Cargo source quality acquisition method and device, computer device, and storage medium

Abstract

The application discloses a kind of goods source quality acquisition method, device, computer equipment and storage medium, belong to internet technical field.Application in computer equipment, this goods source quality acquisition method includes: obtaining the first quality grade corresponding to the absolute value of goods source, absolute value is used to indicate the income level based on the physical attribute information of goods source;Obtain the second quality grade corresponding to the relative value of goods source, relative value is used to indicate the income level based on the market attribute information of goods source;Obtain the third quality grade corresponding to the fusion value of goods source, fusion value is used to indicate the income level based on the physical attribute information of goods source and the market attribute information of goods source is integrated;According to first quality grade, second quality grade and third quality grade, obtain goods source quality.This scheme considers the information of goods source more comprehensive, the actual level of goods source can be reflected by the goods source quality obtained, improve the accuracy of evaluating goods source.

Description

Technical Field

[0001] This application relates to the field of Internet technology, and in particular to a method, apparatus, computer equipment, and storage medium for obtaining the quality of goods. Background Technology

[0002] With the rapid development of science and technology, various applications on the Internet and smart terminals have brought great convenience to people's lives. In terms of freight transportation, cargo owners can use applications on smart terminals to publish the cargo to be transported to a cargo operation platform on a server.

[0003] Currently, to improve order completion rates, the backend server performs quality grading on all goods posted on the goods operation platform and makes different push or scheduling strategies based on the final quality grading. In traditional solutions, single indicators such as freight costs and transportation distances are generally relied upon as the basis for quality grading. The entire process considers too few parameters, resulting in an incomplete quality assessment of the goods and low accuracy in assessing the goods. Summary of the Invention

[0004] To address the problems in related technologies, achieve a more comprehensive assessment of the quality of goods sources, and improve the accuracy of goods source evaluation, this application provides a method, apparatus, computer equipment, and storage medium for obtaining goods source quality. The technical solution is as follows: In one aspect, this application provides a method for obtaining the quality of goods sources, applied to computer equipment, the method comprising: Obtain the first quality level corresponding to the absolute value of the goods, whereby the absolute value is used to indicate the revenue level obtained based on the physical attribute information of the goods; Obtain the second quality grade corresponding to the relative value of the goods, wherein the relative value is used to indicate the profit level obtained based on the market attribute information of the goods; Obtain the third quality level corresponding to the fusion value of the source of goods, wherein the fusion value is used to indicate the revenue level obtained by combining the physical attribute information and the market attribute information of the source of goods; The quality of the goods is obtained based on the first quality level, the second quality level, and the third quality level.

[0005] In another aspect, this application provides a source quality acquisition device, applied to computer equipment, the device comprising: The first acquisition module is used to acquire the first quality level corresponding to the absolute value of the goods, wherein the absolute value is used to indicate the profit level obtained based on the physical attribute information of the goods. The second acquisition module is used to acquire the second quality grade corresponding to the relative value of the goods, wherein the relative value is used to indicate the profit level obtained based on the market attribute information of the goods. The third acquisition module is used to acquire the third quality level corresponding to the fusion value of the goods source. The fusion value is used to indicate the revenue level obtained by combining the physical attribute information and the market attribute information of the goods source. The fourth acquisition module is used to acquire the quality of the goods source based on the first quality level, the second quality level, and the third quality level.

[0006] In another aspect, this application provides a computer device comprising a processor and a memory, wherein the memory stores a computer program that is executed by the processor to implement the source quality acquisition method as described in one aspect above.

[0007] In another aspect, this application provides a computer-readable storage medium storing a computer program that is executed by a processor to implement the source quality acquisition method as described in one aspect above.

[0008] On the other hand, embodiments of this application provide a computer program product that, when run on a computer, causes the computer to execute to implement the source quality acquisition method as described in one aspect above.

[0009] On the other hand, embodiments of this application provide an application publishing platform for publishing computer program products, wherein when the computer program products are run on a computer, the computer executes to implement the source quality acquisition method as described in one aspect above.

[0010] The beneficial effects of the technical solutions provided in this application include at least the following: The method for obtaining the quality of goods provided in this application involves: obtaining a first quality level corresponding to the absolute value of the goods, where the absolute value indicates the profit level based on the physical attributes of the goods; obtaining a second quality level corresponding to the relative value of the goods, where the relative value indicates the profit level based on the market attributes of the goods; and obtaining a third quality level corresponding to the combined value of the goods, where the combined value indicates the profit level based on both the physical and market attributes of the goods. The quality of the goods is then obtained based on these three quality levels. This method combines the first, second, and third quality levels for comprehensive evaluation to obtain the final quality of the goods. It considers more comprehensive information about the goods, and the obtained quality reflects the actual level of the goods, thus improving the accuracy of the goods assessment. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of a scenario architecture for a dedicated APP for cargo transportation, as described in an exemplary embodiment of this application. Figure 2 This is a flowchart of a method for obtaining the quality of goods provided in an exemplary embodiment of this application; Figure 3 This is a flowchart of another method for obtaining the quality of goods provided in an exemplary embodiment of this application; Figure 4 This is a structural block diagram of a source quality acquisition device provided in an exemplary embodiment of this application; Figure 5 This is a schematic diagram of the structure of a computer device provided in an exemplary embodiment of this application. Detailed Implementation

[0013] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0014] In this article, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0015] The solution provided in this application can be used in real-world scenarios where, in the transportation of goods, the back-end computer equipment performs quality assessments on various sources of goods published on the application platform. To facilitate understanding, some terms and application scenarios involved in the embodiments of this application will be briefly introduced below.

[0016] Income Per Mile (IPM) score, or IPM score, reflects how much income a driver earns per mile of transport.

[0017] The Income Per Hour (IPH) score, or IPH score, reflects how much income a driver can earn per hour of transport.

[0018] Mileage score: Reflects the value increase resulting from the increase in transportation distance. It can be calculated using an exponential saturation function and has a diminishing marginal effect.

[0019] Distributed locks are a synchronization mechanism used in distributed systems to control concurrent access to shared resources by multiple nodes (servers / processes). In distributed systems, external middleware (Redis / ZK) can act as "traffic police," queuing requests from all servers in order to ensure that the same goods are not simultaneously claimed by two different users.

[0020] Performance testing data refers to test sample datasets constructed to simulate real business scenarios in order to evaluate the stability, throughput, and response speed of a system under high load before the system goes live.

[0021] Shared cargo refers to cargo that is planned to share the transport space of a truck with other goods and be transported through "carpooling".

[0022] Recall driver information: This refers to the collection of information on drivers who can potentially accept orders from the vicinity of a given freight source; that is, the collection of driver information that can be recalled for that freight source. Each driver's information includes their unique identifier, the distance from the driver to the freight loading point, the type of vehicle used by the driver, and the length of the vehicle used by the driver.

[0023] With the rapid development of science and technology, various applications on the Internet and smart terminals have brought great convenience to people's lives. In order to provide convenience for both freight and cargo, there are more and more applications (APPs) specifically for freight transportation. Shippers can post the goods they need to transport (i.e., cargo sources) on these APPs, and drivers can also search for / view cargo sources that interest them on these APPs.

[0024] Please refer to Figure 1 This illustration shows a schematic diagram of a scenario architecture for cargo transportation using a dedicated APP, as per an exemplary embodiment of this application. Figure 1 As shown, it includes the cargo owner's terminal device 101, server 102, and vehicles 103 for each driver.

[0025] The cargo owner's terminal device 101 can be connected to the server 102 through a communication network, and each driver can also use their own terminal device to connect to the server 102 through the communication network. Optionally, the communication network can be a wired network or a wireless network.

[0026] Optionally, the aforementioned communication network connection uses standard communication technologies and / or protocols. The network is typically the Internet, but can also be any network, including but not limited to any combination of Local Area Network (LAN), Metropolitan Area Network (MAN), Wide Area Network (WAN), mobile, wired or wireless networks, private networks, or Virtual Private Networks. In some embodiments, technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc., are used to represent data exchanged over the network. Furthermore, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), and Internet Protocol Security (IPsec) can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can be used to replace or supplement the aforementioned data communication technologies.

[0027] Optionally, the aforementioned terminal devices may include, but are not limited to, wearable devices (such as wristbands, smartwatches, smart glasses, etc.), mobile phones, tablets, laptops, desktop computers, portable laptops, smart home devices, and other electronic devices with APP interaction functions.

[0028] Optionally, server 102 can be at least one of the following: a single server, multiple servers, a cloud computing platform, and a virtualization center. Server 102 is used to provide backend services to support the aforementioned dedicated APP. Optionally, server 102 can undertake the main computing work, and each terminal device 101 can undertake secondary computing work; or, server 102 can undertake secondary computing work, and each terminal device 101 can undertake primary computing work; or, server 102 and each terminal device 101 can use a distributed computing architecture for collaborative computing.

[0029] For example, a cargo operation platform can be maintained on server 102. The terminal devices 101 of various cargo owners can interact with the cargo operation platform on server 102, and the vehicles of various drivers can also interact with the cargo operation platform on server 102 through driver terminal devices or vehicle terminals. Cargo owners can download the aforementioned dedicated APP on their terminal devices 101. When they want to transport a certain cargo to a certain destination, they can publish the corresponding cargo on the cargo operation platform through the dedicated APP. Similarly, drivers can download the same dedicated APP on their terminal devices and search for various cargo published on the cargo operation platform as drivers, thereby transporting cargo that they are interested in or suitable for their current location.

