Tire transaction method and system, storage medium and electronic equipment

Abstract

The invention provides a tire transaction method and system, a storage medium and electronic equipment, and the method comprises the steps: a user logs in a transaction service platform for buying and selling tires, and browses various tires according to the demands; the transaction service platform recommends an intentional tire to the user through a tire recommendation model; the user selects an individual vehicle in the transaction service platform to carry out ordering qualification evaluation, and displays the credit line of the vehicle; the transaction service platform evaluates the credit line of the vehicle, and after the credit line passes the evaluation, the user fills in a service address and appointment service time, and performs online contract signing; the transaction service platform generates a plurality of shippers for tire delivery personnel to select according to the address of the user; and carrying out installation service for the user by the tire delivery personnel, and then completing the order. The tire transaction efficiency is improved, reasonable allocation of resources is promoted, and the operation cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of tire trading, and in particular to a tire trading method, system, storage medium and electronic equipment. Background Art

[0002] Among existing technologies, most of the current market still uses traditional tire procurement models, which cannot intuitively display important tire information such as inventory, model, price, load capacity, etc., and cannot effectively recommend tires to users based on the information of the user's truck. Information is not transparent, communication efficiency is low, and transactions can only be conducted offline. There is a lack of effective online platform support, the procurement process is cumbersome, the settlement cycle is long, and there is a lack of credit payment solutions for drivers. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a tire trading method, system, storage medium and electronic device, aiming to solve the technical problems mentioned in the background technology.

[0004] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0005] A tire trading method comprises the following steps:

[0006] Users log in to the tire trading service platform and browse various types of tires according to their needs;

[0007] The transaction service platform recommends the intended tires to the user through a tire recommendation model;

[0008] The user selects a personal vehicle on the transaction service platform for order qualification assessment and displays the credit limit of the vehicle;

[0009] The transaction service platform evaluates the credit limit of the vehicle. If approved, the user fills in the service address and appointment time, and signs the contract online;

[0010] The transaction service platform generates multiple shippers based on the user's address for the tire delivery personnel to choose from;

[0011] The tire delivery personnel provide installation services to the user and then complete the order;

[0012] The construction of the tire recommendation model includes:

[0013] Continuous and discrete processing of user information and tire information respectively;

[0014] Mapping the continuous and discretized user information and tire information into low-dimensional vectors;

[0015] Obtain a user browsing behavior sequence, extract a user long-term preference sequence from the user browsing behavior sequence based on an LSTM network, calculate an attention output from the user long-term preference sequence based on a two-layer attention mechanism, concatenate the low-dimensional vectors of the user information and tire information with the attention output, and process the concatenated vectors through a fully connected neural network model to obtain a tire recommendation score, thereby generating a tire recommendation list.

[0016] According to one aspect of the above technical solution, the continuous and discrete processing of the user information and the tire information respectively includes the following steps:

[0017] Obtaining user discrete features and user continuous features based on the user's account information on the transaction service platform, and obtaining a user browsing behavior sequence based on the user's browsing behavior on the transaction service platform;

[0018] Obtaining tire discrete features and tire continuous features based on basic tire information on the trading service platform;

[0019] The user discrete features and the tire discrete features are mapped into natural numbers, and the user continuous features and the tire continuous features are range-divided for discretization.

[0020] According to one aspect of the above technical solution, mapping the continuous and discretized user information and tire information into low-dimensional vectors specifically includes the following steps:

[0021] Converting the user discrete features and user continuous features into user low-dimensional vectors;

[0022] Converting the tire discrete features and tire continuous features into tire low-dimensional vectors;

[0023] A random replacement layer is introduced to randomly transform the user low-dimensional vector and the tire low-dimensional vector in a probabilistic manner during model training.

