Resource supply decision-making method and device, storage medium and program product

Abstract

One or more embodiments of the invention provide a supply resource decision method and device, a storage medium and a program product. The method comprises the following steps: determining target supply resources which do not participate in a preset activity; acquiring historical flow data related to the preset activity, wherein the historical flow data comprises a plurality of historical browsing requests generated in a service area to which the target supply resource belongs, supply resource display queues corresponding to the historical browsing requests and supply resource exposure quantity; playing back the historical browsing request to determine a simulation display queue for the historical browsing request based on the target supply resource and a supply resource display queue corresponding to the historical browsing request; based on the display sequence of the target supply resource in the simulation display queue and the supply resource exposure number corresponding to the historical browsing request, whether the target supply resource can be exposed is evaluated, and an exposure evaluation result is obtained; and based on the exposure evaluation results corresponding to the plurality of historical browsing requests, making a decision on the target supply resource.

Description

Technical Field

[0001] One or more embodiments of this specification relate to the field of computer software technology, and in particular, to a supply resource decision-making method, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] In recent years, with the rapid development of the Internet economy and the continuous integration of online and offline related services, various marketing activities have become an important means to promote merchant brand exposure and sales growth. In the decision-making of various marketing activities, the traditional decision-making method usually uses the change in the order volume before and after the activity as the basis for judging the activity effect. However, this method belongs to post-event statistics, and such post-event statistical means cannot provide timely and scientific guidance for the planning and resource allocation before the activity. Summary of the Invention

[0003] In view of this, one or more embodiments of this specification provide a supply resource decision-making method, an electronic device, a computer-readable storage medium, and a computer program product.

[0004] To achieve the above object, one or more embodiments of this specification provide the following technical solutions: According to the first aspect of one or more embodiments of this specification, a supply resource decision-making method is proposed, including: Determine the target supply resources that have not participated in the preset activity; Obtain historical traffic data related to the preset activity, where the historical traffic data includes a plurality of historical browsing requests generated in the service area to which the target supply resources belong, the supply resource display queue corresponding to the historical browsing requests, and the supply resource exposure quantity; Replay the historical browsing requests to determine a simulation display queue for the historical browsing requests based on the target supply resources and the supply resource display queue corresponding to the historical browsing requests; Based on the display order of the target supply resources in the simulation display queue and the supply resource exposure quantity corresponding to the historical browsing requests, evaluate whether the target supply resources can be exposed to obtain an exposure evaluation result; Based on the exposure evaluation results respectively corresponding to the plurality of historical browsing requests, make a decision on the target supply resources. According to the second aspect of the embodiments of this specification, a supply resource decision-making method is provided, which is applied to an operation client and includes: Determine a plurality of target supply resources that are in the service area to be decided and have not participated in the preset activity; Predict the value indicators of each of the target supply resources when participating in the preset activity, where the value indicators include at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the method described in the first aspect; Based on the value indicators respectively corresponding to the multiple target supply resources, determine the target supply resources to be invited to participate in the preset activity from the multiple target supply resources.

[0005] According to the third aspect of the embodiments of this specification, a supply resource decision-making method is provided, which is applied to a merchant client and includes: Determine multiple preset activities that the target supply resource has not participated in; Predict the value indicators of the target supply resource when participating in each preset activity, where the value indicators include at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the method described in the first aspect; Based on the value indicators of the target supply resource when participating in each preset activity, determine the preset activity that the target supply resource is to participate in from the multiple preset activities.

[0006] According to the fourth aspect of the embodiments of this specification, a supply resource decision-making method is provided, which is applied to a merchant client and includes: Determine multiple target supply resources to participate in the preset activity, where the multiple target supply resources include supply resources with different rights and interests provided by the same resource provider; Predict the value indicators of each of the target supply resources when participating in the preset activity, where the value indicators include at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the method described in the first aspect; Based on the value indicators respectively corresponding to the multiple target supply resources, determine the target supply resources to participate in the preset activity from the multiple target supply resources.

[0007] According to the fifth aspect of the embodiments of this specification, an electronic device is provided, including: A processor; A memory for storing instructions executable by the processor; Wherein, when the processor executes the executable instructions, it is used to implement the method described in the first aspect, the second aspect, the third aspect or the fourth aspect.

[0008] According to a sixth aspect of the embodiments of the present specification, a computer-readable storage medium is provided, on which a computer program is stored. When the program is executed by a processor, the steps of the method described in the first aspect, the second aspect, the third aspect or the fourth aspect are implemented.

[0009] According to a seventh aspect of the embodiments of the present specification, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps of the method described in the first aspect, the second aspect, the third aspect or the fourth aspect are implemented.

[0010] The technical solutions provided by the embodiments of the present specification may include the following beneficial effects: In the embodiments of the present specification, by replaying the historical browsing requests generated in the service area to which the target supply resource belongs, a simulation display queue including the target supply resource can be determined. The simulation display queue simulates the possible display situations of the target supply resource under similar historical conditions, which helps to intuitively evaluate the performance of the target supply resource in the real environment. Furthermore, based on the display order of the target supply resource in the simulation display queue and the exposure quantity of the supply resource corresponding to the historical browsing request, it can be evaluated whether the target supply resource can be exposed, and the exposure potential of the target supply resource can be accurately evaluated before the event, reflecting the adaptation degree between the target supply resource and the preset event, so as to make a precise decision on the target supply resource and provide a scientific basis for the preliminary planning and resource allocation of the marketing event. Different from the traditional ex-post statistical method, based on the replay of the historical browsing requests and the exposure evaluation results, this method can predict the exposure effect of the target resource in the preset event in real time and dynamically, helping the decision maker to optimize the resource selection and allocation before the event, avoid unnecessary resource waste, and improve the overall effect of the event and the utilization efficiency of the resources.

[0011] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a schematic flowchart of a first supply resource decision method provided by an exemplary embodiment.

[0013] Figure 2 is a schematic diagram of an activity page provided by an exemplary embodiment.

[0014] Figure 3 is a schematic diagram of the supply resource display queue corresponding to 3 historical browsing requests and their actual exposure situations provided by an exemplary embodiment.

[0015] Figure 4 is a schematic diagram of the simulation display queue corresponding to 3 historical browsing requests provided by an exemplary embodiment.

[0016] Figure 5 It is a schematic diagram for determining a preset time period provided by an exemplary embodiment.

[0017] Figure 6 It is a schematic flowchart for making an activity decision for a target merchant provided by an exemplary embodiment.

[0018] Figure 7 It is a schematic flowchart of the second supply resource decision method provided by an exemplary embodiment.

[0019] Figure 8 It is a schematic flowchart of the third supply resource decision method provided by an exemplary embodiment.

[0020] Figure 9 It is a schematic flowchart of the fourth supply resource decision method provided by an exemplary embodiment.

[0021] Figure 10 It is a schematic diagram of the structure of an electronic device provided by an exemplary embodiment. Detailed implementation manners

[0022] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with one or more embodiments of this specification. On the contrary, they are only examples of devices and methods consistent with some aspects of one or more embodiments of this specification as detailed in the appended claims.

