data of the vehicles, vehicles identified as being abnormally hoarded, the vehicles identified as being abnormally hoarded being obtained through on-site investigation by marketing personnel;

[0056] The abnormal hoarding vehicle marking submodule is configured to mark the sales feature information and the vehicle condition feature information corresponding to the vehicles identified as being abnormally hoarded in the first training set;

[0057] The vehicle classification submodule is configured to classify, after marking the vehicles identified as being abnormally hoarded, the sales feature information and the vehicle condition feature information of the vehicles in the first training set that are not marked, to obtain all vehicle sales feature information and vehicle condition feature information under abnormal hoarding classification;

[0058] The second training set acquisition submodule is configured to take all the vehicle sales feature information and the vehicle condition feature information under abnormal hoarding classification as a second training set;

[0059] The vehicle detection submodule is configured to train a vehicle detection model using the second training set, and test the vehicle detection model through the vehicle cumulative mileage, the vehicle use time interval, and the vehicle sale time in the first training set, to obtain a trained preset vehicle detection model.

[0060] Further, the vehicle feature information extraction module comprises:

[0061] The sale time acquisition submodule is configured to acquire a sale time of the vehicle according to the sales data of the vehicle.

[0062] The time difference calculation submodule is configured to calculate a time difference between the sale time and a current time.

[0063] The first vehicle feature information extraction submodule is configured to record the time difference as a sold-out time of the vehicle, and take the sold-out time of the vehicle as sales feature information of the vehicle.

[0064] The vehicle condition data acquisition submodule is configured to acquire the vehicle condition data, the vehicle condition data comprising a vehicle cumulative mileage, a vehicle use time interval mean value, a vehicle use time interval minimum value, and a vehicle use time interval maximum value.

[0065] The vehicle condition data classification submodule is configured to classify the vehicle condition data according to the vehicle identification code after data cleaning.

[0066] The second vehicle feature information extraction submodule is configured to extract the vehicle condition feature information from the classified vehicle condition data.

[0067] Further, the vehicle condition data classification submodule comprises:

[0068] The first vehicle condition data classification unit is configured to arrange the accumulated mileage data of the vehicle in ascending order of data collection time after classifying the accumulated mileage data of the vehicle according to the vehicle identification code, calculate the first mileage difference value of adjacent two vehicle accumulated mileage data arranged in ascending order respectively, if the first mileage difference value is less than a preset mileage, retain the first mileage difference value, if the first mileage difference value is greater than or equal to the preset mileage, record the first mileage difference value as zero, and arrange the vehicle accumulated mileage data again with the minimum value of the vehicle accumulated mileage arranged in ascending order as the starting point.

[0069] The second vehicle condition data classification unit is configured to arrange the vehicle condition data in ascending order of vehicle condition data collection time after classifying the vehicle condition data according to the vehicle identification code, calculate the second time difference value of adjacent two vehicle condition data respectively, if the second time difference value is less than or equal to a preset time interval, divide the corresponding journey of the vehicle condition data into the same journey, if the second time difference value is greater than the preset time interval, divide the corresponding journey of the vehicle condition data into different journeys, and calculate the start time of the vehicle journey in each different journey respectively.

[0070] The third vehicle condition data classification unit is configured to arrange the vehicle condition data in ascending order of vehicle journey start time after classifying the vehicle condition data according to the vehicle identification code, calculate the third time difference value of adjacent two vehicle condition data respectively, calculate a plurality of vehicle use time intervals of the vehicle according to the third time difference value, and the vehicle use time intervals comprise the average value of the vehicle use time interval, the minimum value of the vehicle use time interval, the maximum value of the vehicle use time interval, and the variance of the vehicle use time interval.

[0071] Further, the detection data output module is configured to acquire the information of the city and the dealer to which the vehicle belongs if the vehicle is determined as an abnormal hoarding vehicle, and visually display the identification result.

[0072] In another aspect, the embodiments of the present application also provide a vehicle comprising a processor, a memory, and a computer program stored on the memory and capable of running on the processor, wherein the computer program is executed by the processor to implement the steps of the vehicle abnormal hoarding identification method.

