Model generation method, user value limit adjustment information generation method and device

Abstract

The embodiment of the invention discloses a model generation method, a user value limit adjustment information generation method and a user value limit adjustment information generation device. A specific embodiment of the method comprises the steps of obtaining a target user value information set; generating a user value limit state transition information generation model according to the user value limit state information sequence set and the value limit adjustment action information sequence set; generating a user value reward information generation model according to a user value limit state information sequence set, the value limit adjustment action information sequence set and the value acquisition information sequence set; and generating a user value limit adjustment information generation model according to the user value limit state transition information generation model and the user value reward information generation model. The implementation mode is related to artificial intelligence, and the user value limit adjustment information can be accurately generated by utilizing the user value limit adjustment information generation model.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a model generation method, a user value quota adjustment information generation method and a device. Background Art

[0002] At present, the estimation of user value quota has become an important development direction in the current related fields. For high-risk users, the user value quota will be reduced. For low-risk users, the user value quota will be increased. For the prediction of user value quota, the method usually adopted is: through the Markov decision process, the user value quota of the user in the current period is predicted.

[0003] However, the inventors have found that when the above method is used to predict the user value amount, the following technical problems often occur:

[0004] The factors that may cause user default are not fully considered, resulting in inaccurate predictions of user value amounts.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0006] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0007] Some embodiments of the present disclosure propose a model generation method, a user value quota adjustment information generation method and a device.

[0008] In a first aspect, some embodiments of the present disclosure provide a model generation method, comprising: obtaining a target user value information set, wherein the target user value information comprises: a user value quota status information sequence, a value quota adjustment action information sequence and a value acquisition information sequence for the user in each period; generating a user value quota status transfer information generation model based on the obtained user value quota status information sequence set and the value quota adjustment action information sequence set; generating a user value reward information generation model based on the above user value quota status information sequence set, the above value quota adjustment action information sequence set and the obtained value acquisition information sequence set; generating a user value quota adjustment information generation model based on the above user value quota status transfer information generation model and the user value reward information generation model.

[0009] Optionally, the user value quota status information in the above-mentioned user value quota status information sequence set includes: user value behavior scoring information; and the above-mentioned generating a user value quota status transfer information generation model based on the obtained user value quota status information sequence set and the value quota adjustment action information sequence set includes: removing the user value quota status information whose corresponding user value behavior scoring information meets the first scoring information condition from the above-mentioned user value quota status information sequence set to obtain a removed user value quota status information sequence set; selecting the user value quota status information from the above-mentioned removed user value quota status information sequence set as the target user value quota status information, and executing The first training step is as follows: using the above-mentioned target user value quota status information and the corresponding value quota adjustment action information as input data, and the user value quota status information of the next period corresponding to the above-mentioned target user value quota status information as a label, and using the initial user value quota status transfer information generation sub-model based on the second rating information condition to generate user value quota status transfer loss information; according to the above-mentioned user value quota status transfer loss information, determining whether the initial user value quota status transfer information generation sub-model has been trained; in response to the completion of the training, determining the initial user value quota status transfer information generation sub-model as the first user value quota status transfer information generation sub-model.

[0010] Optionally, the above-mentioned generation of a user value limit state transfer information generation model based on the obtained user value limit state information sequence set and value limit adjustment action information sequence set also includes: generating a second user value limit state transfer information generation sub-model based on the above-mentioned first rating information condition; generating a user value limit state transfer information generation model based on the above-mentioned first user value limit state transfer information generation sub-model and the above-mentioned second user value limit state transfer information generation sub-model.

[0011] Optionally, the user value quota status information in the above-mentioned user value quota status information sequence set includes: user value behavior scoring information; and the above-mentioned generation of the user value reward information generation model based on the above-mentioned user value quota status information sequence set, the above-mentioned value quota adjustment action information sequence set and the obtained value acquisition information sequence set includes: removing the user value quota status information whose corresponding user value behavior scoring information meets the first scoring information condition from the above-mentioned user value quota status information sequence set to obtain the removed user value quota status information sequence set; selecting the user value quota status from the above-mentioned removed user value quota status information sequence set. Information is used as the target user value quota status information, and the following second training step is performed: the target user value quota status information and the corresponding value quota adjustment action information are used as input data, the value acquisition information corresponding to the above target user value quota status information is used as a label, and the initial user value reward information generation sub-model under the second scoring information condition is used to generate user value reward loss information; according to the above user value reward loss information, it is determined whether the initial user value reward information generation sub-model has been trained; in response to the completion of the training, the initial user value reward information generation sub-model is determined as the first user value reward information generation sub-model.

[0012] Optionally, the above-mentioned generation of the user value reward information generation model based on the above-mentioned user value quota status information sequence set, the above-mentioned value quota adjustment action information sequence set and the obtained value acquisition information sequence set also includes: generating a second user value reward information generation sub-model based on the above-mentioned first rating information condition; generating a user value reward information generation model based on the above-mentioned first user value reward information generation sub-model and the above-mentioned second user value reward information generation sub-model.

[0013] Optionally, the above-mentioned generation of a user value quota adjustment information generation model based on the above-mentioned user value quota state transfer information generation model and the user value reward information generation model includes: generating a quota value acquisition model based on user value reward information and user value quota transfer information; generating a user value quota adjustment information generation model based on the above-mentioned user value quota state transfer information generation model, the above-mentioned user value reward information generation model and the above-mentioned quota value acquisition model.

[0014] In a second aspect, some embodiments of the present disclosure provide a model generation device, comprising: a first acquisition unit, configured to acquire a target user value information set, wherein the target user value information includes: a user value quota status information sequence for the user in each period, a value quota adjustment action information sequence and a value acquisition information sequence for each period; a first generation unit, configured to generate a user value quota status transition information generation model based on the obtained user value quota status information sequence set and the value quota adjustment action information sequence set; a second generation unit, configured to generate a user value reward information generation model based on the above-mentioned user value quota status information sequence set, the above-mentioned value quota adjustment action information sequence set and the obtained value acquisition information sequence set; and a third generation unit, configured to generate a user value quota adjustment information generation model based on the above-mentioned user value quota status information sequence set, the above-mentioned value quota adjustment action information sequence set and the obtained value acquisition information sequence set.