[0023] It should be noted that: in other embodiments, the steps of the corresponding methods are not necessarily executed in the order shown and described in this specification. In some other embodiments, the steps included in the method may be more or less than those described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may also be combined into a single step for description in other embodiments.

[0024] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this specification are all information and data authorized by the user or fully authorized by all parties. And the collection, use, and processing of relevant data need to comply with relevant laws, regulations, and standards of relevant countries and regions, and corresponding operation entrances are provided for users to choose to authorize or refuse.

[0025] Based on the problems in the related art, the embodiments of this specification provide a supply resource decision-making method, an electronic device, a computer-readable storage medium, and a computer program product, which can comprehensively analyze the target supply resources to be decided before an event and predict the marketing performance in advance. This not only helps merchants more reasonably select the events to participate in, but also provides theoretical support for the supply service platform to identify and preferentially attract high-quality supply resources during the investment promotion process, and has important practical application value.

[0026] Please refer to Figure 1 , which shows a schematic flowchart of a supply resource decision-making method. This method can be executed by an electronic device, which includes but is not limited to physical servers, server clusters, cloud servers, smart phones / mobile phones, tablet computers, personal digital assistants (PDAs), laptop computers, and desktop computers, etc. The supply resource decision-making method includes: In S101, determine the target supply resources that have not participated in the preset event.

[0027] Exemplarily, the supply resources mentioned in the embodiments of this specification can be merchants who have entered the supply service platform. As the basic supply units served by the supply service platform, each entered merchant (such as a catering store, a retail store) itself constitutes an independent supply resource, and its service capabilities (such as fulfillment scope, dish inventory, order acceptance efficiency, etc.) directly affect the supply quality of the platform ecosystem.

[0028] Exemplarily, the supply resources mentioned in the embodiments of this specification can be the goods or services provided by each merchant. That is to say, the goods or services provided by a single merchant (such as the signature set meal of a certain restaurant, the special offer products of a certain supermarket) can also be managed as independent supply resources.

[0029] In this step, the user can select the target supply resources to be decided according to actual needs, and the electronic device determines the target supply resources that have not participated in the preset event based on the user's selection instruction. Alternatively, the electronic device can automatically screen out the target supply resources that have not participated in the preset event from the supply resource pool. This realizes focusing on those potential high-quality supply resources that have not been developed and lays a foundation for subsequent accurate decision-making.

[0030] In S102, obtain historical traffic data related to the preset event. The historical traffic data includes multiple historical browsing requests generated in the service area to which the target supply resources belong, the supply resource display queue corresponding to the historical browsing requests, and the supply resource exposure quantity.

[0031] In this step, the electronic device can collect relevant historical traffic data, which reflects the user behavior and supply resource display situation in the area where the target resources are located in the real scenario.

[0032] Among them, only using historical browsing requests that match the service scope of the target supply resource can exclude noise data brought by irrelevant regions, making the subsequent evaluation closer to the real market environment. Within the same region, factors such as user consumption preferences, delivery radius, and logistics timeliness are relatively consistent. Focusing on historical browsing requests within the service area can more accurately simulate the exposure and conversion of the target supply resource during the event, reduce misjudgment caused by regional differences, and without the need to process browsing request data unrelated to the target supply resource, it also reduces the pressure on data processing and storage, improving the overall operating efficiency of the electronic device.

[0033] The supply resource display queue corresponding to the historical browsing request refers to the list of supply resources actually displayed by the supply service platform when the user accesses the event page. The supply resource display queue contains multiple supply resources and is arranged according to a certain sorting rule.

[0034] The supply resource exposure quantity refers to the number of supply resources actually browsed by the user in the supply resource display queue corresponding to a historical browsing request. In different historical browsing requests, the user may only browse the first few supply resources or scroll down to view more supply resources. The supply resource exposure quantity can reflect the actual browsing depth of the user during this browsing, thereby helping the electronic device evaluate the exposure of a certain target supply resource in different display orders.

[0035] Exemplarily, the supply service platform usually plans and launches multiple different types of marketing activities based on market demand and user preferences, such as full reduction offers, discount promotions, etc. These activities will be integrated into the event pages provided by the supply service platform. For example, different marketing activities provide separate event pages, and users can access the event pages of various types of marketing activities through various terminal devices (such as smartphones, tablets, PCs, etc.) to view and select suitable goods or services. For example, please refer to Figure 2 , which shows the event page corresponding to Activity 1. This event page displays merchants arranged in order, and users can select their favorite merchants and enter the merchant page to view and select suitable goods.

[0036] When a user accesses the activity page of a certain activity, the supply service platform will intelligently screen and display supply resources that meet the user's needs based on the user's basic attributes (such as geographical location, historical purchase records, preference tags, etc.). For example, in the promotion activity of a food delivery platform, if a user is located in a specific business district of a certain city, the food delivery platform will give priority to displaying popular or high-rated dining restaurants and their preferential packages in that area. Another example is that for a retail platform, users may see special offers or limited-time discounts of nearby stores. When a user browses the activity page, their behavioral data (such as clicking on a certain supply resource, dwell time, scrolling, etc.) will be recorded by the platform and become an important basis for subsequent traffic analysis.

[0037] That is to say, the above-mentioned historical browsing requests mainly come from the requests generated when users access the activity page of a preset activity through various terminal devices (such as smartphones, tablets, PCs, etc.). Specifically, when a user is within the service area covered by the target supply resource (for example, a certain specific city or region) and clicks on or browses the activity page related to the preset activity, their behavior will be recorded as a historical browsing request. These historical browsing requests carry geographical location data to verify that the request source is indeed within the service range of the target supply resource. By collecting such targeted browsing data, it can ensure that the subsequent simulation display and exposure evaluation based on historical traffic data are more in line with the actual situation, thus providing an accurate basis for the decision-making process.

[0038] In S103, replay the historical browsing requests to determine a simulation display queue for the historical browsing requests based on the target supply resource and the supply resource display queue corresponding to the historical browsing requests.

[0039] In this step, the electronic device replays each historical browsing request to generate a simulation display queue based on the target supply resource and the supply resource display queue corresponding to the historical browsing requests. This simulation display queue simulates the possible display scenarios of the target supply resource under similar historical conditions. By simulating the actual display effect, it helps to intuitively evaluate the performance of the target supply resource in the real environment, reduce the risk of direct online testing, and provide a basis for subsequent exposure evaluation.

[0040] In S104, based on the display order of the target supply resource in the simulation display queue and the supply resource exposure quantity corresponding to the historical browsing requests, evaluate whether the target supply resource can be exposed to obtain an exposure evaluation result.