[0024] According to one aspect of the above technical solution, the method of obtaining a user browsing behavior sequence, extracting a user long-term preference sequence from the user browsing behavior sequence based on an LSTM network, calculating an attention output from the user long-term preference sequence based on a dual-layer attention mechanism, concatenating the low-dimensional vectors of the user information and tire information with the attention output, and processing the concatenated vectors through a fully connected neural network model to obtain a tire recommendation score, thereby generating a tire recommendation list, specifically includes the following steps:

[0025] Based on the LSTM network, the user browsing behavior sequence is processed and the hidden layer state h from the first time period to the tth time period in the user browsing behavior sequence is calculated. t;

[0026] Based on the hidden layer state h from the first time period to the tth time period in the user browsing behavior sequence t , perform weighted operations on the browsing sequence of the tth time period in the user's browsing behavior sequence, and calculate the user's recent preference sequence P a , and calculate the user's long-term preference sequence P b ;

[0027] From the user's long-term preference sequence P b Extracting the user's evaluation record MT for the tire and mapping the evaluation record into an evaluation low-dimensional vector;

[0028] Calculating the attention score of each tire feature in the evaluation low-dimensional vector based on a dual-layer attention mechanism;

[0029] Based on the attention score, performing weighted summation on the tire features to obtain a weighted tire feature;

[0030] Performing evaluation and record attention calculation on the weighted rear tire features and the time features in the evaluation low-dimensional vector to obtain an attention output;

[0031] Concatenate the attention output, the user low-dimensional vector, and the tire low-dimensional vector to obtain a concatenated vector;

[0032] The concatenated vector is input into a fully connected neural network model for processing to obtain a tire recommendation score, thereby generating a tire recommendation list.

[0033] According to one aspect of the above technical solution, the user discrete features include truck type, tire installation position, cargo type, and driver driving habits; the user continuous features include expected life, load range, and driver daily mileage; the tire discrete features include tire category, pattern type, and compatible vehicle models; and the tire continuous features include tire size, load rating, and performance indicators.

[0034] According to one aspect of the above technical solution, the user selects a personal vehicle in the transaction service platform for order qualification assessment and displays the credit limit of the vehicle, specifically including:

[0035] The user selects one of the multiple pre-stored vehicles in the transaction service platform to place an order;

[0036] The transaction service platform evaluates the credit qualifications of the selected vehicle and displays the credit limit on the operation interface.

[0037] According to one aspect of the above technical solution, the transaction service platform evaluates the credit limit of the vehicle. If the credit limit is approved, the user fills in the service address and appointment time, and signs the online contract, which specifically includes:

[0038] The user fills in the order quantity of the tires to be purchased and clicks "Order Now";

[0039] The user fills in the service address and service time popped up on the transaction service platform to make an appointment;

[0040] The user fills in and signs the online electronic contract that pops up on the transaction service platform to generate an order.

[0041] The present invention also provides a tire trading system, comprising:

[0042] User: Logs into the tire trading service platform and browses various tire types according to needs. The user selects a personal vehicle on the trading service platform for order qualification assessment and displays the credit limit of the vehicle.

[0043] Transaction service platform: recommends the intended tires to the user through the tire recommendation model; evaluates the credit limit of the vehicle, and if approved, the user fills in the service address and appointment time, and signs the contract online; generates multiple shippers based on the user's address for the tire delivery personnel to choose from;

[0044] Tire delivery personnel: provide installation services for the users and complete the orders;

[0045] The construction of the tire recommendation model includes:

[0046] Continuous and discrete processing of user information and tire information respectively;

[0047] Mapping the continuous and discretized user information and tire information into low-dimensional vectors;

[0048] Obtain a user browsing behavior sequence, extract a user long-term preference sequence from the user browsing behavior sequence based on an LSTM network, calculate an attention output from the user long-term preference sequence based on a two-layer attention mechanism, concatenate the low-dimensional vectors of the user information and tire information with the attention output, and process the concatenated vectors through a fully connected neural network model to obtain a tire recommendation score, thereby generating a tire recommendation list.

[0049] The present invention also provides a storage medium storing a computer program, which implements the tire trading method described above when executed by a processor.

