Car owner information intelligent identification system for car washing service

Abstract

The invention relates to the technical field of information identification, and discloses a car owner information intelligent identification system for car washing service. Comprising a data query module for extracting comprehensive vehicle owner information from available data, a first portrait fusion module for fusing user tags into user vehicle washing portraits, a second portrait construction module for constructing real-time service portraits with remark paraphrases, and a package recommendation module for recommending service packages to a user side in sequence. According to the method and the device, the historical car washing service information and the current individual demand service information of the user can be subjected to fusion operation by constructing the car washing portrait and the real-time service portrait of the user, so that the dual portrait construction effect of the user in the past time period and the current time period is realized; the problem that vehicle owner information comprehensiveness is limited during one-time portrait construction is avoided, and it is ensured that the screened service package can keep a high matching association effect with the vehicle owner information.

Description

Technical Field

[0001] The present invention relates to the technical field of information recognition, and more specifically, to an intelligent recognition system for vehicle owner information for car wash services. Background Art

[0002] With the continuous growth of the automobile ownership and the continuous development of the Internet, the operation of integrating car wash service information online and providing reasonable car wash service recommendations to vehicle owners has been gradually accepted by users. In order to ensure that the car wash service platform can recommend reasonable and accurate car wash service information to users, it is necessary to intelligently identify the vehicle owner information and recommend corresponding car wash service packages according to the recognition results.

[0003] The patent application with the publication number CN112579911A discloses a method and system for recommending vehicle car wash information. By determining the license plate information and body color information from the obtained image containing the target vehicle information, then determining the recognition degrees respectively according to the license plate information and body color information and the corresponding new vehicle information, and then judging whether the target vehicle needs car maintenance service according to whether the total recognition degree of the two is greater than a first preset value. If it is not greater than the first preset value, when it is determined that the target vehicle is in a state that needs car maintenance service, then match the corresponding car maintenance service place according to the location of the target vehicle, and push the name and location of the place to the target vehicle, so that the vehicle can be maintained in time, thereby reducing the impact on the service life of the vehicle to a certain extent; When the existing recognition system recognizes the vehicle owner information on the user side, it collects multi-dimensional car wash service data of the vehicle owner in the historical past time period, and constructs a user portrait after analyzing the collected car wash service data. Based on the user portrait, the required car wash service package is screened. However, this method still has deficiencies. Since the user portrait is constructed based on multi-dimensional data in the historical past time period, the user portrait can only summarize and represent the vehicle owner information in the past time period, and cannot effectively combine the personalized car wash needs of the vehicle owner in the current time period, resulting in the problem of data time period limitations in the way of constructing a user portrait once, leading to a low degree of association and matching between the subsequent car wash service package and the vehicle owner information, and thus reducing the screening accuracy of the car wash service package.

[0004] In view of this, the present invention proposes an intelligent recognition system for vehicle owner information for car wash services to solve the above problems. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solution: An intelligent recognition system for vehicle owner information for car wash services, which is applied to a car wash service platform and includes: A data query module, which is used to receive service requests from users within the user terminal, query available data associated with the user from the database in the request state, and extract comprehensive vehicle owner information from the available data; A first portrait fusion module, which is used to correct anomalies in the comprehensive vehicle owner information, extract user tags from the comprehensively vehicle owner information after anomaly correction, and fuse the user tags into a user car wash portrait; A second portrait construction module, which is used to determine the car wash service cycle, collect personalized demand characteristics of the user terminal within the car wash service cycle, and fuse the personalized demand characteristics with the user car wash portrait to construct a real-time service portrait with note interpretations; A package recommendation module, which is used to screen service packages adapted to the real-time service portrait from the database of the service terminal based on a trained package screening model, calculate the adaptation index of the service package, and recommend service packages to the user terminal in order.

[0006] Furthermore, the method for querying available data is as follows: Query the user surname and user contact information corresponding to the original data in the database one by one, and record them as the original surname and the original contact information; Extract the user surname and user contact information corresponding to the input service request respectively, and record them as the request surname and the request contact information; When the original surname and the original contact information of the original data are respectively consistent with the request surname and the request contact information, record the original data as available data.

[0007] Furthermore, the comprehensive vehicle owner information includes service consumption data, vehicle status data, and service environment data; The service consumption data includes the number of car washes, car wash time, service package, consumption amount, and discount amount; The vehicle status data includes vehicle type and the number of paint damages; The service environment data includes weather type and environmental temperature.

[0008] Furthermore, the method for anomaly correction of anomaly data is as follows: Query the recording times of A service consumption data one by one through timestamps, and arrange the A service consumption data in order of time; After summarizing the A vehicle status data and the A service environment data that are at the same recording time as the A service consumption data respectively, generate A data packets; Take the fact that there are no duplicate and missing phenomena in the service consumption data, vehicle status data, and service environment data within the data packet as the standard, and conduct anomaly inspections on the A data packets in turn; The data packets with missing and duplicate phenomena are recorded as abnormal data, and the data corresponding to the missing and duplicate phenomena are respectively recorded as missing data and duplicate data; Eliminate the duplicate data in the abnormal data, trace the original data corresponding to the missing data from the database, and import the traced original data into the missing positions of the abnormal data to obtain the corrected comprehensive vehicle owner information.

