Resource pushing method and apparatus, electronic device, medium, and chip
By acquiring attribute data of target products and candidate objects, and utilizing feedback behavior prediction models and object selection strategies, the problem of low matching degree in resource push was solved, achieving a more efficient resource push effect.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING XIAOMI PAYMENT TECH CO LTD
- Filing Date
- 2024-11-29
- Publication Date
- 2026-05-29
AI Technical Summary
Existing tag-based resource push strategies suffer from low matching rates between resources and objects due to a limited number of object characteristics, thus reducing resource push efficiency.
By acquiring attribute data of target products, resource delivery channels, and candidate objects, the probability of feedback behavior of candidate objects is determined using a feedback behavior prediction model. Combined with object selection strategies, target object groups are selected for resource delivery, including joint training of long-term, medium-term, and short-term feedback behavior prediction models and consideration of effect duration.
It improves the matching accuracy and efficiency of resource push, ensuring that the selected target audience receives relevant resources more accurately, and increasing the probability of resource conversion.
Smart Images
Figure CN122120326A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of computer technology, and in particular to a resource push method, apparatus, electronic device, medium, and chip. Background Technology
[0002] The current resource push strategy is a tag-based resource push strategy. The specific process is as follows: tag each object; when pushing resources to a product, determine the relevant tags; select objects with the tags to push resources to that product.
[0003] Among these, the objects exhibit fewer object characteristics, resulting in a low matching degree between the resources pushed to the objects and the objects themselves, thereby reducing the efficiency of resource push. Summary of the Invention
[0004] This disclosure provides a resource push method, apparatus, electronic device, medium, and chip.
[0005] According to a first aspect of the present disclosure, a resource push method is provided, the method comprising: acquiring a target product, a resource push channel for the target product, and object attribute data of candidate objects; the object attribute data being attribute data related to resource push for the target product; determining the probability of feedback behavior of the candidate objects to resources pushed on the resource push channel based on the object attribute data of the candidate objects and a feedback behavior prediction model; selecting at least one target object from the candidate objects based on the feedback behavior probability of the candidate objects to form a target object group corresponding to the resource push channel; and performing resource push processing of the target product to the target object group on the resource push channel.
[0006] In one embodiment of this disclosure, before determining the probability of a candidate object's feedback behavior towards a resource pushed on the resource push channel based on the candidate object's object attribute data and the feedback behavior prediction model, the method further includes: acquiring multiple candidate feedback behavior prediction models; determining the duration of the effect of the resource pushed on the resource push channel; and selecting the feedback behavior prediction model from the multiple candidate feedback behavior prediction models based on the duration of the effect.
[0007] In one embodiment of this disclosure, the candidate feedback behavior prediction model includes: a long-term feedback behavior prediction model, a medium-term feedback behavior prediction model, and a short-term feedback behavior prediction model; the multiple candidate feedback behavior prediction models are jointly trained by combining the object attribute data of each reference object and the historical feedback behavior records of each reference object on the resources pushed on each resource push channel.
[0008] In one embodiment of this disclosure, selecting at least one target object from the candidate objects based on the feedback behavior probability of the candidate objects to form a target object group corresponding to the resource push channel includes: sorting each of the candidate objects according to the feedback behavior probability to obtain a sorting result; filtering a first candidate object in the sorting result; wherein the feedback behavior probability of the first candidate object is less than or equal to a probability threshold; determining an object selection strategy corresponding to the target product; selecting at least one target object from the sorting result according to the object selection strategy; and combining the at least one target object to obtain the target object group corresponding to the resource push channel.
[0009] In one embodiment of this disclosure, the object selection strategy includes at least one of the following: object blacklist / whitelist, pushed object filtering strategy, and object push frequency control strategy.
[0010] In one embodiment of this disclosure, after the resource push process of the target product is performed to the target object group on the resource push channel, the method further includes: obtaining object attribute data of the target object in the target object group, and feedback behavior records of the target object to the resources pushed on the resource push channel; updating the training data of the feedback behavior prediction model according to the object attribute data of the target object and the feedback behavior records, for fine-tuning the feedback behavior prediction model.
[0011] In one embodiment of this disclosure, the object attribute data includes at least one of the following: basic profile data, historical behavior data, environmental data, and resource text data received during the resource push process.