[0030] Currently, to improve order completion rates, the backend server performs quality grading on all cargo listings on the cargo operation platform and implements different push or scheduling strategies based on the final quality grading. Traditional solutions generally rely on single indicators such as freight costs and transportation distance as the basis for quality grading. For example, the value of a cargo is calculated based on its actual attributes (list price, transportation distance), and different scheduling strategies are applied according to the cargo's value. However, this value calculation method is too simplistic and does not consider driver income, market supply and demand, or the differences between various cargo listings. Furthermore, it ignores key economic indicators such as revenue per unit time, revenue per unit mileage, the impact of the initial fare, and negotiation costs.

[0031] For drivers, a high-cost but long-distance, time-consuming shipment may actually yield far less profit than a short-distance shipment with moderate costs and quick turnover. Therefore, simply assessing the quality of cargo based on listed prices / prices results in a significant discrepancy between the assessment and the driver's perceived actual earnings. Furthermore, cargo with special requirements increases the difficulty of securing orders, cargo requiring express delivery suffers from instability, cargo repeatedly listed indicates unsold inventory, and traffic restrictions at the destination increase transportation costs—all these negative factors directly impact drivers' willingness to accept orders and their actual earnings. However, traditional assessment systems lack a systematic consideration of these factors and deduction mechanisms, leading to incomplete quality assessments and low accuracy in cargo evaluation.

[0032] To address the problems existing in the aforementioned related technologies, achieve a more comprehensive assessment of the quality of goods, and improve the accuracy of the assessment, this application provides a method for obtaining the quality of goods. This method can combine a first quality level, a second quality level, and a third quality level for comprehensive evaluation to obtain the final quality of the goods. This method considers more comprehensive information about the goods, and the obtained quality of the goods can reflect the actual level of the goods.

[0033] Please refer to Figure 2 The illustration shows a flowchart of a method for obtaining the quality of goods according to an exemplary embodiment of this application. This method for obtaining the quality of goods can be applied to a computer device, which may be the aforementioned Figure 1 The server in the middle. For example... Figure 2 As shown, the method for obtaining the quality of the goods may include the following steps: Step 201: Obtain the first quality level corresponding to the absolute value of the goods. The absolute value is used to indicate the profit level obtained based on the physical attribute information of the goods.

[0034] Optionally, in this solution, the computer device can receive a request to perform a quality assessment on one or more sources of goods, and in response to the request, execute the method for obtaining the quality of goods in this solution. For example, for source one posted by a cargo owner, if the computer device wants to obtain the quality of the goods corresponding to source one, it can execute the process of this solution on source one.

[0035] Optionally, the computer equipment can obtain the first quality grade corresponding to the absolute value of the goods in this step. The first quality grade can assess the absolute value of the goods; the higher the first quality grade, the higher the absolute value of the goods.

[0036] Optionally, the computer equipment can first determine the corresponding revenue level (i.e., absolute value) based on the physical attribute information of the goods, and then obtain the first quality level corresponding to the absolute value of the goods. The physical attribute information of the goods can include price tag information, constraint tag information, and shipper tag information. Price tag information can include tags related to price such as the total price of the goods, the unit price of the goods, the starting price of the goods, the negotiated price of the goods, and the standard price of the goods. Constraint tag information can include special requirements of the goods or agreements with the shipper, such as tags indicating that the goods require express delivery, that the goods are already repeatedly listed, that the destination of the goods is a restricted area for a certain type of vehicle, that the payment method is cash on delivery, or that a driver is required to help with unloading. Shipper tag information can include information about the shipper, such as the shipper's fulfillment rate and high-quality shipper tags.

[0037] Step 202: Obtain the second quality level corresponding to the relative value of the goods. The relative value is used to indicate the profit level obtained based on the market attribute information of the goods.

[0038] Optionally, the computer equipment can obtain a second quality level corresponding to the relative value of the goods in this step. Similarly, the second quality level can assess the relative value of the goods; the higher the second quality level, the higher the relative value of the goods.

[0039] Optionally, the computer equipment can first obtain the corresponding revenue level (i.e., relative value) based on the market attribute information of the cargo, and then obtain the second quality level corresponding to the relative value of the cargo. The market attribute information of the cargo may include recalled driver information and market feedback information. The market feedback information includes the deviation between the current actual proportion and the target proportion of cargo quantity included in each grade interval of all cargo.

[0040] Optionally, each grade range for all sources of goods described above can be pre-defined and set by developers in a computer device. The target percentage can be understood as the expected percentage, that is, the percentage of each source of goods expected to be present in each grade range.

[0041] Step 203: Obtain the third quality level corresponding to the fusion value of the source of goods. The fusion value is used to indicate the revenue level obtained by combining the physical attribute information and market attribute information of the source of goods.

[0042] Optionally, the computer equipment can obtain the third quality level corresponding to the blending value of the source in this step. Similarly, the third quality level can assess the blending value of the source; the higher the third quality level, the higher the blending value of the source.

[0043] Optionally, the fusion value is used to indicate the revenue level obtained by combining the physical attribute information and market attribute information of the goods. That is, the computer equipment can perform a fusion evaluation based on the aforementioned physical attribute information and market attribute information of the goods to obtain the desired fusion value.

[0044] It should be noted that the order of steps 201 to 203 above does not represent the actual execution order. In practical applications, steps 201 to 203 can be executed synchronously and in parallel.

[0045] Step 204: Obtain the quality of the goods based on the first quality level, the second quality level, and the third quality level.

[0046] Optionally, after obtaining the first quality level, the second quality level, and the third quality level, the computer equipment can obtain the final quality of the goods based on the first quality level, the second quality level, and the third quality level.

[0047] For example, a computer device can obtain the quality of a source of goods in the following way: For the obtained first quality level, second quality level, and third quality level, the computer device is configured with weight parameters for each of the first, second, and third quality levels. The source quality can be obtained based on the first quality level, second quality level, third quality level, and their respective weight parameters. Alternatively, the computer device can also select the third quality level as the source quality from the first, second, and third quality levels. Alternatively, the computer device can also obtain the priority rules corresponding to the source of goods and obtain the source quality from the first, second, and third quality levels according to these priority rules.

[0048] In summary, the method for obtaining the quality of goods provided in this application obtains a first quality level corresponding to the absolute value of the goods, where the absolute value indicates the profit level based on the physical attribute information of the goods; a second quality level corresponding to the relative value of the goods, where the relative value indicates the profit level based on the market attribute information of the goods; and a third quality level corresponding to the integrated value of the goods, where the relative value indicates the profit level based on the combined physical and market attribute information of the goods. The quality of the goods is then obtained based on these three quality levels. This method combines the first, second, and third quality levels for comprehensive evaluation to obtain the final quality of the goods. It considers more comprehensive information about the goods, and the obtained quality reflects the actual level of the goods, thus improving the accuracy of the goods assessment.

[0049] Below, taking the server receiving a request to assess the quality of one or more sources of goods as an example, the server can prepare the relevant information for this request. After preparing all the necessary information (including the physical attributes and market attributes of the sources of goods), the server will execute the above-mentioned process for obtaining each quality level.

[0050] Please refer to Figure 3 It illustrates a flowchart of another method for obtaining the quality of goods provided by an exemplary embodiment of this application. This method for obtaining the quality of goods can be applied to a computer device, which can be the aforementioned Figure 1 The server in the middle. For example... Figure 3 As shown, the method for obtaining the quality of the goods may include the following steps: Step 301: Receive a request to obtain the quality of the goods.

[0051] Optionally, the computer equipment can perform quality assessments on any one or more cargo sources published on the cargo operation platform. A party requiring cargo quality information can initiate a corresponding quality acquisition request to the computer equipment. This request can include the identifier (ID) of each cargo source to be assessed. The computer equipment can receive this quality acquisition request and thus obtain the IDs of each cargo source to be assessed.

[0052] Optionally, the quality acquisition request can be initiated by back-end staff using management or electronic devices, initiated by the cargo owner, or periodically initiated by other roles in the server architecture (such as quality acquisition strategy / cargo matching strategy).

[0053] Step 302: In response to the quality acquisition request, obtain the physical attribute information and market attribute information of the source of goods.

[0054] Optionally, in response to the aforementioned quality acquisition request, the computer device may obtain the IDs of each source of goods to be evaluated, and based on these IDs, obtain the physical attribute information and market attribute information of each source of goods. If only one source ID is included, the computer device only needs to obtain the corresponding physical attribute information and market attribute information based on that source ID.

[0055] Optionally, before obtaining the physical attribute information and market attribute information of the goods, the computer device can also query whether there is a goods quality corresponding to the identification information in the computer device's cache based on the goods identification information; if there is a goods quality corresponding to the identification information in the computer device's cache, the goods quality corresponding to the identification information is obtained; if there is no goods quality corresponding to the identification information in the computer device's cache, a distributed lock is established based on the identification information, and the computer device's cache is queried again to see if there is a goods quality corresponding to the identification information. If there is no goods quality corresponding to the identification information in the computer device's cache, the step of obtaining the first quality level corresponding to the absolute value of the goods is executed.

[0056] In other words, after receiving the above quality acquisition request, the computer device first checks the cache for the quality of the source based on the source identification information to see if the quality of the source corresponding to the identification information ID already exists. If the quality of the source corresponding to the identification information ID already exists in the cache, it means that the computer device has already executed the method of this scheme for the source ID and has obtained the corresponding quality of the source. The computer device can then retrieve the calculated quality of the source from the cache.

