A car wash service management system based on a car wash information platform

Abstract

This invention relates to the field of service management technology, and discloses a car wash service management system based on a car wash information platform. It includes a merchant profile combination module, which combines the distribution data and discount data of available merchants to form merchant profiles; a profile library conversion module, which constructs a profile library; a user request construction module, which constructs user requests with location features; a target merchant filtering module, which filters target merchants from candidate merchants; and a target merchant recommendation module, which recommends target merchants to users in sequence. This invention can identify and associate different dealers with high similarity and overlap within the profile library. By combining primary filtering based on location features and association channels, and secondary filtering based on user requests and candidate merchants, it enables users to achieve dual independent filtering and querying of target merchants on the car wash information platform. This ensures that the filtered target merchants meet the actual needs of users and improves the effectiveness of car wash service management.

Description

Technical Field

[0001] This invention relates to the field of service management technology, and more specifically, to a car wash service management system based on a car wash information platform. Background Technology

[0002] With the continuous growth of car ownership, the traditional offline car wash appointment and telephone inquiry car wash service management model can no longer meet the increasing demand for car wash services. With the continuous development of the Internet, the car wash service management system, which integrates multi-dimensional data such as merchant information, user reviews and geographical location data, has initially realized the online display, inquiry, appointment and service management of car wash services, and has greatly promoted the intelligent, networked and healthy development of car wash service management.

[0003] Reference patent application CN118396723A discloses a method, system, device, and storage medium for pushing car wash service information to users. This includes extracting historical car wash data and corresponding historical condition data from user vehicles, standardizing the data, performing correlation analysis on the historical car wash data and historical condition data to obtain a user habit set, constructing a user profile, predicting the user's car wash time based on a preset ARIMA prediction model and historical car wash data, generating the optimal push time and personalized recommendation strategy, and pushing car wash service information. Through in-depth analysis of user historical car wash data and conditional factors, as well as the construction of personalized tag sets and user profiles, it can provide car wash service recommendations that are extremely close to the user's personal habits and preferences, increasing the attractiveness of the service and user satisfaction.

[0004] Existing car wash service management systems collect users' historical car wash data and current car wash needs to comprehensively describe the user's car wash service characteristics and filter out merchants that match these characteristics. For example, in the aforementioned patent application, a user profile is constructed by analyzing the correlation between historical car wash data and historical conditional data to provide recommendations for the car wash services the user needs. However, the method of filtering by matching profiles at once results in inconsistent correlations between the selected merchants and the user's service needs. It cannot ensure that all selected merchants maintain a high degree of similarity to the user's service needs, thereby reducing the accuracy of recommending merchants to the user.

[0005] In view of this, the present invention proposes a car wash service management system based on a car wash information platform to solve the above problems. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a car wash service management system based on a car wash information platform, applied to the car wash information platform, comprising:

[0007] The merchant profile combination module collects the distribution data and discount data of registered merchants. Based on service management criteria, it filters out available merchants from the registered merchants and combines the distribution data and discount data of available merchants into a merchant profile. The service management criteria are: to remove registered merchants with the first or second abnormal data.

[0008] The image library conversion module constructs an information library with a circular distribution of image units, imports merchant images into the image units, and establishes association channels between the outline points of image units with related features, thereby converting the information library into an image library.

[0009] The user request building module collects the user's real-time location and historical user tags, and fuses the real-time location and historical user tags to build a user request with location features.

[0010] The target merchant filtering module combines location features with associated channels to filter candidate merchants from available merchants, and matches user requests with the profiles of candidate merchants to filter target merchants from the candidate merchants;

[0011] The target merchant recommendation module determines the recommendation priority based on the car wash recommendation parameters of the target merchants and recommends the target merchants to the user in order.

[0012] Furthermore, the distribution data includes store address, business hours, reputation level, service type, and price; the discount data includes coupon type and quantity.

[0013] The available merchant selection methods are as follows:

[0014] Count the number of distributor data and the number of discount data among the registered merchants one by one, and record them as the first value and the second value respectively.

[0015] When the first value is less than 5, the distribution data is recorded as the first abnormal data;

[0016] When the second value is less than 2, the discount data is recorded as the second abnormal data;

[0017] Remove registered merchants that contain either the first or second abnormal data, and record the remaining registered merchants as available merchants, thus obtaining A available merchants.

[0018] Furthermore, the method for combining merchant profiles is as follows:

[0019] Create A profiles with three vertically parallel tag layers, and label the three tag layers as the basic layer, behavior layer and value layer respectively, from bottom to top.

[0020] S label positions are set in the basic layer, behavior layer and value layer respectively, and label dividing lines are drawn between two adjacent label positions;

[0021] Import the store address, business hours and reputation level of A available merchants into the tag positions of A basic layers one by one; import the service type and type price of A available merchants into the tag positions of A behavior layers one by one; and import the coupon type and coupon quantity of A available merchants into the tag positions of A value layers one by one.