[0012] According to a second aspect of the present disclosure, a resource push device is also provided, the device comprising: a first acquisition module, configured to acquire a target product, a resource push channel for the target product, and object attribute data of candidate objects; the object attribute data being attribute data related to resource push for the target product; a first determination module, configured to determine the probability of feedback behavior of the candidate objects towards resources pushed on the resource push channel based on the object attribute data of the candidate objects and a feedback behavior prediction model; a first selection module, configured to select at least one target object from the candidate objects based on the probability of feedback behavior of the candidate objects to form a target object group corresponding to the resource push channel; and a push processing module, configured to perform resource push processing of the target product to the target object group on the resource push channel.
[0013] In one embodiment of this disclosure, the apparatus further includes: a second acquisition module, a second determination module, and a second selection module; the second acquisition module is used to acquire a plurality of candidate feedback behavior prediction models; the second determination module is used to determine the effect duration of the resources pushed on the resource push channel; and the second selection module is used to select the feedback behavior prediction model from the plurality of candidate feedback behavior prediction models based on the effect duration.
[0014] In one embodiment of this disclosure, the candidate feedback behavior prediction model includes: a long-term feedback behavior prediction model, a medium-term feedback behavior prediction model, and a short-term feedback behavior prediction model; the multiple candidate feedback behavior prediction models are jointly trained by combining the object attribute data of each reference object and the historical feedback behavior records of each reference object on the resources pushed on each resource push channel.
[0015] In one embodiment of this disclosure, the first selection module is specifically configured to: sort each of the candidate objects according to the feedback behavior probability to obtain a sorting result; filter the first candidate objects in the sorting result; ensure that the feedback behavior probability of the first candidate object is less than or equal to a probability threshold; determine the object selection strategy corresponding to the target product; select at least one target object from the sorting result according to the object selection strategy; and combine at least one target object to obtain the target object group corresponding to the resource push channel.
[0016] In one embodiment of this disclosure, the object selection strategy includes at least one of the following: object blacklist / whitelist, pushed object filtering strategy, and object push frequency control strategy.
[0017] In one embodiment of this disclosure, the apparatus further includes: a third acquisition module and an update processing module; the third acquisition module is used to acquire object attribute data of the target objects in the target object group, and feedback behavior records of the target objects to resources pushed on the resource push channel; the update processing module is used to update the training data of the feedback behavior prediction model according to the object attribute data of the target objects and the feedback behavior records, for fine-tuning the feedback behavior prediction model.
[0018] In one embodiment of this disclosure, the object attribute data includes at least one of the following: basic profile data, historical behavior data, environmental data, and resource text data received during the resource push process.
[0019] According to a third aspect of the present disclosure, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the resource push method as described above.
[0020] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium is also provided, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the resource push method as described above.
[0021] According to a fifth aspect of the present disclosure, a chip is also provided, the chip including a processing circuit configured to perform the resource push method as described above.
[0022] The technical solutions provided by the embodiments of this disclosure have at least the following beneficial effects:
[0023] By acquiring target product, target product resource push channels, and candidate object attribute data (object attribute data being attribute data related to resource push for the target product), and based on the candidate object attribute data and feedback behavior prediction model, the probability of the candidate object's feedback behavior towards the resources pushed on the resource push channel is determined. Based on the candidate object's feedback behavior probability, at least one target object is selected from the candidate objects to form a target object group corresponding to the resource push channel. The target product's resource is then pushed to this target object group on the resource push channel. Specifically, combining the candidate object attribute data to determine the candidate object's feedback behavior probability, and then selecting target objects for resource push processing, allows for resource push based on more object characteristics, improving the matching degree between the pushed resources and the object, thereby increasing resource push efficiency.
[0024] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description
[0025] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.
[0026] Figure 1 This is a flowchart of a resource push method according to an embodiment of the present disclosure;
[0027] Figure 2 This is a flowchart of a resource push method according to another embodiment of this disclosure;
[0028] Figure 3 This is a schematic diagram of a resource push process according to an embodiment of the present disclosure;
[0029] Figure 4 This is a schematic diagram of the structure of a resource push device according to an embodiment of the present disclosure;
[0030] Figure 5 This is a structural block diagram of an electronic device according to an exemplary embodiment of the present disclosure;
[0031] Figure 6 This is a schematic diagram of the structure of a chip according to an embodiment of the present disclosure. Detailed Implementation
[0032] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.
[0033] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.
[0034] The current resource push strategy is a tag-based resource push strategy. The specific process is as follows: tag each object; when pushing resources to a product, determine the relevant tags; select objects with the tags to push resources to that product.
[0035] Among these, the objects exhibit fewer object characteristics, resulting in a low matching degree between the resources pushed to the objects and the objects themselves, thereby reducing the efficiency of resource push.