[0041] In this step, the electronic device determines whether the target supply resource can be actually viewed by the user based on the exposure quantity of the supply resources in the historical browsing requests and the display order of the target supply resource in the simulation display queue. The display order of the target supply resource in the simulation display queue directly affects its exposure possibility. The exposure quantity of the supply resources corresponding to the historical browsing requests indicates the number of supply resources that the user actually views in this request. Suppose the exposure quantity of the supply resources in a certain historical browsing request is M, then the supply resources with a display order exceeding M will not be seen by the user, that is, they will not be exposed.

[0042] If the display order of the target supply resource is within the exposure quantity of the supply resources, then the target supply resource will be viewed by the user, that is, the target supply resource can be exposed. If the display order of the target supply resource exceeds the exposure quantity of the supply resources, then the target supply resource will not be viewed by the user, that is, it cannot be exposed. Through the above exposure determination logic, the electronic device can evaluate multiple historical browsing requests respectively and generate exposure evaluation results corresponding to multiple historical browsing requests respectively. Based on the simulation display position of the target supply resource and the exposure quantity of the supply resources in the historical browsing requests, the exposure possibility of the target supply resource in the preset activity is quantified, making the screening of supply resources more accurate, improving the rationality of activity resource allocation, and thus optimizing the overall activity effect.

[0043] In S105, based on the exposure evaluation results corresponding to multiple historical browsing requests respectively, a decision is made on the target supply resource. In this step, the electronic device can comprehensively evaluate the exposure potential of the target supply resource based on the exposure evaluation results of multiple historical browsing requests, and thus make a decision. Compared with the traditional ex-post statistics of marketing effects, this method is more forward-looking, can optimize resource allocation, and provide timely and scientific guidance for the planning and resource allocation before the activity.

[0044] Exemplarily, taking the electronic device as the device belonging to the merchant as an example, the electronic device can, based on the exposure evaluation results corresponding to multiple historical browsing requests respectively, decide whether the target supply resource participates in the preset activity, so as to improve the resource utilization efficiency and enhance the activity response ability.

[0045] Exemplarily, taking the electronic device as the device belonging to the operator as an example, the electronic device can, based on the exposure evaluation results corresponding to multiple historical browsing requests respectively, decide whether to invite the target supply resource to participate in the preset activity, so as to improve the pertinence and accuracy of activity organization.

[0046] In some embodiments, when there are multiple target supply resources to be decided, the multiple target supply resources include at least one of the following: supply resources provided by different resource providers, and supply resources with different rights and interests provided by the same resource provider; wherein, the different rights and interests indicate different preferential policies provided by the resource provider in a preset activity.

[0047] If the multiple target supply resources are from the supply resources provided by different resource providers, for example, products and services provided by different restaurants, retailers or brands respectively. The electronic device can, based on the above method, respectively evaluate the exposure potential of the target supply resources provided by each resource provider in the preset activity, so as to select high-quality supply resources that are more likely to obtain higher exposure in the activity, which can not only ensure the diversity of activity resources, but also optimize the overall resource allocation.

[0048] When the same resource provider participates in a preset activity, it may launch multiple resources according to different marketing strategies or preferential policies. For example, a catering merchant may simultaneously launch multiple preferential strategies, such as offering full reduction discounts, enjoying discounts or giving out coupons, etc. The "different rights and interests" here refers to the different preferential policies provided by the resource provider in the activity. In this case, even if the supply resources come from the same merchant, due to the different rights and interests corresponding to the target supply resources, their abilities to attract users and market competitiveness will also vary. Therefore, it is necessary to separately evaluate the exposure of these target supply resources with different rights and interests to decide whether and how they participate in the activity, so as to achieve more accurate marketing matching and resource allocation.

[0049] In some embodiments, the historical traffic data further includes the evaluation scores of each supply resource in the supply resource display queue corresponding to each historical browsing request; then, during the playback of each historical browsing request by the electronic device, the target supply resource is added to the supply resource display queue corresponding to the historical browsing request to form a simulated recall queue; and the evaluation score of the target supply resource can be determined based on the service information of the target supply resource, and then the supply resources in the simulated recall queue are sorted according to the evaluation score of the target supply resource and the evaluation scores of each supply resource in the supply resource display queue to obtain a simulated display queue, and this simulated display queue simulates the relative position of the target supply resource in the display queue when the user browses in the preset activity.

[0050] In this embodiment, the evaluation scores of each supply resource in the supply resource display queue corresponding to each historical browsing request are pre-recorded. When making a decision on the target supply resource, the existing data can be directly called to quickly sort the display queue, thereby improving the decision-making efficiency. By adding the target supply resource to the historical supply resource display queue based on the evaluation score of the target supply resource, the generated simulation display queue can more realistically reflect the resource sorting that users see on the activity page. Supply resources with high evaluation scores usually mean better service capabilities or more attractive preferential efforts, and their positions in the display queue will be more forward, thus helping to simulate the actual exposure situation.

[0051] In a possible implementation manner, the evaluation scores of each supply resource in the supply resource display queue can be generated by a pre-trained evaluation model based on the service information of each supply resource, and the evaluation score of the target supply resource can also be generated by the evaluation model based on the service information of the target supply resource. In this embodiment, the evaluation scores of each supply resource and the target supply resource in the supply resource display queue are all generated by the same pre-trained evaluation model based on their respective service information. Using the same model to generate evaluation scores ensures that the scoring basis and calculation logic of all supply resources are completely consistent, thus making the comparison between different resources more fair and objective.

[0052] Among them, the evaluation model can be obtained through supervised training based on a number of supply resource samples marked with evaluation scores. During the training process, the evaluation model makes predictions according to the service information of the input supply resource samples, calculates the predicted scores, and compares the predicted scores with the actually marked evaluation scores. With the goal of minimizing the error between the predicted score and the actually marked evaluation score, by comparing the error between the predicted score and the actually marked evaluation score, the internal parameters of the model (such as weights and biases) are continuously adjusted to reduce the prediction error. This process is repeated until the preset conditions are met (such as achieving the optimization goal or reaching the maximum number of iterations), thereby completing the supervised learning.

[0053] In another possible implementation, the evaluation scores of the supply resources in the supply resource display queue are generated by a first evaluation model deployed in the online environment based on the service information of each supply resource, and the evaluation score of the target supply resource is generated by a second evaluation model deployed in the offline environment based on the service information of the target supply resource; wherein, the second evaluation model is a copy of the first evaluation model. In this embodiment, placing the evaluation calculation of the target supply resource in the offline environment avoids affecting the online real-time service process, thereby reducing the system load risk and ensuring the stable operation of the online service. Although the evaluation is performed separately online and offline, the second evaluation model, as a copy of the first model, can ensure the consistency of the evaluation results; at the same time, through environment isolation, it effectively avoids the impact of anomalies or high computational loads during the evaluation process on the online real-time service process. The offline environment supports the deployment of multiple copies of the model, facilitating batch processing and offline evaluation in large-scale data scenarios, providing sufficient data support for resource screening and subsequent policy adjustment, and being able to flexibly expand the computing power according to the scenario requirements.