[0057] If the computer device's cache does not contain the quality of the goods corresponding to the identification information, a distributed lock can be established based on the identification information ID, and the computer device's cache can be queried again to see if the quality of the goods corresponding to the identification information exists. That is, the computer device can acquire a distributed lock based on the goods ID, and after obtaining the distributed lock, it checks the computer device's cache again to see if the quality of the goods corresponding to the identification information exists, avoiding duplicate calculations. Only when the computer device's cache does not contain the quality of the goods corresponding to the identification information will the step of obtaining the first quality level corresponding to the absolute value of the goods (i.e., starting the calculation of the quality level of this scheme) begin. In this step, this is reflected in obtaining the physical attribute information and market attribute information of the goods based on the goods ID.

[0058] In one possible implementation, when there are multiple source IDs, a verification can be performed before obtaining the physical and market attribute information of the sources. For example, the computer equipment can determine the type of source based on each source ID. If it is a regular source, the basic information corresponding to these source IDs is queried and then filtered. If it is a consolidated source, the consolidation information is queried and filtered. This filtering includes directly skipping sources that do not meet preset conditions (including missing information, non-target type, or belonging to stress test data). That is, for each source ID, if the basic information or consolidation information does not meet the preset conditions, it is directly skipped and no quality assessment is performed. For those that meet the preset conditions, the subsequent steps are executed.

[0059] For ease of description, the following content of this solution will be described in terms of the quality acquisition request including a source ID.

[0060] Step 303: Obtain the first quality level corresponding to the absolute value of the goods. The absolute value is used to indicate the profit level obtained based on the physical attribute information of the goods.

[0061] Optionally, the computer device may obtain the first quality level corresponding to the absolute value of the goods in the following manner: obtain the physical attribute information of the goods; obtain the absolute value of the goods based on the physical attribute information of the goods; determine the third quantile of the absolute value of the goods in the third quantile data based on the absolute value of the goods and the third quantile data, wherein the third quantile data is obtained by dividing the absolute value of each goods under a third preset dimension, and the third quantile is used to indicate the proportion of other goods with an absolute value lower than that of the goods in the preset quantile data; and obtain the first quality level corresponding to the third quantile based on the third quantile and the third mapping rule.

[0062] Optionally, obtaining the physical attribute information of the goods can be replaced by step 302 above. The physical attribute information of the goods can be directly obtained from step 302, and then the step of obtaining the absolute value of the goods can be executed. Optionally, the physical attribute information may include price tag information, constraint tag information, and owner tag information. The computer equipment can obtain the absolute value of the goods based on the physical attribute information as follows: Based on the physical attribute information of the goods, obtain the basic score, weighted score, and control score of the goods. The basic score is used to indicate the score corresponding to the price tag information of the goods, the weighted score is used to indicate the score corresponding to the constraint tag information of the goods, and the control score is used to indicate the score corresponding to the owner tag information of the goods; the absolute value of the goods is obtained based on the basic score, weighted score, and control score.

[0063] Optionally, in this scheme, the absolute value is calculated using a three-stage calculation model of "basic score - weighted score + control score". The computer equipment first obtains the basic score, weighted score and control score based on the physical attribute information of the goods, and then calculates the absolute value required for this step according to the formula: absolute value = basic score - weighted score + control score.

[0064] In one possible implementation, the computer device can calculate the base score as follows: Based on the physical attribute information of the cargo, the computer device obtains the price multiplier, IPM score, IPH score, starting price score, negotiation score, and mileage score of the cargo. Based on these scores, the computer device calculates the base score of the cargo. The price multiplier can be calculated as the total price of the cargo divided by the standard price. The price multiplier reflects the degree of advantage of the current total price relative to the standard price. If the current total price is higher than the standard price, it means the cargo is at a premium, and the price multiplier is greater than 1. If the current total price is lower than the standard price, the price multiplier is less than 1.

[0065] Optionally, the IPM score can be calculated using the following formula 1: α (IPM quantiles in the first feature set) Formula 1; The calculation shows that α is a pre-set coefficient (a constant), and the first feature set is the set after the computer device calculates the corresponding IPM for each source of goods. Different IPM quantiles are mapped to this first feature set based on different IPMs. Please refer to Table 1, which shows the correspondence between different IPMs and different IPM quantiles according to an exemplary embodiment of this application.

[0066] Table 1

[0067] As shown in Table 1 above, the computer equipment can map different IPM quantiles to different IPM ranges in the first feature set in advance. When it is necessary to calculate the absolute value of the current cargo, the IPM of the current cargo is obtained by dividing the total price of the cargo by the transportation distance. By querying Table 1 or performing mapping, the IPM quantile of the current cargo's IPM in the first feature set is obtained. The calculation is then performed according to Formula 1 above to obtain the IPM score of the current cargo. Here, the current cargo is the cargo whose quality needs to be calculated in the quality acquisition request.

[0068] Optionally, the IPH score can be calculated using the following formula 2: IPH score = β1 × (IPH quantile in the second feature set) + β2 × (1 - e^(-γ × IPH)) Formula 2; In this context, β1, β2, and γ are pre-set coefficients (constant), α is a weighting coefficient, and the second feature set is the set after the computer equipment calculates the corresponding IPH for each cargo source. Different IPH quantiles are mapped to this second feature set based on different IPH values. This mapping method can be found in Table 1 above and will not be elaborated further here. That is, the computer equipment can pre-map different IPH quantiles to different IPH ranges in the second feature set. When calculating the absolute value of the current cargo source, the IPH of the current cargo source is obtained by dividing the total price of the cargo source by the transportation time. The mapping yields the IPH quantile of the current cargo source in the second feature set, and the calculation is performed according to Formula 2 above to obtain the IPH score of the current cargo source.

[0069] Optionally, for the starting price score, the computer device can determine whether the current item is a starting price item based on its physical attribute information. For example, if the price tag information in the physical attribute information of the current item includes a starting price tag, then the item is a starting price item; if the price tag information does not include a starting price tag, then the item is not a starting price item. If the current item is a starting price item, the starting price score is calculated; if the current item is not a starting price item, the starting price score can be directly set to 0.

[0070] Optionally, the starting price score can be calculated using the following formula 3: Starting price score = a × e^(-transportation distance) + b (Formula 3) Here, a and b are also pre-set coefficients (constants).

[0071] Optionally, for the negotiation score, similar to the starting price score, the computer device can determine whether the current source of goods is a negotiable source based on its physical attribute information. For example, if the price tag information in the physical attribute information of the current source of goods contains a negotiation label, then the source of goods is a negotiable source; if the price tag information in the physical attribute information of the current source of goods does not contain a negotiation label, then the source of goods is not a negotiable source. If the current source of goods is a negotiable source, a negotiation score of y (where y is a constant) can be assigned to it. If the current source of goods is not a negotiable source, the negotiation score can be directly set to 0. Optionally, y is also preset.

[0072] Optionally, the mileage score can be calculated using the following formula 4: Mileage score = c × (1 - e^(-transport distance × d)) Formula 4; Here, c and d are pre-set coefficients (constant). When the transportation distance is short, the mileage score increases rapidly, and after the transportation distance is large enough, it tends to saturate to a value of c, which conforms to the law of diminishing marginal returns in long-distance logistics.

[0073] After obtaining the price multiplier, IPM score, IPH score, starting price score, negotiation score, and mileage score of the current cargo through the above series of calculations, the computer equipment calculates the basic score of the cargo as follows: Basic score = Price multiplier × (IPM score + IPH score + Starting price score + Negotiation score + Mileage score).

[0074] In one possible implementation, the computer device can obtain the weighted score as follows: for the obtained physical attribute information of the current source of goods, query whether the physical attribute information has the triggering conditions corresponding to the weighted dimension. If it does, execute the corresponding deduction. Sum the deductions corresponding to all weighted dimensions to obtain the final weighted score.

[0075] Taking the weight reduction dimensions, including special intent dimension, express dimension, repetitive dimension, traffic restriction dimension, and payment method dimension, as an example, please refer to Table 2, which shows the correspondence between a weight reduction dimension, triggering condition, and deduction method involved in an exemplary embodiment of this application.

[0076] Table 2

[0077] In Table 2 above, z, j, i, k, and l are all pre-set parameters (constants).

[0078] Optionally, the computer equipment can query the physical attribute information of the current goods to see if the current goods meet the above triggering conditions. Whichever condition is met, a corresponding deduction is applied. The deductions for all weighting dimensions are summed to obtain the final weighting score. In Table 2 above, the weighting score = deduction for special intent dimension + deduction for express delivery dimension + deduction for repetition dimension + deduction for traffic restriction dimension + deduction for payment method dimension.

[0079] In one possible implementation, the computer device can calculate the control score as follows: For the shipper's tag information in the physical attribute information of the current cargo source, it queries whether the shipper corresponding to the tag information belongs to the target group. If the shipper belongs to the target group, the control score corresponding to the target group is obtained; if the shipper does not belong to the target group, a preset control score is obtained. For example, the target group consists of people with a credit rate below 50%. This means that for shippers, a low credit rate can affect driver payments or lead to discrepancies between the information on the goods to be transported and the actual goods. Therefore, in this solution, the shipper's tag information is considered to verify the shipper of the current cargo source, thus incorporating this into the calculation of the absolute value. For example, if the shipper belongs to the target group, the control score corresponding to the target group is obtained as m; if the shipper does not belong to the target group, the control score is 0.

[0080] After the computer equipment obtains the basic score, the weighted score, and the adjustment score according to the above process, the absolute value of the goods is calculated by subtracting the weighted score from the basic score and adding the adjustment score.