[0022] By removing redundant tags from A basic layers, A behavioral layers, and A value layers respectively, A profile outlines are generated into A merchant profiles.

[0023] Furthermore, the method for constructing the information database is as follows:

[0024] Establish a database with A circularly arranged blank data positions, and draw the circular bit outlines outside each of the A blank data positions to generate A image units.

[0025] Mark a point on the position contour line, at a position close to the previous position contour line and at a position close to the next position contour line, and denot it as the first contour point and the second contour point.

[0026] Measure the distance from the first contour point to the previous contour line and the distance from the second contour point to the next contour line one by one, and record them as the first distance value and the second distance value respectively.

[0027] The positions of the first and second contour points are continuously adjusted until both the first and second distance values ​​reach their minimum values, at which point the adjustment stops, thus constructing an information database.

[0028] Furthermore, during the identification of related features, natural language processing technology is used to split the same type of distribution data or discount data within the label positions of the two profile units into two element sets.

[0029] When the number of elements with the same meaning in two sets of elements is greater than or equal to half the total number of elements in either set, the elements with the same meaning in the label positions of the two image units are recorded as associated features.

[0030] Furthermore, the method for converting the image database is as follows:

[0031] Import A merchant profiles one by one into A profile units in the information database. Number the A profile units in ascending order according to the import order. Then, group the two profile units with related features into a unit group to obtain C unit groups.

[0032] In the C unit groups, the second contour point of the image unit with the smaller number is recorded as the channel start point, and the first contour point of the image unit with the larger number is recorded as the channel end point. A bidirectional channel is established between the channel start point and the channel end point to obtain C associated channels.

[0033] The associated features of the C unit groups are annotated on the corresponding associated channels, which enables the information database to be transformed into a profile database.

[0034] Furthermore, historical user tags include service price range, service demand type, and lower limit of service reputation;

[0035] The method for constructing a user request is as follows:

[0036] Create a base request with three blank request bits, and label the three blank request bits as the first request bit, the second request bit, and the third request bit, respectively.

[0037] Import the user's service unit price range, service demand type, and service reputation lower limit into the first request position, the second request position, and the third request position, respectively, to generate the unit price request position, the type request position, and the reputation request position.

[0038] The unit price request bit, type request bit, and reputation request bit are respectively assigned the position symbols DJW, LXW, and KBW to facilitate the conversion of basic requests into user requests.

[0039] A bounding box is set on the user request, and the user's real-time location is imported into the bounding box to generate a user request with location features.

[0040] Furthermore, the selection method for candidate merchants is as follows:

[0041] Import the location features requested by the user and the store addresses of A available merchants in the profile database onto the same electronic map to obtain a requested location and A store locations;

[0042] Measure the requested distance from each requested point to A store points, and record the store points whose requested distance is less than the upper limit of the distance as candidate points, thus obtaining D candidate points;

[0043] Mark the corresponding portrait units of D candidate points one by one in the information database, and count the number of associated features on the associated channels connected to the portrait units;

[0044] Remove the image units at both ends of the associated channel with a number of associated features of 1, and record the available merchants corresponding to the remaining image units as candidate merchants to obtain E candidate merchants.

[0045] Furthermore, the method for selecting target merchants is as follows:

[0046] Natural language processing technology is used to identify the textual and numerical parts of the service unit price range, service type, service reputation lower limit, reputation level, service type and type price, and extract the numerical values ​​from the numerical parts.

[0047] When the value of the service unit price range overlaps with the value of the type price, the type price is recorded as the matching data;

[0048] When the value of the service request type overlaps with the value of the service type, the service type is recorded as the matching data;

[0049] When the lower limit of service reputation is less than or equal to the reputation level, the reputation level is recorded as the matching data.

[0050] Count the number of matching data in each of the E candidate merchants, and record the candidate merchants with 3 matching data as target merchants, thus obtaining F target merchants.

[0051] Furthermore, recommended car wash parameters include driving distance, driving time, and remaining wash bays.

[0052] The recommended priority is as follows: the priority of driving time value is higher than the priority of driving distance value, and the priority of driving distance value is higher than the priority of remaining workstation value;

[0053] The method for recommending target merchants to users in order is as follows:

[0054] Based on the driving time value from smallest to largest, F target merchants will be recommended to the user in sequence;

[0055] When there are target merchants with the same driving time value, the target merchants are recommended to the user in order of driving distance value from smallest to largest.

[0056] When there are target merchants with the same driving time and driving distance, the target merchants will be recommended to the user in descending order of remaining workstations.

[0057] When there are target merchants with the same driving distance, driving time, and remaining workstations, the target merchant will be randomly recommended to the user.

