CN117278617A - Push channel selection method, push channel selection device, push channel selection medium and electronic equipment - Google Patents

Push channel selection method, push channel selection device, push channel selection medium and electronic equipment Download PDF

Info

Publication number
CN117278617A
CN117278617A CN202311311241.3A CN202311311241A CN117278617A CN 117278617 A CN117278617 A CN 117278617A CN 202311311241 A CN202311311241 A CN 202311311241A CN 117278617 A CN117278617 A CN 117278617A
Authority
CN
China
Prior art keywords
information
push
gain
pushed
pushing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202311311241.3A
Other languages
Chinese (zh)
Inventor
江佳翼
蒋文瑞
黄伟健
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Douyin Vision Co Ltd
Original Assignee
Douyin Vision Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Douyin Vision Co Ltd filed Critical Douyin Vision Co Ltd
Priority to CN202311311241.3A priority Critical patent/CN117278617A/en
Publication of CN117278617A publication Critical patent/CN117278617A/en
Pending legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/55Push-based network services
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14Network analysis or design
    • H04L41/145Network analysis or design involving simulating, designing, planning or modelling of a network
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/60Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources

Abstract

The disclosure relates to a push channel selection method, a push channel selection device, a push channel selection medium and electronic equipment. The method comprises the following steps: in response to receiving a push request, determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model, wherein the push request comprises the information to be pushed; and determining whether to-be-pushed information is pushed by adopting a preset channel according to the result index gain. Therefore, the result index gain of pushing the information to be pushed by adopting the high-quality resources can be quantified based on causal deduction, and whether the preset channel is allocated to the information to be pushed or not is determined based on the height of the result index gain, so that the limited high-quality resources of the preset channel can be used more reasonably, the coverage time period of the preset channel can be longer in one day, the problem that the awakening effect of important pushing in afternoon and evening is reduced due to the fact that the preset channel amount is consumed in the morning is avoided, the overall information reaching rate is improved, the user experience is improved, and the ineffective occupation and waste of the resources can be avoided.

Description

Push channel selection method, push channel selection device, push channel selection medium and electronic equipment
Technical Field
The disclosure relates to the technical field of communication, and in particular relates to a push channel selection method, a push channel selection device, a push channel selection medium and electronic equipment.
Background
To increase the daily activity of a user, a push approach is typically used to send notifications to the user, thereby reducing the likelihood that the user will miss important activities or messages. The push information needs to be pushed by means of channels, and different channels have different characteristics, for example, some channels have high touch rate but limited quota, and some channels have low touch rate but unlimited quota, so that a proper channel strategy needs to be selected according to push requirements and the like to push the information, so that the touch rate of the whole information is optimized, and user experience is improved.
Disclosure of Invention
This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
In a first aspect, the present disclosure provides a push channel selection method, including:
in response to receiving a push request, determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model, wherein the push request comprises the information to be pushed;
And determining whether to push the information to be pushed by adopting the preset channel according to the result index gain.
In a second aspect, the present disclosure provides a push channel selection device, including:
the first determining module is used for determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model in response to receiving a pushing request, wherein the pushing request comprises the information to be pushed;
and the second determining module is used for determining whether to adopt the preset channel to push the information to be pushed or not according to the result index gain.
In a third aspect, the present disclosure provides a computer readable medium having stored thereon a computer program which, when executed by a processing device, implements the steps of the push channel selection method provided by the first aspect of the present disclosure.
In a fourth aspect, the present disclosure provides an electronic device comprising:
a storage device having a computer program stored thereon;
and the processing device is used for executing the computer program in the storage device to realize the steps of the push channel selection method provided by the first aspect of the disclosure.
In the technical scheme, when a push request is received, firstly, determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model; and then determining whether to-be-pushed information is pushed by adopting a preset channel according to the result index gain. Therefore, the result index gain of pushing the information to be pushed by adopting the high-quality resources (namely the preset channels) can be quantified based on causal deduction, and then whether the preset channels are allocated to the information to be pushed or not is determined based on the height of the result index gain, so that the limited high-quality resources of the preset channels can be used more reasonably, the coverage time period of the preset channels can be longer in one day, the problem that the awakening effect of important pushing in the afternoon and evening is reduced due to the fact that the preset channel limit is consumed in the morning is avoided, the contact rate of the whole information is improved, the user experience is improved, and the ineffective occupation and waste of the resources can be avoided.
Additional features and advantages of the present disclosure will be set forth in the detailed description which follows.