[0036] Figure 1 This is a flowchart of a resource push method according to an embodiment of the present disclosure. It should be noted that the executing entity of the resource push method in this embodiment can be a resource push device, or an electronic device or chip, etc.
[0037] Among them, electronic devices can be any device with computing capabilities, such as personal computers (PCs), mobile terminals, terminal devices, servers, antenna devices, satellite scanning devices, etc. Mobile terminals can be, for example, in-vehicle devices, mobile phones, tablets, personal digital assistants, wearable devices, and other hardware devices with various operating systems, touch screens, and / or displays.
[0038] In addition, the executing entity of the resource push method can also be software in an electronic device used to implement the resource push function. For example, software could be resource push software. The following embodiments will use an electronic device as an example for illustration.
[0039] like Figure 1 As shown, the method includes the following steps:
[0040] Step 101: Obtain the target product, the resource push channel of the target product, and the object attribute data of the candidate objects; the object attribute data is attribute data related to the resource push of the target product.
[0041] In this embodiment of the disclosure, the target product is a product that the object can use or interact with. The target product can be, for example, a virtual product such as an application or service, or a physical product, and can be configured according to actual needs.
[0042] The resource push channels for the target product may include at least one of the following: push channels, short message service (SMS) push channels, advertising push channels, etc.
[0043] In this embodiment of the disclosure, the object attribute data of the candidate object may include at least one of the following: basic profile data, historical behavior data, environmental data, and resource text data received during the resource push process. The basic profile data may include at least one of the following: preference information, consumption characteristics, etc. Historical behavior data may be the interaction behavior between the candidate object and the target product or the target product's pushed resources, such as clicking on videos or pages in the target product, browsing behavior of pushed resources, and usage behavior of functions in the target product, etc., without specific limitations here. Environmental data may be, for example, the location area of the candidate object.
[0044] The use of multiple object attribute data allows for consideration of various characteristics of candidate objects during resource push, thereby enabling the selection of target object groups that match the target product, improving the accuracy of the identified target object groups, and thus increasing the conversion probability of the target object groups for the pushed resources.
[0045] Step 102: Based on the object attribute data of the candidate objects and the feedback behavior prediction model, determine the probability of the candidate objects' feedback behavior towards the resources pushed on the resource push channel.
[0046] In this embodiment of the disclosure, the electronic device may perform step 102 as follows: for each candidate object, input the object attribute data of the candidate object into the feedback behavior prediction model, and obtain the probability of the feedback behavior of the candidate object to the resources pushed on the resource push channel output by the feedback behavior prediction model.
[0047] It should be noted that the feedback behavior prediction model can be used to predict the probability of feedback behavior on multiple resource push channels.
[0048] Among them, the feedback behavior prediction model has a large number of parameters and high calculation accuracy, which can ensure the accuracy of the feedback behavior probability of candidate objects, thereby improving the accuracy of the target object group and further improving the conversion probability of the target object group to the pushed resources.
[0049] Step 103: Based on the feedback behavior probability of the candidate objects, select at least one target object from the candidate objects to form a target object group corresponding to the resource push channel.
[0050] In this embodiment of the disclosure, the electronic device may perform step 103 as follows: sort each candidate object according to the feedback behavior probability to obtain a sorting result; filter the first candidate object in the sorting result; the feedback behavior probability of the first candidate object is less than or equal to a probability threshold; determine the object selection strategy corresponding to the target product; select at least one target object from the sorting result according to the object selection strategy; and combine the at least one target object to obtain the target object group corresponding to the resource push channel.
[0051] The object selection strategy includes at least one of the following: object blacklist / whitelist, pushed object filtering strategy, and object push frequency control strategy. The object blacklist / whitelist includes both a blacklist and a whitelist.
[0052] Taking the object selection strategy, which includes the three strategies mentioned above, as an example, the process by which an electronic device selects a target object from the sorting results according to the object selection strategy can be as follows: For each candidate object in the sorting results, determine whether the candidate object exists in the blacklist; if the candidate object exists in the blacklist, filter the candidate object; if the candidate object does not exist in the blacklist, determine whether the candidate object has received the target product's push resource in the first time period; if so, filter the candidate object; if not, determine whether the number of pushes for the candidate object in the second time period is greater than a preset number threshold; if so, filter the candidate object; and thus obtain the filtered sorting results.
[0053] The second time period is longer than the first time period. The preset threshold for the number of push notifications for the second time period is determined in conjunction with the push frequency.