[0050] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the tire trading method described above when executing the computer program.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] By providing a transaction service platform for users to choose tires, and using a tire recommendation model to recommend possible tires to users based on users' tire evaluation indicators in big data, the previous model of people looking for information has been transformed into a model of information looking for people; after users have selected the tires they want, they place an order based on their credit qualifications, and can conduct risk control judgments on the vehicle to determine the driver's qualifications for obtaining credit sales, thereby improving transaction efficiency, which not only meets the driver's short-term capital shortage problem, but also brings corresponding orders to the tire service provider; finally, by filling in the service address and making an appointment for the service time, the online contract is signed, and then the transaction service platform generates multiple shippers near the user's address for the tire delivery personnel to choose from, which improves delivery efficiency. The tire delivery personnel then perform installation services for the user and complete the order, transforming the existing multi-link offline errands into a one-click online closed loop. In terms of building a tire recommendation model, an LSTM network is used to obtain important information in the user's browsing behavior sequence, and it can be updated over time. The important information of each time period is then combined to obtain the user's recent preference sequence, and the user's long-term preference sequence is deduced. A double-layer attention mechanism is then calculated. The first attention mechanism is used to learn the weights between different feature channels to capture the importance differences of different channels. The second attention mechanism is used to learn the weights of different positions in the same feature channel to capture the importance of different positions. By introducing a double-layer attention mechanism, the model can better capture the importance differences of different feature channels, thereby improving the model's expressiveness and performance.

[0053] The present invention improves tire transaction efficiency, promotes rational allocation of resources, and reduces operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flow chart of the tire trading method in the first embodiment of the present invention;

[0055] Figure 2 This is a structural block diagram of a tire trading system in a second embodiment of the present invention;

[0056] Figure 3 This is a structural block diagram of an electronic device in a third embodiment of the present invention;

[0057] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0058] To facilitate understanding of the present invention, the present invention will be described more fully below with reference to the accompanying drawings. The drawings illustrate several embodiments of the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0059] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be an intermediate element. When an element is referred to as being "connected to" another element, it may be directly connected to the other element or there may be an intermediate element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used in this specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0061] See also Figure 1 , shown is a tire trading method in a first embodiment of the present invention, comprising the following steps:

[0062] S10, the user logs in to the tire trading service platform and browses various types of tires according to needs;

[0063] S20, the transaction service platform recommends the intended tire to the user through the tire recommendation model;

[0064] S30, the user selects a personal vehicle on the transaction service platform for order qualification assessment, and displays the credit limit of the vehicle;

[0065] S40, the transaction service platform evaluates the credit limit of the vehicle. If approved, the user fills in the service address and appointment time, and signs the online contract;

[0066] S50, the transaction service platform generates multiple shippers based on the user's address for the tire delivery personnel to choose from;

[0067] S60: The tire delivery personnel provide installation services to the user and complete the order;

[0068] The construction of the tire recommendation model includes:

[0069] Continuous and discrete processing of user information and tire information respectively;

[0070] Mapping the continuous and discretized user information and tire information into low-dimensional vectors;

[0071] Obtain a user browsing behavior sequence, extract a user long-term preference sequence from the user browsing behavior sequence based on an LSTM network, calculate an attention output from the user long-term preference sequence based on a two-layer attention mechanism, concatenate the low-dimensional vectors of the user information and tire information with the attention output, and process the concatenated vectors through a fully connected neural network model to obtain a tire recommendation score, thereby generating a tire recommendation list.