[0073] In another aspect, the embodiments of the present application also provide a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the vehicle abnormal hoarding identification method.

[0074] The embodiments of the present application have the following advantages: the present application obtains the sales data and the running data of the vehicle, extracts the sales feature information of the vehicle according to the sales data, extracts the vehicle condition feature information according to the running data, inputs the sales feature information and the vehicle condition feature information into a preset vehicle detection model for detection, obtains a detection result, judges whether the vehicle is an abnormal hoarding vehicle according to the detection result and the vehicle running data, and outputs the information of the abnormal hoarding vehicle.

[0075] The embodiments of the present application obtain the sales data and the vehicle condition data of the vehicle, combine the on-site investigation of the vehicle condition by the marketing personnel, select the vehicle identified as the abnormal hoarding from the sales data and the vehicle condition data of the vehicle, identify the feature information of the abnormal hoarding vehicle, train the vehicle detection model according to the feature information, and identify the abnormal hoarding condition of the vehicle in the market by using the vehicle detection model, effectively predict the actual market demand of the vehicle, effectively adjust the inventory, and produce the vehicle in a targeted manner, so that the vehicle better adapts to and meets the market demand, and guarantees the interests of the manufacturer. BRIEF DESCRIPTION OF DRAWINGS

[0076] Figure 1 A step flowchart of a vehicle abnormal hoarding identification method provided by the embodiments of the present application is shown in the figure.

[0077] Figure 2 A flowchart of training a vehicle detection model provided by the embodiments of the present application is shown in the figure.

[0078] Figure 3 A schematic diagram of a vehicle abnormal hoarding identification process provided by the embodiments of the present application is shown in the figure.

[0079] Figure 4 A structural block diagram of a vehicle abnormal hoarding identification device provided by the embodiments of the present application is shown in the figure. DETAILED DESCRIPTION

[0080] In order to make the above objectives, features and advantages of the present application more apparent and understandable, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0081] Tank 300 is a WEY brand under the Great Wall Motor based on the Great Wall Motor intelligent professional off-road platform to create an off-road SUV, some dealers to buy vehicles as a reason to order vehicles, in the form of second-hand vehicles into the market after the vehicle, disrupt the market sales strategy of tank 300. In order to identify these abnormal accumulation of tank 300 vehicles and dealers, the big data platform is based on the tank 300 vehicle networking data, identifies the characteristics of the abnormal accumulation of tank 300 vehicles, and designs a set of tank 300 abnormal accumulation identification method and system based on the characteristics, to guide the marketing department to find the abnormal accumulation of tank 300, and improve the marketing environment of tank 300.

[0082] Figure 1 A step flow chart of a vehicle abnormal accumulation identification method provided by the embodiment of the present application, the method comprising the following steps:

[0083] Step 101, obtaining the sales data and vehicle condition data of the vehicle;

[0084] In the embodiment of the present application, the vehicle is provided with a vehicle terminal, and after networking, the vehicle condition data of the corresponding vehicle is obtained every certain preset time interval, and the vehicle condition data is uploaded to the cloud server through the wireless network. The above-mentioned preset time interval can be 30 seconds or 60 seconds, which can be set according to the actual demand, and the present application does not make any limitation. When the vehicle is manufactured, the sales data of the vehicle is uploaded to the Internet of vehicles, which is convenient for the manufacturer to maintain in the background. The sales data and vehicle condition data of the vehicle include but are not limited to the dealer of the vehicle, the vehicle brand, the driving type, the vehicle type, the sales price, the frame number information, the registration province, the location data, the vehicle speed, the first running time of the vehicle and the like.

[0085] Step 102, extracting the sales feature information of the vehicle according to the sales data, and extracting the vehicle condition feature information of the vehicle according to the vehicle condition data;

[0086] The process of extracting the sales feature information of the vehicle according to the sales data includes: obtaining the sales time of the vehicle according to the sales data of the vehicle, calculating the time difference between the sales time and the current time, taking the time difference as the sold-out time of the vehicle, and taking the sold-out time of the vehicle as the sales feature information of the vehicle.