[0015] Optionally, the user value quota status information in the above-mentioned user value quota status information sequence set includes: user value behavior scoring information; and the first generating unit can be configured to: remove the user value quota status information whose corresponding user value behavior scoring information meets the first scoring information condition from the above-mentioned user value quota status information sequence set to obtain the removed user value quota status information sequence set; select the user value quota status information from the above-mentioned removed user value quota status information sequence set as the target user value quota status information, and perform the following first training step: use the above-mentioned target user value quota status information and the corresponding value quota adjustment action information as input data, and the user value quota status information of the next period corresponding to the above-mentioned target user value quota status information as a label, and use the initial user value quota status transfer information generation sub-model based on the second scoring information condition to generate user value quota status transfer loss information; determine whether the initial user value quota status transfer information generation sub-model is trained according to the above-mentioned user value quota status transfer loss information; in response to the completion of the training, determine the initial user value quota status transfer information generation sub-model as the first user value quota status transfer information generation sub-model.

[0016] Optionally, the first generating unit may be configured to: generate a second user value quota state transfer information generating sub-model based on the first rating information condition; generate a user value quota state transfer information generating model according to the first user value quota state transfer information generating sub-model and the second user value quota state transfer information generating sub-model.

[0017] Optionally, the user value quota status information in the above-mentioned user value quota status information sequence set includes: user value behavior scoring information; and the second generating unit can be configured to: remove the user value quota status information whose corresponding user value behavior scoring information meets the first scoring information condition from the above-mentioned user value quota status information sequence set to obtain the removed user value quota status information sequence set; select the user value quota status information from the above-mentioned removed user value quota status information sequence set as the target user value quota status information, and perform the following second training step: use the target user value quota status information and the corresponding value quota adjustment action information as input data, the value acquisition information corresponding to the above-mentioned target user value quota status information as a label, use the initial user value reward information under the second scoring information condition to generate a sub-model to generate user value reward loss information; determine whether the initial user value reward information generation sub-model is trained according to the above-mentioned user value reward loss information; in response to the completion of the training, determine the initial user value reward information generation sub-model as the first user value reward information generation sub-model.

[0018] Optionally, the second generation unit can be configured to: generate a second user value reward information generation sub-model based on the above-mentioned first rating information condition; generate a user value reward information generation model according to the above-mentioned first user value reward information generation sub-model and the above-mentioned second user value reward information generation sub-model.

[0019] Optionally, the third generation unit can be configured to: generate a quota value acquisition model based on user value reward information and user value quota transfer information; generate a user value quota adjustment information generation model based on the above-mentioned user value quota state transfer information generation model, the above-mentioned user value reward information generation model and the above-mentioned quota value acquisition model.

[0020] In a third aspect, some embodiments of the present disclosure provide a method for generating user value limit adjustment information, including: obtaining current user value limit status information for a target user; using a user value limit adjustment information generation model to generate user value limit adjustment information for the above-mentioned current user value limit status information, wherein the above-mentioned user value limit adjustment information generation model is generated based on the method associated with the first aspect.

[0021] In a fourth aspect, some embodiments of the present disclosure provide a user value limit adjustment information generating device, comprising: a second acquisition unit, configured to acquire current user value limit status information for a target user; a fourth generation unit, configured to generate user value limit adjustment information for the above-mentioned current user value limit status information by using a user value limit adjustment information generation model, wherein the above-mentioned user value limit adjustment information generation model is generated based on the method associated with the first aspect.

[0022] In a fifth aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any one of the implementation methods in the first and third aspects.

[0023] In a sixth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any one of the implementation modes of the first aspect and the third aspect is implemented.

[0024] In a seventh aspect, some embodiments of the present disclosure provide a computer program product, including a computer program, which, when executed by a processor, implements the method described in any one of the implementation modes of the first and third aspects above.

[0025] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the model generation method of some embodiments of the present disclosure, the user value quota adjustment information can be accurately generated by using the user value quota adjustment information generation model. Specifically, the reason why the relevant user value quota adjustment information is not accurate enough is that the factors of user default are not fully considered, resulting in the prediction of the user value quota is not accurate enough, that is, the corresponding user value quota adjustment information is not accurate enough. Based on this, the model generation method of some embodiments of the present disclosure first obtains a target user value information set, wherein the target user value information includes: a user value quota state information sequence for the user in each period, a value quota adjustment action information sequence, and a value acquisition information sequence for each period. Here, the obtained target user value information set can be used as a training data set for the subsequent generation of a user value quota state transition information generation model and a user value reward information generation model. Then, according to the obtained user value quota state information sequence set and the value quota adjustment action information sequence set, a user value quota state transition information generation model is generated. Here, in the process of generating the user value quota state transition information generation model, the development and transformation of the value quota adjustment action information is fully considered, so that the full relationship between the transformation of the value quota adjustment action information and the user's default can be considered. Therefore, based on the transformation of the value quota adjustment action information, a user value quota state transition information generation model that fully considers the relevant conditions of the factors of the user's default is generated. So that the user value quota state transition information generation model can be used to accurately generate the user's user value quota state transition information in a certain period. Then, according to the above-mentioned user value quota state information sequence set, the above-mentioned value quota adjustment action information sequence set and the obtained value acquisition information sequence set, a user value reward information generation model is generated. Similarly, in the process of generating the user value reward information generation model, the development and transformation of the value quota adjustment action information is fully considered, so that the full relationship between the transformation of the value quota adjustment action information and the user's default can be considered. Therefore, based on the transformation of the value quota adjustment action information, a user value reward information generation model that fully considers the relevant conditions of the factors of the user's default is generated. So that the user value reward information generation model can be used to accurately generate the user's user value reward information in a certain period. Finally, through the above-mentioned user value quota state transfer information generation model and user value reward information generation model, a user value quota adjustment information generation model can be generated.In summary, in the generation process of the user value reward information generation model and the user value quota state transfer information generation model, the development and transformation of the value quota adjustment action information is taken into account, so that the full relationship between the transformation of the value quota adjustment action information and the user's default can be taken into account, which not only makes the user value reward information generation model and the user value quota state transfer information generation model more accurate, but also enables the subsequent use of the user value quota adjustment information generation model to accurately generate user value quota adjustment information. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0027] Figure 1 is a schematic diagram of an application scenario of the model generation method according to some embodiments of the present disclosure;