[0009] Furthermore, user tags include car wash frequency value, package preference, price sensitivity, vehicle health, and service environment attributes; The method for fusing user car wash portraits is as follows: Construct a basic portrait with a circular contour having three inner and outer wrapped distributions, and sequentially record the three circular contours as the environment unit, vehicle unit, and user unit from the inside out; Import the service environment attributes into the environment unit, import the vehicle health into the vehicle unit, and import the car wash frequency value, package preference, and price sensitivity into the user unit; Establish data channels between the environment unit and the vehicle unit, and between the vehicle unit and the user unit respectively, so that the basic portrait is converted into a user car wash portrait.

[0010] Furthermore, the method for determining the car wash service cycle is as follows: Query all the original data at the user end through the database, identify the data attributes of the original data one by one, and record the original data with data attributes of access and browsing as valid data; Query the recording times of all the valid data one by one through the timestamp, and in the order of time, record the last recording time as the calibration time, and import the calibration time and the recording times of the valid data onto the same timeline; Record the first recording time after the calibration time as the cycle start point, record the current time as the cycle end point, and record the time period between the cycle start point and the cycle end point as the car wash service cycle.

[0011] Furthermore, personalized demand characteristics include type characteristics, price characteristics, and preferential characteristics; The collection methods for type characteristics, price characteristics, and preferential characteristics are as follows: Identify the text groups and digital groups of B valid data in the car wash service cycle one by one through natural language processing technology; Record the valid data with text groups of fine wash, general wash, and rough wash as type data, record the valid data with text groups of member and non-member as price data, and record the valid data with text groups of coupon, full reduction, and discount as preferential data; Mark the viewing moments of type data, price data, and preferential data one by one through timestamps. After accumulating all the viewing moments, obtain the first duration, the second duration, and the third duration respectively; Combine the first words and numbers corresponding to the maximum value of the first duration, the maximum value of the second duration, and the maximum value of the third duration at the beginning and end respectively to generate a type feature, a price feature, and a preferential feature.

[0012] Furthermore, the construction method of the real-time service portrait is as follows: Set three feature frames at equal angles on the user unit of the user car wash portrait. After sequentially assigning the numbers 1, 2, and 3 to the three feature frames, generate the 1st feature frame, the 2nd feature frame, and the 3rd feature frame; Import the type feature, the price feature, and the preferential feature into the 1st feature frame, the 2nd feature frame, and the 3rd feature frame respectively to generate a feature interpretation; Establish two interpretation positions between the 1st feature frame and the 2nd feature frame, and between the 2nd feature frame and the 3rd feature frame respectively. Import the user's surname and user contact information into the two interpretation positions respectively to generate an identity interpretation, and construct a real-time service portrait with a note interpretation.

[0013] Furthermore, the training method of the package screening model is as follows: Pre-collect multiple groups of real-time service portraits and corresponding service packages, and summarize the car wash frequency values, package preferences, price sensitivities, vehicle health conditions, service environment attributes, type features, price features, and preferential features in the real-time service portraits and mark them as feature vectors to obtain multiple groups of feature vectors; Convert the service packages into labels corresponding to the feature vectors. One feature vector corresponds to one label to form a group of training data. Multiple groups of training data form a training set, and the labeled training features are divided into a training set and a test set. Use the training set to train the package screening model, and use the test set to test the package screening model. Preset an error threshold. When the mean of the prediction errors of all training features in the test set is less than the preset error threshold, obtain the package screening model.

[0014] Furthermore, the calculation method of the adaptation index is as follows: Use the word segmentation technology to split the service items in D service packages one by one, and compare the service items with the real-time service portrait one by one; Record the service items that overlap with the package preference, price sensitivity, vehicle health condition, service environment attribute, type feature, price feature, and preferential feature as valid items, and count the number of valid items to obtain D valid values; Compare the D valid values with the total number of service items in the D service packages respectively to obtain D valid ratios; Query the package amount and actual payment amount of D service packages one by one, calculate the difference between the package amount and the actual payment amount, and then compare them to calculate D preferential ratios; After assigning the D valid ratios and D preferential ratios to the corresponding proportionality coefficients respectively and adding them up, D adaptation indexes are obtained.