Drawings
The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. The same or similar reference numbers will be used throughout the drawings to refer to the same or like elements. It should be understood that the figures are schematic and that elements and components are not necessarily drawn to scale. In the drawings:
fig. 1 is a flow chart illustrating a push channel selection method according to an exemplary embodiment.
Fig. 2 is a schematic diagram illustrating a push channel selection process according to an example embodiment.
FIG. 3 is a flowchart illustrating a causal inference model training method, according to an example embodiment.
Fig. 4 is a block diagram illustrating a push channel selection device according to an example embodiment.
Fig. 5 is a schematic diagram of an electronic device according to an exemplary embodiment.
Detailed Description
As discussed in the background art, the push information needs to be pushed by channels, and different channels have different characteristics, for example, the channel touch rate is high but the limit is limited, and the channel touch rate is low but the limit is unlimited. Specifically, the push channels are generally divided into two types: vendor channels and three-way channels. The manufacturer channels have high touch rate but limited quota, and the three-way channels have low touch rate but unlimited quota. Therefore, when the manufacturer channel is sufficient, the best resources, i.e. the manufacturer channels, should be ensured to be fully used.
However, for the service with more users or more push information, the push amount per day is usually larger than the vendor channel quota, so that the vendor channel quota cannot completely meet the push requirement. The current channel selection strategy is to use the manufacturer channel to push information preferentially, when the residual amount of the manufacturer channel is zero, a three-way channel is used, the information access rate in the later stage of pushing can be greatly reduced, the whole information access rate is influenced, and the number of daily active users (Daily Active User, DAU) is influenced. In addition, the information is pushed according to the channel selection strategy, so that the resource allocation is unreasonable, and the problems of invalid occupation of resources, resource waste and the like are caused.
In view of this, the disclosure provides a push channel selection method, a push channel selection device, a push channel selection medium and an electronic device.
Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure have been shown in the accompanying drawings, it is to be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but 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 for illustration purposes only and are not intended to limit the scope of the present disclosure.
It should be understood that the various steps recited in the method embodiments of the present disclosure may be performed in a different order and/or performed in parallel. Furthermore, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this respect.
The term "including" and variations thereof as used herein are intended to be open-ended, i.e., including, but not limited to. The term "based on" is based at least in part on. The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments. Related definitions of other terms will be given in the description below.
It should be noted that the terms "first," "second," and the like in this disclosure are merely used to distinguish between different devices, modules, or units and are not used to define an order or interdependence of functions performed by the devices, modules, or units.
It should be noted that references to "one", "a plurality" and "a plurality" in this disclosure are intended to be illustrative rather than limiting, and those of ordinary skill in the art will appreciate that "one or more" is intended to be understood as "one or more" unless the context clearly indicates otherwise.
The names of messages or information interacted between the various devices in the embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.
It will be appreciated that prior to using the technical solutions disclosed in the embodiments of the present disclosure, the user should be informed and authorized of the type, usage range, usage scenario, etc. of the personal information related to the present disclosure in an appropriate manner according to the relevant legal regulations.
For example, in response to receiving an active request from a user, a prompt is sent to the user to explicitly prompt the user that the operation it is requesting to perform will require personal information to be obtained and used with the user. Thus, the user can autonomously select whether to provide personal information to software or hardware such as an electronic device, an application program, a server or a storage medium for executing the operation of the technical scheme of the present disclosure according to the prompt information.
As an alternative but non-limiting implementation, in response to receiving an active request from a user, the manner in which the prompt information is sent to the user may be, for example, a popup, in which the prompt information may be presented in a text manner. In addition, a selection control for the user to select to provide personal information to the electronic device in a 'consent' or 'disagreement' manner can be carried in the popup window.
It will be appreciated that the above-described notification and user authorization process is merely illustrative and not limiting of the implementations of the present disclosure, and that other ways of satisfying relevant legal regulations may be applied to the implementations of the present disclosure.
Meanwhile, it can be understood that the data (including but not limited to the data itself, the acquisition or the use of the data) related to the technical scheme should conform to the requirements of the corresponding laws and regulations and related regulations.
Fig. 1 is a flow chart illustrating a push channel selection method according to an exemplary embodiment. The method can be applied to electronic equipment. As shown in fig. 1, the push channel selection method may include S101 and S102.
In S101, in response to receiving the push request, based on the causal inference model, a result indicator gain of pushing the information to be pushed using the preset channel is determined.