[0054] Specifically, for the filtered sorting results, candidate objects that exist in the whitelist in the sorting results are identified as target objects; then, a certain number of target objects are arbitrarily selected from the remaining candidate objects in the sorting results, so that the total number of target objects meets the quantity condition.
[0055] In particular, combining the feedback behavior strategy of candidate objects with the object selection strategy to select target objects can ensure that the selected target objects are the objects that need to be pushed the resources of the target product, thereby further improving the matching degree between the target objects and the resources of the target product, and further improving the accuracy of the identified target object group.
[0056] Step 104: Push the target product to the target audience through the resource push channel.
[0057] The feedback behavior prediction model can be used to predict the probability of feedback behavior across multiple resource push channels. Correspondingly, a target audience can be identified for each resource push channel; and deduplication can be performed on the target audience for each resource push channel, ensuring that each target audience is processed using only one resource push channel, thereby reducing the processing load during resource push and further improving resource push efficiency.
[0058] In the resource push method of this disclosure, the following steps are taken: 1) Obtain the target product, the resource push channel for the target product, and object attribute data of candidate objects. The object attribute data is attribute data related to the resource push of the target product. 2) Determine the probability of feedback behavior of the candidate objects towards the resources pushed on the resource push channel based on the object attribute data of the candidate objects and a feedback behavior prediction model. 3) Select at least one target object from the candidate objects based on the feedback behavior probability of the candidate objects to form a target object group corresponding to the resource push channel. 4) Perform resource push processing of the target product to the target object group on the resource push channel. The method of determining the feedback behavior probability of the candidate objects by combining their object attribute data, and then selecting target objects for resource push processing, can combine more characteristics of the objects for resource push, improving the matching degree between the resources pushed to the objects and the objects themselves, thereby improving resource push efficiency.
[0059] Figure 2 This is a flowchart illustrating a resource push method according to another embodiment of the present disclosure. It should be noted that the executing entity of the resource push method in this embodiment can be a resource push device, or an electronic device or chip, etc.
[0060] Among them, electronic devices can be any device with computing capabilities, such as personal computers (PCs), mobile terminals, terminal devices, servers, antenna devices, satellite scanning devices, etc. Mobile terminals can be, for example, in-vehicle devices, mobile phones, tablets, personal digital assistants, wearable devices, and other hardware devices with various operating systems, touch screens, and / or displays.
[0061] In addition, the executing entity of the resource push method can also be software in an electronic device used to implement the resource push function. For example, software could be resource push software. The following embodiments will use an electronic device as an example for illustration.
[0062] like Figure 2 As shown, the method includes the following steps:
[0063] Step 201: Obtain the target product, the resource push channel of the target product, and the object attribute data of the candidate objects; the object attribute data is attribute data related to the resource push of the target product.
[0064] Step 202: Obtain multiple candidate feedback behavior prediction models.
[0065] In this embodiment of the disclosure, the candidate feedback behavior prediction model may include: a long-term feedback behavior prediction model, a medium-term feedback behavior prediction model, and a short-term feedback behavior prediction model; multiple candidate feedback behavior prediction models are obtained by jointly training them by combining the object attribute data of each reference object and the historical feedback behavior records of each reference object on the resources pushed on each resource push channel.
[0066] The training data used for joint training can include three parts: long-term training data, medium-term training data, and short-term training data. These three types of training data include object attribute data for each reference object and historical feedback records of each reference object to resources pushed through various resource delivery channels. The difference lies in the following: long-term training data includes object attribute data for each reference object and historical feedback records of each reference object to resources pushed through various resource delivery channels within the third time period; medium-term training data includes object attribute data for each reference object and historical feedback records of each reference object to resources pushed through various resource delivery channels within the fourth time period; short-term training data includes object attribute data for each reference object and historical feedback records of each reference object to resources pushed through various resource delivery channels within the fifth time period. The length of the third time period is equal to the length of the fourth time period; the length of the fourth time period is greater than the length of the fifth time period.
[0067] Specifically, the electronic device can input the object attribute data of the reference object from the long-term training data into the long-term feedback behavior prediction model to obtain the predicted first feedback behavior probability; input the object attribute data of the reference object from the mid-term training data into the mid-term feedback behavior prediction model to obtain the predicted second feedback behavior probability; input the object attribute data of the reference object from the short-term training data into the short-term feedback behavior prediction model to obtain the predicted third feedback behavior probability; combine the attention mechanism network to determine the weights of the three feedback behavior prediction models; combine the weights of the three feedback behavior prediction models, the predicted first feedback behavior probability and the corresponding actual feedback behavior probability, the predicted second feedback behavior probability and the corresponding actual feedback behavior probability, and the predicted third feedback behavior probability and the corresponding actual feedback behavior probability to comprehensively determine the value of the loss function; and adjust the parameters of the three feedback behavior prediction models based on the value of the loss function to achieve training.