[0081] Optionally, after obtaining the absolute value of the goods, the computer equipment continues to determine the third quantile of the absolute value of the goods within the third bucket data, based on the absolute value of the goods and the third bucket data. The third bucket data is obtained by dividing the absolute value of each goods under a third preset dimension, which can be a "city + vehicle type" dimension. The third bucket data is the bucket data under the "city + vehicle type" dimension. For example, the computer equipment can pre-calculate the absolute value of each goods and bucket them under the "city + vehicle type" dimension to generate the third bucket data. In this step, the third bucket data is queried using the absolute value of the current goods to obtain the corresponding third quantile. For example, assuming the third bucket data is: 0#2.0#4.5#7.0#9.5#12.0#15.0#18.0#..., if the absolute value of the current goods is 12, falling in the 6th interval (12.0, 15.0) of the third bucket data, then the third quantile is 6.

[0082] Optionally, after obtaining the third quantile, the computer device can obtain the first quality level corresponding to the third quantile based on the third quantile and the third mapping rule. The third mapping rule can be pre-defined. Please refer to Table 3, which shows the correspondence between different quantiles and different quality levels according to an exemplary embodiment of this application.

[0083] Table 3

[0084] Optionally, the computer device can look up Table 3 based on the third quantile to obtain the first quality level corresponding to the third quantile. For example, if the third quantile is 6, it belongs to the quantile range 1. Therefore, the first quality level obtained through the third mapping rule is level D.

[0085] Taking the following cargo as an example, the process of calculating the absolute value above will be illustrated. The physical attributes of the cargo include: departure city: Suzhou, vehicle type: 4.2-meter van, transportation distance: 35 kilometers, total freight: 800 yuan, standard price: 600 yuan, it is a starting price cargo, not a negotiated order, not express, no special requirements, no traffic restrictions at the destination, not cash on delivery, and it has been listed twice; the parameter values ​​in the various formulas set in advance by the backend developers are shown in Table 4: Table 4

[0086] Therefore, when the computer equipment needs to calculate the absolute value of the goods, it uses the above scheme to calculate the IPM score: Assuming the IPM quantile of the first feature set is 12.0, then IPM = 0.4 × 12.0 = 4.8; IPH score: Assuming the IPM quantile of the second feature set is 8.0 and the IPH value is 50.0, the IPH score = 0.3 × 8.0 = 2.4 + 0.15 × (1 - e^(-0.008 × 50)) = 2.4 + 0.050 = 2.45; Starting price score (for starting price goods) = 2.5 × e^(-35) + 0.8 ≈ 0 + 0.8 = 0.8; Negotiation score (for non-negotiated prices) = 0; Mileage score = 3.0 × (1 - e^(-35 × 0.02)) = 1.51; Price multiplier: 800 / 600 = 1.33. The basic score obtained by the computer equipment = 1.33 × (4.8 + 2.45 + 0.8 + 0 + 1.51) = 12.71.

[0087] Optionally, the demotion score obtained by the computer equipment is as follows: Demotion score for special intent (no special needs) = 0; Demotion score for express delivery (non-express) = 0; Demotion score for duplication (listed twice) = 0.15 × 2 = 0.3; Demotion score for traffic restriction (no traffic restriction) = 0; Demotion score for payment method (non-cash on delivery) = 0; Demotion score = 0 + 0 + 0.3 + 0 + 0 = 0.3. Since the owner of this goods does not belong to the target group, its corresponding control score = 0.

[0088] Therefore, for the example above, the absolute value obtained by the computer equipment = 12.71 - 0.3 + 0 = 12.41. Combining this with the data from the third quantile, the absolute value of the current supply is 12.41, which falls in the 6th interval (12.0, 15.0) of the third quantile. Therefore, the third quantile is 6.

[0089] Step 304: Obtain the second quality level corresponding to the relative value of the goods. The relative value is used to indicate the profit level obtained based on the market attribute information of the goods.

[0090] Optionally, the computer device may obtain the second quality level corresponding to the relative value of the cargo source in the following manner: Obtain the market attribute information of the cargo source, including recalled driver information and market feedback information. The recalled driver information indicates the set of drivers that the cargo source can recall. The market feedback information includes the deviation between the current actual proportion and the target proportion of the number of cargo sources included in each grade interval of all cargo sources. Based on the recalled driver information included in the market attribute information of the cargo source, obtain the relative value of the cargo source. Based on the relative value of the cargo source and the second bucket data, determine the second quantile corresponding to the relative value of the cargo source in the second bucket data. The second bucket data is obtained by dividing the relative value of each cargo source under a second preset dimension. The second quantile indicates the proportion of other cargo sources with a relative value lower than that of the cargo source in the second bucket data. Based on the second quantile, market feedback information, and a second mapping rule, obtain the second quality level corresponding to the second quantile.

[0091] Optionally, obtaining the market attribute information of the goods here can be replaced by step 302 above. The market attribute information of the goods can be obtained directly from step 302 above, and then the step of obtaining the absolute value of the goods can be executed.

[0092] Optionally, the computer device may obtain the relative value of a cargo based on the recalled driver information contained in the cargo's market attribute information as follows: Based on the recalled driver information contained in the cargo's market attribute information, query each driver's acceptance rate and completion rate for the cargo; based on each driver's acceptance rate and completion rate, obtain an aggregate score, which indicates the probability that the cargo will be accepted by any driver; based on the cargo's identification information, obtain an order acceptance score, which indicates the difficulty of the cargo being accepted by any driver in a single order; based on the aggregate score, the order acceptance score, and the cargo's price, obtain the relative value of the cargo.

[0093] For example, during the process of obtaining market attribute information of cargo sources in step 2 above, the computer equipment will recall information on drivers in the vicinity who may accept orders for each cargo source. Therefore, each cargo source will have multiple corresponding driver information sets, and these sets of driver information constitute the recalled driver information. In this step, for each driver's identifier included in the recalled driver information for the current cargo source, the pick-up rate and completion rate for each driver for that cargo source are queried. In one possible implementation, the pick-up rate and completion rate for each cargo source can be calculated separately using the pick-up rate model and the completion rate model, respectively. That is, the computer equipment can initiate calculation requests for each driver's pick-up rate and completion rate for that cargo source, allowing the pick-up rate model and completion rate model to perform calculations, thereby obtaining the corresponding calculation results, and thus obtaining each driver's pick-up rate and completion rate for that cargo source.

[0094] Optionally, after obtaining the acceptance rate and completion rate of each driver for each cargo, the computer device then calculates an aggregate score based on these scores. For example, the computer device can calculate the aggregate score as follows: first, calculate the intention intensity for each driver, then aggregate the intention intensities of each driver into a single set to obtain the aggregate score. Here, YIX represents the intention intensity, JEp represents the acceptance rate of the p-th driver, and CJp represents the completion rate of the p-th driver, where p is an integer. Therefore, YIX = (JEp) n ×(CJp) m That is, a driver's intention intensity is equal to the nth power of the driver's pick-up rate multiplied by the mth power of the driver's completion rate. Let JH represent the aggregate score. The process by which the computer device aggregates the intention intensity of each driver into a whole can be as follows: JH = sigmoid(YIX) / q). Among them, YIX To negate the first summation result, which is the sum of the logarithms of the complements of the intention strength of each driver, q represents the total number of all drivers included in the recall driver information for the current freight source.

[0095] Optionally, the computer equipment also needs to obtain the order acceptance score based on the source's identification information. That is, based on the source's ID, the required order acceptance score is obtained. For example, the computer equipment can pre-calculate the corresponding order acceptance score for the source, and when needed, it can query the corresponding order acceptance score based on the source's ID. Alternatively, the computer equipment can also contain a machine learning model for order acceptance scores, which can calculate and obtain the order acceptance score for a specific source in real time based on the source's ID.

[0096] Optionally, after obtaining the aggregation score and order acceptance score, the computer device calculates the relative value of the goods based on these scores and the price of the goods. The price of the goods can refer to the total price of the goods, which can be obtained from the price tag information of the goods mentioned above. In one possible implementation, the computer device can calculate the relative value of the goods as follows: The relative value of the goods is equal to sigmoid(w1 × aggregate score raised to the power of r × price of the goods raised to the power of s + w2 × order acceptance score raised to the power of t). Among them, w1 and w2 are two weight values, and r, s, and t are constants, all of which can be independently configured in advance by developers or staff.

[0097] Optionally, after obtaining the relative value of the goods, the computer equipment continues to determine the second quantile of the relative value of the goods in the second bucket data based on the relative value of the goods and the second bucket data. The second bucket data is similar to the third bucket data, except that it is obtained by dividing the relative value of each goods. For example, the second preset dimension can also be a "city + vehicle type" dimension. The second bucket data is the bucket data corresponding to the relative value of each goods under the "city + vehicle type" dimension. The computer equipment can calculate the relative value of each goods in advance and bucket them under the "city + vehicle type" dimension to generate the second bucket data. In this step, the corresponding second quantile is obtained by querying the second bucket data using the current relative value of the goods. For example, computer equipment performs binary positioning on the relative value of each source of goods in the bucket threshold string of "city + vehicle type" dimension, and obtains bucket numbers from 1 to 100. If the relative value of the current source of goods is between the 56th and 57th thresholds, that is, it falls in the 56th interval, then the second quantile is 56.