[0087] The process of extracting the vehicle condition feature information of the vehicle according to the vehicle condition data includes: obtaining the vehicle condition data, which includes the cumulative mileage of the vehicle, the average value of the vehicle time interval, the minimum value of the vehicle time interval and the maximum value of the vehicle time interval, after data cleaning of the vehicle condition data, classifying the vehicle condition data according to the vehicle identification code, and extracting the vehicle condition feature information of the vehicle according to the classified vehicle condition data.

[0088] In step 103, the sales feature information and the vehicle condition feature information are input into a preset vehicle detection model for detection to obtain a detection result, wherein the vehicle detection model is used to detect abnormal hoarding of vehicles.

[0089] Figure 2 A flowchart for training a vehicle detection model is provided in the embodiment of the present application. In the embodiment, the process of training the vehicle detection model includes: 1. obtaining a first training set of the vehicle detection model, the first training set including sales data and vehicle condition data of a plurality of vehicles, the first training set being obtained by aggregating sales feature information and vehicle condition feature information, the first training set including a vehicle identification code, a cumulative mileage of the vehicle, a use time interval, and a vehicle sale time; 2. selecting vehicles identified as abnormal hoarding from the sales data and the vehicle condition data of the vehicles, the vehicles identified as abnormal hoarding being obtained through on-site investigation by marketing personnel; 3. data labeling the sales feature information and the vehicle condition feature information corresponding to the vehicles identified as abnormal hoarding in the first training set; 4. after data labeling the vehicles identified as abnormal hoarding, classifying the sales feature information and the vehicle condition feature information of unlabeled vehicles in the first training set using a knn method (K-Nearest-Neighbors), extracting a center point value of clustering thereof as a model feature, and labeling the unlabeled vehicles as abnormal hoarding if the sales feature information and the vehicle condition feature information thereof are the same as or similar to the model feature, to obtain all vehicle sales feature information and vehicle condition feature information under abnormal hoarding classification; 5. using all vehicle sales feature information and vehicle condition feature information under abnormal hoarding classification as a second training set; 6. training the vehicle detection model using the second training set, and testing the vehicle detection model through the cumulative mileage of the vehicle, the use time interval, and the vehicle sale time in the first training set, to obtain a preset vehicle detection model after training. In the embodiment, the first training set of the vehicle detection model is obtained by obtaining sales data and vehicle condition data of vehicles, combining on-site investigation of post-sale vehicle conditions by marketing personnel, data labeling the sales feature information and the vehicle condition feature information corresponding to the vehicles identified as abnormal hoarding in the first training set, classifying the sales feature information and the vehicle condition feature information of the vehicles using the knn method, extracting a center point value of clustering thereof as a model feature, training the vehicle detection model according to the feature information, and identifying abnormal hoarding conditions of vehicles in the market using the vehicle detection model.

[0090] In the embodiment, the applicant uses the vehicle tank 300 to test the vehicle detection model, and uses the principal component analysis method to sort the feature importance of all identified abnormal stockpiling tanks 300 vehicles, and the most important features affecting the vehicle detection model are screened out, which are the vehicle use time interval mean, the vehicle sold time and the vehicle cumulative mileage. In the process of training the vehicle detection model using the knn method, using the vehicle use time interval mean, the vehicle sold time and the vehicle cumulative mileage for model training can make the vehicle detection model better identify abnormal stockpiling tanks 300 vehicles.

[0091] Step 104, judging whether the vehicle is an abnormal stockpiling vehicle according to the detection result;

[0092] Step 105, outputting the information of the abnormal stockpiling vehicle;

[0093] Figure 3 A schematic diagram of a vehicle abnormal stockpiling identification process provided by the embodiment of the application is provided, the cloud server obtains the sales data and the vehicle condition data of the vehicle, detects according to the vehicle detection model, judges whether the vehicle is an abnormal stockpiling vehicle, if it is judged that the vehicle is not an abnormal stockpiling vehicle, saves the data in the cloud server, and if it is judged that the vehicle is an abnormal stockpiling vehicle, outputs the information of the abnormal stockpiling vehicle.