In the present disclosure, a push channel refers to a channel that can realize pushing information to be pushed. The pushing channel comprises a preset channel and other channels, wherein the preset channel can be a manufacturer channel, and the manufacturer channel is a channel provided by a manufacturer corresponding to a terminal of a pushing object and belongs to an official channel; the other channels can be three-way channels, and the three-way channels are third-way channels provided by other suppliers besides the manufacturer corresponding to the terminal of the push object, and belong to unofficial channels.
And the upstream business side sends a push request, and the electronic equipment receives the push request and then performs push channel selection service. The push request may include information to be pushed, push objects, etc. The information to be pushed may be, for example, an article, a video, an important notification, etc. pushed by the upstream service direction user, and may also be other types of information.
An object in the present disclosure may refer to a user, and thus, a push object may refer to any user. In one embodiment, the pushing object may specifically refer to an identifier of the pushing object, where the identifier may be used to uniquely identify a terminal of the pushing object to which the information is to be pushed, so that the information to be pushed may be pushed to a terminal corresponding to the identifier (that is, a terminal of the pushing object) later.
The business index can be divided into a result index and a process index, wherein the index reflecting the pushing final target is the result index, and the intermediate index affecting the result is the process index. Illustratively, the result index is a DAU and the process index is a touch rate.
In order to improve the result index brought by pushing, firstly, a problem that a pushing channel is selected as a 'high-quality resource allocation' is defined, and the key point is that a decision is needed to decide which pushing channel resources with high quality are used. Then, starting from the result index, it is defined which pushing should use the better preset channel resources. The ultimate goal of a push is typically to hope that after a push is delivered, the user is more likely to wake up because of the push. If a corresponding user of a push is already active, or if the terminal of the user receives push information through a preset channel, the likelihood that the user is awakened is low, it may be considered that the preset channel is not allocated to the user, but is allocated to other pushes with higher "gain value". The gain value is the result index gain caused by pushing the information to be pushed by adopting a preset channel.
The concept of gain value is introduced above, namely, the high-quality resource of a preset channel is hoped to be allocated to push with higher gain value, and the problem belongs to the problem commonly solved by causal inference. Causal inference means that when an impact (treatment) is applied to something or nothing is done (control), the result (outome) will be different. In the context of information push, the event corresponds to a preset channel, the control corresponds to other channels, the outome directly corresponds to a process index (e.g., a touch rate), and finally affects a result index (e.g., a DAU). Different outomes in causal inference have different values, and gain value is generally defined as the difference between the outomes applying treatment and the outomes performing control. In the present disclosure, the gain value specifically refers to a difference between a result index value of pushing information to be pushed by using a preset channel and a result index value of pushing information to be pushed by using other channels.
In S102, according to the result index gain, it is determined whether to push the information to be pushed by using a preset channel.
In the present disclosure, whether to push information to be pushed by using a preset channel may be determined according to the gain of the result index. If the information to be pushed is determined to be pushed by adopting the preset channel, the information to be pushed is sent by adopting the preset channel; if the information to be pushed is determined not to be pushed by adopting the preset channel, the information to be pushed is sent by adopting other channels.
In the technical scheme, when a push request is received, firstly, determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model; and then determining whether to-be-pushed information is pushed by adopting a preset channel according to the result index gain. Therefore, the result index gain of pushing the information to be pushed by adopting the high-quality resources (namely the preset channels) can be quantified based on causal deduction, and then whether the preset channels are allocated to the information to be pushed or not is determined based on the height of the result index gain, so that the limited high-quality resources of the preset channels can be used more reasonably, the coverage time period of the preset channels can be longer in one day, the problem that the awakening effect of important pushing in the afternoon and evening is reduced due to the fact that the preset channel limit is consumed in the morning is avoided, the contact rate of the whole information is improved, the user experience is improved, and the ineffective occupation and waste of the resources can be avoided.
The following describes in detail the embodiment of determining the result indicator gain of pushing the information to be pushed using the preset channel based on the causal inference model in S101. Specifically, the push request may further include a push object, and in this case, the push request may be implemented by the following steps (1) to (3):
Step (1): and obtaining object characteristics of the push object.
In the present disclosure, object features of a push object may include brand information of a terminal of the push object, historical push conditions of the push object, historical activity information, a plurality of sub-features, and the like. The historical pushing situation may include a plurality of pieces of pushing information which are pushed recently, channel information (used for characterizing which channel is adopted for information pushing) adopted when each piece of pushing information is pushed, and the like, and the historical active information may include a login situation of a pushing object on a product (for example, a target application program or a target browser) of an upstream service side; the plurality of sub-features may be determined based on historical push conditions and historical activity information of the push object.