[0098] Optionally, after obtaining the second quantile, the computer device can obtain the second quality level corresponding to the second quantile based on the second quantile, market feedback information, and the second mapping rule. The second mapping rule can also be pre-defined. In one possible implementation, the computer device obtains the second quality level corresponding to the second quantile based on the second quantile, market feedback information, and the second mapping rule as follows: Based on the second quantile and the second mapping rule, obtain the quality level to be adjusted; if the target deviation in the market feedback information exceeds a preset deviation threshold, adjust the quality level to obtain the target quality level, and use the adjusted target quality level as the second quality level. The target deviation is the deviation between the current actual proportion and the target proportion corresponding to the grade interval of the quality level to be adjusted in the market feedback information; if the target deviation in the market feedback information does not exceed the preset deviation threshold, use the quality level to be adjusted as the second quality level.

[0099] Optionally, the form of the second mapping rule can be referenced in Table 3 above. The computer device queries based on the second quantile to obtain a quality level corresponding to the second quantile (this quality level is temporarily the quality level to be adjusted). For example, if the second quantile is 56, it belongs to the quantile range three. Therefore, the quality level to be adjusted obtained through the second mapping rule is level B.

[0100] In this embodiment, the computer device further determines whether the quality level to be adjusted needs to be adjusted based on the deviation between the current actual proportion and the target proportion corresponding to the level range in the market feedback information where the quality level to be adjusted is located.

[0101] In one possible implementation, market feedback information is obtained based on real-time supply distribution information and target stratification ratio information. The real-time supply distribution information indicates the current actual distribution percentage of supplies within each tier of all supply categories, while the target stratification ratio information indicates the expected percentage of supplies within each tier of all supply categories. For example, the internal system of a computer device maintains statistics on the real-time supply quantity at each quantile from 1 to 100 under the "city + vehicle type" dimension. For a certain city and vehicle type, there are currently 1500 supply items. Of these, 180 fall within the 1-20 quantile, 420 fall within the 21-40 quantile, and so on. This distribution reflects the real-time structure of the current market supply (i.e., real-time supply distribution information). In this example, the percentage of supply items falling within the 1-20 quantile is (180 / 1500) = 12%. Please refer to Table 5, which shows the current actual percentage of supply items within each tier of all supply categories in the cached data of a computer device, according to an exemplary embodiment of this application. Table 5

[0102] Optionally, the internal system of the computer equipment is also configured with target proportions (i.e., target stratification proportion information) under the "city + vehicle type" dimension, that is, the expected proportions of each of the above S / A / B / C / D levels. For example, it may be configured that S level accounts for 10%, A level accounts for 20%, B level accounts for 30%, C level accounts for 20%, and D level accounts for 20%.

[0103] Based on the actual and target proportions of computer equipment in each of the above-mentioned grade ranges, the following market feedback information can be obtained: Grade D: -8% (under-allocated), Grade C: +8% (over-allocated), Grade B: +4% (balanced), Grade A: -2% (balanced), Grade S: -2% (balanced). Please refer to Table 6, which shows the deviation between the actual and target proportions.

[0104] Table 6

[0105] Optionally, taking a preset deviation threshold of 5% as an example, for each of the above levels, if the obtained second quantile is 56, the quality level to be adjusted is level B. In the market feedback information, the target deviation is 4%, which does not exceed the preset deviation threshold of 5%, so the quality level to be adjusted (level B) can be used as the second quality level. If the obtained second quantile is 21, the quality level to be adjusted is level C. In the market feedback information, the target deviation is 8%, which exceeds the preset deviation threshold of 5%, so the quality level to be adjusted is adjusted to obtain the target quality level. For example, the relative value of the source of goods falls in the level C range (barrel number 30), and the current proportion of level C exceeds the target by 8%, which belongs to the "over-allocated" level, while the "under-allocated" level is level D. Since level C and level D are adjacent in order, the source of goods will be adjusted to level D, and the final output quantile is level D. Correspondingly, the second quantile is adjusted to the quantile corresponding to level D, for example, according to a fixed value (10), or according to the value obtained by adding 1 to the left endpoint of the level D range.

[0106] Optionally, the above adjustment process can be performed within a preset adjustment period. For example, if the calculation is performed within the adjustment period, the above process is executed; otherwise, the quality level to be adjusted is used as the second quality level.

[0107] Optionally, after determining the second quantile of the relative value of a source in the second bucket data based on its relative value and the second bucket data, the computer equipment can also update the aforementioned real-time distribution information of the sources based on the second quantile corresponding to the relative value of the sources. That is, for each source's second quantile after calculation, the number of sources corresponding to its level in the real-time distribution information is incremented by 1, for use in subsequent source feedback calculations.

[0108] Taking Suzhou as the departure city and a 4.2-meter van as an example, the following driver recall information was obtained during the relative value calculation: 20 drivers from the surrounding area were recalled for this cargo, of whom 15 had a matching vehicle type and 5 had a partially matching vehicle type. The average distance between the matched drivers was 8 kilometers. Market feedback information is as shown in Tables 5 and 6 above. The pick-up rate index n and the transaction rate index m were calculated using default values ​​of 1 and 3, respectively, and the intention intensity of each group of drivers is shown in Table 7: Table 7

[0109] From the above, the sum of the logarithms is -0.63233, and the exponent is 0.5314. This means the probability that none of the 20 drivers will complete a transaction is 53.1%, and the probability that at least one driver will accept the cargo is 46.9%. The aggregate score JH = sigmoid(0.632326 ÷ 20) = 0.5079. The query shows that the order acceptance score for this cargo is 0.62.

[0110] Optionally, the computer equipment continues to calculate with an exponent r of 1.2, s of 0.15, weights w1 of 1.0, w2 of 0.05, and an exponent t of 0.01. Then, the relative value of the source = sigmoid(w1 × 0.5079 raised to the power of 1.2 × 800 raised to the power of 0.15 + w2 × 0.62 raised to the power of 0.01) = sigmoid(1.2089 + 0.0498) = 0.7788. Since 0.7788 falls between the 56th threshold of 0.7712 and the 57th threshold of 0.7903, the bucket number is 56. Bucket number 56 falls within the B-level interval [40, 70], so the obtained second quality level is B.

[0111] It should be noted that in the examples above of obtaining the first quality level corresponding to absolute value and the second quality level corresponding to relative value, the first and second preset dimensions are both exemplified using the "city + vehicle type" dimension. In practical applications, the "vehicle type" dimension or other bucketed data under dimensions defined by the developers can also be used.

[0112] Step 305: Obtain the third quality level corresponding to the fusion value of the source of goods. The fusion value is used to indicate the revenue level obtained by combining the physical attribute information and market attribute information of the source of goods.

[0113] Optionally, the computer device may obtain the third quality level corresponding to the fusion value of the cargo source in the following manner: obtain the physical attribute information of the cargo source and the recalled driver information, wherein the recalled driver information is used to indicate the set of driver information that the cargo source can recall; obtain the fusion value of the cargo source based on the physical attribute information of the cargo source and the recalled driver information; determine the first quantile corresponding to the fusion value of the cargo source in the first bucket data based on the fusion value of the cargo source and the first bucket data, wherein the first bucket data is obtained by dividing the fusion value of each cargo source under a first preset dimension, and the first quantile is used to indicate the proportion of other cargo sources with a fusion value lower than that of the cargo source in the first bucket data; obtain the third quality level corresponding to the first quantile based on the first quantile and the first mapping rule.

[0114] Optionally, obtaining the physical attribute information of the cargo and the driver recall information can be replaced by step 302 above. That is, the physical attribute information and market attribute information of the cargo can be directly obtained from step 302, and then the step of obtaining the fusion value of the cargo can be executed. In other words, the physical attribute information and market attribute information of the cargo obtained in step 302 can be used for subsequent steps 303 to 305, enabling the computer equipment to perform three-way parallel computing.

[0115] Optionally, the computer equipment may obtain the fusion value of the cargo source based on the physical attribute information of the cargo source and the recalled driver information in the following ways: Based on the recalled driver information, query the pick-up rate and completion rate of each driver for the cargo source; Based on the pick-up rate and completion rate of each driver for the cargo source, obtain scalar aggregation features and distributed aggregation features, whereby the scalar aggregation features are used to indicate the relative value of the cargo source, and the distributed aggregation features are used to indicate the structural relationship between the quantity and quality of drivers supplying the cargo source; Based on the physical attribute information of the cargo source, the scalar aggregation features and / or the distributed aggregation features, obtain the fusion value of the cargo source.

[0116] The method used in this step to query each driver's acceptance rate and transaction rate for cargo is the same as the method used in step 304 above, where the computer device queries each driver's acceptance rate and transaction rate for cargo based on the recalled driver information included in the market attribute information of the cargo. Therefore, it will not be repeated here. That is, the acceptance rate and transaction rate for each driver obtained above can be used in the relative value calculation process in step 304, or in the integrated value calculation process in step 305.

[0117] Optionally, the scalar aggregation feature is used to indicate the relative value of the cargo. In practical applications, the scalar aggregation feature adopts the relative value of the cargo. For the process of obtaining the scalar aggregation feature based on the pick-up rate and transaction rate of each driver, the process can also refer to the description of calculating the relative value in step 305 above, which will not be repeated here.

[0118] Optionally, the computer device can obtain the distribution aggregation features based on each driver's acceptance rate and transaction rate for the cargo as follows: The computer device can classify each driver in the recalled driver information into the same number of intention tiers according to a preset transaction rate threshold range, and count the number of drivers in each tier. Finally, each intention tier, the number of drivers in each tier, and the total number of drivers are used as the obtained distribution aggregation features.