[0094] The embodiment of the application obtains the sales data and the running data of the vehicle, extracts the sales feature information of the vehicle according to the sales data, extracts the vehicle condition feature information of the vehicle according to the running data, inputs the sales feature information and the vehicle condition feature information into a preset vehicle detection model for detection, obtains a detection result, judges whether the vehicle is an abnormal stockpiling vehicle according to the detection result, and outputs the information of the abnormal stockpiling vehicle.

[0095] In an optional embodiment, the process of data cleaning of the vehicle condition data can include: filtering error data in the vehicle condition data and the running data; correcting the offset position data in the running data; correcting the supplement data in the running data according to the time dimension; determining the first running time of the vehicle; and supplementing the registration province, vehicle brand, driving type, vehicle type and frame number information of the vehicle. Wherein, the error data in the running data, such as position information error and driving speed error, is filtered; the offset position data in the running data is corrected, specifically by Fourier filtering method. Through data cleaning of the vehicle condition data, the obtained vehicle sales feature information and vehicle condition feature information are more accurate, and using the data cleaned vehicle sales feature information and vehicle condition feature information for detection can make the detection result more accurate.

[0096] In an optional embodiment, taking the vehicle tank 300 under the Great Wall Motor as an example, the vehicle machine mobile terminal collects the running data of the tank 300 and uploads it to the TSP cloud platform. The TSP cloud platform uploads the sales data and running data of the vehicle to the big data platform after data cleaning. Combined with the vehicle detection model of the tank 300 and the sales data of the tank 300, the information of the suspected hoarding vehicle and the corresponding dealer information are extracted, and the results are transmitted to the large screen for display. The specific implementation process is as follows: 1. Based on the sales data of the tank 300, the tank 300 vehicle information with a sales time greater than 2 months is screened, and the vehicle identification code (VIN code) meeting the condition is recorded; 2. From the tank 300 vehicles with a sales time greater than 2 months, the vehicle information with a recent cumulative mileage less than 1000km is screened, and the VIN code meeting the condition is recorded; 3. In the vehicle meeting the above conditions, the running data of the vehicle in the recent half year is screened, the vehicle condition characteristic information is classified according to the cumulative mileage of the vehicle, the vehicle time interval, the data collection time, and the latitude and longitude, and arranged in ascending order in each category according to the data collection time. The difference value of the adjacent two vehicle condition running data is calculated, when the difference value is greater than 15 minutes, it is recorded as a trip, and a trip number is constructed; 4. According to the VIN code and the trip number, the vehicle running data is classified and summarized. In each trip of each vehicle, the start time, end time and latitude and longitude information of the end position of the trip are extracted; 5. For the trip data, classify according to the VIN code, arrange in ascending order according to the start time of the trip, and make a difference to the start time. When the difference between the current trip start time and the last trip start time is greater than 3 days, record the VIN, the start time of the current trip, and the latitude and longitude of the last trip end. The latitude and longitude information is converted into specific province, city and detailed address information through the Gaode API interface, as the information of the suspected abnormal hoarding vehicle; 6. If the vehicle is judged as an abnormal hoarding vehicle, the information of the city and the dealer to which the vehicle belongs is obtained through the sales system, and the identification result is visualized and displayed. The embodiment of the application obtains the sales data and running data of the tank 300, screens the vehicle information with a recent cumulative mileage less than 1000km from the tank 300 vehicles with a sales time greater than 2 months, records the vehicle identification code meeting the condition, extracts the running data of the vehicle in the recent half year according to the screened vehicle, classifies the vehicle condition characteristic information according to the cumulative mileage of the vehicle, the vehicle time interval, the data collection time, and the latitude and longitude, obtains the vehicle detection model, and identifies the abnormal hoarding condition of the tank 300 in the market by using the vehicle detection model, so as to effectively predict the actual market demand of the vehicle, adjust the inventory accordingly, and produce the vehicle in a targeted manner, so as to better adapt to and meet the market demand and protect the interests of the manufacturer.