[0068] The probability of actual feedback behavior is determined by combining the historical feedback behavior records of the reference object.
[0069] The joint training of multiple candidate feedback behavior prediction models enables these models to learn the expected feedback behavior of an object based on its object attribute data, thereby improving the accuracy of determining the probability of feedback behavior.
[0070] Step 203: Determine the duration of the effect of the resources pushed on the resource push channel.
[0071] In this embodiment of the disclosure, the duration of the effect of the pushed resource refers to the length of time during which the pushed resource can influence the object's feedback behavior. Based on the different durations of the effect, it can be divided into long-term effect, medium-term effect, and short-term effect. A long-term effect refers to the pushed resource influencing the object's feedback behavior over a relatively long period. A short-term effect refers to the pushed resource influencing the object's feedback behavior over a relatively short period.
[0072] Step 204: Select a feedback behavior prediction model from multiple candidate feedback behavior prediction models based on the duration of the effect.
[0073] In this embodiment of the disclosure, the duration of the effect can be divided into three time ranges: a long-term time range, a medium-term time range, and a short-term time range. Correspondingly, the electronic device performing step 204 can, for example, determine the time range to which the duration of the effect belongs; if the time range is a long-term time range, determine the long-term feedback behavior prediction model as the selected feedback behavior prediction model; if the time range is a medium-term time range, determine the medium-term feedback behavior prediction model as the selected feedback behavior prediction model; and if the time range is a short-term time range, determine the short-term feedback behavior prediction model as the selected feedback behavior prediction model.
[0074] One example of a candidate feedback behavior prediction model is the MMOE (Multi-gate Mixture-of-Experts) model. This model is a multi-task learning model that aims to explicitly model the relationships between different tasks and optimize the performance of each task by introducing multiple expert networks and gate networks. Here, the task refers to the model's optimization objective, such as the click-through rate of resources provided on various resource push channels.
[0075] Step 205: Based on the object attribute data of the candidate objects and the feedback behavior prediction model, determine the probability of the candidate objects' feedback behavior towards the resources pushed on the resource push channel.
[0076] Step 206: Based on the feedback behavior probability of the candidate objects, select at least one target object from the candidate objects to form a target object group corresponding to the resource push channel.
[0077] Step 207: Push the target product to the target audience through the resource push channel.
[0078] In this embodiment of the disclosure, in order to ensure the accuracy of the behavior probability prediction of the feedback behavior prediction model, after step 207, the electronic device may also perform the following process: obtain the object attribute data of the target object in the target object group, and the feedback behavior record of the target object to the resources pushed on the resource push channel; update the training data of the feedback behavior prediction model according to the object attribute data of the target object and the feedback behavior record, for fine-tuning of the feedback behavior prediction model.
[0079] In one example, the electronic device can update the training data of the feedback behavior prediction model used to determine the target object group based on the object attribute data of the target object and the feedback behavior record; fine-tune the feedback behavior prediction model separately based on the updated training data; or retrain it.
[0080] In another example, the electronic device can update the training data of multiple candidate feedback behavior prediction models based on the object attribute data of the target object and the feedback behavior records; and then perform joint fine-tuning or retraining on the multiple candidate feedback behavior prediction models based on the updated training data.
[0081] It should be noted that for detailed explanations of steps 201, and steps 205 through 207, please refer to [link / reference needed]. Figure 1 Steps 101 to 104 in the illustrated embodiment will not be described in detail here.
[0082] In the resource push method of this disclosure, the following steps are taken: 1. Obtaining a target product, a resource push channel for the target product, and object attribute data of candidate objects; 2. Obtaining attribute data related to resource push for the target product; 3. Obtaining multiple candidate feedback behavior prediction models; 4. Determining the duration of the effect of the resources pushed on the resource push channel; 5. Selecting a feedback behavior prediction model from the multiple candidate models based on the duration of the effect; 6. Determining the probability of feedback behavior of the candidate objects towards the resources pushed on the resource push channel based on the object attribute data and the feedback behavior prediction model; 7. Selecting at least one target object from the candidate objects based on the probability of feedback behavior to form a target object group corresponding to the resource push channel; 8. Pushing the target product's resources to the target object group on the resource push channel. The method of selecting a feedback behavior prediction model from the multiple candidate models based on the duration of the effect, and then performing feedback behavior probability prediction, can further improve the accuracy of the determined feedback behavior probability, further improve the matching degree between the resources pushed to the objects and the objects, thereby improving resource push efficiency and resource conversion efficiency.