[0054] Among them, the first evaluation model is obtained through supervised training based on a number of supply resource samples annotated with evaluation scores. During the training process, the first evaluation model makes predictions based on the service information of the input supply resource samples, calculates the predicted scores, and compares the predicted scores with the actually annotated evaluation scores. With the goal of minimizing the error between the predicted score and the actually annotated evaluation score, by comparing the error between the predicted score and the actually annotated evaluation score, the internal parameters of the model (such as weights and biases) are continuously adjusted to reduce the prediction error. This process is repeated until the preset conditions are met (such as achieving the optimization goal or reaching the maximum number of iterations), thereby completing the supervised learning.

[0055] Exemplarily, the service information of each supply resource includes but is not limited to: resource provider characteristics, supply resource characteristics, and rights and interests characteristics, where the rights and interests characteristics are used to indicate the preferential information provided by the resource provider in a preset activity.

[0056] The resource provider characteristics refer to the relevant information of the entity (such as merchant, supplier, service provider, etc.) that provides the supply resources, reflecting the basic situation, reputation, scale, etc. of the resource provider. The resource provider characteristics include but are not limited to: (1) Resource provider reputation, such as the evaluation score of the resource provider, historical order situation, service quality, etc. (2) Resource provider resources, such as the inventory and supply chain management capabilities of the resource provider. (3) Resource provider delivery ability, that is, the ability of the resource provider to process orders and complete shipments within a certain period of time. The resource provider characteristics help the model evaluate the stability and reputation of the provider, and affect the overall quality and reliability of the resources. For example, resources provided by a provider with rich historical experience and good reputation are usually more reliable and may obtain higher evaluation scores.

[0057] Supply resource characteristics refer to the relevant information of the resource itself, including its type, quality, quantity, etc. These characteristics are used to describe the specific attributes of the supply resource. Supply resource characteristics include but are not limited to: (1) Category, such as fresh food, electronic products, household goods, etc. (2) Price, the pricing range of the supply resource. (3) Size and weight, the volume and weight of the supply resource, which affect the transportation mode and distribution cost. (4) Packaging type, the packaging of the supply resource (such as bulk commodities, fragile goods, food, etc.). (5) Inventory situation, whether the supply resource has sufficient inventory and whether it is prone to out-of-stock. (6) Brand, the brand of the supply resource, which affects consumer choice. (7) Shelf life, for fresh commodities, etc., the expiration date of the commodity needs to be considered. Supply resource characteristics help the model understand the specific advantages and disadvantages of the supply resource. For example, a supply resource with high-quality resources usually gets a higher evaluation score, while a supply resource with a large quantity or reasonable price may increase its attractiveness, thus affecting the evaluation score.

[0058] Rights and interests characteristics are used to indicate the preferential treatment or added value provided by the resource provider in a preset activity, usually reflecting the additional conditions provided by the provider to attract users. Rights and interests characteristics include but are not limited to: (1) Preferential policies, including activity preferential information such as full reduction, discount, gift, coupon, etc.; (2) Additional services, such as additional after-sales service, technical support, etc.; (3) Loyalty rewards, such as membership points, etc. Rights and interests characteristics mainly reflect the additional attractiveness or competitiveness of the provider in the resource transaction. A provider that offers preferential treatment or additional services may be evaluated as more attractive, thus improving its performance in the evaluation score. For example, the offer of a larger discount or additional services may increase the evaluation score.

[0059] Exemplarily, the above-mentioned evaluation score can be at least one of the click-through rate and the conversion rate. The click-through rate (CTR) is a direct indicator reflecting the attractiveness of the supply resource to users, indicating the proportion of users who click on the supply resource during the display process. A higher click-through rate usually means that the title, picture or preferential information of the resource can arouse more user interest, and a higher score will be given in the evaluation, which helps the resource obtain a better ranking in the display queue. The conversion rate (CR) represents the proportion of users who actually complete the purchase or other target behaviors after clicking on the supply resource. The conversion rate not only reflects the attractiveness of the resource but also reflects the actual marketing effect of the supply resource. In the evaluation, a resource with a higher conversion rate is often considered to have better commercial conversion potential, thus obtaining a higher score. Using the click-through rate, the conversion rate or a combination of both as the evaluation score helps to more comprehensively measure the performance of the supply resource in the preset activity, thus providing data support for subsequent decisions.

[0060] In some embodiments, for each target supply resource, after obtaining the exposure evaluation results corresponding to the target supply resource in multiple historical browsing requests, the electronic device may count the exposure volume of the target supply resource; then, based on the exposure volume of the target supply resource, make a decision on the target supply resource.

[0061] For example, please refer to Figure 3 and Figure 4 , Figure 3 which shows the actual exposure situation of the supply resource display queue corresponding to 3 historical browsing requests. In request 1, the first 4 supply resources were exposed; in request 2, the first 8 supply resources were exposed; in queue 3, the first 2 supply resources were exposed. Figure 4 This is the simulation playback process of these 3 historical browsing requests. After inserting the target supply resource, the exposure situations of the target supply resource in the simulation display queue are as follows: in request 1, the target supply resource was exposed at position 3; in request 2, the target supply resource was exposed at position 7; in request 3, since the target supply resource ranked 6th and did not get exposed, so the target supply resource obtained 2 exposure volumes.

[0062] In a possible implementation, the electronic device may calculate the exposure probability of the target supply resource based on the exposure volume of the target supply resource. Exposure probability = exposure volume of the target supply resource / total number of historical browsing requests. The higher the exposure probability, the more likely the target supply resource is to be seen by users in the actual scenario and the greater the value of participating in the activity. A minimum exposure probability threshold can be set. Assume that the exposure probability of the target supply resource is greater than the minimum exposure probability threshold, then a decision-making suggestion that the target supply resource is suitable for participating in the preset activity can be generated.

[0063] In another possible implementation, an electronic device can obtain an estimated order volume for a target supply resource based on its exposure and evaluation score, where the evaluation score is determined based on the target supply resource's service information. A decision regarding the target supply resource can then be made based on the estimated order volume. For example, if the estimated order volume is high, it indicates that the target supply resource has high sales potential in the event. The electronic device can recommend the resource provider to participate in the event or invite the resource provider to participate. If the estimated order volume is low, it indicates that the target supply resource may be less competitive. The electronic device can recommend the resource provider to select a more suitable event or optimize the target supply resource (such as adjusting its equity strategy or optimizing product information) to improve its competitiveness. In this embodiment, exposure determines how many users the target supply resource can reach, and the evaluation score determines whether the target supply resource can attract users and generate orders. Based on the exposure and evaluation score of the target supply resource, the accuracy of the estimated order volume for the target supply resource can be guaranteed, enabling a data-driven approach to predicting the estimated order volume for the target supply resource, avoiding reliance on experience or subjective judgment, and improving decision-making accuracy.

[0064] Higher exposure means more opportunities for users to see the target supply resource, resulting in a greater probability of clicks and conversions, ultimately leading to more orders. Lower exposure means that even if the target supply resource has a high click-through rate and conversion rate, it may still result in lower orders due to fewer users seeing it. Therefore, the estimated order volume of a target supply resource is positively correlated with its exposure.