[0072] It can be understood that the present invention provides a transaction service platform for users to choose tires, and uses a tire recommendation model to recommend possible tires to users based on the user's evaluation indicators for tires in big data, transforming the previous model of people looking for information to a model of information looking for people; after the user selects the tires he wants, he places an order based on his credit qualifications, and can perform risk control judgment on the vehicle to determine the driver's qualifications for obtaining credit sales, thereby improving transaction efficiency, which not only meets the driver's short-term capital shortage problem, but also brings corresponding orders to the tire service provider; finally, by filling in the service address and making an appointment for the service time, the online contract is signed, and then the transaction service platform generates multiple shippers near the user's address for the tire delivery personnel to choose from, thereby improving delivery efficiency. The tire delivery personnel then perform installation services for the user and complete the order, transforming the existing multi-link offline errands into a one-click online closed loop. In terms of building a tire recommendation model, an LSTM network is used to obtain important information in the user's browsing behavior sequence, and it can be updated over time. The important information of each time period is then combined to obtain the user's recent preference sequence, and the user's long-term preference sequence is deduced. A double-layer attention mechanism is then calculated. The first attention mechanism is used to learn the weights between different feature channels to capture the importance differences of different channels. The second attention mechanism is used to learn the weights of different positions in the same feature channel to capture the importance of different positions. By introducing a double-layer attention mechanism, the model can better capture the importance differences of different feature channels, thereby improving the model's expressiveness and performance.

[0073] The present invention improves tire transaction efficiency, promotes rational allocation of resources, and reduces operating costs.

[0074] Specifically, in this embodiment, the user information and tire information are respectively processed into a continuous form and a discrete form, which specifically includes the following steps:

[0075] Obtaining user discrete features and user continuous features based on the user's account information on the transaction service platform, and obtaining a user browsing behavior sequence based on the user's browsing behavior on the transaction service platform;

[0076] Obtaining tire discrete features and tire continuous features based on basic tire information on the trading service platform;

[0077] The user discrete features and the tire discrete features are mapped into natural numbers, and the user continuous features and the tire continuous features are range-divided for discretization.

[0078] The user discrete features include truck type, tire installation location, cargo type, and driver's driving habits; the user continuous features include expected lifespan, load range, and driver's daily mileage; the tire discrete features include tire category, pattern type, and compatible vehicle models; the tire continuous features include tire size, load rating, and performance indicators.

[0079] It is understandable that by obtaining the user's account information on the transaction service platform through big data, the user's discrete features and user continuous features are obtained, and then the user's basic information can be known, such as the user's truck type, what goods are usually transported, daily mileage, etc. These will affect the type of tire the user ultimately chooses. Then, based on the specific types of all tires in the transaction service platform, the tire discrete features and tire continuous features are obtained, and then the basic information of the tire, such as the tire pattern type, tire size, etc., can be known. These data are used to match the user's basic information to recommend a more suitable tire to the user, and then obtain the user's basic information. The user's browsing behavior sequence on the transaction service platform is obtained by analyzing the user's browsing situation (such as click-through rate, dwell time, etc.). It should be noted that the tire recommendation model used in this application can be based on the context attribute perception model, and can obtain more accurate personalized recommendations by collecting discrete features and continuous features for analysis. In addition, user discrete features and tire discrete features cannot be directly input into the context attribute perception model, and need to be mapped into natural numbers. For example, the types of trucks include heavy trucks, medium trucks and small trucks, which are mapped to 1, 2, and 3 respectively. User continuous features and tire continuous features need to be range-divided before being input into the model to avoid dimensionality explosion.

[0080] Furthermore, mapping the continuous and discretized user information and tire information into low-dimensional vectors specifically includes the following steps:

[0081] Converting the user discrete features and user continuous features into user low-dimensional vectors;

[0082] Converting the tire discrete features and tire continuous features into tire low-dimensional vectors;

[0083] A random replacement layer is introduced to randomly transform the user low-dimensional vector and the tire low-dimensional vector in a probabilistic manner during model training.