[0015] Technical effects and advantages of an intelligent identification system for vehicle owner information for car wash services according to the present invention: By extracting comprehensive vehicle owner information, the present invention can collect multi-dimensional comprehensive car wash service information of users in historical past periods, expand the data dimension for intelligent identification of vehicle owner information, and by correcting anomalies in the comprehensive vehicle owner information, abnormal data with missing and duplicate phenomena in the comprehensive vehicle owner information can be eliminated and supplemented, improving the integrity and uniqueness of the comprehensive vehicle owner information. And by constructing user car wash portraits and real-time service portraits, the historical car wash service information and current personalized demand service information of users can be integrated into one, thus achieving the double construction effect of user portraits in the past and current time periods, avoiding the problem of limitations in the comprehensiveness of vehicle owner information when constructing portraits once, greatly improving the accuracy of intelligent identification of vehicle owner information, and combined with the intelligent screening process of the package screening model, it can quickly and accurately screen out service packages that match and are associated with the vehicle owner information, ensuring that the selected service packages can maintain a high matching and association effect with the vehicle owner information, and providing reasonable and accurate recommendation results for users' subsequent car wash services. Brief Description of the Drawings

[0016] Figure 1 It is a schematic structural diagram of an intelligent identification system for vehicle owner information for car wash services provided in Embodiment 1 of the present invention; Figure 2 It is a schematic flow diagram of an intelligent identification method for vehicle owner information for car wash services provided in Embodiment 2 of the present invention. Detailed Embodiments

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0018] Embodiment 1: Please refer to Figure 1 As shown, an intelligent identification system for vehicle owner information for car wash services described in this embodiment is applied to a car wash service platform and includes: The data query module receives the service requests of users in the user terminal. In the request state, it queries the available data associated with the users from the database and extracts the comprehensive vehicle owner information from the available data. The comprehensive vehicle owner information includes service consumption data, vehicle status data, and service environment data. The user terminal refers to the client port that can provide data input, data feedback, and result reception for users, used to achieve the two-way data interaction effect between users and the car wash service platform. Specifically, the user terminal includes, but is not limited to, ports such as the APP installed on the user's mobile phone, WeChat official account, and WeChat mini-program.

[0019] The service request refers to the request data containing information related to car wash requirements sent by the user terminal, which can be used as the data identification basis for the user to send car wash requirements to the car wash service platform. Specifically, the service request includes, but is not limited to, car wash service item query requirements, car wash service price query requests, car wash service discount query requirements, etc.; when inputting the user request to the user terminal, the user can input it through methods such as text editing, voice input, and project button selection.

[0020] The request state refers to the real-time working state of the car wash service platform after receiving the service request of the user, which can distinguish the current working state and mode of the car wash service platform, ensuring that the car wash service platform is in the real-time open request state only after receiving the service request from the user terminal. This enables the car wash service platform to be turned on when needed and remain silent or closed at other times, which can not only avoid the risk of increased data information leakage due to the long-term open state of the user terminal but also prevent the car wash service platform from sending pop-up advertisements to the user terminal irregularly, causing unnecessary negative interference to the user.

[0021] When the car wash service platform is in the request state, it is necessary to query the diverse original data associated with the users from the database at this time. The diverse original data is recorded as available data, and all car wash service situations of the users saved and recorded in the car wash service platform are represented by the available data. When judging whether the data is associated with the user, it is necessary to judge the consistency between the attributes of the original data in the database and the attributes of the user; the attribute is a specific representation of the corresponding type of data. Exemplarily, the attributes include the user's surname, user contact information, user address, user vehicle model, etc. Specifically, query the user's surname and user contact information corresponding to the original data in the database one by one, and record them as the original surname and original contact information. Extract the user's surname and user contact information corresponding to the input service request respectively, and record them as the request surname and request contact information. When the original surname and the original contact information are respectively consistent with the requested surname and the requested contact information, it indicates that the original data recorded and saved in the database is associated with the user. Then, the original data with the original surname and the original contact information being respectively consistent with the requested surname and the requested contact information is recorded as available data.

[0022] The comprehensive vehicle owner information refers to the data in the available data that can represent the car wash service situation generated by the user within the historical time period. Specifically, the comprehensive vehicle owner information includes service consumption data, vehicle status data, and service environment data. The service consumption data refers to the specific representation of the relevant dynamic situations of the user during the car wash service process in the available data, and it is the data exclusive to the user personal in the vehicle owner information. Specifically, the service consumption data includes the number of car washes, car wash time, service package, consumption amount, and discount amount. Among them, the number of car washes refers to the total number of all car wash services of the user in the past time period recorded and saved in the car wash service platform. The car wash time refers to the specific time corresponding to each car wash of the user recorded and saved in the car wash service platform. The service package refers to the type of car wash package corresponding to each car wash of the user recorded and saved in the car wash service platform. The consumption amount refers to the actual payment price corresponding to each car wash of the user recorded and saved in the car wash service platform. The discount amount refers to the discount rate corresponding to each car wash of the user recorded and saved in the car wash service platform.

[0023] The vehicle status data refers to the specific data in the available data that can represent the relevant supply platform situation of the user's vehicle during the car wash service process, and it is the data exclusive to the user's vehicle in the vehicle owner information. The vehicle status data includes the vehicle type and the number of paint surface damages. Among them, the vehicle type refers to the specific type of the vehicle driven by the user recorded and saved in the car wash service platform. The vehicle type includes, but is not limited to, sedans, urban SUVs, small trucks, etc. The number of paint surface damages refers to the number of damaged positions on the vehicle's exterior paint surface when the user washes the car each time, recorded and saved in the car wash service platform.