[0065] A supply resource with a higher evaluation score—that is, one with a better click-through rate (CTR) or conversion rate (CR)—indicates that users are more likely to click on and ultimately purchase the resource, leading to more orders. If a supply resource has a lower evaluation score, even if it receives high exposure, users are less likely to click on it. Or, even if they do click, they may not ultimately purchase, affecting the actual order volume. Therefore, the estimated order volume of a target supply resource is positively correlated with its evaluation score.

[0066] Exemplarily, when determining the estimated order volume of the target supply resource, relying solely on the exposure volume and the evaluation score may have certain deviations. Because the evaluation score is a theoretical value calculated based on the evaluation model, while the actual order volume may be affected by factors such as user behavior, market environment, and competition relationship. Therefore, in order to improve the accuracy of the estimation, a preset calibration coefficient is introduced to compensate for the error of the model. Then, the estimated order volume of the target supply resource is the product of the exposure volume of the target supply resource, the evaluation score of the target supply resource, and the preset calibration coefficient, that is, estimated order volume = exposure volume × evaluation score × preset calibration coefficient. In this embodiment, considering that the actual performance of different supply resources may deviate from the model prediction, the introduction of the calibration coefficient can compensate for the overall error, making the estimated order volume closer to the actual order volume. The calculation of the estimated order volume of the target supply resource not only depends on the exposure volume and the evaluation score, but also combines the calibration coefficient determined based on historical data, which can more accurately reflect the estimated sales volume of the supply resource in a specific preset activity, thereby improving the decision-making accuracy.

[0067] Among them, the preset calibration coefficient is determined based on the relationship between the actual order volume, the actual exposure volume, and the evaluation score of multiple reference supply resources that have participated in the preset activity. The core goal of the preset calibration coefficient is to calculate a correction factor based on historical data to make the estimated order volume closer to the actual order volume. In the process of calculating the preset calibration coefficient by the electronic device, first, the actual order volume, the actual exposure volume, and the evaluation score of multiple reference supply resources that have participated in the preset activity are obtained from the historical data. Then, the product of the actual exposure volume and the evaluation score of the reference resource can be calculated. Furthermore, the preset calibration coefficient is determined based on the ratio between the actual order volume of the reference supply resource and this product.

[0068] Exemplarily, since the browsing requests and order volumes are affected by time periods, seasonal factors, or unexpected events, there may be significant fluctuations in single - time data. Therefore, it is crucial to select an appropriate time window for supply resource decision - making. In the embodiments of this specification, the historical traffic data includes multiple historical browsing requests, which include: historical browsing requests generated in the service area of the target supply resource within a preset time period. After the electronic device obtains the estimated order volume of the target supply resource, it can determine the order volume of the target supply resource per unit time based on the estimated order volume of the target supply resource and the preset time period, and then generate a decision - making suggestion on whether the target supply resource participates in the preset activity based on the order volume of the target supply resource per unit time. In this embodiment, by collecting historical traffic data within a preset time period and calculating the order volume of the target supply resource per unit time, short - term outliers can be smoothed, and the performance of the target supply resource under normal service cycles can be more realistically reflected; the order volume per unit time, as a normalized indicator, facilitates horizontal comparison between different target supply resources, can more objectively evaluate the performance of each target supply resource under the same time conditions, and generate more reliable investment promotion and resource screening decision - making suggestions.

[0069] Exemplarily, please refer to Figure 5 , by statistically analyzing historical traffic and order data, the fluctuation period of the data can be identified, and thus a preset time period that can smooth short - term fluctuations and truly reflect the performance of supply resources can be determined. For example, the preset time period can be one week, two weeks, or one month, etc., and the unit time can be one day. For example, takeaway and retail services may have obvious daily peaks and troughs. Selecting a preset time period covering a complete working week or including weekends helps to reflect the real situation.

[0070] In an exemplary embodiment, please refer to Figure 6 , taking the target supply resource as a merchant as an example for exemplary illustration, the user can input specific merchant IDs and right IDs according to actual needs to form an input multi - tuple <merchant ID, right ID>. Generally, they are new merchant IDs that have not participated in the activity, and the target merchant to be decided can be determined based on the merchant ID.

[0071] In a simulation environment, the electronic device can obtain historical traffic data related to the preset activity. This historical traffic data includes multiple historical browsing requests generated in the service area of the target merchant, the merchant display queues corresponding to each historical browsing request, the merchant exposure quantities corresponding to each merchant display queue, and the evaluation scores of all merchants in each merchant display queue.

[0072] An offline simulation playback is performed for each historical browsing request, and the electronic device adds the target merchant to the merchant display queue corresponding to the historical browsing request to obtain a simulated recall queue for the historical browsing request. The electronic device then uses a second evaluation model deployed in an offline environment to generate an evaluation score for the target merchant based on the service information of the target merchant. Then, based on the evaluation score of the target merchant and the evaluation scores of all merchants in the merchant display queue corresponding to the historical browsing request, all merchants in the simulated recall queue are sorted from large to small according to the evaluation score to obtain a simulated display queue for the historical browsing request. Then, based on the display order of the target merchant in the simulated display queue and the number of merchant exposures corresponding to the historical browsing request, the electronic device evaluates whether the target merchant can be exposed to obtain an exposure evaluation result.

[0073] In the decision-making environment, after the electronic device obtains the exposure evaluation results corresponding to multiple historical browsing requests, it statistically obtains the exposure of the target merchant based on the exposure evaluation results corresponding to the multiple historical browsing requests, and then calculates the product of the exposure of the target merchant, the evaluation score of the target merchant and the preset calibration coefficient, which is the estimated order volume of the target merchant. Finally, based on the estimated order volume of the target merchant, it generates a decision recommendation on whether the target supply resources should participate in the preset activity.

[0074] In one exemplary application scenario, the aforementioned supply resource decision-making method can be applied to intelligent merchant selection in a business recruitment scenario. In this scenario, the supply service platform hopes to attract more high-quality merchants to participate in an event to improve the overall supply quality and user experience. Traditional business recruitment methods rely on manual judgment of merchant potential, such as screening based on historical sales volume, brand awareness, etc., but this approach can be biased and has limited coverage. This solution utilizes intelligent merchant circle formation to implement a data-driven business recruitment decision-making process: the supply service platform identifies multiple target merchants that are not currently participating in the event and, based on the aforementioned method, determines a value metric for each target merchant. This value metric includes at least one of the following: the exposure evaluation result, exposure volume, estimated order volume, and order volume per unit time of the target supply resource in the pre-set event. For example, using estimated order volume, multiple merchants can be ranked based on their respective estimated order volumes, with merchants whose estimated sales volume exceeds a preset threshold or whose rankings are high being selected as priority targets for business recruitment. Based on the results of data analysis, high-potential merchants can be precisely invited to participate in the event, improving the overall effectiveness of the event. Compared with manual merchant circle selection, the use of intelligent merchant circle selection has increased the order production effect of merchant activities by more than 40%, indicating that data-driven intelligent merchant circle selection can more accurately match high-potential merchants and improve activity output efficiency.