[0028] Figure 2 is a flow chart of some embodiments of the model generation method according to the present disclosure;

[0029] Figure 3 are flow charts of other embodiments of the model generation method according to the present disclosure;

[0030] Figure 4 is a flow chart of some embodiments of the method for generating user value quota adjustment information according to the present disclosure;

[0031] Figure 5 is a schematic diagram of the structure of some embodiments of the model generation device according to the present disclosure;

[0032] Figure 6 It is a structural schematic diagram of some embodiments of the user value quota adjustment information generating device according to the present disclosure;

[0033] Figure 7 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0034] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0035] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0036] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0037] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0038] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0039] With regard to the collection, storage, and use of user personal information (such as target user value information) involved in this disclosure, before performing the corresponding operations, the relevant organizations or individuals shall fulfill their obligations, including conducting personal information security impact assessments, fulfilling the obligation to inform the personal information subject, and obtaining the authorization and consent of the personal information subject in advance.

[0040] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0041] Figure 1 It is a schematic diagram of an application scenario of the model generation method according to some embodiments of the present disclosure.

[0042] exist Figure 1 In the application scenario, first, the electronic device 101 can obtain the target user value information set 102. Among them, the target user value information includes: the user value quota state information sequence for the user in each period, the value quota adjustment action information sequence and the value acquisition information sequence for each period. Then, the electronic device 101 can generate a user value quota state transition information generation model 106 according to the obtained user value quota state information sequence set 103 and the value quota adjustment action information sequence set 104. Next, the electronic device 101 can generate a user value reward information generation model 107 according to the above-mentioned user value quota state information sequence set 103, the above-mentioned value quota adjustment action information sequence set 104 and the obtained value acquisition information sequence set 105. Finally, the electronic device 101 can generate a user value quota adjustment information generation model 108 according to the above-mentioned user value quota state transition information generation model 106 and the user value reward information generation model 107.

[0043] It should be noted that the electronic device 101 can be hardware or software. When the electronic device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or it can be implemented as a single server or a single terminal device. When the electronic device is embodied as software, it can be installed in the hardware devices listed above. It can be implemented as multiple software or software modules for providing distributed services, for example, or it can be implemented as a single software or software module. No specific limitation is made here.

[0044] It should be understood that Figure 1 The number of electronic devices in the embodiment is only for illustration. Any number of electronic devices may be provided according to implementation requirements.

[0045] Continue to refer Figure 2 , shows a process 200 of some embodiments of the model generation method according to the present disclosure. The model generation method comprises the following steps:

[0046] Step 201, obtaining a target user value information set.

[0047] In some embodiments, the execution entity of the above model generation method (for example Figure 1The electronic device 101 shown can obtain the target user value information set through a wired connection method or a wireless connection method. Among them, the target user value information includes: a user value quota state information sequence for the user in each period, a value quota adjustment action information sequence and a value acquisition information sequence for each period. Among them, the target user value information in the target user value information set and the user information in the user information set have a one-to-one correspondence. That is, the target user value information can be the historical value transformation data of the corresponding user. In practice, the above-mentioned target user value information can be the historical loan quota data of the user. The user value quota state information sequence can be a sequence for the corresponding user that characterizes the user value quota state transformation. Specifically, each period can be each historical period. The time period corresponding to each period can be pre-set. For example, each historical period can be each day in the previous month. That is, the duration corresponding to each period is one day. There is a one-to-one correspondence between the period in each period and the user value quota state information in the user value quota state information sequence. That is, the user value quota state information can be the user value quota state corresponding to a certain period. In practice, the user value quota state information can be the quota information that the user can borrow. Correspondingly, the user value limit state information sequence of each period can represent the change of the amount of loan that the user can borrow in the historical time period. For example, each period is {January 1, January 2, January 3, January 4, January 5, January 6, January 7}. The corresponding user value limit state information sequence is {"10,000", "10,000", "15,000", "20,000", "30,000", "40,000", "100,000"}. The value limit adjustment action information in the value limit adjustment action information sequence has a one-to-one correspondence with the period in each period. The value limit adjustment action information can represent the amount adjustment situation in the current period. That is, the value limit adjustment action information can be the action information of the value limit adjustment action to be executed in the current period. The value limit adjustment action can be an action to adjust the value limit. In practice, the action information can be identification information. Similarly, in actual situations, the value limit adjustment action information can execute the action identification of the loan limit adjustment. For example, the value limit adjustment action information can be an action identification to adjust the current loan limit to 200,000 yuan. The value acquisition information in the value acquisition information sequence may have a one-to-one correspondence with the period in each period. Specifically, the value acquisition information may be the benefit information obtained from the user in the current period. Specifically, the benefit information may be loan interest information. The action adjustment trend corresponding to the value amount adjustment action information may be preset. Specifically, the action adjustment trend corresponding to the value amount adjustment action information may include: a value amount increase trend, and a value amount unchanged trend.

[0048] It should be noted that the target user value information set may be an unbiased data set. The target user value information set may be generated by the following steps:

[0049] In the first step, some user information sets are randomly selected as experimental samples, and the corresponding quotas are randomly adjusted.

[0050] The second step is to count the value information of the selected experimental samples in each performance period to obtain the target user value information set.

[0051] Here, by constructing a target user value information set in the form of an unbiased dataset, the prediction bias caused by selection bias can be encoded.

[0052] Step 202: Generate a user value quota state transition information generation model based on the obtained user value quota state information sequence set and value quota adjustment action information sequence set.