[0084] It can be understood that after processing the user discrete features, user continuous features, tire discrete features, and tire continuous features, they are converted into a form that can be processed by a computer through an encoding operation. Then, the embedding layer in the context attribute perception model is used to convert the user discrete features and user continuous features into user low-dimensional vectors, and the tire discrete features and tire continuous features into tire low-dimensional vectors. This step can be implemented using the Embedding technology in the embedding layer, mapping the user's own characteristics into the user low-dimensional vector, and mapping the tire's own characteristics into the tire low-dimensional vector; then a random replacement layer is introduced, and the user low-dimensional vector and the tire low-dimensional vector are input into the random replacement layer. The random replacement layer is a new processing layer in this application, which is used to perform the steps (taking the tire low-dimensional vector as an example) when training the model: randomly replacing a random small part of a tire low-dimensional vector with another tire low-dimensional vector. The replacement between the two will not cause a significant change in the loss distribution. During model training, the random replacement layer will randomly replace part of the low-dimensional vector with a certain probability, further reducing overfitting and bringing good generalization effect.

[0085] Furthermore, the method of obtaining a user browsing behavior sequence, extracting a user long-term preference sequence from the user browsing behavior sequence based on an LSTM network, calculating an attention output from the user long-term preference sequence based on a double-layer attention mechanism, concatenating the low-dimensional vectors of the user information and tire information with the attention output, and processing the concatenated vectors through a fully connected neural network model to obtain a tire recommendation score, thereby generating a tire recommendation list, specifically includes the following steps:

[0086] Based on the LSTM network, the user browsing behavior sequence is processed and the hidden layer state h from the first time period to the tth time period in the user browsing behavior sequence is calculated. t ;

[0087] h t =LSTM(c t ,h t-1 );

[0088] Among them, c t Represents the original data of the user's browsing behavior sequence in the current time period; h t-1 Represents the hidden layer state of the previous step;

[0089] Based on the hidden layer state h from the first time period to the tth time period in the user browsing behavior sequence t, perform weighted operations on the browsing sequence of the tth time period in the user's browsing behavior sequence, and calculate the user's recent preference sequence P a , and calculate the user's long-term preference sequence P b ;

[0090]

[0091] Among them, n represents the total number of time periods in the user browsing behavior sequence, T represents transposition, and P n The vector representing the user browsing behavior sequence (i.e., the original sequence), W h represents a trainable weight matrix;

[0092] From the user's long-term preference sequence P b Extracting the user's evaluation record MT for the tire and mapping the evaluation record into an evaluation low-dimensional vector;

[0093] Where, M=(M1,M2,……,M q ),M q represents the evaluation record of the qth tire label (such as the tire pattern type, tire size, etc. mentioned above), T represents the time period, and the low-dimensional evaluation vector to which M is mapped is recorded as m i ', the low-dimensional evaluation vector to which T is mapped is denoted as t';

[0094] Calculating the attention score of each tire feature in the evaluation low-dimensional vector based on a dual-layer attention mechanism;

[0095]

[0096] Among them, ξ a and W m Represents the parameters of the first layer attention mechanism, W m represents the weight, ξ a Represents bias, σ represents activation function, T represents transposition, and G i 'Perform normalization to obtain the attention score G i ;

[0097]

[0098] Based on the attention score, performing weighted summation on the tire features to obtain a weighted tire feature;

[0099]

[0100] The weighted rear tire features and the time features in the evaluation low-dimensional vector are evaluated and recorded for attention calculation to obtain the attention output z a ;

[0101] z a =G m Y a +G t t';

[0102] in,

[0103]

[0104] W z and ξ b is the parameter in the second layer attention mechanism, W z represents the weight, ξ b represents bias, t represents the number of hidden layers, and T represents transpose;

[0105] Concatenate the attention output, the user low-dimensional vector, and the tire low-dimensional vector to obtain a concatenated vector;

[0106] The concatenated vector is input into a fully connected neural network model for processing to obtain a tire recommendation score, thereby generating a tire recommendation list.