[0024] The service environment data refers to the specific data in the available data that can represent the external environment status when the user washes the car each time. The service environment data includes the weather type and the environmental temperature. Among them, the weather type refers to the weather conditions corresponding to each car wash of the user recorded and saved in the car wash service platform. The weather type includes, but is not limited to, sunny days, rainy days, snowy days, etc. The environmental temperature refers to the external environmental temperature corresponding to each car wash of the user recorded and saved in the car wash service platform.

[0025] The first image fusion module corrects anomalies in the comprehensive vehicle owner information, extracts user tags from the corrected comprehensive vehicle owner information, and fuses the user tags into a user car wash image; After obtaining the comprehensive vehicle owner information, it is necessary to ensure the integrity, accuracy, and reasonableness of the comprehensive vehicle owner information at this time, and avoid abnormal data such as abnormal numerical sizes and duplicate data types in the comprehensive vehicle owner information. Therefore, it is necessary to perform anomaly correction operations on the comprehensive vehicle owner information, that is, operations such as eliminating, supplementing, and arranging the abnormal data existing in the comprehensive vehicle owner information; In order to ensure the pertinence and accuracy of the anomaly correction operation, it is necessary to first identify the service consumption data, vehicle status data, and service environment data in the comprehensive vehicle owner information respectively, and perform anomaly correction operations on the existing abnormal data according to the identified results; The anomaly correction method for abnormal data is: Query the recording times of A service consumption data one by one through the time stamp, and arrange the A service consumption data in chronological order; After summarizing the A vehicle status data and A service environment data that are at the same recording time as the A service consumption data respectively, generate A data packets; A data packet is a representation for summarizing different types of available data, so that there is exactly one unique service consumption data, vehicle status data, and service environment data in one data packet; Taking the fact that there are no duplicate and missing phenomena in the service consumption data, vehicle status data, and service environment data in the data packet as the standard, check for anomalies in the A data packets in turn; Mark the data packets with missing and duplicate phenomena as abnormal data, and mark the corresponding data of the missing and duplicate phenomena as missing data and duplicate data respectively; Eliminate the duplicate data in the abnormal data, trace the original data corresponding to the missing data from the database, and import the traced original data into the missing position of the abnormal data to obtain the corrected comprehensive vehicle owner information. The missing position refers to the specific position of the missing data in the data packet corresponding to the abnormal data, and is used as the supplementary import position limit for the subsequently traced original data.

[0026] It should be noted that the missing data refers to all data that do not include service consumption data, vehicle status data, and service environment data in the data packet, and the duplicate data refers to two identical service consumption data, vehicle status data, and service environment data existing in the data packet. When tracing and supplementing data from the database, it is necessary to take the recording time corresponding to the data packet as the standard, and assist in analyzing other service consumption data, vehicle status data, and service environment data in the data packet, so as to judge the specific data corresponding to the missing data, and thus achieve the integrity of each different type of data in the comprehensive vehicle owner information.

[0027] After the abnormal correction of the comprehensive vehicle owner information, it is necessary to extract user tags from the corrected comprehensive vehicle owner information, so that the user tags can intuitively, concisely and accurately represent the car washing service conditions recorded and saved by the user in the car washing service platform; User tags include car washing frequency value, package preference, price sensitivity, vehicle health, and service environment attributes; Specifically, the car washing frequency value refers to the car washing frequency of the user in the comprehensive vehicle owner information within a unit time, which can represent the frequency of the user's car washing within a unit time; in this embodiment, the moment when the first service consumption data is generated is recorded as the starting moment, the moment when the last service consumption data is generated is recorded as the ending moment, and after comparing the number of car washes with the duration between the starting moment and the ending moment, the car washing frequency value is obtained.

[0028] The package preference refers to the service package with the largest number of times when the user washes the car in the comprehensive vehicle owner information, which can represent the user's preference degree for the specific type of car washing; in this embodiment, the package preference is obtained by counting the quantities corresponding to all service packages, and the service package corresponding to the maximum quantity is recorded as the package preference.

[0029] The price sensitivity refers to the proportion of the number of times the user uses car washing coupons when washing the car in the comprehensive vehicle owner information, which can represent the degree of the user's sensitivity to the car washing price; in this embodiment, the price sensitivity is obtained by counting the number of times of using coupons when washing the car and comparing the number of times of using coupons with the number of car washes.

[0030] The vehicle health refers to the representation of the severity of the damage to the vehicle appearance when the user washes the car in the comprehensive vehicle owner information, which can represent the health degree of the vehicle's exterior paint; in this embodiment, it is obtained by counting the number of damages to the vehicle paint during each car wash and taking the average value.

[0031] The service environment attribute refers to the representation of the user's preference for the external environment when washing the car in the comprehensive vehicle owner information, which can represent the type of weather environment when the user washes the car; in this embodiment, the service environment attribute is obtained by querying the weather type and environmental temperature during each car wash, combining them into different service environment attributes, counting the quantities corresponding to each service environment attribute, and selecting the service environment attribute corresponding to the maximum quantity; for example, when the weather type is sunny and the environmental temperature is 30 degrees Celsius, the service environment attribute formed at this time is "sunny: 30".

