Recommendation method and related device

Abstract

The embodiment of the invention provides a recommendation method and a related device.The recommendation method comprises the steps that resource recommendation probability prediction is conducted according to user data, a first probability is obtained, and a first parameter and a second parameter corresponding to the first probability are inquired; inputting the user data and the resource value of the first resource into a model for transaction growth prediction to obtain a third parameter; and determining recommendation information in the first resource according to at least one of the first parameter and the second parameter, the third parameter and the resource value. By adopting the embodiment of the invention, the accuracy of information recommendation can be improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a recommendation method and related devices. Background Art

[0002] With the continuous development of Internet technology, more and more trading parties are gradually providing related trading services online, and the competition among trading parties is also becoming increasingly fierce. In order to expand their influence and increase the number of users, trading parties will carry out some online marketing activities. In the process of carrying out marketing activities, resource recommendation is crucial, such as the recommendation of resources such as coupons, red envelopes for full discounts, and gift coupons. Through resource recommendation, the competitiveness of the trading parties themselves can be improved. In this process, each trading party also faces greater pressure and challenges in resource recommendation. Summary of the Invention

[0003] In a first aspect, an embodiment of the present application provides a recommendation method, comprising: Predicting the probability of resource recommendation based on user data to obtain a first probability, and querying a first parameter and a second parameter corresponding to the first probability, where the first parameter is used to represent the transaction amount, and the second parameter is used to represent the utilization rate of the resource after the resource recommendation; Inputting the user data and the resource value of the first resource into a model to perform transaction growth prediction, thereby obtaining a third parameter, wherein the third parameter is used to represent the growth rate of the transaction amount; Recommendation information is determined in the first resource according to at least one of the first parameter and the second parameter, the third parameter, and the resource value.

[0004] It can be seen that in the embodiment of the present application, the resource recommendation probability is predicted based on the user data, and the first parameter and the second parameter corresponding to the predicted first probability are queried. Since the resource recommendation probability prediction is predicted from the entire resource dimension, the first parameter and the second parameter corresponding to the predicted first probability are parameters determined from a coarse-grained level. On this basis, the user data and the resource value of the first resource are input into the model, and the transaction growth prediction is performed through the model to obtain the third parameter. In this way, by specifying the resource to the first resource, the third parameter is determined from a fine-grained level. Finally, based on at least one of the first parameter and the second parameter, the third parameter and the resource value of the first resource, the recommended information is determined in the first resource. In this way, the comprehensiveness of the recommended information is improved by at least one of the first parameter and the second parameter at the coarse-grained level, the third parameter at the fine-grained level, and the resource value of the first resource, thereby improving the accuracy of the information recommendation. In addition, since the first parameter represents the transaction amount, the second parameter represents the utilization rate of the resource after the resource recommendation, and the third parameter represents the growth rate of the transaction amount, in the process of determining the recommended information, data from multiple different dimensions can be combined, which helps to improve the effectiveness and accuracy of the information recommendation.

[0005] In a second aspect, an embodiment of the present application provides a recommendation device, including: A probability prediction module is used to predict the probability of resource recommendation based on user data, obtain a first probability, and query a first parameter and a second parameter corresponding to the first probability, where the first parameter is used to represent the transaction amount and the second parameter is used to represent the utilization rate of the resource after the resource recommendation; a prediction module, configured to input the user data and the resource value of the first resource into a model to perform transaction growth prediction, thereby obtaining a third parameter, wherein the third parameter is used to represent a growth rate of the transaction amount; An information determination module is configured to determine recommended information in the first resource based on at least one of the first parameter and the second parameter, the third parameter, and the resource value.

[0006] In a third aspect, an embodiment of the present application provides a computer device comprising: a processor; and a memory configured to store computer-executable instructions, wherein the computer-executable instructions, when executed, cause the processor to execute the recommendation method described in the first aspect.

[0007] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium for storing computer-executable instructions, which, when executed by a processor, implement the recommendation method as described in the first aspect.

[0008] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the recommendation method as described in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments described in this specification. Those skilled in the art can also derive other drawings based on these drawings without inventive work. Figure 1 A flowchart of a recommended method provided in an embodiment of the present application; Figure 2 A schematic diagram of a sample grouping generation process provided in an embodiment of the present application; Figure 3 A schematic diagram of a third parameter provided in an embodiment of the present application; Figure 4 A flowchart of a recommendation method for a resource recommendation scenario provided in an embodiment of the present application; Figure 5 A schematic diagram of a recommended device provided in an embodiment of the present application; Figure 6 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0010] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0011] This specification provides a recommended method embodiment: Currently, when recommending information to users, we often only focus on the user's purchase probability. We select resources from the resource set to be recommended based on the purchase probability and recommend information to users. As a result, there is less reference data in the resource selection process and the accuracy of information recommendation is also low.

[0012] To this end, the recommendation method provided in this embodiment, on the one hand, predicts the resource recommendation probability based on user data, and queries the first parameter and second parameter corresponding to the predicted first probability, so as to determine the first parameter and the second parameter from a coarse-grained level; on the other hand, the user data and the resource value of the first resource are input into the model, and the transaction growth prediction is performed through the model to obtain the third parameter, so as to determine the third parameter from a fine-grained level by specifying the resource to the first resource; finally, the recommendation information is determined in the first resource based on at least one of the first parameter and the second parameter, the third parameter and the resource value of the first resource, so as to improve the comprehensiveness of the recommended information determined by at least one of the first parameter and the second parameter at the coarse-grained level, the third parameter at the fine-grained level, and the resource value of the first resource, thereby improving the accuracy of the information recommendation; in addition, the transaction amount represented by the first parameter, the resource utilization rate after the resource recommendation represented by the second parameter, and the growth rate of the transaction amount represented by the third parameter are used to comprehensively determine the recommended information, which helps to improve the effectiveness and accuracy of the information recommendation.