[0107] This step can calculate the most suitable tires corresponding to different user features. Since the LSTM network is used to obtain important information in the user browsing behavior sequence and can be updated over time, the important information of each time period is combined to obtain the user's recent preference sequence and the user's long-term preference sequence. Then, the user's browsing behavior (including browsing, clicking, and positive reviews) can be used to obtain the user's true rating of the tire. Because there are many users who change tires midway, but later repurchase the tires they used for the first time because they are not good, or the evaluation of tires in each time period is different based on the usage experience, we cannot only consider recent or past evaluation records, but must consider all time periods comprehensively. LSTM can capture this information and then take the user's long-term preference sequence as the main body of the research, performing a double-layer attention mechanism calculation. The first attention mechanism is used to learn the weights between different feature channels to capture the importance differences of different channels. The second attention mechanism is used to learn the weights of different positions of the same feature channel to capture the importance of different positions. By introducing the double-layer attention mechanism, the model can better capture the importance differences of different feature channels, thereby improving the model's expression ability and performance. The first-layer attention mechanism is used to distinguish the impact of different tire labels on a tire. The second-layer attention mechanism is used to distinguish the impact of tire labels from users who are not aware of the label. These two factors are then used to weight the impact of different tire labels on a tire and to weight the evaluations of different users. Finally, the user low-dimensional vector, the tire low-dimensional vector, and the feature vector output by the attention mechanism are concatenated. The multi-layer perception mechanism in a fully connected neural network model compresses and abstracts the data, improving representation capabilities. Finally, a probability distribution is calculated to generate a list of recommended tires.

[0108] The tire recommendation model in this application matches user features with tire features, and fully utilizes data information through LSTM and attention mechanisms to improve the accuracy of recommendations. By introducing a random replacement layer, it can not only further reduce overfitting and bring about a good generalization effect, but also further enrich contextual information during self-attention mechanism calculation to enhance the model's representation capabilities.

[0109] Furthermore, the user selects a personal vehicle in the transaction service platform for order qualification assessment and displays the credit limit of the vehicle, specifically including:

[0110] The user selects one of the multiple pre-stored vehicles in the transaction service platform to place an order;

[0111] The transaction service platform evaluates the credit qualifications of the selected vehicle and displays the credit limit on the operation interface.

[0112] It can be understood that this application adopts a model of one credit qualification for each vehicle. In this way, in actual operations, a user can bind multiple trucks, which can be owned by different people. The credit qualifications of the trucks are related to their owners. The user can help colleagues purchase and use the credit qualifications of colleagues to directly purchase, thereby improving transaction efficiency.

[0113] Furthermore, the transaction service platform evaluates the credit limit of the vehicle. If approved, the user fills in the service address and appointment time, and signs the online contract, which specifically includes:

[0114] The user fills in the order quantity of the tires to be purchased and clicks "Order Now";

[0115] The user fills in the service address and service time popped up on the transaction service platform to make an appointment;

[0116] The user fills in and signs the online electronic contract that pops up on the transaction service platform to generate an order.

[0117] Finally, the tire delivery staff will choose a shipper that is convenient for pickup to pick up the goods, and then go to the address filled in by the user to install the tires and complete the order.

[0118] In summary, the tire trading method in the above-mentioned embodiment of the present invention improves tire trading efficiency, promotes rational allocation of resources, and reduces operating costs.

[0119] Please refer to Figure 2 , shown is a tire trading system in a second embodiment of the present invention, comprising:

[0120] User 11: Logs into a tire trading service platform and browses various tire types according to their needs. The user selects a vehicle on the trading service platform for order eligibility assessment and displays the credit limit for the vehicle.

[0121] Transaction service platform 12: recommends the intended tires to the user through the tire recommendation model; evaluates the credit limit of the vehicle, and if approved, the user fills in the service address and reservation time, and signs the contract online; generates multiple shippers based on the user's address for the tire delivery personnel to choose from;

[0122] Tire delivery personnel 13: provide installation services for the user and complete the order;

[0123] The construction of the tire recommendation model includes:

[0124] Continuous and discrete processing of user information and tire information respectively;

[0125] Mapping the continuous and discretized user information and tire information into low-dimensional vectors;

[0126] Obtain a user browsing behavior sequence, extract a user long-term preference sequence from the user browsing behavior sequence based on an LSTM network, calculate an attention output from the user long-term preference sequence based on a two-layer attention mechanism, concatenate the low-dimensional vectors of the user information and tire information with the attention output, and process the concatenated vectors through a fully connected neural network model to obtain a tire recommendation score, thereby generating a tire recommendation list.