[0058] The technical effects and advantages of the car wash service management system based on a car wash information platform of the present invention are as follows:

[0059] (1): By combining distribution data and discount data into merchant profiles and merging merchant profiles with profile units, a profile library with associated channels can be constructed. This allows for multi-dimensional and accurate collection, fusion, and summary display of car wash service information of qualified registered merchants within the car wash information platform. Furthermore, by combining associated channels to perform associated guidance operations on profile units with associated characteristics, different dealers with high similarity and overlap in the profile library can be associated and identified. This avoids negative interference caused by different dealers with low similarity and overlap in the subsequent screening and identification process, and facilitates accurate query operations for dealers with associated characteristics in the future.

[0060] (2): By integrating real-time location with historical user tags into user requests with location features, the real-time nature and characteristics of user requests can be highlighted, improving the matching efficiency and accuracy of user requests when querying profile units in the profile database. Combined with the first-level filtering of location features and associated channels, and the second-level filtering of user requests and candidate merchants, users can achieve dual independent filtering and querying of target merchants on the car wash information platform. This avoids the problem of inaccurate filtering results that exists in the single-level filtering method. At the same time, it effectively combines the factors of the similarity between different target merchants and user requests, ensuring that the filtered target merchants can meet the actual needs of users and effectively improving the management effect of car wash services.

[0061] (3): By setting recommendation priorities and recommending target merchants in an orderly manner, the recommendation operation of target merchants can be carried out in an orderly, accurate and reasonable manner based on the degree of similarity between the target merchants and the user's request. This ensures that the car wash service management system can quickly and orderly recommend relevant car wash information that meets the user's needs in the first time, and further improves the effectiveness of car wash service management. Attached Figure Description

[0062] Figure 1 This is a schematic diagram of the architecture of a car wash service management system based on a car wash information platform, provided in Embodiment 1 of the present invention.

[0063] Figure 2 This is a flowchart illustrating a car wash service management method based on a car wash information platform, as provided in Embodiment 2 of the present invention. Detailed Implementation

[0064] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] Example 1: Please refer to Figure 1 As shown in this embodiment, a car wash service management system based on a car wash information platform is applied to the car wash information platform and includes:

[0066] The merchant profile combination module collects the distribution data and discount data of registered merchants, filters out available merchants from the registered merchants based on service management principles, and combines the distribution data and discount data of available merchants into a merchant profile.

[0067] Registered merchants refer to car wash dealerships that have completed the registration of car wash service management information on the car wash information platform as of the current time and have obtained the right to conduct normal car wash information management on the car wash information platform. In this embodiment, the car wash dealership is a car wash store.

[0068] Distribution data refers to the store registration information inherent to registered merchants when they register on the car wash information platform, enabling distribution data to represent the inherent information of registered merchants;

[0069] Distribution data includes store address, business hours, reputation level, service type, and price per type;

[0070] Specifically, "store address" refers to the location of a registered car wash store on the car wash information platform, used to accurately represent the address of a registered merchant's car wash. "Business hours" refers to the normal business hours of a registered merchant's store on the car wash information platform. "Reputation level" refers to the service reputation rating of a registered merchant on the car wash information platform. "Service type" refers to the types of car washes available to a registered merchant on the car wash information platform. "Type price" refers to the single-use price for each type of car wash offered by a registered merchant on the car wash information platform. The store address, business hours, reputation level, service type, and type price are obtained by querying the merchant database in the registered database of the car wash information platform.

[0071] The registration database is a database within the car wash information platform used to record and save various types of data. The registration database includes a merchant database and a user database. The merchant database is used to record and save data on dealers, and the user database is used to record and save data on users.

[0072] It should be noted that by accurately collecting information such as store address, business hours, reputation level, service type, and price, we can obtain multi-dimensional information about the registered merchants. At the same time, when there are changes or updates to the store address, business hours, reputation level, service type, and price of registered merchants, we also need to update and replace these information in real time.

[0073] Discount data refers to the current discount information of registered merchants on the car wash information platform, which is used to accurately represent the current car wash discount situation of registered merchants;

[0074] The discount data includes coupon type and coupon quantity;

[0075] Specifically, the coupon type refers to the type of car wash discount coupon issued by a registered merchant at the current moment, which can indicate the specific type of car wash discount. The coupon quantity refers to the number of coupons issued by a registered merchant at the current moment. The coupon type and coupon quantity are obtained by querying the merchant database of the car wash information platform.

[0076] After obtaining the distribution and discount data of registered merchants, it is necessary to identify the obtained data and, under the constraints of the service management guidelines, identify and screen whether the registered merchants meet the conditions for subsequent car wash service management, and record the registered merchants that meet the conditions for subsequent car wash service management as available merchants.

[0077] The service management principle is to remove registered merchants with the first or second abnormal data. This ensures that the number of distribution and discount data of the remaining available merchants is reasonable, avoids registered merchants lacking distribution and discount data from participating in subsequent car wash service management operations, and thus raises the screening threshold for available merchants.

[0078] The available merchant selection methods are as follows:

[0079] Count the number of distributor data and the number of discount data among the registered merchants one by one, and record them as the first value and the second value respectively.