[0013] The recommendation method provided in this embodiment can be executed by a recommendation device, which can be a terminal or a server, wherein the terminal can include a mobile phone, a laptop computer, an intelligent interactive device, etc., and the server can include an independent physical server, a server cluster composed of multiple servers, or a cloud server capable of cloud computing.

[0014] Reference Figure 1 The recommendation method provided in this embodiment specifically includes steps S102 to S106.

[0015] Step S102 : predicting resource recommendation probability based on user data to obtain a first probability, and querying a first parameter and a second parameter corresponding to the first probability.

[0016] The user data in this embodiment refers to the relevant data of the user, and the user data may include user attribute data and / or user behavior data, such as user attribute data such as age, gender, occupation and / or income, and user behavior data such as user browsing data, user click data and / or user login data. The examples of user attribute data and user behavior data here are only illustrative, and can be determined according to the actual application scenario; optionally, the first parameter is used to represent the transaction amount, and the transaction amount can specifically be the transaction amount when no resource recommendation is made, that is, it can be the transaction amount of the user when no resource recommendation is made to the user, such as the transaction amount between the user and the transaction party when no resource recommendation is made to the user, that is, the consumption amount or purchase amount; optionally, the second parameter is used to represent the utilization rate of the resource after the resource recommendation, that is, it can be the utilization rate or utilization ratio of the resource by the user after the resource is recommended to the user; this embodiment can be applied to an information recommendation system, and the information recommendation system can be deployed on a recommendation device.

[0017] In this embodiment, the resource may be an equity resource or an equity, specifically any type of equity resource, such as a coupon, phone bill voucher, gift voucher, and / or a red envelope with a discount. Equity resources may also be other types of equity resources. The first probability may be used to represent the probability of a resource being recommended, specifically the probability of a resource being recommended to a user.

[0018] In a specific implementation, in order to improve the comprehensiveness and accuracy of information recommendation to users, a resource recommendation probability prediction can be performed based on user data to obtain a first probability, which is used to predict the probability of resource recommendation to the user. After obtaining the first probability, a first parameter and a second parameter corresponding to the first probability are queried. Specifically, in the process of performing resource recommendation probability prediction based on user data to obtain the first probability, the user data can be input into a probability prediction model to perform resource recommendation probability prediction to obtain the first probability. Among them, the probability prediction model can be a classification model, specifically a binary classification model, for example, the model structure of the probability prediction model is DNN (Deep Neural Network); optionally, the probability prediction model can be obtained by model training based on positive samples and / or negative samples, where the positive samples can be user samples of recommended resources and the negative samples can be user samples of non-recommended resources.

[0019] In actual applications, users may have different first probabilities, and transaction amounts and resource usage rates after resource recommendation may also be different. Therefore, in order to improve the flexibility and convenience of determining the first parameter and the second parameter, an optional implementation provided in this embodiment performs the following operations during the process of querying the first parameter and the second parameter corresponding to the first probability: determining a probability interval in which the first probability lies; The first parameter and the second parameter corresponding to the probability interval are searched in the parameter mapping table.

[0020] The parameter mapping table may be a table consisting of a plurality of probability intervals and a first parameter and a second parameter corresponding to each probability interval.

[0021] Specifically, the multiple probability intervals in the parameter mapping table can be obtained in the following manner: clustering the fourth training sample according to the first probability of the fourth training sample to obtain the second training sample, and constructing multiple probability intervals based on the largest first probability and the smallest first probability among the first probabilities of the second training samples.

[0022] Among them, the first probability of the fourth training sample can be obtained after the fourth training sample is input into the probability prediction model to predict the resource recommendation probability; in the process of clustering the fourth training sample according to the first probability of the fourth training sample to obtain the second training sample, the fourth training sample can be sorted according to the first probability of the fourth training sample to obtain the fifth training sample, and the fifth training sample can be split into second training samples according to a preset number; optionally, the second resource contained in the second training sample matches the preset resource, and the number of the second resource contained in the second training sample exceeds the preset number threshold.

[0023] In the above process of splitting the fifth training sample into the second training sample according to the preset number, the fifth training sample can be split according to the first preset number to obtain a sample bucket. If the resources contained in the training sample in the sample bucket match the preset resources, and the number of training samples corresponding to the resources exceeds the preset number threshold, the sample bucket is determined to be the second training sample. If the resources contained in the training sample in the sample bucket do not match the preset resources, or the number of training samples corresponding to the resources does not exceed the preset number threshold, the first preset number is reduced to obtain the second preset number, and the first preset number is updated based on the second preset number until the resources contained in the training sample in the obtained sample bucket match the preset resources and the number of training samples of the resource exceeds the preset number threshold, the sample bucket can be used as the second training sample.

[0024] For example, Figure 2 As shown, the resource recommendation probability is predicted for the fourth training sample to obtain the first probability of the fourth training sample, the fourth training sample is sorted based on the first probability of the fourth training sample to obtain the fifth training sample, and the fifth training sample is split according to a preset number into m1 sample buckets, and the m1 sample buckets are the second training samples.

[0025] For another example, the fourth training sample is sorted according to the first probability of the fourth training sample to obtain the fifth training sample, and the fifth training sample is divided into m1 equal parts according to the first preset number m1 to obtain m1 sample buckets. Assuming that the preset resources are Q resources, if the resources contained in the training samples in each sample bucket all contain the preset resources, and the number of training samples corresponding to the resources exceeds the preset number threshold, that is, the resources contained in the training samples in each sample bucket all contain Q resources, and the number of training samples corresponding to each resource in the Q resources exceeds the preset number threshold, then the m1 sample bucket is used as the second training sample. Otherwise, reduce the first preset number m1 to obtain the second preset number m2, and update the first preset number m1 based on the second preset number m2 until the resources contained in the training samples in the obtained sample buckets all contain the preset resources, and the number of training samples corresponding to the resources exceeds the preset number threshold. The obtained sample buckets are used as the second training samples. Assuming that the second training samples are m1 sample buckets, based on the minimum first probability and the maximum first probability of the first probabilities of the training samples in each sample bucket, construct multiple probability intervals, that is, the minimum first probability is the lower limit of the interval, and the maximum first probability is the upper limit of the interval.