[0097] It should be noted that, for the method embodiments, the series of acts combined is described for simplicity, but those skilled in the art should know that the embodiments of the present application are not limited to the order of the acts described, because according to the embodiments of the present application, certain steps can be performed in other orders or at the same time. Secondly, those skilled in the art should know that the embodiments described in the specification are all preferred embodiments, and the acts involved are not necessarily essential to the embodiments of the present application.

[0098] In order to realize the vehicle abnormal hoarding identification method, the embodiments of the present application also provide a vehicle abnormal hoarding identification device, Figure 4 The structure block diagram of the vehicle abnormal hoarding identification device provided by the embodiments of the present application is shown in the figure, and the device comprises:

[0099] The vehicle data acquisition module 401 is configured to acquire the sales data and the vehicle condition data of the vehicle.

[0100] The vehicle feature information extraction module 402 is configured to extract the sales feature information of the vehicle according to the sales data, and extract the vehicle condition feature information of the vehicle according to the vehicle condition data.

[0101] The vehicle abnormal hoarding detection module 403 is configured to input the sales feature information and the vehicle condition feature information into a preset vehicle detection model for detection to obtain a detection result, wherein the vehicle detection model is used to detect the abnormal hoarding of the vehicle.

[0102] The vehicle abnormal hoarding judgment module 404 is configured to judge whether the vehicle is an abnormal hoarding vehicle according to the detection result.

[0103] The detection data output module 405 outputs the information of the abnormal hoarding vehicle.

[0104] In an optional embodiment, the vehicle abnormal hoarding detection module 403 can comprise:

[0105] The first training set acquisition submodule is configured to acquire a first training set of the vehicle detection model, wherein the first training set comprises the sales data and the vehicle condition data of a plurality of vehicles, the first training set is obtained by aggregating the sales feature information and the vehicle condition feature information, and the first training set comprises a vehicle identification code, a vehicle cumulative mileage, a vehicle use time interval, and a vehicle sale time.

[0106] The abnormal hoarding vehicle identification submodule is configured to select a vehicle identified as abnormal hoarding from the sales data and the vehicle condition data of the vehicle, and the vehicle identified as abnormal hoarding is obtained by field investigation of a marketing personnel.

[0107] an abnormal hoarding vehicle label sub-module, configured to label sales feature information and vehicle condition feature information corresponding to vehicles identified as abnormal hoarding in the first training set;

[0108] a vehicle classification sub-module, configured to classify sales feature information and vehicle condition feature information of vehicles in the first training set that have not been labeled after the vehicles identified as abnormal hoarding are labeled, to obtain all sales feature information and vehicle condition feature information of vehicles under abnormal hoarding classification;

[0109] a second training set acquisition sub-module, configured to take all sales feature information and vehicle condition feature information of vehicles under abnormal hoarding classification as a second training set;

[0110] a vehicle detection sub-module, configured to train a vehicle detection model using the second training set, and test the vehicle detection model by using the cumulative mileage of the vehicle, the use time interval, and the vehicle sale time in the first training set, to obtain a preset vehicle detection model after training.

[0111] In an optional embodiment, the vehicle feature information extraction module 402 can include:

[0112] a sales time acquisition sub-module, configured to acquire a sales time of a vehicle according to sales data of the vehicle;

[0113] a time difference calculation sub-module, configured to calculate a time difference between the sales time and a current time;

[0114] a first vehicle feature information extraction sub-module, configured to record the time difference as a sold-out time of the vehicle, and take the sold-out time of the vehicle as sales feature information of the vehicle.