Illustratively, the result index gain is a daily active user number gain, i.e. a DAU gain; the plurality of sub-features may include a first sub-feature for characterizing whether the push object is a daily active user, a second sub-feature for characterizing whether the terminal of the push object receives over-push information through a preset channel on the same day, and a third sub-feature for characterizing whether the push object is a long-term inactive user. If the push object is in an inactive state within a preset period of time (for example, 30 days), the push object is considered to be a long-term inactive user.
In the present disclosure, when the object feature of the push object is acquired, a prompt message may be sent to the push object to explicitly prompt the push object that the operation requested to be performed will need to acquire and use the personal information to the push object. Therefore, the push object can autonomously select whether to provide personal information for software or hardware such as electronic equipment, application programs, servers or storage media for executing the operation of the technical scheme according to the prompt information.
Specifically, as shown in fig. 2, the object features of the push object may be obtained by reading online data in an online database, where the online data includes history push information, object features of the push object of the history push information, and the like, and in particular, the online data may include real-time data and offline data. The real-time data can comprise historical push information of the current day, object characteristics of push objects of the historical push information and the like, and can be written into the online database according to the log information of the current day; the offline data may include historical push information for a last preset period of time (e.g., last 30 days), and information such as object characteristics of push objects of the historical push information.
Step (2): and determining the process index gain of pushing the information to be pushed by adopting the preset channel based on the causal inference model according to the object characteristics.
Step (3): and determining a result index gain according to the object characteristics and the process index gain.
In the present disclosure, the process index gain refers to a gain on a process index caused by pushing information to be pushed by using a preset channel.
The causal inference model may be a process indicator prediction model, e.g., a reachability prediction model, that is used to model a pushed process indicator (e.g., a reachability). The reason for modeling the process index here is that: the process indexes (such as the touch rate) of different channels are relatively large in difference, modeling is relatively convenient, and meanwhile, the channel selection decision-making result is also a channel; and, although the final objective of the service is a result index (e.g., DAU), the result of the result index is related to the push content, and if the result index is directly modeled, the influence of the channel is weakened, so that the model is hard to learn the difference of the channels.
The following describes in detail the specific embodiment of determining the process index gain of pushing the information to be pushed by using the preset channel based on the causal inference model according to the object characteristics in the step (2). Specifically, the object characteristics of the push object and the preset channel can be input into the causal inference model to obtain a first process index predicted value of pushing information to be pushed by adopting the preset channel, and the object characteristics of the push object and other channels are input into the causal inference model to obtain a second process index predicted value of pushing information to be pushed by adopting other channels; then, a difference between the first process indicator predicted value and the second process indicator predicted value is determined as a process indicator gain.
In the present disclosure, as shown in fig. 2, two model requests may be constructed for the push request, and the two model requests are respectively sent to a causal inference model, so as to obtain a fruit indicator predicted value (i.e., model output) through the causal inference model; one model request is used for requesting a causal inference model to predict a process index of pushing information to be pushed by adopting a preset channel according to the input object characteristics of a pushing object and the preset channel so as to obtain a first process index predicted value; the other model request is used for requesting the causal inference model to predict the process index of pushing the information to be pushed by adopting other channels according to the input object characteristics of the pushing object and other channels so as to obtain a second process index predicted value.
The following describes in detail the specific embodiment of determining the result index gain according to the object feature and the process index gain in the step (3). Specifically, this can be achieved by the following steps (31) and (32):
step (31): and determining the feature gain corresponding to the object feature of the pushing object.
Step (32): and determining the result index gain of pushing the information to be pushed by adopting the preset channel according to the characteristic gain and the process index gain.
In the present disclosure, the feature gain corresponding to the object feature of the push object refers to the gain brought by the object feature of the push object.
Because the causal inference model models the process index, the object feature of the pushing object may not be completely learned, and the final result index gain of pushing the information to be pushed by adopting the preset channel needs to be determined, in order to link the process index gain with the result index gain, the process index gain may be adjusted by using the object feature of the pushing object to obtain the result index gain.
In one embodiment, the sum of the characteristic gain corresponding to the object characteristic of the pushing object and the process index gain may be determined as the result index gain of pushing the information to be pushed by using the preset channel.
The following describes in detail a specific embodiment of the feature gain corresponding to the feature to be determined in the step (31). Specifically, the object feature may include a plurality of sub-features, and in this case, for each sub-feature, a weight corresponding to the sub-feature may be determined according to a feature value of the sub-feature; and then, according to the weight corresponding to each sub-feature, calculating the weighted sum of the sub-feature gains corresponding to each sub-feature, and obtaining the feature gain corresponding to the object feature of the pushing object.