[0080] When the first value is less than 5, it indicates that there is a missing quantity in the distribution data, and the distribution data is recorded as the first abnormal data.

[0081] When the second value is less than 2, it indicates that there is a missing quantity of discount data, and the discount data is recorded as the second abnormal data.

[0082] Remove registered merchants that contain either the first or second abnormal data, and record the remaining registered merchants as available merchants, thus obtaining A available merchants.

[0083] After filtering out available merchants, the distribution data and discount data of available merchants can be combined and summarized, and the combined and summarized data can be used to build a merchant profile. This merchant profile can comprehensively represent the car wash service situation of available merchants in the car wash information platform, and serve as the basis for users to select available merchants as car wash options.

[0084] The method for combining merchant profiles is as follows:

[0085] Create A profile profiles with three vertically parallel tag layers, and label the three tag layers as the base layer, behavior layer and value layer respectively, from bottom to top. The profile profile refers to a merchant profile without any substantial content. The tag layer is the smallest unit that makes up the profile profile and provides the import location for different types of data such as distribution data and discount data.

[0086] S label positions are set in the basic layer, behavior layer and value layer respectively, and label dividing lines are drawn between adjacent label positions. The label position is used to identify the data position of a specific data in the distribution data and discount data to ensure the independence between each data. The label dividing line is used to separate the data in adjacent label positions to avoid the data in adjacent label positions from intersecting and being arranged in a disordered manner.

[0087] Import the store address, business hours and reputation level of A available merchants into the tag positions of A basic layers one by one; import the service type and type price of A available merchants into the tag positions of A behavior layers one by one; and import the coupon type and coupon quantity of A available merchants into the tag positions of A value layers one by one.

[0088] By removing redundant tags from A basic layers, A behavioral layers, and A value layers respectively, A profile outlines are generated into A merchant profiles.

[0089] It should be noted that each available merchant has one and only one corresponding merchant profile, ensuring that all relevant information of each available merchant on the car wash information platform can be completely and comprehensively integrated, thereby facilitating subsequent identification and matching operations.

[0090] The profile conversion module constructs an information database with a circular distribution of profile units, imports merchant profiles into the profile units, and establishes association channels between profile units with related features, thereby converting the information database into a profile database.

[0091] The information database refers to a database that has not imported any substantial merchant profiles, and enables the information database to serve as the foundation for building the subsequent profile database. The profile unit is the smallest component of the information database and provides accurate location constraints for importing merchant profiles, so that a merchant profile can correspond to a profile unit.

[0092] The method for constructing the information database is as follows:

[0093] Establish a database with A circularly arranged blank data bits, and draw circular bit outlines on the outside of each of the A blank data bits to generate A portrait units. Blank data bits refer to data bits without any substantial content. By drawing bit outlines, each blank data bit can be sealed and protected to ensure that the data within each bit will not be leaked, thereby improving data security.

[0094] Mark a point on the bit contour line at a position close to the previous bit contour line and at a position close to the next bit contour line, and denoted as the first contour point and the second contour point; the first contour point and the second contour point are the two endpoints used to construct the associated channel and provide accurate positional constraints for the establishment of the associated channel.

[0095] Measure the distance from the first contour point to the previous contour line and the distance from the second contour point to the next contour line one by one, and record them as the first distance value and the second distance value respectively.

[0096] The positions of the first and second contour points are continuously adjusted until both the first and second distance values ​​reach their minimum values, at which point the adjustment stops, thus constructing an information database.

[0097] After the information database is built, the merchant profiles can be imported one by one into each profile unit of the information database, so that each profile unit has meaningful data. After the merchant profiles are imported into the profile unit, there will be similar or identical distribution data or discount data between different merchant profiles, so there will be correlation between different profile units. Therefore, the corresponding data of similar or identical distribution data or discount data are recorded as correlation features.

[0098] Specifically, when determining whether there are related features between any two profile units, it is necessary to compare the similarity and overlap of the distribution data or discount data in the label positions of any two profile units, and determine whether there are related features between two profile units whose similarity and overlap is greater than the standard overlap.

[0099] Similarity overlap is a numerical representation of the degree of similarity between distribution data or promotional data with the same meaning in the label positions of two profile units. Specifically, it can be judged by comparing the number of similar elements between distribution data or promotional data with the same meaning.

[0100] For example, natural language processing technology can be used to split the same type of distribution data or discount data in the tag positions of two profile units into two element sets. When the number of elements with the same meaning in the two element sets is greater than or equal to half of the total number of elements in either element set, it is determined that there is a correlation feature between the two profile units, and the elements with the same meaning in the tag positions of the two profile units are recorded as correlation features.

[0101] The association channel is a data channel used to transmit distribution data or verification data bidirectionally between two profile units with related characteristics, and to achieve the effect of association query for car wash service management between profile units with related characteristics.