[0083] The following example illustrates this. For example... Figure 3 The diagram shown is a schematic representation of a resource push process according to an embodiment of this disclosure. Figure 3 This may include the following steps.
[0084] Step 301 involves real-time online traffic feedback, specifically obtaining the object's attribute data and its feedback behavior records regarding resources pushed through the resource push channel. The pushed resources can be resources of the target product, such as a wallet application.
[0085] Step 302 involves combining feedback behavior records to perform statistical processing of feedback behavior and evaluation of resource push effectiveness to obtain an evaluation result. Statistical indicators include, for example, wallet video user push exposure / clicks; wallet video user page exposure / clicks; wallet credit user SMS exposure / clicks; wallet credit user activated credit limit; wallet credit user credit application approval; wallet credit user application approval, etc.
[0086] Step 303: Based on the feedback behavior records, determine a unified offline performance conversion sample library. The unified offline performance conversion sample library may include various feedback behavior records.
[0087] Step 304: Collect user data (object attribute data of the object), such as basic user profile data, user historical behavior data, user environmental context data (environmental data), user marketing copy information (resource copy data received during resource push), etc.
[0088] Step 305: Combining the data in the unified offline effect conversion sample library and user data, an automated offline training engine is used to train the long-term conversion multi-objective model (long-term feedback behavior prediction model), the medium-term conversion multi-objective model (medium-term feedback behavior prediction model), the short-term conversion multi-objective model (short-term feedback behavior prediction model), and the institution pass rate model (optional, used for specific products, such as determining the pass rate when users apply to institutions in financial products), to obtain the offline model.
[0089] Step 306: Combining the offline model, the established delivery strategy, and the target audience selection strategy (e.g., frequency control for delivery services, blacklists and whitelists for delivery services, and filtering strategies for different services), determine the final audience package (target group) and automatically connect to the audience delivery system for delivery processing. Then, re-execute step 301.
[0090] Figure 4 This is a schematic diagram of the structure of a resource push device according to an embodiment of the present disclosure.
[0091] like Figure 4 As shown, the resource push device may include: a first acquisition module 401, a first determination module 402, a first selection module 403, and a push processing module 404.
[0092] The system includes a first acquisition module 401, used to acquire a target product, a resource push channel for the target product, and object attribute data of candidate objects; the object attribute data is attribute data related to the resource push of the target product; a first determination module 402, used to determine the probability of feedback behavior of the candidate objects to the resources pushed on the resource push channel based on the object attribute data of the candidate objects and a feedback behavior prediction model; a first selection module 403, used to select at least one target object from the candidate objects based on the probability of feedback behavior of the candidate objects to form a target object group corresponding to the resource push channel; and a push processing module 404, used to perform resource push processing of the target product to the target object group on the resource push channel.
[0093] In one embodiment of this disclosure, the apparatus further includes: a second acquisition module, a second determination module, and a second selection module; the second acquisition module is used to acquire a plurality of candidate feedback behavior prediction models; the second determination module is used to determine the effect duration of the resources pushed on the resource push channel; and the second selection module is used to select the feedback behavior prediction model from the plurality of candidate feedback behavior prediction models based on the effect duration.
[0094] In one embodiment of this disclosure, the candidate feedback behavior prediction model includes: a long-term feedback behavior prediction model, a medium-term feedback behavior prediction model, and a short-term feedback behavior prediction model; the multiple candidate feedback behavior prediction models are jointly trained by combining the object attribute data of each reference object and the historical feedback behavior records of each reference object on the resources pushed on each resource push channel.
[0095] In one embodiment of this disclosure, the first selection module 403 is specifically configured to: sort each of the candidate objects according to the feedback behavior probability to obtain a sorting result; filter the first candidate objects in the sorting result; ensure that the feedback behavior probability of the first candidate object is less than or equal to a probability threshold; determine the object selection strategy corresponding to the target product; select at least one target object from the sorting result according to the object selection strategy; and combine at least one target object to obtain the target object group corresponding to the resource push channel.
[0096] In one embodiment of this disclosure, the object selection strategy includes at least one of the following: object blacklist / whitelist, pushed object filtering strategy, and object push frequency control strategy.