[0032] After obtaining the user tags, the user tags at this time can comprehensively represent the user's car washing situation in the historical past time period, and through the orderly integration of the user tags, a user car washing portrait that conforms to the user's past car washing performance is constructed to ensure that the user car washing portrait can comprehensively represent the user's car washing personal preferences, car washing vehicle conditions, and car washing environmental factors; The fusion method of the user car washing portrait is as follows: Construct a basic portrait with three annular contours distributed inside and outside, and in the order from inside to outside, the three annular contours are sequentially recorded as the environmental unit, the vehicle unit, and the user unit; the basic portrait refers to a blank portrait without any substantial content, and the annular contour is a component unit used to construct the basic portrait, and can also provide an independent position limit for the subsequent import of user tags. Combining the structure of the annular contour with internal and external wrapping distribution can ensure the independence of each annular contour and prevent the phenomenon of mutual cross-interference between the data in different annular contours; Import the service environment attributes into the environmental unit, import the vehicle health into the vehicle unit, and import the car washing frequency value, package preference, and price sensitivity into the user unit; Establish data channels between the environmental unit and the vehicle unit, and between the vehicle unit and the user unit respectively, so that the basic portrait is converted into a user car washing portrait. The data channel is a representation used to provide an associated data transmission channel between two adjacent units, so as to keep the user tags between two adjacent units in a dynamically associated state.

[0033] It should be noted that the user car washing portrait is not immutable. The user car washing portrait can update and replace the data in the environmental unit, vehicle unit, and user unit in the user car washing portrait in real time according to the different car washing frequency values, package preferences, price sensitivities, vehicle health, and service environment attributes generated by the user in the car washing service platform in real time, so as to ensure that the user car washing portrait can be consistent with the real-time car washing service related data of the user, thereby avoiding the negative impact brought by the untimely update of the user car washing portrait data.

[0034] The second portrait construction module determines the car washing service cycle, collects the personalized demand characteristics of the user end during the car washing service cycle, and integrates the personalized demand characteristics with the user car washing portrait to construct a real-time service portrait with note interpretations; The car washing service cycle refers to the maximum time span for recording and storing the user-related data of the user end in the car washing service platform, that is, it can provide a time limit for collecting the relevant data generated and recorded by the user in the car washing service platform, so as to play a time limit role for the subsequent collection of personalized demand characteristics; The determination method of the car washing service cycle is as follows: Query all the original data of the client through the database, identify the data attributes of the original data one by one, and record the original data with the data attributes of access and browsing as valid data; Query the recording times of all valid data one by one through timestamps. In the order of time, record the last recording time as the calibration time, and import the calibration time and the recording times of valid data onto the same timeline; Record the first recording time after the calibration time as the cycle start point, record the current time as the cycle end point, and record the time period between the cycle start point and the cycle end point as the car wash service cycle.

[0035] After determining the car wash service cycle, it is necessary to collect the relevant car wash service information that the user has concerned and accessed during the car wash service cycle, so as to simply and accurately represent the personalized car wash needs of the user during the car wash service cycle; The personalized demand characteristics include type characteristics, price characteristics, and preferential characteristics; specifically, the type characteristics are used to represent the car wash package service types that the user is concerned about during the car wash service cycle, the price characteristics are used to represent the actual paid prices of the car wash packages that the user is concerned about during the car wash service cycle, and the preferential characteristics are used to represent the preferential intensities of the car wash packages that the user is concerned about during the car wash service cycle; The collection methods of type characteristics, price characteristics, and preferential characteristics are as follows: Use natural language processing technology to identify the text groups and digital groups of B valid data in the car wash service cycle one by one; the text groups and digital groups are respectively used to simply represent the text part and the digital part of each service item in the valid data, and serve as the smallest components for constructing type characteristics, price characteristics, and preferential characteristics later; Record the valid data with the text groups of fine wash, general wash, and rough wash as type data, record the valid data with the text groups of member and non-member as price data, and record the valid data with the text groups of coupon, full reduction, and discount as preferential data; Mark the access and browsing times of type data, price data, and preferential data one by one through timestamps. After accumulating all the access and browsing times, obtain the first duration, the second duration, and the third duration respectively; Combine the head and tail of the text group and digital group corresponding to the maximum value of the first duration to generate type characteristics; Combine the head and tail of the text group and digital group corresponding to the maximum value of the second duration to generate price characteristics; Combine the head and tail of the text group and digital group corresponding to the maximum value of the third duration to generate preferential characteristics.

[0036] After obtaining the type feature, price feature, and preferential feature, the type feature, price feature, and preferential feature can accurately represent relevant data such as the type of car wash service package that the user has been concerned about and interested in recently, and serve as the construction basis for comprehensively representing the user's car wash service needs subsequently, so as to construct a real-time service portrait that can represent the information related to the car wash service that the user has been concerned about and interested in recently; When constructing the real-time service portrait, it is necessary to effectively integrate the type feature, price feature, and preferential feature with the user's car wash portrait, so that the user's car wash portrait can represent the user's car wash preferences in the historical past time period, and the type feature, price feature, and preferential feature can represent the user's recent car wash attention preferences, thereby improving the personalization and accuracy of the real-time service portrait.