[0102] A profile database refers to an information database that contains complete profile units and associated channels, enabling it to comprehensively integrate car wash information from all available merchants and lay the foundation for subsequent user queries and management of car wash information.

[0103] The method for converting the image database is as follows:

[0104] Import A merchant profiles one by one into A profile units in the information database. Number the A profile units in ascending order according to the import order. Then, group the two profile units with related features into a unit group to obtain C unit groups.

[0105] In the C unit groups, the second contour point of the image unit with the smaller number is recorded as the channel start point, and the first contour point of the image unit with the larger number is recorded as the channel end point. A bidirectional channel is established between the channel start point and the channel end point to obtain C associated channels.

[0106] The associated features of the C unit groups are annotated on the corresponding associated channels, which enables the information database to be transformed into a profile database.

[0107] It should be noted that annotating the associated features on the associated channels can clearly and explicitly identify the type and attributes of each associated channel, and facilitate accurate querying and service management of car wash information in the profile database.

[0108] The user request building module collects the user's real-time location and historical user tags, and fuses the real-time location and historical user tags to build a user request with location features.

[0109] Real-time location refers to the current geographical location of the user who needs to query car wash information and manage services. It can provide a comparison of the distance between the location and the available merchants for subsequent filtering and querying. Real-time location is obtained through real-time positioning query through electronic map.

[0110] Historical user tags refer to specific information generated when a user performs car wash services and other operations in the past within the car wash information platform, which can comprehensively display the user's specific car wash service consumption in the past;

[0111] Historical user tags include service price range, service demand type, and lower limit of service reputation;

[0112] Specifically, the service price range refers to the minimum to maximum price for each car wash within a past time period on the car wash information platform; the service demand type refers to all types of car washes the user has experienced within a past time period on the car wash information platform; and the service reputation lower limit refers to the lowest reputation level of the corresponding car wash store the user has experienced within a past time period on the car wash information platform. The service price range, service demand type, and service reputation lower limit are obtained by querying the user database of the car wash information platform.

[0113] Once the user's real-time location and historical user tags are obtained, the real-time location and historical user tags can be fused together to effectively and accurately combine different types of information and ultimately generate a user request that matches the user's behavior in the past time period on the car wash information platform.

[0114] After constructing the user request, the user request can only be used to represent the user's historical information as a whole. In order to highlight the real-time nature of the user request and improve the efficiency of subsequent query and matching with the profile database, it is necessary to add location features to the user request so that the location features can accurately represent the specific location of the car wash service management in the user request and become the key point for subsequent matching and filtering with merchant profiles.

[0115] The method for constructing a user request is as follows:

[0116] Create a base request with three blank request bits, and label the three blank request bits as the first request bit, the second request bit, and the third request bit, respectively.

[0117] Import the user's service unit price range, service demand type, and service reputation lower limit into the first request position, the second request position, and the third request position, respectively, to generate the unit price request position, the type request position, and the reputation request position.

[0118] The unit price request position, type request position, and reputation request position are respectively assigned the position symbols DJW, LXW, and KBW to facilitate the conversion of basic requests into user requests; the position symbols are used to distinguish and identify the unit price request position, type request position, and reputation request position, and maintain the uniqueness of each request position.

[0119] A feature box is created on the user request, and the user's real-time location is imported into the feature box to generate a user request with location features. The feature box is a blank data group used to provide annotations for the location features of the user request, ensuring that the real-time location can be correlated with and annotated with the user request.

[0120] It should be noted that the user requests with location characteristics are constructed as the basis for subsequent queries to the profile database for car wash service management, and provide a filtering basis with two-layer filtering conditions for subsequent queries and filtering of required merchants.

[0121] The target merchant filtering module combines location features with associated channels to filter candidate merchants from available merchants, and matches user requests with the profiles of candidate merchants to filter target merchants from the candidate merchants;

[0122] Candidate merchants refer to available merchants corresponding to the merchant profiles initially selected from the profile database based on the location features of the user's request and in conjunction with the associated channels. At this stage, the candidate merchants are only the result of the initial screening, resulting in a large number of candidate merchants, and all candidate merchants can maintain a basic association with the user's request.

[0123] When screening candidate merchants, it is necessary to conduct the screening in an orderly manner based on location characteristics, associated channels, and candidate merchants.

[0124] Specifically, firstly, the location features requested by the user and the store addresses of A available merchants in the profile database are all imported onto the same electronic map to obtain a requested location and A store locations;

[0125] Then, the requested distance values ​​from the requested point to each of the A store points are measured one by one, and the store points whose requested distance values ​​are less than the distance upper limit are recorded as candidate points, thus obtaining D candidate points. The distance upper limit refers to the maximum distance between the user's real-time location and the store address of the candidate merchant on the electronic map, thereby providing a numerical limit on the distance between the candidate merchants and the location that meets the subsequent user service needs.