[0026] On this basis, in an optional implementation manner provided by this embodiment, the first parameter and the second parameter corresponding to any (any) probability interval in the parameter mapping table are calculated based on the sample group corresponding to the any probability interval and the control sample of the sample group. Specifically, the calculation can be performed as follows: Based on the sample number and total transaction amount of the control samples of the sample group corresponding to any probability interval, the average transaction amount of the control samples is calculated as the first parameter corresponding to any probability interval; According to the number of samples in the sample group and the number of samples using resources in the sample group, the resource usage rate after resource recommendation is calculated as the second parameter corresponding to any probability interval.

[0027] Among them, the sample group refers to a group composed of one or more samples, each sample in the sample group can be a user sample of recommended resources, and the control samples of the sample group can also be one or more. The control samples can be samples of non-recommended resources, that is, they can be user samples of non-recommended resources, such as user samples of users who have not been issued coupons; the sample number of control samples refers to the sample number of control samples, and the total transaction amount of the control samples can be the total purchase amount or total consumption amount of the control samples; the sample number of the sample group refers to the sample number of samples included in the sample group, and the number of samples using resources in the sample group can be the number of samples in the sample group that use recommended resources.

[0028] Optionally, the sample grouping is obtained after grouping the second training samples according to the resource value of the second resource contained in the second training samples; after grouping the second training samples according to the resource value of the second resource contained in the second training samples, a control sample of the sample grouping can also be obtained, for example, Figure 2 As shown, the preset resources are Q coupons of different denominations, and the second training samples are m1 sample buckets. Taking sample bucket 1 as an example, after grouping sample bucket 1, Q sample groups and 1 control sample are obtained, that is, for each sample bucket, training samples with the same coupons are divided into the same sample group, and training samples without coupon recommendations are divided into control samples.

[0029] Specifically, in the process of calculating the average transaction amount of the control samples based on the sample number and total transaction amount of the control samples of the sample grouping corresponding to any probability interval, the ratio of the total transaction amount of the control samples of the sample grouping corresponding to any probability interval to the sample number can be calculated as the average transaction amount of the control samples; in the process of calculating the resource utilization rate after resource recommendation based on the sample number of the sample grouping and the number of samples using the resources in the sample grouping, the ratio of the number of samples using the resources in the sample grouping to the number of samples in the sample grouping can be calculated as the resource utilization rate after resource recommendation; in this way, the first parameter and the second parameter corresponding to any probability interval can be conveniently calculated through the sample grouping and the control sample. At the same time, the division of the probability interval is equivalent to the division of homogeneous users. The same transaction amount and resource utilization rate after resource recommendation can be flexibly set to homogeneous users through the parameter mapping table, which is helpful for subsequent resource recommendation to users.

[0030] Step S104: Input the user data and the resource value of the first resource into the model to perform transaction growth prediction to obtain a third parameter.

[0031] The above-mentioned resource recommendation probability prediction based on user data obtains the first probability, and queries the corresponding first and second parameters based on the first probability. In this step, the user data and the resource value of the first resource are input into the model, and the transaction growth prediction is performed through the model to obtain the third parameter.

[0032] In this embodiment, the first resource can be the resource to be recommended. The first resource can be one or more. For example, the first resource can be multiple coupons of different amounts, such as x1 yuan, x2 yuan, x3 yuan, etc. Another example is that the first resource can be multiple different types of resources, such as a x1 yuan coupon, a x2 yuan gift coupon, a x3 yuan phone coupon, etc. The resource value of the first resource can be the equity amount of the first equity resource, specifically the face value or value of the first equity resource. For example, a x1 yuan coupon represents a resource value of x1 yuan for the first resource, while a x3 yuan phone coupon represents a resource value of x3 yuan for the first resource. The model in this embodiment can be used to predict transaction growth, and the model structure can be a DNN (deep learning network).

[0033] Optionally, the third parameter is used to represent the growth rate of the transaction amount; the growth rate of the transaction amount can specifically represent the growth rate of the transaction amount when the user is recommended the first resource relative to when the first resource is not recommended; the third parameter can specifically be the third parameter of each resource of the user in the first resource, that is, each resource of the user in the first resource can have a corresponding third parameter, for example, Figure 3 As shown, there are q resources in the first resource, and the user data and the resource values ​​of the q resources are input into the model for transaction growth prediction to obtain the third parameter of the user under the q resources. Alternatively, the user data and resources 1, 2...q in the first resource can be input into the model in sequence for transaction growth prediction to obtain the third parameter of the user under resources 1, 2...q.

[0034] During specific implementation, user data and the resource value of each resource in the first resource can be directly input into the model for transaction growth prediction to obtain the third parameter of the user under each resource. User data and the resource value of each resource in the first resource can also be input into the model in turn for transaction growth prediction to obtain the third parameter of the user under each resource. For example, user data and resource 1 in the first resource are input for the first time, user data and resource 2 in the first resource are input for the second time, and so on until user data and resource q in the first resource are input to obtain the third parameter of the user under each resource in the first resource.

[0035] In practical applications, in order to improve the accuracy and effectiveness of transaction growth prediction, a model can be trained in advance to predict transaction growth through the model. At the same time, the second resource included in the second training sample may not be suitable for the user and cannot meet the user's actual consumption needs. In order to improve the adaptability of the second resource included in the second training sample to the user, the effectiveness and accuracy of the second training sample are improved. In an optional implementation manner provided in this embodiment, the model is trained based on the first training sample; The first training sample is generated as follows: Grouping the second training samples according to the resource value of the second resource included in the second training samples to obtain sample groups, and determining the transaction amount and transaction rate of the sample groups; A resource is selected from the second resources according to the transaction amount and transaction rate of the sample group, and a resource value of the second resource included in the second training sample is updated based on the selected resource to obtain a first training sample.