Illustratively, the result index gain is a daily active user number gain, i.e. a DAU gain; the object features may include a first sub-feature for characterizing whether the push object is a daily active user, a second sub-feature for characterizing whether the terminal of the push object receives push information through a preset channel on the same day, and a third sub-feature for characterizing whether the push object is a long-term inactive user. At this time, for the first sub-feature, a weight corresponding to the first sub-feature may be determined according to a feature value of the first sub-feature. Specifically, if the feature value of the first sub-feature is that the push object is a daily active user, it indicates that the push object is already an active user, and no user wake-up is necessary, at this time, the weight corresponding to the first sub-feature may be reduced, for example, the weight corresponding to the first sub-feature may be a first preset weight within the range of (0, 1); if the feature value of the first sub-feature is that the push object is a non-daily active user, it indicates that the push object is not an active user, and at this time, the user needs to be awakened by push, and at this time, the weight corresponding to the first sub-feature may be increased, for example, the weight corresponding to the first sub-feature may be 1.
Meanwhile, for the second sub-feature, determining the weight corresponding to the second sub-feature according to the feature value of the second sub-feature. Specifically, if the feature value of the second sub-feature is that the terminal of the push object has received push information through the preset channel on the same day, the likelihood that the push object has been awakened or is awakened is low, and at this time, the weight corresponding to the second sub-feature may be reduced, for example, the weight corresponding to the second sub-feature may be a second preset weight within the range of (0, 1); if the feature value of the second sub-feature is that the terminal of the pushing object does not receive the push information through the preset channel on the same day, the user may be awakened through pushing, and at this time, the weight corresponding to the second sub-feature may be increased, for example, the weight corresponding to the second sub-feature may be 1.
In addition, for the third sub-feature, the weight corresponding to the third sub-feature may be determined according to the feature value of the third sub-feature. Specifically, if the feature value of the third sub-feature is that the push object is a long-term inactive user, it indicates that the likelihood that the push object is awakened is low, and at this time, the weight corresponding to the third sub-feature may be reduced, for example, the weight corresponding to the third sub-feature may be a third preset weight within the range of (0, 1); if the feature value of the third sub-feature is not a long-term inactive user, an attempt may be made to wake up the user by pushing, and at this time, the weight corresponding to the third sub-feature may be increased, for example, the weight corresponding to the third sub-feature may be 1.
The following describes in detail the specific implementation manner of determining whether to push the information to be pushed by using the preset channel according to the result index gain in S102. Specifically, the method can be realized by the following steps [1] to [3]:
step [1]: a current gain threshold is obtained.
Step [2]: if the result index gain is larger than the current gain threshold, determining to-be-pushed information pushed by adopting a preset channel.
Step [3]: if the gain of the result index is smaller than or equal to the current gain threshold, determining to push the information to be pushed by adopting other channels.
The following describes in detail the specific embodiment for obtaining the current gain threshold in the step [1 ]. Specifically, it may be implemented in various embodiments, and in one embodiment, the preset gain threshold may be determined as the current gain threshold, that is, the gain threshold is a fixed preset value.
In another embodiment, the current gain threshold is a dynamic threshold, specifically, for each piece of second history push information in M pieces of recently pushed second history push information, a history result index gain of pushing the second history push information by using a preset channel may be obtained; the median of the largest K historical result index gains of the M historical result index gains is then determined as the current gain threshold, where K is less than M, illustratively,
In the embodiment, the current gain threshold value is determined in real time by adopting the dynamic sliding window mode, so that the change amplitude of the current gain threshold value is smoother, the ratio of the information pushing quantity of different time periods in a day to the information pushing quantity pushed through the preset channel is closer, thus the preset channel limit can be consumed more smoothly, the coverage time period of the preset channel which can be used in one day is longer, the problem that the awakening effect of important pushing in afternoon and evening is reduced due to the fact that the preset channel limit is consumed in the morning is avoided,
the following describes the training method of the causal inference model in detail. Specifically, the causal inference model described above may be trained by S301 and S302 shown in fig. 3.
In S301, a history push channel of the first history push information, a history object feature of the history push object, and an actual process index value are acquired.
In the present disclosure, as shown in fig. 2, a training sample may be obtained from offline data, where the training sample may include a history pushing channel of the first history pushing information, a history object feature of the history pushing object, and an actual process index value.
The actual process index value is the actual process index value when the history pushing channel is adopted to push the first history pushing information to the history pushing object. For example, the actual process index value is an actual reaching rate, if the history pushing object receives the first history pushing information, the actual reaching rate is 1, otherwise, the actual reaching rate is 0.