[0037] The note interpretation is a specific representation for comprehensively noting the user's personalized demand characteristics, user contact information, and user surname, that is, it can accurately and comprehensively display the actual situation of each user on the user side; specifically, the note interpretation includes feature interpretation and identity interpretation; among them, the feature interpretation is used to represent the user's personalized demand characteristics, and the identity interpretation is used to represent the user's car owner identity information; The construction method of the real-time service portrait is as follows: Set three feature frames at equal angles on the user unit of the user's car wash portrait. After sequentially numbering the three feature frames as 1, 2, and 3, generate the 1st feature frame, 2nd feature frame, and 3rd feature frame; the feature frame is used to provide a position limit for importing personalized demand characteristics, ensuring that the type feature, price feature, and preferential feature can be associated and integrated with the user's car wash portrait in an orderly and independent manner; Import the type feature, price feature, and preferential feature into the 1st feature frame, 2nd feature frame, and 3rd feature frame respectively to generate feature interpretation; Establish two interpretation positions between the 1st feature frame and the 2nd feature frame, and between the 2nd feature frame and the 3rd feature frame respectively, and import the user's surname and user contact information into the two interpretation positions respectively to generate identity interpretation, and construct a real-time service portrait with note interpretation. The interpretation position is used to provide a position limit for importing the user's surname and user contact information, and to provide an intuitive and accurate display effect for the car owner identity information corresponding to the real-time service portrait.

[0038] The package recommendation module, based on the trained package screening model, screens out the service packages adapted to the real-time service portrait from the database of the service side, calculates the adaptation index of the service packages, and recommends the service packages to the user side in order; The server refers to a customer port that provides users with car wash service packages and other related data inquiries, and is used to achieve two-way data interaction between merchants and the car wash service platform. Specifically, the server includes but is not limited to APPs, WeChat public accounts, and WeChat mini-programs installed by merchants on mobile phones.

[0039] A service package refers to a related car wash service package stored in the server that can be associated with and adapted to the real-time service profile, and is used as a package for subsequent car wash service recommendations to users, so that the service package can meet the user's car wash service needs as much as possible; the target package is obtained by collecting a large number of different real-time service profiles in history and selecting the car wash packages.

[0040] When screening service packages, a pre-trained package screening model can be used for intelligent identification and screening, enabling the package screening model to accurately screen service packages based on real-time service profiles. The training method of the package screening model is: Collect multiple sets of real-time service profiles and corresponding service packages in advance, and aggregate the car wash frequency values, package preferences, price sensitivity, vehicle health, service environment attributes, type characteristics, price characteristics, and discount characteristics in the real-time service profiles and mark them as feature vectors to obtain multiple sets of feature vectors; The service packages are converted into labels corresponding to feature vectors. One feature vector corresponds to one label, forming a set of training data. Multiple sets of training data constitute a training set. The labeled training features are then divided into a training set and a test set. Use 70% of the training features as the training set and 30% of the training features as the test set. Use the training set to train the package screening model, and use the test set to test the package screening model. Preset the error threshold. When the mean of the prediction errors of all training features in the test set is less than the preset error threshold, the package screening model is obtained.

[0041] Exemplarily, the package screening model adopts either a support vector machine model or a random forest model; the preset error threshold is pre-set according to the actual accuracy required by the package screening model.

[0042] After the package screening model is trained, the real-time service profile obtained can be input into the package screening model, and then the service package that is compatible with the real-time service profile can be screened out from the server's database.

[0043] Since the number of filtered service packages is not unique, and the degree of matching between each service package and the real-time service profile varies, it is necessary to numerically calculate the degree of matching between each service package and the real-time service profile, and numerically represent the matching degree through the adaptation index; The calculation method of the adaptation index is as follows: The service items in D service packages are split one by one through the word segmentation technology, and the service items are compared with the real-time service portrait one by one for coincidence; The service items that are repeated with the package preference, price sensitivity, vehicle health, service environment attributes, type characteristics, price characteristics, and preferential characteristics are recorded as valid items, and the number of valid items is counted to obtain D valid values; After comparing the D valid values with the total number of service items in the D service packages respectively, D valid ratios are obtained; The expression of the valid ratio is: ; In the formula, is the valid ratio of the th service package, = 1, 2,..., D, is the valid value of the th service package, is the total number of service items of the th service package; The package amount and the actual payment amount of the D service packages are queried one by one, and after subtracting the package amount and the actual payment amount and then comparing, D preferential ratios are calculated; The expression of the preferential ratio is: ; In the formula, is the preferential ratio of the th service package, is the package amount of the th service package, is the actual payment amount of the th service package; The D valid ratios and the D preferential ratios are respectively assigned corresponding proportional coefficients and then added together to obtain D adaptation indexes; The expression of the adaptation index is: ; In the formula, is the adaptation index of the th service package, , are respectively the proportional coefficients of the valid ratio and the preferential ratio, and , are both greater than 0.