[0075] In some embodiments, see Figure 7, embodiments of this specification also provide a supply resource decision-making method, which is applied to an operation client and includes: In S701, determine multiple target supply resources within the service area to be decided and not participating in a preset activity.

[0076] In S702, predict the value indicators of each target supply resource when participating in the preset activity. The value indicators include at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the above method.

[0077] In S703, based on the value indicators respectively corresponding to the multiple target supply resources, decide the target supply resources to be invited to participate in the preset activity from the multiple target supply resources.

[0078] Through the above supply resource decision-making method, the operation client can, in the activity planning stage, scientifically evaluate the potential contributions of each resource in the activity based on the prediction results of the value indicators of multiple target supply resources that have not participated in the activity within the service area, and accordingly screen out target resources with greater exposure potential or conversion value to participate in the preset activity. This method realizes data-driven forward-looking decision-making, effectively improves the accuracy and efficiency of activity resource allocation, and thus helps to improve the overall activity effect and maximize the resource utilization benefit.

[0079] In another exemplary embodiment, the above supply resource decision-making method can be applied to the activity selection scenario of merchants to help merchants choose whether to participate in an activity and how to participate in the activity. When merchants choose to participate in a promotional activity, they usually lack accurate data support and are easily in a "blind selection" state, resulting in the selected activity not maximizing the conversion rate and sales volume. This solution helps merchants intelligently select the most suitable activity to improve the marketing ROI (return on investment).

[0080] For example, taking a certain target merchant as an example, the merchant plans to participate in multiple preset marketing activities initiated by the platform during the upcoming promotion period, but its resources are limited and it cannot participate in all activities simultaneously. To improve the input-output ratio, the merchant client uses the supply resource decision-making method to make supply resource decisions. First, identify multiple preset activities that the target merchant has not participated in, such as "Weekend Flash Sale", "Holiday Big Promotion", "Exclusive New User Discount", etc. Subsequently, for each preset activity, evaluate the value indicators of the target supply resources (such as specific goods or services) provided by the merchant under different activities. The value indicators include at least one of the following: the exposure evaluation result of the target supply resources in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time. Taking the estimated order volume as an example, for instance, based on the estimated order volume of the target merchant in each preset activity, sort the multiple preset activities. Through this prediction and sorting logic, the merchant client can obtain clear and data-driven participation suggestions, thereby avoiding resource waste caused by empiricism or blind selection, and effectively improving the hit rate of promotional activities and the overall operation efficiency.

[0081] In some embodiments, please refer to Figure 8 , the embodiment of the present specification provides a supply resource decision-making method, which is applied to the merchant client and includes: In S801, determine multiple preset activities that the target supply resources have not participated in.

[0082] In S802, predict the value indicators of the target supply resources when participating in each preset activity. The value indicators include at least one of the following: the exposure evaluation result of the target supply resources in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the above method.

[0083] In S803, based on the value indicators of the target supply resources when participating in each preset activity, decide the preset activities that the target supply resources are to participate in from multiple preset activities.

[0084] Through the above supply resource decision-making method, the merchant client can, among multiple unparticipated preset activities, based on the prediction of the value indicators of the target supply resources in different activity scenarios, intelligently identify the most potential and beneficial activities as the participation targets. This method realizes precise matching and optimal allocation based on data driving, significantly improves the use efficiency of supply resources and the conversion effect of participating activities, thereby helping the merchant achieve higher operation benefits and more effective activity participation strategies.

[0085] For another example, the target merchant can determine a variety of different rights and interests according to actual needs, such as (1) full discount (such as 20 off for purchases over 100); (2) discount (such as 10% off, 20% off); (3) coupons; (4) cash back and other rights and interests schemes; but not limited to these. The electronic device will calculate the value index for all different rights and interests schemes of the same target merchant, and the value index includes at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, exposure volume, estimated order volume and order volume per unit time; taking the estimated order volume as an example, the various rights and interests schemes are ranked based on the estimated order volume, and the ranking result directly reflects the potential of various rights and interests schemes in increasing order volume. The rights and interests schemes with higher rankings indicate that they are more likely to generate more orders under the same exposure conditions. Decision recommendations can be generated based on the rights and interests schemes with estimated sales exceeding the preset threshold or with higher rankings, for example, it is recommended that merchants participate in the activity based on the rights and interests schemes with sales exceeding the preset threshold or with higher rankings. By comparing the estimated order volumes under multiple rights and interests, merchants can obtain data-based activity recommendations, avoid blind selection, and improve the accuracy and effectiveness of promotional activities. This method calculates and ranks the estimated order volumes under different benefits, providing merchants with a data-driven promotional activity selection plan. It can not only scientifically predict the order growth potential brought by various benefits, but also provide merchants with a clear decision-making basis on how to allocate promotional resources when participating in activities.

[0086] In other embodiments, see Figure 9 The embodiment of this specification provides a supply resource decision method, which is applied to a merchant client, including: In S901 , a plurality of target supply resources to be involved in a preset activity are determined. The plurality of target supply resources include supply resources with different rights and interests provided by the same resource supplier.

[0087] In S902, the value indicators of each target supply resource when participating in the preset activity are predicted, and the value indicators include at least one of the following: exposure evaluation results, exposure volume, estimated order volume, and order volume per unit time of the target supply resource in the preset activity; wherein the value indicators are determined based on the above method.

[0088] In S903 , based on the value indicators respectively corresponding to the multiple target supply resources, a target supply resource to be involved in the preset activity is determined from the multiple target supply resources.

[0089] Through the above supply resource decision-making method, the merchant client can intelligently select the most potential resources for exposure or conversion from multiple target supply resources with different rights and interests provided by the same resource provider based on the value indicators when participating in the preset activity. This method enables merchants to fully explore the value differences of resources with different rights and interests under the condition of limited resource allocation, implement a more refined and efficient resource selection strategy, and thus improve the overall effect of the preset activity and the merchant's operation return. The various technical features in the above embodiments can be combined arbitrarily as long as there is no conflict or contradiction between the features. However, due to space limitations, they are not described one by one. Therefore, any combination of the various technical features in the above embodiments also belongs to the scope disclosed in this specification.

[0090] In some embodiments, the embodiments of this specification also provide an electronic device, including: a processor; a memory for storing executable instructions of the processor; wherein, the processor realizes the method described in any one of the above by running the executable instructions.