In the present disclosure, when the history object feature of the history push object is acquired, a prompt message may be sent to the push object to explicitly prompt the history push object that the operation requested to be performed will need to acquire and use the personal information of the history push object. Therefore, the history push object can autonomously select whether to provide personal information for software or hardware such as electronic equipment, application programs, servers or storage media for executing the operation of the technical scheme of the disclosure according to the prompt information.
In S302, model training is performed by using the history object feature and the history push channel as inputs to the causal inference model and using the actual process index value as a target output of the causal inference model, so as to obtain the causal inference model.
Specifically, the history object features and the history pushing channel can be input into a causal inference model to obtain a process index predicted value of pushing the first history pushing information to the history pushing object by adopting the history pushing channel; then, according to the difference between the predicted value of the process index and the actual process index value, updating the model parameters of the causal inference model; then, new training data (i.e., the history pushing channel of the new first history pushing information, the history object feature of the history pushing object, and the actual process index value) is acquired again, and model training is continued, i.e., the above S301 is returned until the training deadline condition is reached.
In one embodiment, the training cutoff condition may be that the number of times of training reaches a preset number of times, where the preset number of times may be set according to an actual usage scenario, and when the number of times of training reaches the preset number of times, it may be determined that the number of times of training is sufficient, so that the causal inference model may learn characteristics of the channel and the historical push object.
In another embodiment, the training cutoff condition may be that the target loss of the causal inference model is less than a preset threshold, which may be set according to the actual usage scenario. In the case that the target loss of the causal inference model is smaller than the preset threshold, the accuracy of the process index prediction of the causal inference model can be considered to meet the accuracy requirement.
Fig. 4 is a block diagram illustrating a push channel selection device according to an example embodiment. As shown in fig. 4, the apparatus 400 includes:
a first determining module 401, configured to determine, in response to receiving a push request, a result indicator gain of pushing information to be pushed using a preset channel based on a causal inference model, where the push request includes the information to be pushed;
a second determining module 402, configured to determine whether to push the information to be pushed using the preset channel according to the result indicator gain.
In the technical scheme, when a push request is received, firstly, determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model; and then determining whether to-be-pushed information is pushed by adopting a preset channel according to the result index gain. Therefore, the result index gain of pushing the information to be pushed by adopting the high-quality resources (namely the preset channels) can be quantified based on causal deduction, and then whether the preset channels are allocated to the information to be pushed or not is determined based on the height of the result index gain, so that the limited high-quality resources of the preset channels can be used more reasonably, the coverage time period of the preset channels can be longer in one day, the problem that the awakening effect of important pushing in the afternoon and evening is reduced due to the fact that the preset channel limit is consumed in the morning is avoided, the contact rate of the whole information is improved, the user experience is improved, and the ineffective occupation and waste of the resources can be avoided.
Optionally, the push request further includes a push object;
the first determining module 401 includes:
the first acquisition sub-module is used for acquiring object characteristics of the pushing object;
the first determining submodule is used for determining the process index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model according to the object characteristics;
And the second determining submodule is used for determining the result index gain according to the object characteristics and the process index gain.
Optionally, the first determining submodule includes:
the prediction sub-module is used for inputting the object characteristics and the preset channel into the causal inference model to obtain a first process index predicted value of pushing the information to be pushed by adopting the preset channel, and inputting the object characteristics and other channels into the causal inference model to obtain a second process index predicted value of pushing the information to be pushed by adopting the other channels;
and a third determining sub-module, configured to determine a difference between the first process indicator predicted value and the second process indicator predicted value as the process indicator gain.
Optionally, the causal inference model is trained by a model training module, wherein the model training module comprises:
the acquisition module is used for acquiring a history pushing channel of the first history pushing information, history object characteristics of a history pushing object and an actual process index value;
and the training model is used for carrying out model training by taking the historical object characteristics and the historical push channel as inputs of the causal inference model and taking the actual process index value as a target output of the causal inference model so as to obtain the causal inference model.
Optionally, the second determining submodule includes:
a fourth determining submodule, configured to determine a feature gain corresponding to the object feature;
and a fifth determining submodule, configured to determine the result indicator gain according to the characteristic gain and the process indicator gain.
Optionally, the object feature comprises a plurality of sub-features;
the fourth determination submodule includes:
a sixth determining sub-module, configured to determine, for each sub-feature, a weight corresponding to the sub-feature according to a feature value of the sub-feature;
and the calculating sub-module is used for calculating the weighted sum of the sub-feature gains corresponding to each sub-feature according to the weight corresponding to each sub-feature to obtain the feature gain.