[0126] Next, the image units corresponding to the D candidate points are marked one by one in the information database, and the number of associated features on the associated channels connected to the image units is counted.

[0127] Finally, the image units at both ends of the associated channel with a number of associated features of 1 are removed, and the available merchants corresponding to the remaining image units are recorded as candidate merchants, resulting in E candidate merchants.

[0128] After selecting candidate merchants, these merchants can maintain a basic connection with the user's request in terms of location distance. However, the candidate merchants may not meet the requirements of other dimensions of data in the user's request. A second filtering process is needed to obtain the target merchant, which can then serve as the final distributor that is matched with the user's request.

[0129] When conducting a second screening of target merchants, it is necessary to perform profile matching between user requests and specific data from candidate merchants, thereby representing the degree of matching between user requests and candidate merchants and improving the accuracy of target merchant screening.

[0130] The selection method for target merchants is as follows:

[0131] The user's requested service unit price range, service demand type, and service reputation lower limit are respectively matched with the reputation level, service type, and type price of E candidate merchants to create a profile.

[0132] Natural language processing technology is used to identify the textual and numerical parts of the service unit price range, service type, service reputation lower limit, reputation level, service type and type price, and extract the numerical values ​​from the numerical parts. The textual and numerical parts are used to concisely represent the textual and numerical content of each data point, respectively, and the numerical values ​​are used to directly represent the magnitude of the numerical part.

[0133] When the value of the service unit price range overlaps with the value of the type price, it indicates that there is a match between the user's request and the candidate merchant for the single price of car washing. In this case, the type price is recorded as the matching data.

[0134] When the value of the service request type overlaps with the value of the service type, it indicates that there is a match between the user request and the candidate merchant for the type of car wash, and the service type is recorded as the matching data.

[0135] When the lower limit of service reputation is less than or equal to the reputation level, it means that there is a match between the user's request and the candidate merchant's reputation rating of the car wash store, and the reputation level is recorded as the matching data.

[0136] Count the number of matching data in each of the E candidate merchants, and record the candidate merchants with 3 matching data as target merchants, thus obtaining F target merchants.

[0137] It is important to note that the selected target businesses can be directly recommended to users as car wash service dealers, allowing users to choose any one of them as their car wash destination. In practice, the number of target businesses is usually no less than one to ensure that users have multiple different car wash dealers to choose from.

[0138] The target merchant recommendation module obtains car wash recommendation parameters from target merchants, sets recommendation priorities, and recommends target merchants to users in order according to the recommendation priorities;

[0139] Car wash recommendation parameters refer to multi-dimensional data that influences the order in which users choose target businesses as the best car wash dealers, and serve as the basis for determining the recommendation priority of target businesses to users.

[0140] Recommended car wash parameters include driving distance, driving time, and remaining wash bays.

[0141] Specifically, the driving distance value refers to the distance traveled from the user's real-time location to the target merchant's store address on the electronic map. The larger the driving distance value, the lower the probability that the user will choose the target merchant, and the lower the recommendation priority of the target merchant. The driving time value refers to the time taken for the user to travel from the user's real-time location to the target merchant's store address on the electronic map. The larger the driving time value, the lower the probability that the user will choose the target merchant, and the lower the recommendation priority of the target merchant. Both the driving distance value and the driving time value are obtained after querying the electronic map.

[0142] The remaining workstation value refers to the number of available car wash workstations that the target merchant can provide at the current moment. The larger the remaining workstation value, the greater the probability that users will choose the target merchant, and the higher the recommendation priority of the target merchant. The remaining workstation value is obtained by querying the merchant database within the car wash information platform.

[0143] When prioritizing recommendations, three dimensions need to be considered: the driving distance to the target business, the driving time, and the number of remaining car wash bays. Specifically, since the driving time directly affects the time it takes for the user to reach the target business, it has the greatest impact on the user's car wash recommendation experience, so the driving time has the highest priority. The number of remaining car wash bays does not directly affect the time it takes for the user to reach the target business, so it has the least impact on the user's car wash recommendation experience, so the number of remaining car wash bays has the lowest priority.

[0144] In summary, the recommended priority is as follows: the priority of driving time is higher than the priority of driving distance, and the priority of driving distance is higher than the priority of remaining workstations.

[0145] After determining the recommendation priority, the F target merchants need to be recommended to the user in order according to the recommendation priority. This will achieve an orderly, accurate and reasonable recommendation effect for the F target merchants, and meet the user's needs for car wash service management as much as possible.

[0146] Specifically, the method for recommending target merchants to users in sequence is as follows:

[0147] Compare the driving time values ​​of the F target merchants one by one, and recommend the F target merchants to the user in order of increasing driving time value;

[0148] When there are target merchants with the same driving time value, the target merchants are recommended to the user in order of driving distance value from smallest to largest.

[0149] When there are target merchants with the same driving time and driving distance, the target merchants will be recommended to the user in descending order of remaining workstations.