[0053] In some embodiments, the above-mentioned execution subject can generate a user value quota state transfer information generation model based on the obtained user value quota state information sequence set and value quota adjustment action information sequence set. Among them, the user value quota state transfer information generation model can be a model that generates probability information of the user value quota state information in the current period being transferred to the user value quota state information in the next period. In practice, the user value quota state transfer information generation model can be a neural network model. For example, the user value quota state transfer information generation model can be a recurrent neural network (RNN) model.

[0054] As an example, the user value quota state transition information generation model may be a recurrent neural network, with the above target user value quota state information and the corresponding value quota adjustment action information as input data, and the user value quota state information of the next period corresponding to the above target user value quota state information as a label, and the initial user value quota state transition information generation model is trained to obtain the user value quota state transition information generation model. Among them, the initial user value quota state transition information generation model may be a recurrent neural network model that has not yet been trained.

[0055] Step 203: Generate a user value reward information generation model based on the user value quota state information sequence set, the value quota adjustment action information sequence set and the obtained value acquisition information sequence set.

[0056] In some embodiments, the execution subject may generate a user value reward information generation model based on the user value quota status information sequence set, the value quota adjustment action information sequence set and the obtained value acquisition information sequence set. Among them, the user value reward information generation model may be a model for generating user value reward information in the current period. The user value reward information may be the value acquisition information that can be brought under the current user quota status information. That is, the user value reward information may be the environmental reward brought by executing the corresponding value quota adjustment action under the current user quota status information. Specifically, the relevant definition of user value reward information can refer to the reward in the Markov decision process.

[0057] In practice, the user value reward information generation model may be a neural network model. For example, the user value reward information generation model may be a long short-term memory network (LSTM, Long Short-Term Memory) model.

[0058] As an example, if the user value reward information generation model is a long short-term memory network model, the above target user value quota status information and the corresponding value quota adjustment action information are used as input data, and the value acquisition information corresponding to the above target user value quota status information is used as a label, and the initial user value reward information generation model is trained to obtain the user value reward information generation model. Among them, the initial user value reward information generation model can be a long short-term memory network model that has not yet been trained.

[0059] In some optional implementations of some embodiments, the user value limit status information in the above-mentioned user value limit status information sequence set includes: user value behavior scoring information. The user value behavior scoring information can represent the user's credit level. The user value behavior scoring information can include: "no credit identification" and a value behavior score set. "No credit identification" can represent that the corresponding user is a user with no credit. The value behavior score can represent the specific good or bad degree information of the user's credit. Specifically, the higher the value behavior score, the better the credit of the corresponding user.

[0060] Optionally, the generating of the user value reward information generating model according to the user value quota state information sequence set, the value quota adjustment action information sequence set and the obtained value acquisition information sequence set may include the following steps:

[0061] In the first step, the execution subject may remove the user value limit status information whose corresponding user value behavior scoring information satisfies the first scoring information condition from the user value limit status information sequence set, and obtain the removed user value limit status information sequence set. Among them, the first scoring information condition may be that the user value limit status information is "no credit mark" for the corresponding user value behavior scoring information. The user value limit status information is "no credit mark" for the corresponding user value behavior scoring information, which may include two situations: the user value limit status information corresponds to the user value behavior scoring information in the period of "no credit mark", and the user value limit status information corresponds to the user value behavior scoring information in the next period of "no credit mark".

[0062] In the second step, the execution subject may select user value quota status information from the above-mentioned removed user value quota status information sequence set as the target user value quota status information, and execute the following second training step:

[0063] Sub-step 1, the execution subject may use the target user value quota status information and the corresponding value quota adjustment action information as input data, the value acquisition information corresponding to the target user value quota status information as a label, and use the initial user value reward information generation sub-model under the second scoring information condition to generate user value reward loss information. Among them, the second scoring information condition may be that the user value quota status information is not information that the corresponding user value behavior scoring information is "no credit mark", that is, the corresponding user value behavior scoring information is a value behavior score. The user value quota status information is not "no credit mark" for the corresponding user value behavior scoring information, which may include two situations: the user value behavior scoring information in the period corresponding to the user value quota status information is not "no credit mark", and the user value behavior scoring information in the next period corresponding to the user value quota status information is not "no credit mark". The value acquisition information corresponding to the target user value quota status information may be the value acquisition information in the same period as the target user value quota status information. The initial user value reward information generation sub-model under the second scoring information condition may be a model that generates user value reward information on the premise that the corresponding user value behavior scoring information is not "no credit mark". Specifically, the initial user value reward information generation sub-model may be a user value reward information model that has not yet been trained. The user value reward information model may be a machine learning model. In practice, the user value reward information model may be a decision tree model. The user value reward loss information may represent the difference between the user's predicted value acquisition information and the value acquisition information.

[0064] As an example, first, the execution subject may input the target user value quota state information and the corresponding value quota adjustment action information into the initial user value reward information generation sub-model to generate the initial user value reward information. Then, based on the initial user value reward information and the corresponding value acquisition information, the target loss function is used to generate the user value reward loss information.

[0065] Sub-step 2: Determine whether the initial user value reward information generation sub-model has been trained based on the above user value reward loss information.

[0066] As an example, in response to determining that the user value reward loss information is less than the first value, it is determined that the training of the initial user value reward information generation sub-model is completed. In response to determining that the user value reward loss information is greater than or equal to the first value, it is determined that the training of the initial user value reward information generation sub-model is not completed.

[0067] Sub-step 3: in response to the training being completed, determining the initial user value reward information generation sub-model as the first user value reward information generation sub-model.

[0068] Optionally, the steps further include:

[0069] In response to determining that the training is not completed, reselecting user value quota status information from the above-mentioned removed user value quota status information sequence set as the target user value quota status information, and performing the following second training step again. Wherein, the reselected user value quota status information has not been used to train the initial user value reward information generation sub-model.

[0070] Optionally, the method of generating a user value reward information generation model according to the user value quota status information sequence set and the obtained value acquisition information sequence set further includes the following steps:

[0071] The first step is to generate a sub-model for generating the second user value reward information based on the above-mentioned first scoring information condition. The sub-model for generating the second user value reward information based on the above-mentioned first scoring information condition may be a user value reward information generation model under the first scoring information condition. In practice, the sub-model for generating the second user value reward information based on the above-mentioned first scoring information condition may be {for the user value behavior scoring information in the current period is "no credit mark", the user value reward information in the current period is set to "0", for the user value behavior scoring information in the next period of the current period is "no credit mark", the user value reward information in the current period is set to the product value of the quota information in the current period and the target profit coefficient}.