[0115] a vehicle condition data acquisition sub-module, configured to acquire vehicle condition data, the vehicle condition data including a cumulative mileage of the vehicle, a mean value of a use time interval, a minimum value of the use time interval, and a maximum value of the use time interval;

[0116] a vehicle condition data classification sub-module, configured to classify the vehicle condition data according to a vehicle identification code after data cleaning of the vehicle condition data;

[0117] a second vehicle feature information extraction sub-module, configured to extract vehicle condition feature information of the vehicle according to the classified vehicle condition data.

[0118] In an optional embodiment, the vehicle condition data classification sub-module includes:

[0119] The first vehicle condition data classification unit is configured to sort the vehicle cumulative mileage data in ascending order of data collection time after classifying the vehicle cumulative mileage data according to the vehicle identification code, calculate a first mileage difference value of adjacent two vehicle cumulative mileage data in the ascending order respectively, retain the first mileage difference value if the first mileage difference value is less than a preset mileage, mark the first mileage difference value as zero if the first mileage difference value is greater than or equal to the preset mileage, and rearrange the vehicle cumulative mileage data with the minimum value as a starting point.

[0120] The second vehicle condition data classification unit is configured to sort the vehicle condition data in ascending order of vehicle condition data collection time after classifying the vehicle condition data according to the vehicle identification code, calculate a second time difference value of adjacent two vehicle condition data respectively, divide the corresponding journey of the vehicle condition data into the same journey if the second time difference value is less than or equal to a preset time interval, divide the corresponding journey of the vehicle condition data into different journeys if the second time difference value is greater than the preset time interval, and calculate the start time of the vehicle journey in each different journey.

[0121] The third vehicle condition data classification unit is configured to sort the vehicle condition data in ascending order of vehicle journey start time after classifying the vehicle condition data according to the vehicle identification code, calculate a third time difference value of adjacent two vehicle condition data respectively, and calculate a plurality of vehicle use time intervals of the vehicle according to the third time difference value, wherein the vehicle use time intervals include an average value of the vehicle use time intervals, a minimum value of the vehicle use time intervals, a maximum value of the vehicle use time intervals, and a variance of the vehicle use time intervals.

[0122] In an optional embodiment, the detection data output module is configured to, if it is judged that the vehicle is an abnormal hoarding vehicle, acquire information of a city and a dealer to which the vehicle belongs, and visually display the identification result.

[0123] The embodiments of the present application have the following advantages: the present application acquires sales data and vehicle condition data of a vehicle, extracts sales feature information of the vehicle according to the sales data, extracts vehicle condition feature information of the vehicle according to the vehicle condition data, inputs the sales feature information and the vehicle condition feature information into a preset vehicle detection model for detection to obtain a detection result, judges whether the vehicle is an abnormal hoarding vehicle according to the detection result and vehicle operation data, and outputs information of the abnormal hoarding vehicle. The beneficial effects of the present application include: based on the Internet of Vehicles data of the vehicle, the feature information of the abnormal hoarding vehicle is identified, the vehicle detection model is trained according to the feature information, the abnormal hoarding condition of the vehicle in the market is identified by using the vehicle detection model, the actual market demand of the vehicle is effectively predicted, the inventory is effectively adjusted, the vehicle production is targeted, the market demand is better met, and the interests of the manufacturer are protected.

[0124] The embodiment of the present application also provides a vehicle, which comprises a processor, a memory, and a computer program stored in the memory and capable of running on the processor, and when the computer program is executed by the processor, each process of the vehicle abnormal accumulation identification method embodiment described above is implemented, and the same technical effects can be achieved. To avoid repetition, details are not described here.

[0125] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and when the computer program is executed by the processor, each process of the vehicle abnormal accumulation identification method embodiment described above is implemented, and the same technical effects can be achieved. To avoid repetition, details are not described here.

[0126] Each embodiment in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same or similar parts between each embodiment are referred to each other.

[0127] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, device, or computer program product. Therefore, the embodiments of the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0128] The embodiments of the present application are described with reference to flowcharts and / or block diagrams according to the method, terminal device (system), and computer program product of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the computer or other programmable data processing terminal device produce a device for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The device for implementing the functions specified in one flow or multiple flows and / or blocks. Figure 1 The device for implementing the functions specified in one flow or multiple flows and / or blocks.