[0091] Figure 10 It is a schematic structural diagram of a device provided by an exemplary embodiment. Please refer to Figure 10 , at the hardware level, the device includes a processor 1002, an internal bus 1004, a network interface 1006, a memory 1008, and a non-volatile memory 1010. Of course, there may also be other hardware required for other functions. One or more embodiments of this specification can be implemented in a software manner. For example, the processor 1002 reads the corresponding computer program from the non-volatile memory 1010 into the memory 1008 and then runs it. Of course, in addition to the software implementation manner, one or more embodiments of this specification do not exclude other implementation manners, such as a logic device or a combination of software and hardware. That is to say, the execution subject of the following processing flow is not limited to each logic unit, and can also be hardware or a logic device.

[0092] In some embodiments, the supply resource decision-making device can be applied to a device as shown in Figure 10 to implement the technical solution of this specification. Among them, the supply resource decision-making device can include: A target supply resource determination module, configured to determine target supply resources that have not participated in the preset activity; A historical traffic data acquisition module, configured to acquire historical traffic data related to the preset activity, where the historical traffic data includes a plurality of historical browsing requests generated in the service area to which the target supply resources belong, a supply resource display queue corresponding to the historical browsing requests, and the supply resource exposure quantity; A traffic playback module, configured to playback the historical browsing requests to determine a simulation display queue for the historical browsing requests based on the target supply resource and the supply resource display queue corresponding to the historical browsing requests; An exposure evaluation module, configured to evaluate whether the target supply resource can be exposed based on the display order of the target supply resource in the simulation display queue and the exposure quantity of the supply resource corresponding to the historical browsing request, so as to obtain an exposure evaluation result; A decision-making module, configured to make a decision on the target supply resource based on the exposure evaluation results respectively corresponding to a plurality of the historical browsing requests.

[0093] Exemplarily, the historical traffic data further includes the evaluation scores of each supply resource in the supply resource display queue; Specifically, the traffic playback module is configured to add the target supply resource to the supply resource display queue corresponding to the historical browsing requests to form a simulation recall queue; determine the evaluation score of the target supply resource based on the service information of the target supply resource; sort the supply resources in the simulation recall queue according to the evaluation score of the target supply resource and the evaluation scores of each supply resource in the supply resource display queue, so as to obtain the simulation display queue.

[0094] Exemplarily, the evaluation scores of each supply resource in the supply resource display queue are generated by a first evaluation model deployed in an online environment based on the service information of each supply resource; the evaluation score of the target supply resource is generated by a second evaluation model deployed in an offline environment based on the service information of the target supply resource; wherein, the second evaluation model is a copy of the first evaluation model; the first evaluation model is obtained through supervised training based on a number of supply resource samples marked with evaluation scores.

[0095] Exemplarily, the service information includes at least one of the following: resource provider characteristics, supply resource characteristics, and rights and interests characteristics, and the rights and interests characteristics are used to indicate the preferential information provided by the resource provider in the preset activity; Exemplarily, the evaluation score includes click-through rate and / or conversion rate.

[0096] Exemplarily, the decision-making module is specifically configured to count the exposure quantity of the target supply resource based on the exposure evaluation results respectively corresponding to the plurality of historical browsing requests; and make a decision on whether the target supply resource participates in the preset activity based on the exposure quantity of the target supply resource.

[0097] Exemplarily, the decision-making module is specifically configured to obtain the estimated order volume of the target supply resource based on the exposure volume of the target supply resource and the evaluation score of the target supply resource; wherein, the evaluation score of the target supply resource is determined based on the service information of the target supply resource; and generate a decision-making suggestion on whether the target supply resource participates in the preset activity based on the estimated order volume of the target supply resource.

[0098] Exemplarily, the estimated order volume of the target supply resource has a positive correlation with the exposure volume of the target supply resource; and the estimated order volume of the target supply resource has a positive correlation with the evaluation score of the target supply resource.

[0099] Exemplarily, the estimated order volume of the target supply resource is the product of the exposure volume of the target supply resource, the evaluation score of the target supply resource, and a preset calibration coefficient; wherein, the preset calibration coefficient is determined based on the relationship between the actual order volume, actual exposure volume, and evaluation score of multiple reference supply resources that have participated in the preset activity.

[0100] Exemplarily, the multiple historical browsing requests included in the historical traffic data include: historical browsing requests generated in the service area of the target supply resource within a preset time period; the decision-making module is specifically configured to determine the order volume of the target supply resource per unit time based on the estimated order volume of the target supply resource and the preset time period; and generate a decision-making suggestion on whether the target supply resource participates in the preset activity based on the order volume of the target supply resource per unit time.

[0101] Exemplarily, in the case where there are multiple target supply resources to be decided, the multiple target supply resources include: supply resources provided by different resource providers, and / or supply resources with different rights and interests provided by the same resource provider; wherein, the different rights and interests indicate different preferential strategies provided by the resource provider in the preset activity.

[0102] In some embodiments, the supply resource decision-making device can be applied to a device as Figure 10 shown to implement the technical solutions of this specification. Among them, the supply resource decision-making device can include: A target supply resource determination module, configured to determine multiple target supply resources that are in the service area to be decided and have not participated in the preset activity.

[0103] A prediction module, configured to predict the value indicators of each target supply resource when participating in a preset activity, where the value indicators include at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the above method.

[0104] A decision-making module, configured to, based on the value indicators respectively corresponding to multiple target supply resources, decide the target supply resources to be invited to participate in the preset activity from the multiple target supply resources.

[0105] In some embodiments, the supply resource decision-making device can be applied to a device as Figure 10 shown, to implement the technical solutions of this specification. Wherein, the supply resource decision-making device may include: A preset activity determination module, configured to determine multiple preset activities that the target supply resources have not participated in.

[0106] A prediction module, configured to predict the value indicators of the target supply resources when participating in each preset activity, where the value indicators include at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the above method.

[0107] A decision-making module, configured to, based on the value indicators of the target supply resources when participating in each preset activity, decide the preset activities that the target supply resources are to participate in from the multiple preset activities.

[0108] In some embodiments, the supply resource decision-making device can be applied to a device as Figure 10 shown, to implement the technical solutions of this specification. Wherein, the supply resource decision-making device may include: A target supply resource determination module, configured to determine multiple target supply resources to participate in the preset activity, where the multiple target supply resources include supply resources with different rights and interests provided by the same resource provider.

[0109] A prediction module, configured to predict the value indicators of each target supply resource when participating in the preset activity, where the value indicators include at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the above method.

[0110] A decision-making module, configured to, based on the value indicators respectively corresponding to the multiple target supply resources, decide the target supply resources to participate in the preset activity from the multiple target supply resources.

[0111] The implementation processes of the functions and roles of each module in the above device are specifically detailed in the implementation processes of the corresponding steps in the above method, and will not be elaborated here.

[0112] Based on the same concept as the above method, this specification also provides an electronic device, including: a processor; a memory for storing instructions executable by the processor; wherein, the processor runs the executable instructions to implement the steps of the method as described in any of the above embodiments.

[0113] Based on the same concept as the above method, this specification also provides a computer-readable storage medium, on which computer instructions are stored, and when the instructions are executed by a processor, the steps of the method as described in any of the above embodiments are implemented.