[0072] In the second step, a user value reward information generation model is generated according to the first user value reward information generation sub-model and the second user value reward information generation sub-model.

[0073] As an example, the execution entity may combine the first user value reward information generation sub-model and the second user value reward information generation sub-model to generate a user value reward information generation model.

[0074] Step 204: Generate a user value quota adjustment information generation model based on the user value quota state transfer information generation model and the user value reward information generation model.

[0075] In some embodiments, the above-mentioned execution entity may generate a user value quota adjustment information generation model based on the above-mentioned user value quota state transfer information generation model and the user value reward information generation model. Among them, the user value quota adjustment information generation model may be a model for generating user value quota adjustment information. The user value quota adjustment information may be quota adjustment information of the user value quota. In practice, the user value quota adjustment information may be the quota adjustment of the loan quota. For example, the user value quota adjustment information may be information for increasing the loan quota. The user value quota adjustment information may be information for keeping the loan quota unchanged.

[0076] As an example, the execution entity may combine the user value quota state transition information generation model, the user value reward information generation model and the user value quota adjustment information output layer to generate a user value quota adjustment information generation model. The user value quota adjustment information output layer may be a convolutional neural network. The input of the user value quota adjustment information output layer may be the user value quota state transition information and the user value reward information, and the output is the user value quota adjustment information.

[0077] In some optional implementations of some embodiments, the above-mentioned generation of the user value quota adjustment information generation model according to the above-mentioned user value quota state transfer information generation model and the user value reward information generation model may include the following steps:

[0078] The first step is to generate a quota value acquisition model based on user value reward information and user value quota transfer information. The quota value acquisition model can be a model that generates maximum benefit information based on the transformation of user value quota state information for the next period under the current period. The user value quota adjustment information generation model can be a model generated based on the Bellman equation. The quota value acquisition model is a model that generates maximum benefit information based on user value reward information for the corresponding period and user value quota transfer information for the corresponding period.

[0079] As an example, the quota value acquisition model can be reflected in the following formula:

[0080]

[0081] V(s) may be the profit information brought by the user value quota status information S in period t. V() may be a profit determination function. The details are not repeated here. a may be the value quota adjustment action information corresponding to period t. S may be the user value quota status information corresponding to period t. r t (s, a) can be the user value reward information corresponding to period t. A can be the set of all value spaces of the value quota adjustment action information. γ can be the discount coefficient. s′ can be the user value quota status information of the next period corresponding to period t. p t (s′|s, a) may be the user value quota transfer information corresponding to period t. S may be each period. V(s′) may be the profit information brought by the user value quota state information s′ in the next period of period t.

[0082] The second step is to generate a user value quota adjustment information generation model based on the user value quota state transfer information generation model, the user value reward information generation model and the quota value acquisition model.

[0083] As an example, the execution entity may combine the user value quota state transfer information generation model, the user value reward information generation model and the quota value acquisition model to generate a user value quota adjustment information generation model.

[0084] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the model generation method of some embodiments of the present disclosure, the user value quota adjustment information can be accurately generated by using the user value quota adjustment information generation model. Specifically, the reason why the relevant user value quota adjustment information is not accurate enough is that the factors of user default are not fully considered, resulting in the prediction of the user value quota is not accurate enough, that is, the corresponding user value quota adjustment information is not accurate enough. Based on this, the model generation method of some embodiments of the present disclosure first obtains a target user value information set, wherein the target user value information includes: a user value quota state information sequence for the user in each period, a value quota adjustment action information sequence, and a value acquisition information sequence for each period. Here, the obtained target user value information set can be used as a training data set for the subsequent generation of a user value quota state transition information generation model and a user value reward information generation model. Then, according to the obtained user value quota state information sequence set and the value quota adjustment action information sequence set, a user value quota state transition information generation model is generated. Here, in the process of generating the user value quota state transition information generation model, the development and transformation of the value quota adjustment action information is fully considered, so that the full relationship between the transformation of the value quota adjustment action information and the user's default can be considered. Therefore, based on the transformation of the value quota adjustment action information, a user value quota state transition information generation model that fully considers the relevant conditions of the factors of the user's default is generated. So that the user value quota state transition information generation model can be used to accurately generate the user's user value quota state transition information in a certain period. Then, according to the above-mentioned user value quota state information sequence set, the above-mentioned value quota adjustment action information sequence set and the obtained value acquisition information sequence set, a user value reward information generation model is generated. Similarly, in the process of generating the user value reward information generation model, the development and transformation of the value quota adjustment action information is fully considered, so that the full relationship between the transformation of the value quota adjustment action information and the user's default can be considered. Therefore, based on the transformation of the value quota adjustment action information, a user value reward information generation model that fully considers the relevant conditions of the factors of the user's default is generated. So that the user value reward information generation model can be used to accurately generate the user's user value reward information in a certain period. Finally, through the above-mentioned user value quota state transfer information generation model and user value reward information generation model, a user value quota adjustment information generation model can be generated.In summary, in the generation process of the user value reward information generation model and the user value quota state transfer information generation model, the development and transformation of the value quota adjustment action information is taken into account, so that the full relationship between the transformation of the value quota adjustment action information and the user's default can be taken into account, which not only makes the user value reward information generation model and the user value quota state transfer information generation model more accurate, but also enables the subsequent use of the user value quota adjustment information generation model to accurately generate user value quota adjustment information.

[0085] Further references Figure 3 , shows a process 300 of some other embodiments of the model generation method according to the present disclosure. The model generation method comprises the following steps:

[0086] Step 301, obtaining a target user value information set.

[0087] Step 302: remove the user value limit status information whose corresponding user value behavior rating information meets the first rating information condition from the above user value limit status information sequence set to obtain a removed user value limit status information sequence set.