[0036] Optionally, the second training sample is obtained by clustering the fourth training sample based on the first probability of the fourth training sample. The implementation process of the clustering process has been described in detail above and will not be repeated here. The implementation process of grouping the second training samples according to the resource value of the second resource contained in the second training sample to obtain the sample grouping is also described in detail above and will not be repeated here.

[0037] Specifically, in the process of determining the transaction amount and transaction rate of the sample group, the transaction rate of the sample group can be calculated based on the number of samples in the sample group and the number of transaction samples in the sample group, and the average transaction amount of the sample group can be calculated based on the number of samples in the sample group and the total transaction amount of the sample group as the transaction amount of the sample group; the number of transaction samples here can be the number of training samples for transactions or consumption in the sample group, and the total transaction amount of the sample group can be the total consumption amount or the total purchase amount of the sample group, such as the total transaction amount of users in the sample group; the average transaction amount of the sample group can be the average consumption amount or the average purchase amount, and the transaction rate can be the transaction ratio.

[0038] In the process of calculating the transaction rate of the sample group based on the number of samples in the sample group and the number of transaction samples in the sample group, the ratio of the number of transaction samples in the sample group to the number of samples can be calculated as the transaction rate of the sample group; in the process of calculating the average transaction amount of the sample group based on the number of samples in the sample group and the total transaction amount of the sample group, the ratio of the total transaction amount of the sample group to the number of samples in the sample group can be calculated as the average transaction amount of the sample group.

[0039] On this basis, in an optional implementation provided by this embodiment, in the process of selecting resources from the second resource based on the transaction amount and transaction rate of the sample grouping, the following operations are performed: Calculate the third and fifth parameters of the sample group based on the transaction amount and transaction rate of the sample group and the transaction amount and transaction rate of the control sample; optionally, the fifth parameter is used to represent the growth value of the transaction rate; A resource is selected from the second resources based on the third parameter and the fifth parameter of the sample grouping.

[0040] Among them, the control sample can be the control sample of the sample group. The control sample can also be obtained after grouping the second training samples according to the resource values of the second resources included in the second training samples. The control sample can be the training samples that do not include the second resources in the second training samples, that is, the training samples or user samples that are not recommended by resources in the second training samples.

[0041] Specifically, in the process of calculating the third parameter and the fifth parameter of the sample group based on the transaction amount and transaction rate of the sample group and the transaction amount and transaction rate of the control sample, the difference between the transaction rate of the sample group and the transaction rate of the control sample of the sample group can be calculated as the fifth parameter. In addition, calculate the ratio of the transaction amount of the sample group to the transaction amount of the control sample, and calculate the difference between this ratio and a preset value (such as 1) as the third parameter. In this way, the growth value of the transaction rate and the growth rate of the transaction amount of the sample group when recommended the second resource compared to when not recommended the second resource can be obtained conveniently, and the impact degree of the second resource on the transaction of the sample group can be mined, which is convenient for subsequent selection of resources in the second resources.

[0042] Furthermore, in order to ensure that there is a positive transaction profit amount in the subsequent first training samples while reducing the input resource cost, making the first training samples more in line with the transaction profit amount and cost requirements of the trading parties, and thus improving the accuracy of the first training samples. In an optional implementation manner provided in this embodiment, in the process of selecting resources from the second resources based on the third parameter and the fifth parameter of the sample group, the following operations are performed: Screen the sample groups in which both the third parameter and the fifth parameter are greater than the preset value from the sample groups; Among the second resources of the screened sample groups, select the second resource with the smallest resource value as the selected resource.

[0043] For example, in this embodiment, the preset value can be set to 0, and there are 3 sample groups. The third parameter and the fifth parameter of the sample groups are respectively: for sample group 1 (coupon 1, x1 yuan), the growth rate of the transaction amount is 0.8% and the growth value of the transaction rate is -0.3%; for sample group 2 (coupon 2, x2 yuan), the growth rate of the transaction amount is 0.6% and the growth value of the transaction rate is 0.4%; for sample group 3 (coupon 3, x3 yuan), the growth rate of the transaction amount is 0.7% and the growth value of the transaction rate is 0.6%. Screen the sample groups in which both the third parameter and the fifth parameter are greater than 0 from the sample groups (sample group 1, sample group 2, sample group 3) as sample group 2 and sample group 3. Among the second resources of sample group 2 - coupon 2 and the second resource of sample group 3 - coupon 3, determine the resource with the smallest resource value. Assuming x2 < x3, then select coupon 2 as the selected resource.

[0044] In the process of updating the resource value of the second resource contained in the second training sample based on the selected resource, the resource value of the second resource contained in the second training sample can be replaced with the resource value of the selected resource; continuing with the above example, if the selected resource is Coupon 2, the resource value of the second resource contained in the second training sample is replaced with the face value x2 of Coupon 2.

[0045] It should be added that, in the process of training the model based on the first training sample to obtain the model, the first training sample can be input into the model to be trained to predict transaction growth, and the third parameter of the first training sample can be obtained. The loss is calculated based on the third parameter of the first training sample and the sample label of the first training sample, and the parameters of the model to be trained are adjusted based on the calculated loss. The loss calculation here can be calculated based on the loss function, and the loss can be the mean square error loss. The model to be trained can be iteratively trained with reference to the above process until the loss converges to obtain a model that can be used for actual transaction growth prediction.

[0046] Step S106 : determining recommendation information in the first resource according to at least one of the first parameter and the second parameter, the third parameter, and the resource value.

[0047] The above model predicts transaction growth based on user data and the resource value of the first resource to obtain the third parameter. In this step, the recommended information is determined in the first resource in combination with at least one of the first parameter and the second parameter, the third parameter and the resource value of the first resource. The recommended information can be recommended to the user to improve the accuracy of the recommended information.