[0114] Computer-readable media includes both permanent and non-permanent, removable and non-removable media and can be implemented by any method or technology for information storage. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic disk storage, quantum memory, graphene-based storage media, or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0115] Based on the same concept as the above method, this specification also provides a computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the method as described in any of the above embodiments are implemented.

[0116] The above description is only the preferred embodiment of one or more embodiments of this specification, and is not intended to limit one or more embodiments of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of one or more embodiments of this specification shall be included within the protection scope of one or more embodiments of this specification.

Claims

1. A method for making a supply resource decision, comprising: Determining a target supply resource that has not participated in a preset activity; Obtaining historical traffic data related to the preset activity, where the historical traffic data includes a plurality of historical browsing requests generated in the service area to which the target supply resource belongs, a supply resource display queue corresponding to the historical browsing requests, and the number of supply resource exposures; Replaying the historical browsing requests to determine a simulation display queue for the historical browsing requests based on the target supply resource and the supply resource display queue corresponding to the historical browsing requests; Evaluating whether the target supply resource can be exposed based on the display order of the target supply resource in the simulation display queue and the number of supply resource exposures corresponding to the historical browsing requests, to obtain an exposure evaluation result; Making a decision on the target supply resource based on the exposure evaluation results respectively corresponding to the plurality of historical browsing requests.

2. The method according to claim 1, wherein the historical traffic data further includes the evaluation scores of each supply resource in the supply resource display queue; The determining a simulation display queue for the historical browsing requests based on the target supply resource and the supply resource display queue corresponding to the historical browsing requests includes: Adding the target supply resource to the supply resource display queue corresponding to the historical browsing requests to form a simulation recall queue; Determining the evaluation score of the target supply resource based on the service information of the target supply resource; Sorting the supply resources in the simulation recall queue according to the evaluation score of the target supply resource and the evaluation scores of each supply resource in the supply resource display queue to obtain the simulation display queue.

3. The method according to claim 2, wherein the evaluation scores of each supply resource in the supply resource display queue are generated by a first evaluation model deployed in an online environment based on the service information of each supply resource; The evaluation score of the target supply resource is generated by a second evaluation model deployed in an offline environment based on the service information of the target supply resource; Among them, The second evaluation model is a copy of the first evaluation model; The first evaluation model is obtained through supervised training based on a number of supply resource samples marked with evaluation scores.

4. The method according to claim 2 or 3, wherein the service information includes at least one of the following: resource provider characteristics, supply resource characteristics, and rights and interests characteristics, and the rights and interests characteristics are used to indicate the preferential information provided by the resource provider in the preset activity; The evaluation score includes click-through rate and / or conversion rate.

5. The method according to claim 1, wherein the making a decision on the target supply resource based on the exposure evaluation results respectively corresponding to the plurality of historical browsing requests includes: Making a decision on whether the target supply resource participates in the preset activity based on the exposure evaluation results respectively corresponding to the plurality of historical browsing requests; Or, Making a decision on whether to invite the target supply resource to participate in the preset activity based on the exposure evaluation results respectively corresponding to the plurality of historical browsing requests; And / or, In the case where there are multiple target supply resources to be decided, the multiple target supply resources include at least one of the following: supply resources provided by different resource providers, and supply resources with different rights and interests provided by the same resource provider; wherein, the different rights and interests indicate different preferential strategies provided by the resource provider in the preset activity.

6. The method according to any one of claims 1 to 3, wherein the making a decision on the target supply resources based on the exposure evaluation results respectively corresponding to the multiple historical browsing requests includes: Counting the exposure volume of the target supply resources based on the exposure evaluation results respectively corresponding to the multiple historical browsing requests; Making a decision on the target supply resources based on the exposure volume of the target supply resources.

7. The method according to claim 6, wherein the making a decision on the target supply resources based on the exposure volume of the target supply resources includes: Obtaining the estimated order volume of the target supply resources based on the exposure volume of the target supply resources and the evaluation score of the target supply resources; wherein, the evaluation score of the target supply resources is determined based on the service information of the target supply resources; the estimated order volume of the target supply resources has a positive correlation with the exposure volume of the target supply resources; and the estimated order volume of the target supply resources has a positive correlation with the evaluation score of the target supply resources; Making a decision on the target supply resources based on the estimated order volume of the target supply resources.

8. The method according to claim 7, wherein the estimated order volume of the target supply resources is the product of the exposure volume of the target supply resources, the evaluation score of the target supply resources, and a preset calibration coefficient; Among them, The preset calibration coefficient is determined based on the relationship among the actual order volume, actual exposure volume, and evaluation score of multiple reference supply resources that have participated in the preset activity.

9. The method according to claim 7, wherein the plurality of historical browsing requests included in the historical traffic data comprise: Historical browsing requests generated in the service area to which the target supply resources belong within a preset time period; The making a decision on the target supply resources based on the estimated order volume of the target supply resources includes: Determining the order volume per unit time of the target supply resources based on the estimated order volume of the target supply resources and the preset time period; Making a decision on the target supply resources based on the order volume per unit time of the target supply resources.

10. A supply resource decision method, comprising: Determining multiple target supply resources within the service area to be decided and not participating in the preset activity; Predicting the value indicators of each of the target supply resources when participating in the preset activity, the value indicators including at least one of the following: the exposure evaluation result, exposure volume, estimated order volume, and order volume per unit time of the target supply resources in the preset activity; wherein, the value indicators are determined according to the method of any one of claims 1 to 9; Making a decision on the target supply resources to be invited to participate in the preset activity from the multiple target supply resources based on the value indicators respectively corresponding to the multiple target supply resources.

11. A supply resource decision method, comprising: Determine multiple preset activities that the target supply resource does not participate in; Predict the value indicators of the target supply resource when participating in each preset activity, where the value indicators include at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the method described in any one of claims 1 to 9; Based on the value indicators of the target supply resource when participating in each preset activity, decide the preset activity that the target supply resource is to participate in from the multiple preset activities.

12. A supply resource decision-making method, comprising: Determine multiple target supply resources to participate in a preset activity, where the multiple target supply resources include supply resources with different rights and interests provided by the same resource provider; Predict the value indicators of each of the target supply resources when participating in the preset activity, where the value indicators include at least one of the following: the exposure evaluation result of the target supply resource in the preset activity, the exposure volume, the estimated order volume, and the order volume per unit time; wherein, the value indicators are determined based on the method described in any one of claims 1 to 9; Based on the value indicators respectively corresponding to the multiple target supply resources, decide the target supply resource to participate in the preset activity from the multiple target supply resources.

13. An electronic device, comprising: A processor; A memory for storing instructions executable by the processor; wherein, the processor realizes the steps of the method described in any one of claims 1 to 12 by running the executable instructions.

14. A computer-readable storage medium, on which computer instructions are stored, and when the instructions are executed by a processor, the steps of the method described in any one of claims 1 to 12 are realized.

15. A computer program product, comprising a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the method described in any one of claims 1 to 12 are realized.

Images