[0088] In some embodiments, an execution entity (e.g. Figure 1 The electronic device 101 shown can remove the user value limit status information whose corresponding user value behavior rating information meets the first rating information condition from the above user value limit status information sequence set to obtain the removed user value limit status information sequence set. The user value limit status information in the above user value limit status information sequence set includes: user value behavior rating information.

[0089] Step 303, selecting user value quota status information from the above-mentioned user value quota status information sequence set as target user value quota status information, and performing the following first training step:

[0090] Step 3031, taking the above-mentioned target user value limit status information and the corresponding value limit adjustment action information as input data, and the user value limit status information of the next period corresponding to the above-mentioned target user value limit status information as a label, using the initial user value limit status transfer information based on the second scoring information condition to generate a sub-model, and generating user value limit status transfer loss information.

[0091] In some embodiments, the execution subject may use the target user value quota state information and the corresponding value quota adjustment action information as input data, the user value quota state information of the next period corresponding to the target user value quota state information as a label, and generate user value quota state transfer loss information using the initial user value quota state transfer information generation sub-model based on the second rating information condition. Among them, the next period corresponding to the target user value quota state information may be the next period of the period corresponding to the target user value quota state information. The initial user value quota state transfer information generation sub-model based on the second rating information condition may be a model that generates user value quota state transfer information on the premise that the corresponding user value behavior rating information is not "no credit mark". The initial user value quota state transfer information generation sub-model may be a user value quota state transfer information generation sub-model that has not yet been trained. In practice, the user value quota state transfer information generation sub-model may be a decision tree model. The user value quota state transfer loss information may represent the difference information between the user value quota state transfer information and the actual user value quota state transfer information.

[0092] Step 3032: Determine whether the initial user value quota state transfer information generation sub-model has been trained based on the user value quota state transfer loss information.

[0093] In some embodiments, the execution entity may determine whether the initial user value quota state transfer information generation sub-model has been trained based on the user value quota state transfer loss information.

[0094] As an example, in response to determining that the user value quota state transfer loss information is less than the second value, it is determined that the initial user value quota state transfer information generation sub-model training is completed. In response to determining that the user value quota state transfer loss information is greater than or equal to the second value, it is determined that the initial user value quota state transfer information generation sub-model training is not completed.

[0095] Step 3033: In response to the training being completed, the initial user value quota state transfer information generation sub-model is determined as the first user value quota state transfer information generation sub-model.

[0096] In some embodiments, in response to the training being completed, the execution entity may determine the initial user value quota state transfer information generation sub-model as the first user value quota state transfer information generation sub-model.

[0097] Optionally, after step 303, the steps further include: reselecting user value limit status information from the above-mentioned removed user value limit status information sequence set as target user value limit status information, and performing the following first training step again.

[0098] Step 304: Generate a sub-model for generating the second user value quota state transition information based on the first rating information condition.

[0099] In some embodiments, the above-mentioned execution entity may generate a second user value quota state transfer information generation sub-model based on the above-mentioned first rating information condition. Among them, the second user value quota state transfer information generation sub-model based on the above-mentioned first rating information condition may be a user value quota state transfer information model under the first rating information condition. In practice, the second user value quota state transfer information generation sub-model based on the above-mentioned first rating information condition may be {for the user value behavior rating information in the current period is "no credit mark", the user value quota state transfer information in the current period is set to "0", for the user value behavior rating information in the next period of the current period is "no credit mark", the user value quota state transfer information in the current period is set to "0"}.

[0100] Step 305: Generate a user value quota state transfer information generation model based on the first user value quota state transfer information generation sub-model and the second user value quota state transfer information generation sub-model.

[0101] In some embodiments, the execution entity may generate a user value quota state transfer information generation model based on the first user value quota state transfer information generation sub-model and the second user value quota state transfer information generation sub-model.

[0102] As an example, the execution entity may combine the first user value quota state transfer information generation sub-model and the second user value quota state transfer information generation sub-model to generate a user value quota state transfer information generation model.

[0103] Step 306: Generate a user value reward information generation model based on the user value quota state information sequence set, the value quota adjustment action information sequence set and the obtained value acquisition information sequence set.

[0104] Step 307: Generate a user value quota adjustment information generation model based on the user value quota state transfer information generation model and the user value reward information generation model.

[0105] In some embodiments, the specific implementation of steps 301, 306-307 and the technical effects thereof can be referred to in Figure 2 The steps 201, 203-204 in the corresponding embodiment will not be described in detail here.

[0106] from Figure 3 It can be seen that Figure 2 Compared with the description of some corresponding embodiments, Figure 3 The process 300 of the model generation method in some corresponding embodiments describes the generation process of the user value quota state transfer information generation model. Thus, the first user value quota state transfer information generation sub-model and the second user value quota state transfer information generation sub-model can be used to generate more accurate user value quota state transfer information.

[0107] Continue to refer Figure 4 , shows a process 400 of some embodiments of the method for generating user value quota adjustment information according to the present disclosure. The method for generating user value quota adjustment information includes the following steps:

[0108] Step 401, obtaining the current user value quota status information for the target user.

[0109] In some embodiments, the execution subject (e.g., electronic device) of the above-mentioned method for generating user value quota adjustment information can obtain the current user value quota status information for the target user by wired or wireless means. Among them, the current user value quota status information can be the user value quota status information of the user in the corresponding period at the current moment. The specific concept of user value quota status information is not repeated here.

[0110] Step 402: using a user value quota adjustment information generation model, generate user value quota adjustment information for the above current user value quota status information.

[0111] In some embodiments, the execution subject may generate user value quota adjustment information for the current user value quota status information using a user value quota adjustment information generation model. The user value quota adjustment information generation model is generated based on a model generation method. The user value quota adjustment information may be information for adjusting the quota information included in the current user value quota status information.

[0112] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the user value quota adjustment information generation method of some embodiments of the present disclosure using the user value quota adjustment information generation model, the user value quota adjustment information for the above-mentioned current user value quota status information can be accurately generated.