[0127] The present invention also provides an electronic device, see Figure 3 , shown is an electronic device in the third embodiment of the present invention, including a memory 10, a processor 20, and a computer program 30 stored in the memory 10 and executable on the processor 20. When the processor 20 executes the computer program 30, the above-mentioned tire trading method is implemented.

[0128] Among them, the memory 10 includes at least one type of storage medium, and the storage medium includes a flash memory, a hard disk, a multimedia card, a card-type memory (for example, an SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 10 can be an internal storage unit of an electronic device, such as a hard disk of the electronic device. In other embodiments, the memory 10 can also be an external storage device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Further, the memory 10 can also include both an internal storage unit of an electronic device and an external storage device. The memory 10 can be used not only to store application software and various types of data installed in the electronic device, but also to temporarily store data that has been output or is to be output.

[0129] In some embodiments, the processor 20 can be an electronic control unit (ECU, also known as a vehicle computer), a central processing unit (CPU), a controller, a microcontroller, a microprocessor or other data processing chip, used to run the program code stored in the memory 10 or process data, such as executing access restriction programs.

[0130] It should be pointed out that Figure 3 The structure shown does not constitute a limitation to the electronic device. In other embodiments, the electronic device may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0131] An embodiment of the present invention further provides a readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the tire trading method described above is implemented.

[0132] Those skilled in the art will appreciate that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0133] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0134] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement the hardware: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0135] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0136] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A tire trading method, characterized in that: The steps include: Users log in to the tire trading service platform and browse various types of tires according to their needs; The transaction service platform recommends the intended tires to the user through a tire recommendation model; The user selects a personal vehicle on the transaction service platform for order qualification assessment and displays the credit limit of the vehicle; The transaction service platform evaluates the credit limit of the vehicle. If approved, the user fills in the service address and appointment time, and signs the contract online; The transaction service platform generates multiple shippers based on the user's address for the tire delivery personnel to choose from; The tire delivery personnel provide installation services to the user and then complete the order; The construction of the tire recommendation model includes: Continuous and discrete processing of user information and tire information respectively; Mapping the continuous and discretized user information and tire information into low-dimensional vectors; Obtain a user browsing behavior sequence, extract a user long-term preference sequence from the user browsing behavior sequence based on an LSTM network, calculate an attention output from the user long-term preference sequence based on a two-layer attention mechanism, concatenate the low-dimensional vectors of the user information and tire information with the attention output, and process the concatenated vectors through a fully connected neural network model to obtain a tire recommendation score, thereby generating a tire recommendation list.

2. The tire trading method according to claim 1, characterized in that: The user information and tire information are respectively processed in a continuous and discrete manner, specifically comprising the following steps: Obtaining user discrete features and user continuous features based on the user's account information on the transaction service platform, and obtaining a user browsing behavior sequence based on the user's browsing behavior on the transaction service platform; Obtaining tire discrete features and tire continuous features based on basic tire information on the trading service platform; The user discrete features and the tire discrete features are mapped into natural numbers, and the user continuous features and the tire continuous features are range-divided for discretization.

3. The tire trading method according to claim 2, characterized in that: Mapping the continuous and discretized user information and tire information into low-dimensional vectors specifically includes the following steps: Converting the user discrete features and user continuous features into user low-dimensional vectors; Converting the tire discrete features and tire continuous features into tire low-dimensional vectors; A random replacement layer is introduced to randomly transform the user low-dimensional vector and the tire low-dimensional vector in a probabilistic manner during model training.