[0097] In one embodiment of this disclosure, the apparatus further includes: a third acquisition module and an update processing module; the third acquisition module is used to acquire object attribute data of the target objects in the target object group, and feedback behavior records of the target objects to resources pushed on the resource push channel; the update processing module is used to update the training data of the feedback behavior prediction model according to the object attribute data of the target objects and the feedback behavior records, for fine-tuning the feedback behavior prediction model.
[0098] In one embodiment of this disclosure, the object attribute data includes at least one of the following: basic profile data, historical behavior data, environmental data, and resource text data received during the resource push process.
[0099] In the resource push device of this embodiment, the following steps are taken: 1. Obtaining target product, resource push channel for target product, and object attribute data of candidate objects; 2. Object attribute data is attribute data related to resource push of target product; 3. Determining the probability of feedback behavior of candidate objects towards resources pushed on the resource push channel based on the object attribute data of candidate objects and a feedback behavior prediction model; 4. Selecting at least one target object from the candidate objects based on the feedback behavior probability of candidate objects to form a target object group corresponding to the resource push channel; 5. Pushing target product resources to the target object group on the resource push channel; 6. Determining the probability of feedback behavior of candidate objects by combining the object attribute data of candidate objects, and then selecting target objects for resource push processing, can combine more characteristics of objects for resource push, improving the matching degree between the resources pushed to the objects and the objects, thereby improving resource push efficiency.
[0100] According to a third aspect of the present disclosure, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the resource push method as described above.
[0101] To implement the above embodiments, this disclosure also proposes a non-transitory computer-readable storage medium.
[0102] When the computer program in the storage medium is executed by the processor, the processor is able to execute the resource push method described above.
[0103] To implement the above embodiments, this disclosure also provides a computer program product.
[0104] When the computer program product is executed by the processor of the electronic device, it enables the electronic device to perform the above-described method.
[0105] Figure 5This is a structural block diagram illustrating an electronic device according to an exemplary embodiment. For example, the electronic device 500 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.
[0106] Reference Figure 5 The electronic device 500 may include one or more of the following components: processing component 502, memory 504, power component 506, multimedia component 508, audio component 510, input / output (I / O) interface 512, sensor component 514, and communication component 516.
[0107] Processing component 502 typically controls the overall operation of electronic device 500, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 502 may include one or more processors 520 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 502 may include one or more modules to facilitate interaction between processing component 502 and other components. For example, processing component 502 may include a multimedia module to facilitate interaction between multimedia component 508 and processing component 502.
[0108] Memory 504 is configured to store various types of data to support the operation of electronic device 500. Examples of such data include instructions for any application or method operating on electronic device 500, contact data, phonebook data, messages, pictures, videos, etc. Memory 504 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.
[0109] Power component 506 provides power to various components of electronic device 500. Power component 506 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 500.
[0110] Multimedia component 508 includes a screen that provides an output interface between the electronic device 500 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 508 includes a front-facing camera and / or a rear-facing camera. When the electronic device 500 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.
[0111] Audio component 510 is configured to output and / or input audio signals. For example, audio component 510 includes a microphone (MIC) configured to receive external audio signals when electronic device 500 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 504 or transmitted via communication component 516. In some embodiments, audio component 510 also includes a speaker for outputting audio signals.
[0112] I / O interface 512 provides an interface between processing component 502 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.
[0113] Sensor assembly 514 includes one or more sensors for providing state assessments of various aspects of electronic device 500. For example, sensor assembly 514 may detect the on / off state of electronic device 500, the relative positioning of components such as the display and keypad of electronic device 500, changes in position of electronic device 500 or a component of electronic device 500, the presence or absence of user contact with electronic device 500, orientation or acceleration / deceleration of electronic device 500, and temperature changes of electronic device 500. Sensor assembly 514 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 514 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 514 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.
[0114] Communication component 516 is configured to facilitate wired or wireless communication between electronic device 500 and other devices. Electronic device 500 can access wireless networks based on communication standards, such as WiFi, 4G, or 5G, or combinations thereof. In one exemplary embodiment, communication component 516 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 516 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
[0115] In an exemplary embodiment, the electronic device 500 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.
[0116] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 504 including instructions, which can be executed by a processor 520 of an electronic device 500 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.
[0117] To implement the above embodiments, this application also proposes a chip, including: the chip includes a processing circuit configured to perform the methods provided in the foregoing embodiments.
[0118] Figure 6 This is a schematic diagram of the structure of a chip according to an embodiment of this disclosure. See also... Figure 6 The diagram shown is a schematic representation of the structure of chip 600, but it is not limited to this.