[0048] The recommendation information in this embodiment may include a third resource determined in the first resource, which may specifically be a resource recommended to the user. For example, if the first resource is four coupons of x1 yuan, x2 yuan, x3 yuan, and x4, the third resource may be any one or more coupons selected from x1 yuan, x2 yuan, x3 yuan, and x4.

[0049] In actual applications, during the resource recommendation process, the cost may be fixed. To achieve the highest possible transaction profit while meeting cost requirements and improve the utilization rate of resources recommended to users, in an optional implementation provided by this embodiment, in the process of determining recommendation information in the first resource based on at least one of the first parameter and the second parameter, the third parameter, and the resource value, the following operations are performed: Calculating a fourth parameter corresponding to each resource in the first resource based on at least one of the first and third parameters and the resource value; optionally, the fourth parameter is used to represent a transaction profit amount of the transaction party after the user uses each resource to trade with the transaction party; The first resource with the largest transaction profit is selected from the first resources as the recommended information.

[0050] The transaction profit amount may be the actual profit amount of the transaction parties.

[0051] Specifically, the fourth parameter corresponding to each resource in the first resource can be calculated based on at least one, the third parameter, the resource value of the first resource and the resource recommendation expenditure; the resource recommendation expenditure here may include the resource recommendation cost, that is, it can be the cost of making resource recommendations, such as the total cost of resource recommendation set in advance by the transaction parties.

[0052] To be more specific, in the process of calculating the fourth parameter corresponding to each resource in the first resource based on at least one, the third parameter and the resource value of the first resource, the sixth parameter can be calculated based on the transaction amount and the second parameter in the first parameter, and the seventh parameter can be calculated based on the resource value of each resource in the first resource and the utilization rate of the resource after the resource recommendation in the first parameter, and the fourth parameter corresponding to each resource in the first resource is calculated based on the sixth parameter and the seventh parameter; wherein the sixth parameter can be used to represent the transaction amount of user growth after the recommendation of each resource in the first resource; and the seventh parameter can be used to represent the cost of each resource in the first resource after being used.

[0053] It should be supplemented that the above process of determining recommendation information in the first resource based on at least one of the first parameter and the second parameter, the third parameter, and the resource value of the first resource can be performed based on a resource allocation model. Specifically, the first parameter, the second parameter, the third parameter, the resource value of the first resource, and the resource recommendation expenditure can be input into the resource allocation model to perform resource allocation, obtain the user's allocation status for each resource in the first resource, and determine recommendation information in the first resource based on the allocation status; the allocation status can be used to indicate whether each resource in the first resource is recommended to the user, for example, an allocation status of 1 indicates that the resource is recommended to the user, and an allocation status of 0 indicates that the resource is not recommended to the user; For example, resource allocation models include:

[0054] and is a positive integer

[0055] in, represents the allocation status of user i for each resource q in the first resource, {0,1|Recommend each resource q to user i and take 1, otherwise take 0}; represents the resource value of each resource q in the first resource; represents the third parameter of user i under each resource q in the first resource; The first parameter representing user i; The second parameter representing user i; Represents the total number of users; Represents resource recommendation expenditure, which can be the resource recommendation cost.

[0056] It should be noted that the above step S106 can be replaced by determining the recommended information in the first resource based on at least one of the first parameter and the second parameter, the third parameter and / or the resource value of the first resource. Here, the unused data in the first parameter, the second parameter, the third parameter and the resource value of the first resource may not be obtained or calculated in the above.

[0057] In summary, the recommendation method provided in this embodiment first predicts the probability of resource recommendation based on user data to obtain a first probability, and determines the probability interval in which the first probability lies. The first parameter and the second parameter corresponding to the probability interval are queried in the parameter mapping table. The first parameter is used to represent the transaction amount, and the second parameter is used to represent the utilization rate of the resource after the resource recommendation. In this way, based on the resource recommendation probability prediction, the first parameter and the second parameter are conveniently queried through the parameter mapping table, and the first parameter and the second parameter are determined at a coarse-grained level without distinguishing specific resources. Secondly, the user data and the resource value of the first resource are input into the model, and the transaction growth prediction is performed through the model to obtain a third parameter for each resource of the user in the first resource. The third parameter is used to represent the growth rate of the transaction amount. In this way, the resource is specific to the first resource, that is, the third parameter is determined at a fine-grained level. Finally, based on at least one of the first parameter and the second parameter, the third parameter and the resource value of the first resource, the fourth parameter corresponding to each resource in the first resource is calculated. The fourth parameter is used to represent the transaction profit of the transaction party after the user uses each resource to trade with the transaction party. The first resource with the largest transaction profit is selected as the recommended information among the first resources. In this way, the recommended information recommended to the user is determined by combining at least one of the first parameter and the second parameter at the coarse-grained level, the third parameter at the fine-grained level and the resource value of the first resource, thereby improving the accuracy of information recommendation.

[0058] The following uses the application of a recommendation method provided by this embodiment in a rights resource recommendation scenario as an example to further illustrate the recommendation method provided by this embodiment. Figure 4 ,The recommendation method applied to the equity resource recommendation scenario specifically includes the following steps.

[0059] Step S402 : predicting the recommendation probability of the equity resource based on the user data to obtain a first probability.

[0060] Step S404: determine the probability interval in which the first probability is located, and search the parameter mapping table for the first parameter and the second parameter corresponding to the probability interval.

[0061] Optionally, the first parameter is used to indicate the transaction amount, and the second parameter is used to indicate the resource usage rate after the resource recommendation.

[0062] Step S406: Input the user data and the resource value of each equity resource in the first equity resource into the model to perform transaction growth prediction, and obtain the third parameter of the user for each equity resource.

[0063] Optionally, the third parameter is used to indicate the growth rate of the transaction amount.

[0064] Step S408 : Calculate a fourth parameter corresponding to each equity resource according to the first parameter, the second parameter, the third parameter, and the resource value of the first equity resource.