4. The tire trading method according to claim 3, characterized in that: The method of obtaining a user browsing behavior sequence, extracting a user long-term preference sequence from the user browsing behavior sequence based on an LSTM network, calculating an attention output from the user long-term preference sequence based on a double-layer attention mechanism, concatenating the low-dimensional vectors of the user information and tire information with the attention output, and processing the concatenated vectors through a fully connected neural network model to obtain a tire recommendation score, thereby generating a tire recommendation list, specifically includes the following steps: Based on the LSTM network, the user browsing behavior sequence is processed and the hidden layer state h from the first time period to the tth time period in the user browsing behavior sequence is calculated. t ; Based on the hidden layer state h from the first time period to the tth time period in the user browsing behavior sequence t , perform weighted operations on the browsing sequence of the tth time period in the user's browsing behavior sequence, and calculate the user's recent preference sequence P a , and calculate the user's long-term preference sequence P b ; From the user's long-term preference sequence P b Extracting the user's evaluation record MT for the tire and mapping the evaluation record into an evaluation low-dimensional vector; Calculating the attention score of each tire feature in the evaluation low-dimensional vector based on a dual-layer attention mechanism; Based on the attention score, performing weighted summation on the tire features to obtain a weighted tire feature; Performing evaluation and record attention calculation on the weighted rear tire features and the time features in the evaluation low-dimensional vector to obtain an attention output; Concatenate the attention output, the user low-dimensional vector, and the tire low-dimensional vector to obtain a concatenated vector; The concatenated vector is input into a fully connected neural network model for processing to obtain a tire recommendation score, thereby generating a tire recommendation list.

5. The tire trading method according to claim 4, characterized in that: The user discrete features include truck type, tire installation location, cargo type, and driver's driving habits; the user continuous features include expected lifespan, load range, and driver's daily mileage; the tire discrete features include tire category, pattern type, and compatible vehicle models; the tire continuous features include tire size, load rating, and performance indicators.

6. The tire trading method according to claim 1, characterized in that: The user selects a personal vehicle on the transaction service platform for order qualification assessment and displays the credit limit of the vehicle, including: The user selects one of the multiple pre-stored vehicles in the transaction service platform to place an order; The transaction service platform evaluates the credit qualifications of the selected vehicle and displays the credit limit on the operation interface.

7. The tire trading method according to claim 1, characterized in that: The transaction service platform evaluates the credit limit of the vehicle. If approved, the user fills in the service address and appointment time, and signs the online contract, which specifically includes: The user fills in the order quantity of the tires to be purchased and clicks "Order Now"; The user fills in the service address and service time popped up on the transaction service platform to make an appointment; The user fills in and signs the online electronic contract that pops up on the transaction service platform to generate an order.

8. A tire trading system, characterized in that: include: User: Log in to the tire trading service platform and browse various types of tires according to needs; The user selects a personal vehicle on the transaction service platform for order qualification assessment and displays the credit limit of the vehicle; Transaction service platform: recommends the intended tires to the user through the tire recommendation model; evaluates the credit limit of the vehicle, and if approved, the user fills in the service address and appointment time, and signs the contract online; generates multiple shippers based on the user's address for the tire delivery personnel to choose from; Tire delivery personnel: provide installation services for the users and complete the orders; The construction of the tire recommendation model includes: Continuous and discrete processing of user information and tire information respectively; Mapping the continuous and discretized user information and tire information into low-dimensional vectors; Obtain a user browsing behavior sequence, extract a user long-term preference sequence from the user browsing behavior sequence based on an LSTM network, calculate an attention output from the user long-term preference sequence based on a two-layer attention mechanism, concatenate the low-dimensional vectors of the user information and tire information with the attention output, and process the concatenated vectors through a fully connected neural network model to obtain a tire recommendation score, thereby generating a tire recommendation list.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the tire transaction method according to any one of claims 1 to 7 is implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the tire trading method according to any one of claims 1 to 7 is implemented.

Images