[0150] If the driving distance, driving time, and remaining workstations of a target merchant are all the same, then the target merchant will be randomly recommended to the user.

[0151] It should be noted that, under normal circumstances, due to the influence of external traffic environment factors and random entry factors of car wash stores, it is almost impossible for the driving distance value, driving time value and remaining work bay value of the target merchant to be the same at the same time. This embodiment analyzes the driving distance value, driving time value and remaining work bay value that are the same under ideal conditions, so as to ensure the rigor and comprehensiveness of the target merchant's recommendation to the user.

[0152] In this embodiment, by combining distribution data and discount data into merchant profiles and merging merchant profiles with profile units, a profile library with associated channels is constructed. This allows for multi-dimensional and accurate collection, fusion, and summary display of car wash service information from qualified registered merchants within the car wash information platform. Furthermore, by combining associated channels to perform association guidance operations on profile units with related characteristics, different dealers with high similarity and overlap within the profile library can be associated and identified. This avoids negative interference from dealers with low similarity and overlap in subsequent screening and identification processes, facilitating accurate subsequent queries on associated dealers.

[0153] By fusing real-time location with historical user tags to create location-featured user requests, the real-time nature and distinctiveness of user requests can be highlighted. This improves the matching efficiency and accuracy of user requests when querying profile units in the profile database. Combining primary filtering based on location features and associated channels with secondary filtering based on user requests and candidate merchants, users can achieve dual independent filtering of target merchants on the car wash information platform. This avoids the inaccuracy issues of primary filtering and effectively incorporates the similarity between different target merchants and user requests, ensuring that the filtered target merchants meet the actual needs of users and effectively improving the management of car wash services.

[0154] By prioritizing recommendations for target businesses, the system can make orderly, accurate, and reasonable recommendations based on the degree of similarity between the target business and the user's request. This ensures that the car wash service management system can quickly and systematically recommend relevant car wash information that meets the user's needs, further improving the effectiveness of car wash service management.

[0155] Example 2: Please refer to Figure 2 As shown, parts not described in detail in this embodiment are described in Embodiment 1. This embodiment provides a car wash service management method based on a car wash information platform, applied to the car wash information platform, and implemented based on a car wash service management system based on the car wash information platform, including:

[0156] S1: Collect the distribution data and discount data of registered merchants, filter out available merchants from the registered merchants based on service management principles, and combine the distribution data and discount data of available merchants into a merchant profile;

[0157] S2: Construct an information database with a circular distribution of portrait units, import merchant portraits into the portrait units, and establish association channels between the outline points of portrait units with related features to facilitate the transformation of the information database into a portrait database.

[0158] S3: Collect users' real-time location and historical user tags, fuse the real-time location and historical user tags to construct user requests with location features;

[0159] S4: Combine location features with associated channels to filter out candidate merchants from available merchants, and match user requests with candidate merchant profiles to filter out target merchants from candidate merchants;

[0160] S5: Based on the car wash recommendation parameters of the target merchants, determine the recommendation priority and recommend the target merchants to the user in order.

[0161] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A car wash service management system based on a car wash information platform, applied to the car wash information platform, characterized in that, include: The merchant profile combination module collects the distribution data and discount data of registered merchants. Based on service management criteria, it filters out available merchants from the registered merchants and combines the distribution data and discount data of available merchants into a merchant profile. The service management criteria are: to remove registered merchants with the first or second abnormal data. Distribution data includes store address, business hours, reputation level, service type, and price; discount data includes coupon type and quantity. The available merchant selection methods are as follows: Count the number of distributor data and the number of discount data among the registered merchants one by one, and record them as the first value and the second value respectively. When the first value is less than 5, the distribution data is recorded as the first abnormal data; When the second value is less than 2, the discount data is recorded as the second abnormal data; Remove registered merchants that have the first or second abnormal data, and record the remaining registered merchants as available merchants to obtain A available merchants; The image library conversion module constructs an information library with a circular distribution of image units, imports merchant images into the image units, and establishes association channels between the outline points of image units with related features, thereby converting the information library into an image library. The user request building module collects the user's real-time location and historical user tags, and fuses the real-time location and historical user tags to build a user request with location features. The target merchant filtering module combines location features with associated channels to filter candidate merchants from available merchants, and matches user requests with the profiles of candidate merchants to filter target merchants from the candidate merchants; The target merchant recommendation module determines the recommendation priority based on the car wash recommendation parameters of the target merchants and recommends the target merchants to the user in order.

2. The car wash service management system based on a car wash information platform according to claim 1, characterized in that, The method for combining merchant profiles is as follows: Create A profiles with three vertically parallel tag layers, and label the three tag layers as the basic layer, behavior layer and value layer respectively, from bottom to top. S label positions are set in the basic layer, behavior layer and value layer respectively, and label dividing lines are drawn between two adjacent label positions; Import the store address, business hours and reputation level of A available merchants into the tag positions of A basic layers one by one; import the service type and type price of A available merchants into the tag positions of A behavior layers one by one; and import the coupon type and coupon quantity of A available merchants into the tag positions of A value layers one by one. By removing redundant tags from A basic layers, A behavioral layers, and A value layers respectively, A profile outlines are generated into A merchant profiles.