[0065] Optionally, the fourth parameter is used to represent the transaction profit amount of the transaction party after the user uses each equity resource to trade with the transaction party.

[0066] Step S410 : Selecting the first equity resource with the largest transaction profit from the first equity resources as the second equity resource, and recommending the second equity resource to the user.

[0067] It should be noted that any one of steps S402 to S410 or any combination of multiple steps can be combined with any one of steps S102 to S106 to form a new implementation method according to the needs of implementation deployment; in addition, according to the needs of actual deployment, any one or multiple technical features can be selected from steps S402 to S410 and combined with any one or multiple technical features provided by steps S102 to S106 to form a new implementation method; or, any one or multiple technical features in steps S402 to S410 can be replaced with any one or multiple technical features provided by steps S102 to S106 to form a new implementation method according to the needs of actual deployment, which will not be repeated here.

[0068] A recommended device embodiment provided in this specification is as follows: In the above embodiment, a recommendation method is provided, and correspondingly, a recommendation device is also provided, which will be described below with reference to the accompanying drawings.

[0069] Reference Figure 5 , which shows a schematic diagram of a recommended device provided by this embodiment.

[0070] Since the device embodiment corresponds to the method embodiment, the description is relatively simple. For the relevant parts, please refer to the corresponding description of the method embodiment provided above. The device embodiment described below is only illustrative.

[0071] This embodiment provides a recommendation device, which includes: Probability prediction module 502, configured to predict resource recommendation probabilities based on user data, obtain a first probability, and query a first parameter and a second parameter corresponding to the first probability, where the first parameter represents the transaction amount and the second parameter represents the resource usage rate after the resource recommendation; Prediction module 504, configured to input the user data and the resource value of the first resource into a model to perform transaction growth prediction, thereby obtaining a third parameter, wherein the third parameter is used to represent a growth rate of the transaction amount; The information determination module 506 is configured to determine recommendation information in the first resource according to at least one of the first parameter and the second parameter, the third parameter, and the resource value.

[0072] In one embodiment, determining recommendation information in the first resource based on at least one of the first parameter and the second parameter, the third parameter, and the resource value is implemented as follows: Calculating a fourth parameter corresponding to each of the resources based on the at least one parameter, the third parameter, and the resource value, wherein the fourth parameter is used to represent a transaction profit amount of the transaction party after the user uses the resource to trade with the transaction party; The first resource with the largest transaction profit is selected from the first resources as the recommendation information.

[0073] In one embodiment, querying the first parameter and the second parameter corresponding to the first probability is implemented in the following manner: determining a probability interval in which the first probability lies; The first parameter and the second parameter corresponding to the probability interval are searched in a parameter mapping table.

[0074] In one embodiment, the first parameter and the second parameter corresponding to any probability interval in the parameter mapping table are calculated as follows: Calculate the average transaction amount of the control samples based on the sample number and total transaction amount of the control samples of the sample group corresponding to the any probability interval as the first parameter corresponding to the any probability interval; The resource usage rate after resource recommendation is calculated according to the number of samples in the sample group and the number of samples using resources in the sample group, and is used as the second parameter corresponding to any probability interval.

[0075] In one embodiment, the model is obtained by performing model training based on the first training sample; The first training sample is generated in the following manner: Grouping the second training samples according to the resource value of the second resource included in the second training samples to obtain sample groups, and determining the transaction amount and transaction rate of the sample groups; Resources are selected from the second resources according to the transaction amounts and transaction rates of the sample groups, and resource values ​​of the second resources included in the second training sample are updated based on the selected resources to obtain the first training sample.

[0076] In one embodiment, selecting a resource from the second resource based on the transaction amount and transaction rate of the sample group is implemented as follows: Calculating a third parameter and a fifth parameter of the sample group based on the transaction amount and transaction rate of the sample group and the transaction amount and transaction rate of the control sample, wherein the fifth parameter is used to represent the growth value of the transaction rate; A resource is selected from the second resources based on the third parameter and the fifth parameter of the sample grouping.

[0077] In one embodiment, selecting a resource from the second resource based on the third parameter and the fifth parameter of the sample group is implemented as follows: Selecting a sample group from the sample group whose third parameter and fifth parameter are both greater than a preset value; Among the second resources of the sample group obtained by screening, the second resource with the smallest resource value is selected as the selected resource.

[0078] In the recommendation device provided in this embodiment, resource recommendation probability prediction is performed based on user data, and the first parameter and second parameter corresponding to the predicted first probability are queried. Since the resource recommendation probability prediction is predicted from the entire resource dimension, the first parameter and second parameter corresponding to the predicted first probability are parameters determined from a coarse-grained level. On this basis, the user data and the resource value of the first resource are input into the model, and the transaction growth prediction is performed through the model to obtain the third parameter. In this way, the third parameter is determined from a fine-grained level by specifying the resource to the first resource. Finally, the recommendation information is determined in the first resource based on at least one of the first parameter and the second parameter, the third parameter and the resource value of the first resource. In this way, the comprehensiveness of the recommended information is improved by at least one of the first parameter and the second parameter at the coarse-grained level, the third parameter at the fine-grained level, and the resource value of the first resource, thereby improving the accuracy of the information recommendation.

[0079] An embodiment of a computer device provided in this specification is as follows: Corresponding to the recommendation method described above, based on the same technical concept, an embodiment of the present application further provides a computer device, which is used to execute the recommendation method provided above. Figure 6A schematic diagram of the structure of a computer device provided in an embodiment of the present application.

[0080] This embodiment provides a computer device, including: like Figure 6 As shown, computer devices can vary significantly depending on their configuration or performance. They may include one or more processors 601 and memory 602. Memory 602 may store one or more applications or data. Memory 602 may be either ephemeral or persistent. Applications stored in memory 602 may include one or more modules (not shown), each of which may include a series of computer-executable instructions within the computer device. Furthermore, processor 601 may be configured to communicate with memory 602 to execute the series of computer-executable instructions within memory 602 on the computer device. The computer device may also include one or more power supplies 603, one or more wired or wireless network interfaces 604, one or more input / output interfaces 605, one or more keyboards 606, and the like.