[0119] Chip 600 includes processing circuitry 601, which is configured to perform any of the above methods.
[0120] In some embodiments, chip 600 further includes one or more interface circuits 602. Optionally, the interface circuit 602 is connected to memory 603, and the interface circuit 602 can be used to receive signals from memory 603 or other devices, and the interface circuit 602 can be used to send signals to memory 603 or other devices. For example, the interface circuit 602 can read instructions stored in memory 603 and send the instructions to processing circuit 601.
[0121] In some embodiments, the interface circuit 602 performs at least one of the communication steps such as sending and / or receiving in the above method, and the processing circuit 601 performs other steps.
[0122] In some embodiments, the terms interface circuit, interface, transceiver pin, transceiver, etc., can be used interchangeably.
[0123] In some embodiments, chip 600 further includes one or more memories 603 for storing instructions. Optionally, all or part of the memories 603 may be located outside of chip 600.
[0124] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0125] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0126] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.
[0127] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.
[0128] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.
[0129] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0130] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.
[0131] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.
[0132] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.
Claims
1. A resource push method, characterized in that, The method includes: Obtain the target product, the resource push channel of the target product, and the object attribute data of the candidate objects; the object attribute data is attribute data related to the resource push of the target product. Based on the object attribute data of the candidate objects and the feedback behavior prediction model, determine the probability of the feedback behavior of the candidate objects in response to the resources pushed on the resource push channel; Based on the feedback behavior probability of the candidate objects, at least one target object is selected from the candidate objects to form a target object group corresponding to the resource push channel; The resource push process for the target product is carried out to the target audience through the resource push channel.
2. The method according to claim 1, characterized in that, Before determining the probability of a candidate object's feedback behavior towards resources pushed on the resource push channel based on the candidate object's object attribute data and feedback behavior prediction model, the method further includes: Obtain multiple candidate feedback behavior prediction models; Determine the duration of the effect of the resources pushed on the resource push channel; Based on the duration of the effect, the feedback behavior prediction model is selected from the plurality of candidate feedback behavior prediction models.
3. The method according to claim 2, characterized in that, The candidate feedback behavior prediction model includes: a long-term feedback behavior prediction model, a medium-term feedback behavior prediction model, and a short-term feedback behavior prediction model. Multiple candidate feedback behavior prediction models are jointly trained by combining the object attribute data of each reference object and the historical feedback behavior records of each reference object to the resources pushed on each resource push channel.
4. The method according to claim 1, characterized in that, The step of selecting at least one target object from the candidate objects based on the feedback behavior probability of the candidate objects to form a target object group corresponding to the resource push channel includes: The candidate objects are sorted according to the feedback behavior probability to obtain the sorting result. The first candidate object in the sorting result is filtered; the probability of the feedback behavior of the first candidate object is less than or equal to the probability threshold. Determine the target product's corresponding object selection strategy; At least one of the target objects is selected from the sorting results according to the object selection strategy; At least one of the target objects is combined to obtain the target object group corresponding to the resource push channel.
5. The method according to claim 4, characterized in that, The object selection strategy includes at least one of the following: object blacklist / whitelist, pushed object filtering strategy, and object push frequency control strategy.
6. The method according to claim 1, characterized in that, After pushing the target product to the target audience through the resource push channel, the method further includes: Obtain object attribute data of the target object in the target object group, as well as the feedback behavior records of the target object to the resources pushed on the resource push channel; Based on the object attribute data of the target object and the feedback behavior record, the training data of the feedback behavior prediction model is updated for fine-tuning of the feedback behavior prediction model.
7. The method according to claim 1, characterized in that, The object attribute data includes at least one of the following: basic profile data, historical behavior data, environmental data, and resource text data received during the resource push process.
8. A resource delivery device, characterized in that, The device includes: The first acquisition module is used to acquire the target product, the resource push channel of the target product, and the object attribute data of the candidate objects; the object attribute data is attribute data related to the resource push of the target product. The first determining module is used to determine the probability of the feedback behavior of the candidate object to the resource pushed on the resource push channel based on the object attribute data of the candidate object and the feedback behavior prediction model. The first selection module is used to select at least one target object from the candidate objects based on the feedback behavior probability of the candidate objects, so as to form a target object group corresponding to the resource push channel; The push processing module is used to push the target product to the target group through the resource push channel.
9. An electronic device, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the resource push method as described in any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the resource push method as described in any one of claims 1 to 7.
11. A chip, characterized in that, The chip includes a processing circuit configured to perform the resource push method as described in any one of claims 1 to 7.