[0119] For example, the preset conversion rate threshold ranges include seven intervals: (0, 0.4634], (0.4634, 0.6466], (0.6466, 0.7772], (0.7772, 0.8333], (0.8333, 0.8867], (0.8867, 0.9157], and (0.9157, 1.0]. Therefore, there are seven corresponding intention levels. Please refer to Table 8, which shows the correspondence between an exemplary embodiment of this application, the corresponding conversion rate threshold ranges, and the number of drivers in each level.

[0120] Table 8

[0121] Optionally, for each row of data in Table 8 above, the computer device can combine it with the total number of drivers to obtain the distribution aggregation features that need to be acquired.

[0122] After the computer device obtains the required scalar aggregation features and distribution aggregation features, it can acquire the fusion value of the goods based on the physical attribute information, scalar aggregation features, and / or distribution aggregation features of the goods. In one possible implementation, the computer device includes a fusion learning model whose input is either the physical attribute information and scalar aggregation features of the goods, or the physical attribute information and distribution aggregation features of the goods. The output of the fusion learning model is the required fusion value. That is, the computer device inputs the physical attribute information and scalar aggregation features of the goods as feature parameters to the fusion learning model and obtains its output fusion value; or, the computer device inputs the physical attribute information and distribution aggregation features of the goods as feature parameters to the fusion learning model and obtains its output fusion value.

[0123] Optionally, after obtaining the fusion value, the computer device determines the first quantile of the fusion value of the cargo in the first bucket data based on the fusion value of the cargo and the first bucket data. Then, based on the first quantile and the first mapping rule, it obtains the third quality level corresponding to the first quantile. The first bucket data is similar to the third and second bucket data mentioned above, except that it is obtained by dividing the fusion value of each cargo. For example, the first preset dimension can also be a "city + vehicle type" dimension. The first bucket data is the bucket data corresponding to the fusion value of each cargo under the "city + vehicle type" dimension. The computer device can pre-calculate the fusion value of each cargo and bucket it under the "city + vehicle type" dimension to generate the first bucket data. In this step, the first bucket data is queried using the fusion value of the current cargo to obtain the corresponding first quantile. For example, computer equipment can perform binary positioning on the bin threshold string of "city + vehicle type" to determine the fusion value of each source of goods, and obtain bin numbers from 1 to 100. If the fusion value of the current source of goods is between the 42nd and 43rd thresholds, that is, it falls in the 42nd interval, then the first quantile is 42.

[0124] Optionally, after obtaining the first quantile, the computer device can obtain the first quality level corresponding to the first quantile based on the first quantile and the first mapping rule. The first mapping rule can also be pre-defined; its implementation can be found in Table 3, and will not be elaborated here. Optionally, the computer device can look up the first quantile in Table 3 to obtain the third quality level corresponding to the first quantile. For example, if the first quantile is 42, it falls within the quantile range three; therefore, the first quality level obtained through the first mapping rule is level B.

[0125] In one possible implementation, before determining the first quantile of the cargo's fusion value in the first bucket data based on the cargo's fusion value and the first bucket data, the computer device can also obtain the cargo owner's historical behavior data within a preset time period based on the cargo's identification information. This historical behavior data includes historical transportation distances. Based on the cargo owner's historical behavior data within the preset time period, mileage segment information is generated. This mileage segment information indicates the correspondence between different historical transportation distances and different mileage segment codes. The mileage segment codes in the mileage segment information are used as one dimension in a first preset dimension to divide the fusion value of each cargo, thus obtaining the first bucket data. The preset time period can be pre-set by the developers.

[0126] For example, a computer device, based on the cargo's identification information, queries the cargo owner and then queries several offline characteristics based on that cargo owner: the cargo owner's average call duration over the past three days, the shipment completion rate over the past three days, and the transportation distance of cargo posted over the past three days. Here, the past three days refers to a preset time period. These offline characteristics can be historical behavioral data that needs to be obtained. Based on the cargo owner's historical behavioral data within the preset time period, mileage segment information is generated. Please refer to Table 9, which shows the correspondence between different historical transportation distances and different mileage segment codes involved in an exemplary embodiment of this application.

[0127] Table 9

[0128] Table 9 above can be seen as the generated mileage segment information. The computer equipment uses the mileage segment code in the mileage segment information as one of the first preset dimensions, and divides the fusion value of each cargo source to obtain the first bucket data. That is, in this scheme, in addition to the above-mentioned "city + vehicle type" dimension, the first preset dimension can also be the mileage segment code as a dimension, "city + vehicle type + mileage segment code" dimension.

[0129] Taking Suzhou as the departure city and 4.2-meter van as an example, in calculating the fusion value, the recalled drivers remain the same 20. In this embodiment, the pick-up rate index n and the transaction rate index m are calculated as scalar aggregation features using default values ​​of 1 and 1, respectively. The 20 drivers are divided into seven tiers based on the transaction rate (refer to Table 8) to obtain the corresponding distribution aggregation features, which are then combined with the mileage segment information generated from the cargo owner's data for the past three days (refer to Table 9). The final fusion value output by the fusion model is 0.7150. This score is located in the binary search of the bucket threshold string between the 42nd threshold of 0.7108 and the 43rd threshold of 0.7163, resulting in a first quantile of 42, corresponding to a grade of B.

[0130] Step 306: Obtain the quality of the goods based on the first quality level, the second quality level, and the third quality level.

[0131] Optionally, after obtaining the first, second, and third quality levels, the computer device determines the source quality based on these levels. For example, the computer device can obtain a priority rule corresponding to the source, which indicates the priority order of the first, second, and third quality levels. Based on the priority rule, the highest priority quality level among the first, second, and third quality levels is selected as the source quality.

[0132] In one possible implementation, different priority rules can be set for each source of goods based on its platform or type. In this solution, the computer equipment sets different priority rules for different types of sources of goods. For example, for a first-type source of goods, the priority rule is that the priority of the second quality level is greater than the priority of the first quality level, which is greater than the priority of the third quality level. Therefore, if the current source of goods is a first-type source, the final obtained quality level will be second quality. For a second-type source of goods, the priority rule is that the priority of the first quality level is greater than the priority of the third quality level, which is greater than the priority of the second quality level. Therefore, if the current source of goods is a second-type source, the final obtained quality level will be first quality. For a third-type source of goods, the priority rule is that the priority of the first quality level is greater than the priority of the second quality level, which is greater than the priority of the third quality level. Therefore, if the current source of goods is a third-type source, the final obtained quality level will be first quality.

[0133] Optionally, taking Suzhou as the departure city and a 4.2-meter van as an example, the third quantile of the absolute value calculated above is 6, corresponding to the first quality level D; the second quantile of the relative value is 65, corresponding to the second quality level B; and the third quantile of the combined value is 42, corresponding to the second quality level B. Since the cargo type is the third type, the final determined quality level is B.

[0134] Optionally, in addition to selecting according to the priority order set above, if the current source of goods does not belong to any type, a fallback strategy will be adopted, that is, the priority order is: the first quality level has a higher priority than the second quality level, which has a higher priority than the third quality level, and the highest priority quality level will be selected as the source quality.

[0135] In summary, the method for obtaining the quality of goods provided in this application obtains a first quality level corresponding to the absolute value of the goods, whereby the absolute value indicates the profit level based on the physical attribute information of the goods; a second quality level corresponding to the relative value of the goods, whereby the relative value indicates the profit level based on the market attribute information of the goods; and a third quality level corresponding to the integrated value of the goods, whereby the integrated value indicates the profit level based on the combined physical and market attribute information of the goods. The quality of the goods is then obtained based on these three quality levels. This method combines the first, second, and third quality levels for comprehensive evaluation to obtain the final quality of the goods, taking into account more comprehensive information about the goods. The obtained quality of the goods reflects the actual level of the goods, thus improving the accuracy of the evaluation of the goods.

[0136] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0137] Please refer to Figure 4 This illustration shows a structural block diagram of a source quality acquisition device 400 provided in an exemplary embodiment of this application. The source quality acquisition device 400 can be applied to a computer device and can execute all or part of the steps performed by the computer device in the methods provided in the above-described method embodiments. The source quality acquisition device 400 includes: The first acquisition module 401 is used to acquire the first quality level corresponding to the absolute value of the goods, wherein the absolute value is used to indicate the profit level obtained based on the physical attribute information of the goods. The second acquisition module 402 is used to acquire the second quality grade corresponding to the relative value of the goods, wherein the relative value is used to indicate the profit level obtained based on the market attribute information of the goods. The third acquisition module 403 is used to acquire the third quality level corresponding to the fusion value of the source of goods. The fusion value is used to indicate the revenue level obtained by combining the physical attribute information and the market attribute information of the source of goods. The fourth acquisition module 404 is used to acquire the quality of the goods source based on the first quality level, the second quality level and the third quality level.

[0138] In summary, the method for obtaining the quality of goods provided in this application obtains a first quality level corresponding to the absolute value of the goods, whereby the absolute value indicates the profit level based on the physical attribute information of the goods; a second quality level corresponding to the relative value of the goods, whereby the relative value indicates the profit level based on the market attribute information of the goods; and a third quality level corresponding to the integrated value of the goods, whereby the integrated value indicates the profit level based on the combined physical and market attribute information of the goods. The quality of the goods is then obtained based on these three quality levels. This method combines the first, second, and third quality levels for comprehensive evaluation to obtain the final quality of the goods, taking into account more comprehensive information about the goods. The obtained quality of the goods reflects the actual level of the goods, thus improving the accuracy of the evaluation of the goods.