[0081] In a specific embodiment, a computer device includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions in the computer device, and the one or more programs are configured to be executed by one or more processors, including computer-executable instructions for performing the following: Predicting the probability of resource recommendation based on user data to obtain a first probability, and querying a first parameter and a second parameter corresponding to the first probability, where the first parameter is used to represent the transaction amount, and the second parameter is used to represent the utilization rate of the resource after the resource recommendation; Inputting the user data and the resource value of the first resource into a model to perform transaction growth prediction, thereby obtaining a third parameter, wherein the third parameter is used to represent the growth rate of the transaction amount; Recommendation information is determined in the first resource according to at least one of the first parameter and the second parameter, the third parameter, and the resource value.

[0082] In one embodiment, determining recommendation information in the first resource based on at least one of the first parameter and the second parameter, the third parameter, and the resource value is implemented as follows: Calculating a fourth parameter corresponding to each of the resources based on the at least one parameter, the third parameter, and the resource value, wherein the fourth parameter is used to represent a transaction profit amount of the transaction party after the user uses the resource to trade with the transaction party; The first resource with the largest transaction profit is selected from the first resources as the recommendation information.

[0083] In one embodiment, querying the first parameter and the second parameter corresponding to the first probability is implemented in the following manner: determining a probability interval in which the first probability lies; The first parameter and the second parameter corresponding to the probability interval are searched in a parameter mapping table.

[0084] In one embodiment, the first parameter and the second parameter corresponding to any probability interval in the parameter mapping table are calculated as follows: Calculate the average transaction amount of the control samples based on the sample number and total transaction amount of the control samples of the sample group corresponding to the any probability interval as the first parameter corresponding to the any probability interval; The resource usage rate after resource recommendation is calculated according to the number of samples in the sample group and the number of samples using resources in the sample group, and is used as the second parameter corresponding to any probability interval.

[0085] In one embodiment, the model is obtained by performing model training based on the first training sample; The first training sample is generated in the following manner: Grouping the second training samples according to the resource value of the second resource included in the second training samples to obtain sample groups, and determining the transaction amount and transaction rate of the sample groups; Resources are selected from the second resources according to the transaction amounts and transaction rates of the sample groups, and resource values ​​of the second resources included in the second training sample are updated based on the selected resources to obtain the first training sample.

[0086] In one embodiment, selecting a resource from the second resource based on the transaction amount and transaction rate of the sample group is implemented as follows: Calculating a third parameter and a fifth parameter of the sample group based on the transaction amount and transaction rate of the sample group and the transaction amount and transaction rate of the control sample, wherein the fifth parameter is used to represent the growth value of the transaction rate; A resource is selected from the second resources based on the third parameter and the fifth parameter of the sample grouping.

[0087] In one embodiment, selecting a resource from the second resource based on the third parameter and the fifth parameter of the sample group is implemented as follows: Selecting a sample group from the sample group whose third parameter and fifth parameter are both greater than a preset value; Among the second resources of the sample group obtained by screening, the second resource with the smallest resource value is selected as the selected resource.

[0088] In the computer device provided in this embodiment, resource recommendation probability prediction is performed based on user data, and the first parameter and second parameter corresponding to the predicted first probability are queried. Since the resource recommendation probability prediction is predicted from the entire resource dimension, the first parameter and second parameter corresponding to the predicted first probability are parameters determined from a coarse-grained level. On this basis, the user data and the resource value of the first resource are input into the model, and transaction growth prediction is performed through the model to obtain a third parameter. By specifying the resource to the first resource, the third parameter is determined from a fine-grained level. Finally, based on at least one of the first parameter and the second parameter, the third parameter and the resource value of the first resource, the recommended information is determined in the first resource. By using at least one of the first parameter and the second parameter at the coarse-grained level, the third parameter at the fine-grained level, and the resource value of the first resource, the comprehensiveness of the recommended information is improved, thereby improving the accuracy of the information recommendation.

[0089] An embodiment of a computer-readable storage medium provided in this specification is as follows: Corresponding to the recommendation method described above, based on the same technical concept, an embodiment of the present application also provides a computer-readable storage medium.

[0090] The computer-readable storage medium provided in this embodiment is used to store computer-executable instructions. When the computer-executable instructions are executed by a processor, the following process is implemented: Predicting the probability of resource recommendation based on user data to obtain a first probability, and querying a first parameter and a second parameter corresponding to the first probability, where the first parameter is used to represent the transaction amount, and the second parameter is used to represent the utilization rate of the resource after the resource recommendation; Inputting the user data and the resource value of the first resource into a model to perform transaction growth prediction, thereby obtaining a third parameter, wherein the third parameter is used to represent the growth rate of the transaction amount; Recommendation information is determined in the first resource according to at least one of the first parameter and the second parameter, the third parameter, and the resource value.

[0091] In the computer-readable storage medium provided in this embodiment, resource recommendation probability prediction is performed based on user data, and the first parameter and second parameter corresponding to the predicted first probability are queried. Since the resource recommendation probability prediction is predicted from the entire resource dimension, the first parameter and second parameter corresponding to the predicted first probability are parameters determined from a coarse-grained level. On this basis, the user data and the resource value of the first resource are input into the model, and the transaction growth prediction is performed through the model to obtain the third parameter. In this way, the third parameter is determined from a fine-grained level by specifying the resource to the first resource. Finally, the recommendation information is determined in the first resource based on at least one of the first parameter and the second parameter, the third parameter and the resource value of the first resource. In this way, the comprehensiveness of the recommended information is improved by at least one of the first parameter and the second parameter at the coarse-grained level, the third parameter at the fine-grained level, and the resource value of the first resource, thereby improving the accuracy of the information recommendation.