[0129] These computer program instructions can also be stored in a computer readable storage medium, which can guide the computer or other programmable data processing terminal device to work in a specific way, so that the instructions stored in the computer readable storage medium produce a product including instruction devices, which implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The device for implementing the functions specified in one flow or multiple flows and / or blocks. Figure 1the function specified in the one or more blocks.

[0130] These computer program instructions can also be loaded into computer or other programmable data processing terminal devices, so that a series of operation steps are performed on the computer or other programmable terminal devices to generate a computer implemented process, so that the instructions executed on the computer or other programmable terminal devices provide a process for implementing the flow Figure 1 the flow or flows and / or blocks Figure 1 the function specified in the one or more blocks.

[0131] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic inventive concept. Therefore, the appended claims are intended to cover all changes and modifications falling within the scope of the embodiments of the present application.

[0132] Finally, it should be noted that the relational terms herein, such as first and second, and the like, are used solely to distinguish one from another entity or action, without necessarily requiring or implying any such actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or terminal device. Without further limitation, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or terminal device that comprises the recited element.

[0133] The above describes in detail a vehicle abnormal accumulation identification method and a vehicle abnormal accumulation identification device provided by the present application. The principles and implementation manners of the present application are described by using specific examples. The above description of the embodiments is only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges can be changed. In summary, the content of the present description should not be understood as a limitation of the present application.

Claims

1. A vehicle abnormal hoarding identification method characterized by comprising: The method comprises the following steps: obtaining sales data and vehicle condition data of a vehicle; extracting sales feature information of the vehicle according to the sales data and extracting vehicle condition feature information of the vehicle according to the vehicle condition data; inputting the sales feature information and the vehicle condition feature information into a preset vehicle detection model for detection to obtain a detection result, wherein the vehicle detection model is used to detect abnormal hoarding of a vehicle; judging whether the vehicle is an abnormal hoarding vehicle according to the detection result; outputting information of the abnormal hoarding vehicle; wherein the process of training the vehicle detection model comprises: obtaining a first training set for model training, wherein the first training set comprises sales data and vehicle condition data of a plurality of vehicles, selecting a vehicle identified as abnormal hoarding from the sales data and vehicle condition data of the vehicle, and marking the sales feature information and the vehicle condition feature information corresponding to the vehicle identified as abnormal hoarding in the first training set; classifying the unmarked vehicle sales feature information and vehicle condition feature information in the first training set to obtain all vehicle sales feature information and vehicle condition feature information under abnormal hoarding classification as a second training set; training a vehicle detection model using the second training set and testing the vehicle detection model through the vehicle condition data in the first training set to obtain a preset trained vehicle detection model.

2. The method of claim 1, wherein, The first training set is obtained by aggregating the sales feature information and the vehicle condition feature information; The vehicle identified as abnormal hoarding is obtained through on-site investigation by a marketing personnel.

3. The method of claim 2, wherein, The first training set comprises a vehicle identification code, a vehicle cumulative mileage, a vehicle use time interval, and a vehicle sale time; The classification of the unmarked vehicle sales feature information and vehicle condition feature information in the first training set comprises: obtaining a first set of vehicle identification codes with a sales time greater than a preset time interval and a second set of vehicle identification codes with a vehicle cumulative mileage less than a preset mileage according to the unmarked vehicle sales feature information in the first training set; obtaining vehicle operation data corresponding to the vehicle identification code from the vehicle condition feature information corresponding to the first set and the second set, wherein the operation data comprises the vehicle cumulative mileage, the vehicle use time interval, the data collection time, and the latitude and longitude during the vehicle operation; classifying the vehicle condition feature information according to the vehicle cumulative mileage, the vehicle use time interval, the data collection time, and the latitude and longitude.

4. The method of claim 1, wherein, The extraction of the sales feature information of the vehicle according to the sales data comprises: obtaining the sales time of the vehicle according to the sales data of the vehicle; calculating the time difference between the sales time and the current time; marking the time difference as the sold-out time of the vehicle and taking the sold-out time of the vehicle as the sales feature information of the vehicle.