[0044] After calculating the adaptation indexes of all service packages, at this time, the adaptation degree between the service packages and the user's car washing service needs can be represented according to the size of the adaptation indexes, and used as the basis for recommending service packages to users in the future; Specifically, when the adaptation index is larger, the degree of adaptation between the service package and the real-time service profile is higher, and the degree of correlation between the service package and the user's actual car wash service needs is higher, and vice versa. Therefore, when recommending service packages to users, D service packages are recommended to users in order according to the adaptation index from large to small, so as to ensure that the car wash service platform can recommend service packages to users with a degree of correlation with the car owner's information from high to low, and help users accurately, quickly and reasonably choose service packages that meet their own car wash service needs.

[0045] In this embodiment, by extracting comprehensive car owner information, the comprehensive car wash service information of the user in the past historical time periods can be collected in multiple dimensions, which expands the data dimension of the car owner information for intelligent identification. By correcting the abnormalities of the comprehensive car owner information, the abnormal data with missing and repeated phenomena in the comprehensive car owner information can be eliminated and supplemented, thereby improving the integrity and uniqueness of the comprehensive car owner information. By constructing the user car wash portrait and the real-time service portrait, the user's historical car wash service information and the current personalized demand service information can be combined into one, thereby achieving the dual construction effect of the user's portrait in the past time period and the current time period, avoiding the problem of limitations in the comprehensiveness of the car owner information when constructing a portrait at one time, greatly improving the accuracy of intelligent identification of the car owner information, and combined with the intelligent screening processing of the package screening model, it can quickly and accurately screen out service packages that match and are associated with the car owner information, ensuring that the screened service packages can maintain a high matching and correlation effect with the car owner information, and providing reasonable and accurate recommendation results for the user's subsequent car wash services.

[0046] Example 2: Please refer to Figure 2 As shown, for parts not described in detail in this embodiment, please refer to the description of the first embodiment. A method for intelligently identifying vehicle owner information for a car wash service is provided, which is applied to a car wash service platform and is implemented based on an intelligent vehicle owner information identification system for a car wash service, including: S1: Receive a service request from a user in the user terminal. In the request state, query the available data associated with the user from the database and extract comprehensive vehicle owner information from the available data; S2: Correct the abnormal data in the comprehensive car owner information, extract the user label from the abnormally corrected comprehensive car owner information, and fuse the user label into the user's car washing profile; S3: Determine the car wash service cycle, collect the user's individual demand characteristics during the car wash service cycle, and integrate the individual demand characteristics with the user's car wash profile to construct a real-time service profile with annotations; S4: Based on the trained service package screening model, screen out the service packages that match the real-time service profile from the database of the server, calculate the adaptation index of the service packages, and recommend the service packages to the user side in order.

[0047] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. An intelligent vehicle owner information recognition system for car wash services, which is applied to a car wash service platform, and is characterized in that, Including: A data query module, which is used to receive service requests from users in the user terminal. In the request state, it queries the available data associated with the user from the database and extracts comprehensive vehicle owner information from the available data; A first portrait fusion module, which is used to correct anomalies in the comprehensive vehicle owner information, extract user tags from the comprehensively corrected comprehensive vehicle owner information, and fuse the user tags into a user car wash portrait; A second portrait construction module, which is used to determine the car wash service cycle, collect the personalized demand characteristics of the user terminal within the car wash service cycle, and fuse the personalized demand characteristics with the user car wash portrait to construct a real-time service portrait with note interpretations; A package recommendation module, which is used to screen out service packages adapted to the real-time service portrait from the database of the service terminal based on the trained package screening model, calculate the adaptation index of the service package, and recommend service packages to the user terminal in order; 2. The intelligent identification system for vehicle owner information used in car wash service according to claim 1, characterized in that, The query method for available data is as follows: Query the user surname and user contact information corresponding to the original data in the database one by one, and record them as the original surname and the original contact information; Extract the user surname and user contact information corresponding to the input service request respectively, and record them as the request surname and the request contact information; When the original surname and the original contact information of the original data are respectively consistent with the request surname and the request contact information, record the original data as available data.

3. The intelligent vehicle owner information recognition system for car wash service according to claim 2, wherein The comprehensive vehicle owner information includes service consumption data, vehicle status data, and service environment data; The service consumption data includes the number of car washes, car wash time, service package, consumption amount, and discount amount; The vehicle status data includes vehicle type and the number of paint damages; The service environment data includes weather type and environmental temperature.

4. The intelligent vehicle owner information recognition system for car wash service according to claim 3, characterized in that, The anomaly correction method for anomaly data is as follows: Query the recording times of A service consumption data one by one through timestamps, and arrange the A service consumption data in sequence according to the chronological order; After summarizing the A vehicle status data and the A service environment data at the same recording time as the A service consumption data respectively, generate A data packets; Taking the fact that there are no duplicate and missing phenomena in the service consumption data, vehicle status data, and service environment data in the data packet as the standard, check the anomalies of the A data packets in sequence; Record the data packets with missing and duplicate phenomena as anomaly data, and record the corresponding data of the missing and duplicate phenomena as missing data and duplicate data respectively; Delete the duplicate data in the anomaly data, trace the original data corresponding to the missing data from the database, and import the traced original data into the missing position of the anomaly data to obtain the corrected comprehensive vehicle owner information.