[0139] Optionally, the fourth acquisition module 404 includes: a first acquisition unit and a second acquisition unit; The first acquisition unit is used to acquire the priority rule corresponding to the source of goods, and the priority rule is used to indicate the priority order of the first quality level, the second quality level and the third quality level respectively; The second acquisition unit is used to acquire the highest priority quality level from the first quality level, the second quality level and the third quality level as the quality of the goods, according to the priority rules corresponding to the goods source.

[0140] Optionally, the third acquisition module 403 includes: a third acquisition unit, a fourth acquisition unit, a first determination unit, and a fifth acquisition unit; The third acquisition unit is used to acquire the physical attribute information of the cargo source and the driver recall information, wherein the driver recall information is used to indicate the set of driver information that can be recalled by the cargo source. The fourth acquisition unit is used to acquire the fusion value of the cargo source based on the physical attribute information of the cargo source and the recalled driver information; The first determining unit is used to determine the first quantile of the fusion value of the source of goods in the first bucket data based on the fusion value of the source of goods and the first bucket data. The first bucket data is obtained by dividing the fusion value of each source of goods under a first preset dimension. The first quantile is used to indicate the proportion of other sources of goods in the first bucket data that have a lower fusion value than the source of goods. The fifth acquisition unit is used to acquire the third quality level corresponding to the first quantile according to the first quantile and the first mapping rule.

[0141] Optionally, the fourth acquisition unit is specifically used for: Based on the recalled driver information, query the pick-up rate and transaction rate of each driver for the cargo; Based on each driver's acceptance rate and completion rate for the cargo, scalar aggregation features and distribution aggregation features are obtained. The scalar aggregation features are used to indicate the relative value of the cargo, and the distribution aggregation features are used to indicate the structural relationship between the quantity and quality of drivers supplying the cargo. Based on the physical attribute information of the goods, the scalar aggregation feature and / or the distribution aggregation feature, the fusion value of the goods is obtained.

[0142] Optionally, the device further includes: The fifth acquisition module is used to acquire historical behavior data of each driver within a preset time period based on the recalled driver information before determining the first quantile of the fusion value of the cargo source in the first bucket data according to the fusion value of the cargo source and the first bucket data. The historical behavior data includes historical transportation distance. The first generation module is used to generate mileage segment information based on the historical behavior data of each driver within a preset time period. The mileage segment information is used to indicate the correspondence between different historical transportation distances and different mileage segment codes. The sixth acquisition module is used to take the mileage segment code in the mileage segment information as one of the first preset dimensions, and to divide the fusion value of each cargo source to obtain the first bucket data.

[0143] Optionally, the second acquisition module 402 includes: a sixth acquisition unit, a seventh acquisition unit, a second determination unit, and an eighth acquisition unit; The sixth acquisition unit is used to acquire the market attribute information of the cargo source. The market attribute information includes recalled driver information and market feedback information. The recalled driver information is used to indicate the set of driver information that the cargo source can recall. The market feedback information includes the deviation between the current actual proportion and the target proportion of the number of cargo sources included in each level interval of all cargo sources. The seventh acquisition unit is used to acquire the relative value of the cargo source based on the recall driver information contained in the market attribute information of the cargo source; The second determining unit is used to determine the second quantile of the relative value of the goods in the second quantile data based on the relative value of the goods and the second quantile data. The second quantile data is obtained by dividing the relative value of each goods in the second preset dimension. The second quantile is used to indicate the proportion of other goods in the second quantile data that are lower than the relative value of the goods. The eighth acquisition unit is used to acquire the second quality level corresponding to the second quantile based on the second quantile, the market feedback information, and the second mapping rule.

[0144] Optionally, the seventh acquisition unit is specifically used for: Based on the recall driver information contained in the market attribute information of the cargo source, query the pick-up rate and transaction rate of each driver for the cargo source; An aggregate score is obtained based on each driver's acceptance rate and completion rate for the cargo. The aggregate score is used to indicate the probability that the cargo will be accepted by any driver. Based on the identification information of the cargo source, an order acceptance score is obtained, which is used to indicate the difficulty of any driver accepting the order for the cargo source in one go; The relative value of the goods is obtained based on the aggregation score, the order acceptance score, and the price of the goods.

[0145] Optionally, the eighth acquisition unit is specifically used for: Based on the second quantile and the second mapping rule, obtain the quality level to be adjusted; If the target deviation exceeds a preset deviation threshold in the market feedback information, the quality level to be adjusted is adjusted to obtain the target quality level, and the adjusted target quality level is used as the second quality level. The target deviation is the deviation between the current actual proportion and the target proportion corresponding to the level interval in the market feedback information where the quality level to be adjusted is located. If the target deviation does not exceed the preset deviation threshold in the market feedback information, the quality level to be adjusted will be used as the second quality level.

[0146] Optionally, the market feedback information is obtained based on real-time distribution information of goods and target stratification ratio information; the real-time distribution information of goods is used to indicate the current actual distribution ratio of goods contained in each level interval of all goods, and the target stratification ratio information is used to indicate the expected ratio of goods contained in each level interval of all goods. The device further includes: The first update module is used to update the real-time distribution information of the goods based on the second quantile corresponding to the relative value of the goods in the second quantile data after determining the second quantile corresponding to the relative value of the goods based on the relative value of the goods and the second quantile data.

[0147] Optionally, the first acquisition module 401 includes: a ninth acquisition unit, a tenth acquisition unit, a third determination unit, and an eleventh acquisition unit; The ninth acquisition unit is used to acquire the physical attribute information of the goods. The tenth acquisition unit is used to acquire the absolute value of the goods based on the physical attribute information of the goods. The third determining unit is used to determine the third quantile of the absolute value of the goods in the third quantile data based on the absolute value of the goods and the third quantile data. The third quantile data is obtained by dividing the absolute value of each goods in the third preset dimension. The third quantile is used to indicate the proportion of other goods in the preset quantile data that are lower than the absolute value of the goods. The eleventh acquisition unit is used to acquire the first quality level corresponding to the third quantile according to the third quantile and the third mapping rule.

[0148] Optionally, the physical attribute information includes price tag information, constraint tag information, and cargo owner tag information; the tenth acquisition unit is specifically used for: Based on the physical attribute information of the goods, the basic score, the weighted score and the control score of the goods are obtained. The basic score is used to indicate the score corresponding to the price tag information of the goods. The weighted score is used to indicate the score corresponding to the constraint tag information of the goods. The control score is used to indicate the score corresponding to the owner tag information of the goods. The absolute value of the goods is obtained based on the basic score, the weighted score, and the adjustment score.

[0149] Optionally, the device further includes: The first query module is used to query whether there is a quality of goods in the cache of the computer device that corresponds to the identification information, based on the identification information of the goods, before obtaining the first quality level corresponding to the absolute value of the goods. The seventh acquisition module is used to acquire the quality of goods corresponding to the identification information if the cache of the computer device contains the quality of goods corresponding to the identification information. The first execution module is configured to, if the cache of the computer device does not contain the quality of the goods corresponding to the identification information, establish a distributed lock based on the identification information, and query the cache of the computer device again to see if the quality of the goods corresponding to the identification information exists. If the cache of the computer device does not contain the quality of the goods corresponding to the identification information, execute the step of obtaining the first quality level corresponding to the absolute value of the goods.

[0150] Please refer to Figure 5 This illustration shows a schematic diagram of a computer device provided in an exemplary embodiment of this application. The computer device 500 includes a Central Processing Unit (CPU) 501, a system memory 504 including Random Access Memory (RAM) 502 and Read-Only Memory (ROM) 503, and a system bus 505 connecting the system memory 504 and the CPU 501. The computer device 500 also includes a Basic Input / Output System (I / O System) 506 that facilitates information transfer between various devices within the computer, and a mass storage device 508 for storing an operating system 513, application programs 514, and other program modules 515.

[0151] The basic input / output system 506 includes a display 507 for displaying information and an input device 509 for user input, such as a mouse or keyboard. Both the display 507 and the input device 509 are connected to the central processing unit 501 via an input / output controller 510 connected to the system bus 505. The basic input / output system 506 may also include the input / output controller 510 for receiving and processing input from multiple other devices such as a keyboard, mouse, or electronic stylus. Similarly, the input / output controller 510 also provides output to a display screen, printer, or other types of output devices.

[0152] The mass storage device 508 is connected to the central processing unit 501 via a mass storage controller (not shown) connected to the system bus 505. The mass storage device 508 and its associated computer-readable media provide non-volatile storage for the computer device 500. That is, the mass storage device 508 may include computer-readable media (not shown) such as a hard disk or a CD-ROM (Compact Disc Read-Only Memory) drive.

[0153] The computer-readable medium may include computer storage media and communication media. Computer storage media include volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include RAM, ROM, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other solid-state storage technologies, CD-ROM, digital versatile disc (DVD) or other optical storage, magnetic tape cassettes, magnetic tape, disk storage, or other magnetic storage devices. Of course, those skilled in the art will recognize that the computer storage media are not limited to the above-mentioned types. The system memory 504 and mass storage device 508 described above can be collectively referred to as memory.

[0154] Computer device 500 can be connected to the Internet or other network devices via network interface unit 511 connected to the system bus 505.

[0155] The memory also includes one or more programs stored in the memory, and the central processing unit 501 executes the one or more programs to implement all or part of the steps.

[0156] It should be understood that, in the embodiments of this application, the processor may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0157] Optionally, embodiments of this application also provide a computer device, which includes a processor and a memory. The memory stores a computer program, which is executed by the processor to implement all or part of the steps performed by the computer device in the source quality acquisition methods of the above embodiments.