[0092] It should be noted that the embodiment of a computer-readable storage medium in this specification and the embodiment of a recommended method in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the corresponding method mentioned above, and the repeated parts will not be repeated.

[0093] An embodiment of a computer program product provided in this specification is as follows: Corresponding to the recommendation method described above, based on the same technical concept, an embodiment of the present application also provides a computer program product.

[0094] The computer program product provided in this embodiment includes a computer program. When the computer program is executed by a processor, the following process is implemented: Predicting the probability of resource recommendation based on user data to obtain a first probability, and querying a first parameter and a second parameter corresponding to the first probability, where the first parameter is used to represent the transaction amount, and the second parameter is used to represent the utilization rate of the resource after the resource recommendation; Inputting the user data and the resource value of the first resource into a model to perform transaction growth prediction, thereby obtaining a third parameter, wherein the third parameter is used to represent the growth rate of the transaction amount; Recommendation information is determined in the first resource according to at least one of the first parameter and the second parameter, the third parameter, and the resource value.

[0095] In the computer program product provided in this embodiment, resource recommendation probability prediction is performed based on user data, and the first parameter and second parameter corresponding to the predicted first probability are queried. Since the resource recommendation probability prediction is predicted from the entire resource dimension, the first parameter and second parameter corresponding to the predicted first probability are parameters determined from a coarse-grained level. On this basis, the user data and the resource value of the first resource are input into the model, and transaction growth prediction is performed through the model to obtain a third parameter. By specifying the resource to the first resource, the third parameter is determined from a fine-grained level. Finally, based on at least one of the first parameter and the second parameter, the third parameter and the resource value of the first resource, the recommended information is determined in the first resource. By using at least one of the first parameter and the second parameter at the coarse-grained level, the third parameter at the fine-grained level, and the resource value of the first resource, the comprehensiveness of the recommended information is improved, thereby improving the accuracy of the information recommendation.

[0096] It should be noted that the embodiment of a computer program product in this specification and the embodiment of a recommended method in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the corresponding method mentioned above, and the repeated parts will not be repeated.

[0097] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0098] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the embodiments of the present application may take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware. Furthermore, this specification may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0099] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable test equipment to produce a machine, so that the instructions executed by the processor of the computer or other programmable test equipment generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0100] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable test equipment to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device that implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0101] These computer program instructions may also be loaded onto a computer or other programmable test device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide for implementing the process described in the flow. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0102] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0103] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0104] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. 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 RAM (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 cassettes, magnetic disk storage 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.

[0105] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0106] The embodiments of the present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the present specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0107] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0108] The foregoing description is merely an example of the present invention and is not intended to limit the present invention. Persons skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be included within the scope of the claims herein.

Claims

1. A recommendation method, characterized in that: The method comprises: Predicting the probability of resource recommendation based on user data to obtain a first probability, and querying a first parameter and a second parameter corresponding to the first probability, where the first parameter is used to represent the transaction amount, and the second parameter is used to represent the utilization rate of the resource after the resource recommendation; Inputting the user data and the resource value of the first resource into a model to perform transaction growth prediction, thereby obtaining a third parameter, wherein the third parameter is used to represent the growth rate of the transaction amount; Recommendation information is determined in the first resource according to at least one of the first parameter and the second parameter, the third parameter, and the resource value.

2. The method according to claim 1, characterized in that The determining the recommendation information in the first resource according to at least one of the first parameter and the second parameter, the third parameter, and the resource value includes: Calculating a fourth parameter corresponding to each of the resources based on the at least one parameter, the third parameter, and the resource value, wherein the fourth parameter is used to represent a transaction profit amount of the transaction party after the user uses the resource to trade with the transaction party; The first resource with the largest transaction profit is selected from the first resources as the recommendation information.

3. The method according to claim 1, characterized in that The querying of the first parameter and the second parameter corresponding to the first probability includes: determining a probability interval in which the first probability lies; The first parameter and the second parameter corresponding to the probability interval are searched in a parameter mapping table.

4. The method according to claim 3, characterized in that The first parameter and the second parameter corresponding to any probability interval in the parameter mapping table are calculated as follows: Calculate the average transaction amount of the control samples based on the sample number and total transaction amount of the control samples of the sample group corresponding to the any probability interval as the first parameter corresponding to the any probability interval; The resource usage rate after resource recommendation is calculated according to the number of samples in the sample group and the number of samples using resources in the sample group, and is used as the second parameter corresponding to any probability interval.

5. The method according to claim 1, wherein The model is obtained by performing model training based on the first training sample; The first training sample is generated in the following manner: Grouping the second training samples according to the resource value of the second resource included in the second training samples to obtain sample groups, and determining the transaction amount and transaction rate of the sample groups; Resources are selected from the second resources according to the transaction amounts and transaction rates of the sample groups, and resource values ​​of the second resources included in the second training sample are updated based on the selected resources to obtain the first training sample.

6. The method according to claim 5, characterized in that The selecting of resources from the second resources according to the transaction amounts and transaction rates of the sample groups includes: Calculating a third parameter and a fifth parameter of the sample group based on the transaction amount and transaction rate of the sample group and the transaction amount and transaction rate of the control sample, wherein the fifth parameter is used to represent the growth value of the transaction rate; A resource is selected from the second resources based on the third parameter and the fifth parameter of the sample grouping.

7. The method according to claim 6, characterized in that The selecting a resource from the second resource based on the third parameter and the fifth parameter of the sample grouping includes: Selecting a sample group from the sample group whose third parameter and fifth parameter are both greater than a preset value; Among the second resources of the sample group obtained by screening, the second resource with the smallest resource value is selected as the selected resource.

8. A computer device, characterized in that: The device comprises: A processor; and a memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the recommendation method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store computer-executable instructions, and when the computer-executable instructions are executed by a processor, the recommendation method according to any one of claims 1 to 7 is implemented.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the recommendation method according to any one of claims 1 to 7 when the computer program is executed by a processor.

Images