Optionally, the second determining module 402 includes:
the second acquisition submodule is used for acquiring the current gain threshold value;
and a seventh determining submodule, configured to determine to push the information to be pushed by using the preset channel if the result indicator gain is greater than the current gain threshold.
Optionally, the second acquisition submodule includes:
a third obtaining sub-module, configured to obtain, for each of M pieces of second history push information that are recently pushed, a history result indicator gain of pushing the second history push information by using the preset channel;
And an eighth determining submodule, configured to determine a median of the maximum K historical result index gains among the M historical result index gains as a current gain threshold, where K is smaller than M.
In addition, the present disclosure also provides a computer readable medium having stored thereon a computer program which, when executed by a processing device, implements the steps of the push channel selection method provided by the present disclosure.
Referring now to fig. 5, a schematic diagram of an electronic device (e.g., a terminal device or server) 600 suitable for use in implementing embodiments of the present disclosure is shown. The terminal devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 5 is merely an example and should not be construed to limit the functionality and scope of use of the disclosed embodiments.
As shown in fig. 5, the electronic device 600 may include a processing means (e.g., a central processing unit, a graphics processor, etc.) 601, which may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 602 or a program loaded from a storage means 608 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic apparatus 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
In general, the following devices may be connected to the I/O interface 605: input devices 606 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, and the like; an output device 607 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 608 including, for example, magnetic tape, hard disk, etc.; and a communication device 609. The communication means 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. While fig. 5 shows an electronic device 600 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead.
In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a non-transitory computer readable medium, the computer program comprising program code for performing the method shown in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via communication means 609, or from storage means 608, or from ROM 602. The above-described functions defined in the methods of the embodiments of the present disclosure are performed when the computer program is executed by the processing device 601.
It should be noted that the computer readable medium described in 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 can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present disclosure, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.
In some implementations, the clients, servers 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"), the internet (e.g., the internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed networks.
The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device.
The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: in response to receiving a push request, determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model, wherein the push request comprises the information to be pushed; and determining whether to push the information to be pushed by adopting the preset channel according to the result index gain.
Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including, but not limited to, an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules described in the embodiments of the present disclosure may be implemented in software or hardware. The name of the module is not limited to the module itself in some cases, for example, the second determining module may also be described as "a module for determining whether to push the information to be pushed using the preset channel according to the result indicator gain".
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: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a Complex Programmable Logic Device (CPLD), and the like.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
According to one or more embodiments of the present disclosure, example 1 provides a push channel selection method, including:
in response to receiving a push request, determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model, wherein the push request comprises the information to be pushed;
and determining whether to push the information to be pushed by adopting the preset channel according to the result index gain.
Example 2 provides the method of example 1, according to one or more embodiments of the present disclosure, the push request further comprising a push object;
the determining, based on the causal inference model, a result index gain of pushing information to be pushed by using a preset channel includes:
acquiring object characteristics of the push object;
determining a process index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model according to the object characteristics;
and determining the result index gain according to the object characteristics and the process index gain.
According to one or more embodiments of the present disclosure, example 3 provides the method of example 2, the determining, based on the causal inference model, a process indicator gain to push information to be pushed using a preset channel according to the object feature, including:
Inputting the object characteristics and the preset channel into the causal inference model to obtain a first process index predicted value of pushing the information to be pushed by adopting the preset channel, and inputting the object characteristics and other channels into the causal inference model to obtain a second process index predicted value of pushing the information to be pushed by adopting the other channels;
and determining a difference value between the first process index predicted value and the second process index predicted value as the process index gain.
Example 4 provides the method of example 3, according to one or more embodiments of the present disclosure, the causal inference model being trained by:
acquiring a history pushing channel of the first history pushing information, history object characteristics of a history pushing object and an actual process index value;
model training is performed by taking the historical object features and the historical push channel as inputs of the causal inference model and taking the actual process index value as a target output of the causal inference model to obtain the causal inference model.
Example 5 provides the method of example 2, according to one or more embodiments of the present disclosure, the determining the result indicator gain from the object feature and the process indicator gain comprising:
Determining a feature gain corresponding to the object feature;
and determining the result index gain according to the characteristic gain and the process index gain.
Example 6 provides the method of example 5, the object feature comprising a plurality of sub-features, according to one or more embodiments of the present disclosure;
the determining the feature gain corresponding to the object feature comprises the following steps:
for each sub-feature, determining the weight corresponding to the sub-feature according to the feature value of the sub-feature;
and calculating the weighted sum of the sub-feature gains corresponding to each sub-feature according to the weight corresponding to each sub-feature to obtain the feature gain.