[0158] Optionally, embodiments of this application also provide a computer-readable medium storing a computer program that is executed by a processor to implement all or part of the steps performed by a computer device in the source quality acquisition methods of the various embodiments described above.

[0159] Optionally, this application also provides a computer program product that, when run on a computer, causes the computer to execute all or part of the steps performed by the computer device to implement the source quality acquisition method of the above embodiments.

[0160] Optionally, embodiments of this application also provide an application publishing platform for publishing computer program products. When the computer program product is run on a computer, it causes the computer to execute all or part of the steps performed by the computer device to implement the source quality acquisition method of the above embodiments.

[0161] It should be noted that the apparatus provided in the above embodiments is only illustrated by the division of the above functional modules when controlling computer equipment. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiments, which will not be repeated here.

[0162] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0163] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0164] The above description is merely an optional embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for obtaining the quality of goods, characterized in that, Applied to a computer device, the method includes: Obtain the first quality level corresponding to the absolute value of the goods, whereby the absolute value is used to indicate the revenue level obtained based on the physical attribute information of the goods; Obtain the second quality grade corresponding to the relative value of the goods, wherein the relative value is used to indicate the profit level obtained based on the market attribute information of the goods; Obtain the third quality level corresponding to the fusion value of the source of goods, wherein the fusion value is used to indicate the revenue level obtained by combining the physical attribute information and the market attribute information of the source of goods; The quality of the goods is obtained based on the first quality level, the second quality level, and the third quality level.

2. The method according to claim 1, characterized in that, The step of obtaining the quality of the goods based on the first quality level, the second quality level, and the third quality level includes: Obtain the priority rule corresponding to the source of goods, the priority rule is used to indicate the priority order of the first quality level, the second quality level and the third quality level respectively; According to the priority rules corresponding to the source of goods, the highest priority quality level among the first quality level, the second quality level and the third quality level is obtained as the quality of the source of goods.

3. The method according to claim 1, characterized in that, The process of obtaining the third quality level corresponding to the fusion value of the source of goods includes: Obtain the physical attribute information of the cargo source and the driver recall information, wherein the driver recall information is used to indicate the set of driver information that can be recalled from the cargo source; Based on the physical attribute information of the cargo and the information of the recalled drivers, the fusion value of the cargo is obtained; Based on the fusion value of the source of goods and the first bucket data, the first quantile corresponding to the fusion value of the source of goods in the first bucket data is determined. The first bucket data is obtained by dividing the fusion value of each source of goods under a first preset dimension. The first quantile is used to indicate the proportion of other sources of goods in the first bucket data that have a fusion value lower than that of the source of goods. Based on the first quantile and the first mapping rule, obtain the third quality level corresponding to the first quantile.

4. The method according to claim 3, characterized in that, The step of obtaining the fusion value of the cargo source based on its physical attribute information and the recalled driver information includes: Based on the recalled driver information, query the pick-up rate and transaction rate of each driver for the cargo; Based on each driver's acceptance rate and completion rate for the cargo, scalar aggregation features and distribution aggregation features are obtained. The scalar aggregation features are used to indicate the relative value of the cargo, and the distribution aggregation features are used to indicate the structural relationship between the quantity and quality of drivers supplying the cargo. Based on the physical attribute information of the goods, the scalar aggregation feature and / or the distribution aggregation feature, the fusion value of the goods is obtained.

5. The method according to claim 3, characterized in that, Before determining the first quantile of the fusion value of the cargo source in the first bucket data based on the fusion value of the cargo source and the first bucket data, the method further includes: Based on the identification information of the cargo source, obtain the historical behavior data of the cargo owner within a preset time period, including historical transportation distance; Based on the historical behavior data of the cargo owner within a preset time period, mileage segment information is generated. The mileage segment information is used to indicate the correspondence between different historical transportation distances and different mileage segment codes. The mileage segment code in the mileage segment information is used as one of the first preset dimensions, and the fusion value of each cargo source is divided to obtain the first bucket data.

6. The method according to claim 1, characterized in that, The step of obtaining the second quality level corresponding to the relative value of the goods includes: Obtain market attribute information of the cargo source, including recall driver information and market feedback information. The recall driver information is used to indicate the set of driver information that the cargo source can recall. The market feedback information includes the deviation between the current actual proportion and the target proportion of the number of cargo sources included in each level interval of all cargo sources. The relative value of the cargo is obtained based on the recall driver information contained in the market attribute information of the cargo. Based on the relative value of the goods and the second bucket data, the second quantile corresponding to the relative value of the goods in the second bucket data is determined. The second bucket data is obtained by dividing the relative value of each goods under a second preset dimension. The second quantile is used to indicate the proportion of other goods in the second bucket data that are lower than the relative value of the goods. Based on the second quantile, the market feedback information, and the second mapping rule, obtain the second quality level corresponding to the second quantile.

7. The method according to claim 6, characterized in that, The step of obtaining the relative value of the cargo based on the recalled driver information contained in the market attribute information of the cargo includes: Based on the recall driver information contained in the market attribute information of the cargo source, query the pick-up rate and transaction rate of each driver for the cargo source; An aggregate score is obtained based on each driver's acceptance rate and completion rate for the cargo. The aggregate score is used to indicate the probability that the cargo will be accepted by any driver. Based on the identification information of the cargo source, an order acceptance score is obtained, which is used to indicate the difficulty of any driver accepting the order for the cargo source in one go; The relative value of the goods is obtained based on the aggregation score, the order acceptance score, and the price of the goods.

8. The method according to claim 6, characterized in that, The step of obtaining the second quality level corresponding to the second quantile based on the second quantile, the market feedback information, and the second mapping rule includes: Based on the second quantile and the second mapping rule, obtain the quality level to be adjusted; If the target deviation exceeds a preset deviation threshold in the market feedback information, the quality level to be adjusted is adjusted to obtain the target quality level, and the adjusted target quality level is used as the second quality level. The target deviation is the deviation between the current actual proportion and the target proportion corresponding to the level interval in the market feedback information where the quality level to be adjusted is located. If the target deviation does not exceed the preset deviation threshold in the market feedback information, the quality level to be adjusted will be used as the second quality level.

9. The method according to claim 6, characterized in that, The market feedback information is obtained based on real-time distribution information of goods and target stratification ratio information; the real-time distribution information of goods is used to indicate the current actual distribution ratio of goods contained in each level interval of all goods, and the target stratification ratio information is used to indicate the expected ratio of goods contained in each level interval of all goods. After determining the second quantile of the relative value of the goods in the second quantile data based on the relative value of the goods and the second quantile data, the method further includes: The real-time distribution information of the goods is updated based on the second quantile corresponding to the relative value of the goods.

10. The method according to claim 1, characterized in that, The process of obtaining the first quality grade corresponding to the absolute value of the goods includes: Obtain the physical attribute information of the goods; Based on the physical attribute information of the goods, obtain the absolute value of the goods; Based on the absolute value of the goods and the third bucket data, the third quantile of the absolute value of the goods in the third bucket data is determined. The third bucket data is obtained by dividing the absolute value of each goods under a third preset dimension. The third quantile is used to indicate the proportion of other goods in the preset bucket data that are lower than the absolute value of the goods. Based on the third quantile and the third mapping rule, obtain the first quality level corresponding to the third quantile.

11. The method according to claim 10, characterized in that, The physical attribute information includes price tag information, constraint tag information, and cargo owner tag information; obtaining the absolute value of the cargo source based on the physical attribute information includes: Based on the physical attribute information of the goods, the basic score, the weighted score and the control score of the goods are obtained. The basic score is used to indicate the score corresponding to the price tag information of the goods. The weighted score is used to indicate the score corresponding to the constraint tag information of the goods. The control score is used to indicate the score corresponding to the owner tag information of the goods. The absolute value of the goods is obtained based on the basic score, the weighted score, and the adjustment score.

12. The method according to any one of claims 1 to 11, characterized in that, Before obtaining the first quality grade corresponding to the absolute value of the goods, the method further includes: Based on the identification information of the goods, query whether there is a goods quality corresponding to the identification information in the cache of the computer device; If the cache of the computer device contains the quality of goods corresponding to the identification information, obtain the quality of goods corresponding to the identification information; If the cache of the computer device does not contain the quality of the goods corresponding to the identification information, a distributed lock is established based on the identification information, and the cache of the computer device is queried again to see if the quality of the goods corresponding to the identification information exists. If the cache of the computer device does not contain the quality of the goods corresponding to the identification information, the step of obtaining the first quality level corresponding to the absolute value of the goods is executed.

13. A device for obtaining the quality of goods, characterized in that, Applied to computer equipment, the device includes: The first acquisition module is used to acquire the first quality level corresponding to the absolute value of the goods, wherein the absolute value is used to indicate the profit level obtained based on the physical attribute information of the goods. The second acquisition module is used to acquire the second quality grade corresponding to the relative value of the goods, wherein the relative value is used to indicate the profit level obtained based on the market attribute information of the goods. The third acquisition module is used to acquire the third quality level corresponding to the fusion value of the goods source. The fusion value is used to indicate the revenue level obtained by combining the physical attribute information and the market attribute information of the goods source. The fourth acquisition module is used to acquire the quality of the goods source based on the first quality level, the second quality level, and the third quality level.

14. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing a computer program that is executed by the computer device to implement the source quality acquisition method as described in any one of claims 1 to 12.

15. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which is executed by a processor to implement the source quality acquisition method as described in any one of claims 1 to 12.

Images