5. The intelligent vehicle owner information recognition system for car wash service according to claim 4, wherein User tags include car wash frequency value, package preference, price sensitivity, vehicle health, and service environment attributes; The fusion method of the user car wash portrait is as follows: Construct a basic portrait with a circular contour with three inner and outer wrapped distributions, and record the three circular contours as the environment unit, vehicle unit, and user unit in sequence from the inside out; Import the service environment attributes into the environment unit, import the vehicle health into the vehicle unit, and import the car wash frequency value, package preference, and price sensitivity into the user unit; A data channel is established between the environment unit and the vehicle unit, and between the vehicle unit and the user unit respectively, so as to promote the conversion of the basic portrait into the user car wash portrait.

6. The intelligent vehicle owner information recognition system for car wash service according to claim 5, characterized in that, The method for determining the car wash service cycle is as follows: All the original data at the user end is queried through the database, the data attributes of the original data are identified one by one, and the original data with the data attributes of access and browsing is recorded as valid data; The recording times of all the valid data are queried one by one through the time stamp. In the order of time, the recording time at the last position is recorded as the calibration time, and the calibration time and the recording time of the valid data are imported onto the same time line; The first recording time after the calibration time is recorded as the cycle start point, the current time is recorded as the cycle end point, and the time period between the cycle start point and the cycle end point is recorded as the car wash service cycle.

7. An intelligent vehicle owner information recognition system for car wash services according to claim 6, characterized in that, The personalized demand characteristics include type characteristics, price characteristics and preferential characteristics; The acquisition methods of the type characteristics, price characteristics and preferential characteristics are as follows: The text groups and digital groups of B valid data in the car wash service cycle are identified one by one through natural language processing technology; The valid data with the text groups of fine wash, general wash and rough wash are recorded as type data, the valid data with the text groups of member and non-member are recorded as price data, and the valid data with the text groups of coupon, full reduction and discount are recorded as preferential data; The access and browsing times of the type data, price data and preferential data are marked one by one through the time stamp. After adding up all the access and browsing times, the first duration, the second duration and the third duration are obtained respectively; The text groups and digital groups corresponding to the maximum values of the first duration, the second duration and the third duration are combined at the head and tail to generate the type characteristics, price characteristics and preferential characteristics respectively.

8. The intelligent vehicle owner information recognition system for car wash service according to claim 7, characterized in that The construction method of the real-time service portrait is as follows: Three feature frames are set at equal angles on the user unit of the user car wash portrait. After the three feature frames are sequentially assigned the numbers 1, 2 and 3, the 1st feature frame, the 2nd feature frame and the 3rd feature frame are generated; The type characteristics, price characteristics and preferential characteristics are respectively imported into the 1st feature frame, the 2nd feature frame and the 3rd feature frame to generate feature interpretations; Two interpretation positions are established between the 1st feature frame and the 2nd feature frame, and between the 2nd feature frame and the 3rd feature frame respectively. The user's surname and user contact information are respectively imported into the two interpretation positions to generate identity interpretations, and a real-time service portrait with note interpretations is constructed.

9. The intelligent vehicle owner information recognition system for car wash service according to claim 8, characterized in that, The training method of the package screening model is as follows: Multiple groups of real-time service portraits and corresponding service packages are collected in advance, and the car wash frequency values, package preferences, price sensitivities, vehicle health conditions, service environment attributes, type characteristics, price characteristics and preferential characteristics in the real-time service portraits are summarized and marked as feature vectors to obtain multiple groups of feature vectors; Convert the service packages into labels corresponding to the feature vectors, with one feature vector corresponding to one label, to form a set of training data. Multiple sets of training data constitute a training set, and the labeled training features are divided into a training set and a test set. Use the training set to train the package screening model, and use the test set to test the package screening model. Preset an error threshold. When the mean of the prediction errors of all training features in the test set is less than the preset error threshold, obtain the package screening model.

10. The intelligent identification system for vehicle owner information used in car wash services according to claim 9, characterized in that, The calculation method of the adaptation index is as follows: Use the word segmentation technology to split the service items in D service packages one by one, and compare each service item with the real-time service portrait one by one; Record the service items that overlap with the package preference, price sensitivity, vehicle health, service environment attributes, type features, price features, and preferential features as valid items, and count the number of valid items to obtain D valid values; After comparing the D valid values with the total number of service items in the D service packages respectively, obtain D valid ratios; Query the package amount and the actual payment amount of the D service packages one by one, and compare them after taking the difference between the package amount and the actual payment amount to calculate D preferential ratios; Add the D valid ratios and the D preferential ratios after assigning corresponding proportional coefficients respectively to obtain D adaptation indexes.

Images