[0113] Further references Figure 5 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a model generation device, and these device embodiments are Figure 2 Corresponding to the method embodiments shown, the model generation device can be specifically applied to various electronic devices.

[0114] like Figure 5 As shown, a model generation device 500 includes: a first acquisition unit 501, a first generation unit 502, a second generation unit 503 and a third generation unit 504. The first acquisition unit 501 is configured to acquire a target user value information set, wherein the target user value information includes: a user value quota state information sequence, a value quota adjustment action information sequence and a value acquisition information sequence for each period for the user; the first generation unit 502 is configured to generate a user value quota state transition information generation model according to the obtained user value quota state information sequence set and the value quota adjustment action information sequence set; the second generation unit 503 is configured to generate a user value reward information generation model according to the above user value quota state information sequence set, the above value quota adjustment action information sequence set and the obtained value acquisition information sequence set; the third generation unit 504 is configured to generate a user value quota adjustment information generation model according to the above user value quota state transition information generation model and the user value reward information generation model.

[0115] In some optional implementations of some embodiments, the user value quota status information in the user value quota status information sequence set includes: user value behavior scoring information; and the first generating unit 502 can be further configured to: remove the user value quota status information whose corresponding user value behavior scoring information meets the first scoring information condition from the above user value quota status information sequence set to obtain the removed user value quota status information sequence set; select the user value quota status information from the above removed user value quota status information sequence set as the target user value quota status information, and perform the following first training step: The user value quota status information and the corresponding value quota adjustment action information are used as input data, the user value quota status information of the next period corresponding to the above-mentioned target user value quota status information is used as a label, and the initial user value quota status transfer information generation sub-model based on the second rating information condition is used to generate user value quota status transfer loss information; according to the above-mentioned user value quota status transfer loss information, it is determined whether the initial user value quota status transfer information generation sub-model has been trained; in response to the completion of the training, the initial user value quota status transfer information generation sub-model is determined as the first user value quota status transfer information generation sub-model.

[0116] In some optional implementations of some embodiments, the first generation unit 502 can be further configured to: generate a second user value limit state transfer information generation sub-model based on the above-mentioned first rating information condition; generate a user value limit state transfer information generation model according to the above-mentioned first user value limit state transfer information generation sub-model and the above-mentioned second user value limit state transfer information generation sub-model.

[0117] In some optional implementations of some embodiments, the user value quota status information in the above-mentioned user value quota status information sequence set includes: user value behavior scoring information; and the second generating unit 503 can be further configured to: remove the user value quota status information whose corresponding user value behavior scoring information meets the first scoring information condition from the above-mentioned user value quota status information sequence set to obtain the removed user value quota status information sequence set; select the user value quota status information from the above-mentioned removed user value quota status information sequence set as the target user value quota status information, and perform the following second training step: use the target user value quota status information and the corresponding value quota adjustment action information as input data, the value acquisition information corresponding to the above-mentioned target user value quota status information as a label, use the initial user value reward information under the second scoring information condition to generate a sub-model to generate user value reward loss information; determine whether the initial user value reward information generation sub-model is trained according to the above-mentioned user value reward loss information; in response to the completion of the training, determine the initial user value reward information generation sub-model as the first user value reward information generation sub-model.

[0118] In some optional implementations of some embodiments, the second generation unit 503 may be further configured to: generate a second user value reward information generation sub-model based on the above-mentioned first rating information condition; generate a user value reward information generation model according to the above-mentioned first user value reward information generation sub-model and the above-mentioned second user value reward information generation sub-model.

[0119] In some optional implementations of some embodiments, the third generation unit 504 can be further configured to: generate a quota value acquisition model based on user value reward information and user value quota transfer information; generate a user value quota adjustment information generation model based on the above-mentioned user value quota state transfer information generation model, the above-mentioned user value reward information generation model and the above-mentioned quota value acquisition model.

[0120] It can be understood that the units recorded in the model generation device 500 are similar to those in the reference Figure 2Therefore, the operations, features and beneficial effects described above for the method are also applicable to the model generation device 500 and the units included therein, and will not be described in detail here.

[0121] Further references Figure 6 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a user value quota adjustment information generating device. These device embodiments are similar to Figure 4 Corresponding to the method embodiments shown, the user value limit adjustment information generating device can be specifically applied to various electronic devices.

[0122] like Figure 6 As shown, a user value quota adjustment information generating device 600 includes: a second acquisition unit 601 and a fourth generation unit 602. The second acquisition unit 601 is configured to acquire the current user value quota status information for the target user; the fourth generation unit 602 is configured to generate user value quota adjustment information for the current user value quota status information using a user value quota adjustment information generation model, wherein the user value quota adjustment information generation model is generated based on a model generation method.

[0123] It is understandable that the various units recorded in the user value quota adjustment information generating device 600 are similar to the reference Figure 4 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the user value quota adjustment information generating device 600 and the units included therein, and will not be described in detail here.

[0124] Reference below Figure 7 , which shows an electronic device (eg, Figure 1 Schematic diagram of the structure of the electronic device 101)700. Figure 7 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0125] like Figure 7 As shown, the electronic device 700 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory 702 or a program loaded from a storage device 708 into a random access memory 703. Various programs and data required for the operation of the electronic device 700 are also stored in the random access memory 703. The processing device 701, the read-only memory 702, and the random access memory 703 are connected to each other via a bus 704. An input / output interface 705 is also connected to the bus 704.

[0126] Typically, the following devices may be connected to the input / output interface 705: an input device 706 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 707 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 708 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 709. The communication device 709 may allow the electronic device 700 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 7 The electronic device 700 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 7 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0127] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from a network through a communication device 709, or installed from a storage device 708, or installed from a read-only memory 702. When the computer program is executed by the processing device 701, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0128] It should be noted that the computer-readable medium in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0129] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0130] The computer-readable medium may be included in the electronic device; or it may exist independently without being installed in the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains a target user value information set, wherein the target user value information includes: a user value quota state information sequence, a value quota adjustment action information sequence, and a value acquisition information sequence for each period for the user; generates a user value quota state transition information generation model based on the obtained user value quota state information sequence set and the value quota adjustment action information sequence set; generates a user value reward information generation model based on the above user value quota state information sequence set, the above value quota adjustment action information sequence set, and the obtained value acquisition information sequence set; generates a user value quota adjustment information generation model based on the above user value quota state information sequence set, the above value quota adjustment action information sequence set, and the obtained value acquisition information sequence set; generates a user value quota adjustment information generation model based on the above user value quota state transition information generation model and the user value reward information generation model. Obtain current user value quota status information for the target user; use a user value quota adjustment information generation model to generate user value quota adjustment information for the above current user value quota status information, wherein the above user value quota adjustment information generation model is generated based on the method associated with the first aspect.