5. The method of claim 1, wherein, The extraction of the vehicle condition feature information of the vehicle according to the vehicle condition data comprises: obtaining the vehicle condition data, wherein the vehicle condition data comprises the vehicle cumulative mileage, the average vehicle use time interval, the minimum vehicle use time interval, and the maximum vehicle use time interval; The vehicle condition data is classified according to the vehicle identification code after data cleaning of the vehicle condition data; Vehicle condition feature information of the vehicle is extracted according to the classified vehicle condition data.

6. The method of claim 5, wherein, The vehicle condition feature information of the vehicle is extracted according to the classified vehicle condition data, including: The vehicle cumulative mileage data is included in the vehicle condition data, and the vehicle cumulative mileage data is arranged in ascending order of data collection time after classification of the vehicle cumulative mileage data according to the vehicle identification code; A first mileage difference value of adjacent two vehicle cumulative mileage data arranged in ascending order is calculated respectively; If the first mileage difference value is less than a preset mileage, the first mileage difference value is retained; If the first mileage difference value is greater than or equal to the preset mileage, the first mileage difference value is recorded as zero; The minimum value of the vehicle cumulative mileage arranged in ascending order is taken as a starting point, and the vehicle cumulative mileage data is rearranged.

7. The method of claim 6, wherein, The vehicle condition feature information of the vehicle is extracted according to the classified vehicle condition data, including: The vehicle condition data is arranged in ascending order of vehicle condition data collection time; A second time difference value of adjacent two vehicle condition data is calculated respectively; If the second time difference value is less than or equal to a preset time interval, a corresponding trip of the vehicle condition data is divided into the same trip; If the second time difference value is greater than the preset time interval, the corresponding trip of the vehicle condition data is divided into different trips; The start time of the vehicle trip is calculated in each different trip.

8. A vehicle abnormal hoarding recognition device characterized by comprising: Including: A vehicle data acquisition module is configured to acquire sales data and vehicle condition data of a vehicle; A vehicle feature information extraction module is configured to extract sales feature information of the vehicle according to the sales data and vehicle condition feature information of the vehicle according to the vehicle condition data; A vehicle abnormal accumulation detection module is configured to input the sales feature information and the vehicle condition feature information into a preset vehicle detection model for detection to obtain a detection result, wherein the vehicle detection model is configured to detect abnormal accumulation of the vehicle; A vehicle abnormal accumulation judgment module is configured to judge whether the vehicle is an abnormal accumulation vehicle according to the detection result; A detection data output module is configured to output information of the abnormal accumulation vehicle; The vehicle abnormal accumulation detection module includes: A first training set acquisition submodule is configured to acquire a first training set of the vehicle detection model, and the first training set includes sales data and vehicle condition data of a plurality of vehicles; An abnormal accumulation vehicle identification submodule is configured to select a vehicle identified as abnormal accumulation from the sales data and vehicle condition data of the vehicle; An abnormal accumulation vehicle marking submodule is configured to mark sales feature information and vehicle condition feature information corresponding to the vehicle identified as abnormal accumulation in the first training set; A vehicle classification submodule is configured to classify sales feature information and vehicle condition feature information of a vehicle not marked in the first training set after marking the vehicle identified as abnormal accumulation to obtain all vehicle sales feature information and vehicle condition feature information under abnormal accumulation classification; The second training set obtaining submodule is configured to take all vehicle sales feature information and vehicle condition feature information under the abnormal accumulation classification as a second training set. The vehicle detection submodule is configured to train a vehicle detection model by using the second training set, and test the vehicle detection model by using the vehicle condition data in the first training set to obtain a trained preset vehicle detection model.

9. A vehicle characterized by comprising: The computer program is stored on the computer readable storage medium and is executed by the processor to implement the steps of the method according to any one of claims 1-7. The computer program is stored on the computer readable storage medium and is executed by the processor to implement the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, ​

Images