3. A car wash service management system based on a car wash information platform according to claim 2, characterized in that, The method for constructing the information database is as follows: Establish a database with A circularly arranged blank data positions, and draw the circular bit outlines outside each of the A blank data positions to generate A image units. Mark a point on the position contour line, at a position close to the previous position contour line and at a position close to the next position contour line, and denot it as the first contour point and the second contour point. Measure the distance from the first contour point to the previous contour line and the distance from the second contour point to the next contour line one by one, and record them as the first distance value and the second distance value respectively. The positions of the first and second contour points are continuously adjusted until both the first and second distance values ​​reach their minimum values, at which point the adjustment stops, thus constructing an information database.

4. A car wash service management system based on a car wash information platform according to claim 3, characterized in that, When identifying related features, natural language processing technology is used to split the same type of distribution data or discount data in the tag positions of the two profile units into two element sets. When the number of elements with the same meaning in two sets of elements is greater than or equal to half the total number of elements in either set, the elements with the same meaning in the label positions of the two image units are recorded as associated features.

5. A car wash service management system based on a car wash information platform according to claim 4, characterized in that, The method for converting the image database is as follows: Import A merchant profiles one by one into A profile units in the information database. Number the A profile units in ascending order according to the import order. Then, group the two profile units with related features into a unit group to obtain C unit groups. In the C unit groups, the second contour point of the image unit with the smaller number is recorded as the channel start point, and the first contour point of the image unit with the larger number is recorded as the channel end point. A bidirectional channel is established between the channel start point and the channel end point to obtain C associated channels. The associated features of the C unit groups are annotated on the corresponding associated channels, which enables the information database to be transformed into a profile database.

6. A car wash service management system based on a car wash information platform according to claim 5, characterized in that, Historical user tags include service price range, service demand type, and lower limit of service reputation; The method for constructing a user request is as follows: Create a base request with three blank request bits, and label the three blank request bits as the first request bit, the second request bit, and the third request bit, respectively. Import the user's service unit price range, service demand type, and service reputation lower limit into the first request position, the second request position, and the third request position, respectively, to generate the unit price request position, the type request position, and the reputation request position. The unit price request bit, type request bit, and reputation request bit are respectively assigned the position symbols DJW, LXW, and KBW to facilitate the conversion of basic requests into user requests. A bounding box is set on the user request, and the user's real-time location is imported into the bounding box to generate a user request with location features.

7. A car wash service management system based on a car wash information platform according to claim 6, characterized in that, The selection method for candidate merchants is as follows: Import the location features requested by the user and the store addresses of A available merchants in the profile database onto the same electronic map to obtain a requested location and A store locations; Measure the requested distance from each requested point to A store points, and record the store points whose requested distance is less than the upper limit of the distance as candidate points, thus obtaining D candidate points; Mark the corresponding portrait units of D candidate points one by one in the information database, and count the number of associated features on the associated channels connected to the portrait units; Remove the image units at both ends of the associated channel with a number of associated features of 1, and record the available merchants corresponding to the remaining image units as candidate merchants to obtain E candidate merchants.

8. A car wash service management system based on a car wash information platform according to claim 7, characterized in that, The selection method for target merchants is as follows: Natural language processing technology is used to identify the textual and numerical parts of the service unit price range, service type, service reputation lower limit, reputation level, service type and type price, and extract the numerical values ​​from the numerical parts. When the value of the service unit price range overlaps with the value of the type price, the type price is recorded as the matching data; When the value of the service request type overlaps with the value of the service type, the service type is recorded as the matching data; When the lower limit of service reputation is less than or equal to the reputation level, the reputation level is recorded as the matching data. Count the number of matching data in each of the E candidate merchants, and record the candidate merchants with 3 matching data as target merchants, thus obtaining F target merchants.

9. A car wash service management system based on a car wash information platform according to claim 8, characterized in that, Recommended car wash parameters include driving distance, driving time, and remaining wash bays. The recommended priority is as follows: the priority of driving time value is higher than the priority of driving distance value, and the priority of driving distance value is higher than the priority of remaining workstation value; The method for recommending target merchants to users in order is as follows: Based on the driving time value from smallest to largest, F target merchants will be recommended to the user in sequence; When there are target merchants with the same driving time value, the target merchants are recommended to the user in order of driving distance value from smallest to largest. When there are target merchants with the same driving time and driving distance, the target merchants will be recommended to the user in descending order of remaining workstations. When there are target merchants with the same driving distance, driving time, and remaining workstations, the target merchant will be randomly recommended to the user.

Images