[0131] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0132] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0133] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The units described may also be provided in a processor, for example, may be described as: a processor includes a first acquisition unit, a first generation unit, a second generation unit, and a third generation unit. The names of these units do not, in some cases, constitute limitations on the units themselves, for example, the first acquisition unit may also be described as a "unit for acquiring a target user value information set".

[0134] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0135] Some embodiments of the present disclosure further provide a computer program product, including a computer program, which implements any of the above-mentioned model generation methods or user value quota adjustment information generation methods when executed by a processor.

[0136] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A model generation method, comprising: Obtaining a target user value information set, wherein the target user value information includes: a user value quota state information sequence for the user in each period, a value quota adjustment action information sequence, and a value acquisition information sequence for each period; Generate a user value quota state transition information generation model according to the obtained user value quota state information sequence set and value quota adjustment action information sequence set; Generate a user value reward information generation model according to the user value quota state information sequence set, the value quota adjustment action information sequence set and the obtained value acquisition information sequence set; A user value quota adjustment information generation model is generated according to the user value quota state transfer information generation model and the user value reward information generation model.

2. The method according to claim 1, wherein: The user value quota status information in the user value quota status information sequence set includes: user value behavior score information; and The step of generating a user value quota state transfer information generation model based on the obtained user value quota state information sequence set and value quota adjustment action information sequence set includes: Removing user value quota status information whose corresponding user value behavior rating information satisfies the first rating information condition from the user value quota status information sequence set, to obtain a removed user value quota status information sequence set; Selecting user value quota status information from the removed user value quota status information sequence set as target user value quota status information, and performing the following first training step: The target user value quota state information and the corresponding value quota adjustment action information are used as input data, the user value quota state information of the next period corresponding to the target user value quota state information is used as a label, and the initial user value quota state transition information based on the second rating information condition is used to generate a sub-model to generate user value quota state transition loss information; Determining whether the initial user value quota state transfer information generation sub-model has been trained according to the user value quota state transfer loss information; In response to the training being completed, the initial user value limit state transfer information generation sub-model is determined as the first user value limit state transfer information generation sub-model.

3. The method according to claim 2, wherein: The step of generating a user value quota state transfer information generation model based on the obtained user value quota state information sequence set and value quota adjustment action information sequence set also includes: Generate a second user value quota state transfer information generation sub-model based on the first rating information condition; A user value quota state transfer information generation model is generated according to the first user value quota state transfer information generation sub-model and the second user value quota state transfer information generation sub-model.

4. The method according to claim 1, wherein: The user value quota status information in the user value quota status information sequence set includes: user value behavior score information; and The generating a user value reward information generating model according to the user value quota state information sequence set, the value quota adjustment action information sequence set and the obtained value acquisition information sequence set includes: Removing user value quota status information whose corresponding user value behavior rating information satisfies the first rating information condition from the user value quota status information sequence set, to obtain a removed user value quota status information sequence set; Selecting user value quota status information from the removed user value quota status information sequence set as target user value quota status information, and performing the following second training step: The target user value quota state information and the corresponding value quota adjustment action information are used as input data, the value acquisition information corresponding to the target user value quota state information is used as a label, and the initial user value reward information under the second scoring information condition is used to generate a sub-model to generate user value reward loss information; Determining whether the initial user value reward information generation sub-model has been trained according to the user value reward loss information; In response to the training being completed, the initial user value reward information generation sub-model is determined as the first user value reward information generation sub-model.

5. The method according to claim 4, wherein: The generating a user value reward information generating model according to the user value quota state information sequence set, the value quota adjustment action information sequence set and the obtained value acquisition information sequence set also includes: Generate a second user value reward information generation sub-model based on the first rating information condition; A user value reward information generation model is generated according to the first user value reward information generation sub-model and the second user value reward information generation sub-model.

6. The method according to claim 1, wherein: The generating a user value quota adjustment information generating model according to the user value quota state transfer information generating model and the user value reward information generating model comprises: Generate a quota value acquisition model based on user value reward information and user value quota transfer information; A user value quota adjustment information generation model is generated according to the user value quota state transfer information generation model, the user value reward information generation model and the quota value acquisition model.

7. A method for generating user value quota adjustment information, comprising: Obtain the current user value quota status information for the target user; The user value quota adjustment information generation model is used to generate user value quota adjustment information for the current user value quota status information, wherein the user value quota adjustment information generation model is generated based on the method described in any one of claims 1-6.

8. A model generation device, comprising: A first acquisition unit is configured to acquire a target user value information set, wherein the target user value information includes: a user value quota state information sequence for the user in each period, a value quota adjustment action information sequence, and a value acquisition information sequence for each period; A first generating unit is configured to generate a user value quota state transition information generation model according to the obtained user value quota state information sequence set and value quota adjustment action information sequence set; A second generating unit is configured to generate a user value reward information generating model according to the user value quota state information sequence set, the value quota adjustment action information sequence set and the obtained value acquisition information sequence set; The third generating unit is configured to generate a user value quota adjustment information generating model according to the user value quota state transfer information generating model and the user value reward information generating model.

9. A user value quota adjustment information generating device, comprising: A second acquisition unit is configured to acquire current user value quota status information for the target user; The fourth generating unit is configured to generate user value limit adjustment information for the current user value limit status information by using a user value limit adjustment information generating model, wherein the user value limit adjustment information generating model is generated based on the method described in one of claims 1-6.

10. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

11. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

12. A computer program product, comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images