According to one or more embodiments of the present disclosure, example 7 provides the method of example 1, wherein the determining whether to push the information to be pushed using the preset channel according to the result indicator gain includes:
acquiring a current gain threshold;
and if the result index gain is larger than the current gain threshold, determining to push the information to be pushed by adopting the preset channel.
In accordance with one or more embodiments of the present disclosure, example 8 provides the method of example 7, the obtaining the current gain threshold comprising:
Acquiring a history result index gain of pushing the second history push information by adopting the preset channel aiming at each piece of second history push information in M pieces of recently pushed second history push information;
and determining the median of the maximum K historical result index gains in the M historical result index gains as a current gain threshold, wherein K is smaller than M.
According to one or more embodiments of the present disclosure, example 9 provides a push channel selection apparatus, including:
the first determining module is used for determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model in response to receiving a pushing request, wherein the pushing request comprises the information to be pushed;
and the second determining module is used for determining whether to adopt the preset channel to push the information to be pushed or not according to the result index gain.
According to one or more embodiments of the present disclosure, example 10 provides a computer-readable medium having stored thereon a computer program which, when executed by a processing device, implements the steps of the method of any of examples 1-8.
Example 11 provides an electronic device according to one or more embodiments of the present disclosure, comprising:
A storage device having a computer program stored thereon;
processing means for executing said computer program in said storage means to carry out the steps of the method according to any one of claims 1-8.
The foregoing description is only of the preferred embodiments of the present disclosure and description of the principles of the technology being employed. It will be appreciated by persons skilled in the art that the scope of the disclosure referred to in this disclosure is not limited to the specific combinations of features described above, but also covers other embodiments which may be formed by any combination of features described above or equivalents thereof without departing from the spirit of the disclosure. Such as those described above, are mutually substituted with the technical features having similar functions disclosed in the present disclosure (but not limited thereto).
Moreover, although operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are example forms of implementing the claims. The specific manner in which the various modules perform the operations in the apparatus of the above embodiments have been described in detail in connection with the embodiments of the method, and will not be described in detail herein.

Claims (11)

1. The push channel selection method is characterized by comprising the following steps:
in response to receiving a push request, determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model, wherein the push request comprises the information to be pushed;
and determining whether to push the information to be pushed by adopting the preset channel according to the result index gain.
2. The method of claim 1, wherein the push request further comprises a push object;
the determining, based on the causal inference model, a result index gain of pushing information to be pushed by using a preset channel includes:
Acquiring object characteristics of the push object;
determining a process index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model according to the object characteristics;
and determining the result index gain according to the object characteristics and the process index gain.
3. The method of claim 2, wherein determining, based on the causal inference model, a process indicator gain for pushing information to be pushed using a preset channel according to the object characteristics, comprises:
inputting the object characteristics and the preset channel into the causal inference model to obtain a first process index predicted value of pushing the information to be pushed by adopting the preset channel, and inputting the object characteristics and other channels into the causal inference model to obtain a second process index predicted value of pushing the information to be pushed by adopting the other channels;
and determining a difference value between the first process index predicted value and the second process index predicted value as the process index gain.
4. A method according to claim 3, wherein the causal inference model is trained by:
acquiring a history pushing channel of the first history pushing information, history object characteristics of a history pushing object and an actual process index value;
Model training is performed by taking the historical object features and the historical push channel as inputs of the causal inference model and taking the actual process index value as a target output of the causal inference model to obtain the causal inference model.
5. The method of claim 2, wherein said determining said result indicator gain based on said object characteristics and said process indicator gain comprises:
determining a feature gain corresponding to the object feature;
and determining the result index gain according to the characteristic gain and the process index gain.
6. The method of claim 5, wherein the object features comprise a plurality of sub-features;
the determining the feature gain corresponding to the object feature comprises the following steps:
for each sub-feature, determining the weight corresponding to the sub-feature according to the feature value of the sub-feature;
and calculating the weighted sum of the sub-feature gains corresponding to each sub-feature according to the weight corresponding to each sub-feature to obtain the feature gain.
7. The method of claim 1, wherein the determining whether to push the information to be pushed using the preset channel according to the result indicator gain comprises:
Acquiring a current gain threshold;
and if the result index gain is larger than the current gain threshold, determining to push the information to be pushed by adopting the preset channel.
8. The method of claim 7, wherein the obtaining the current gain threshold comprises:
acquiring a history result index gain of pushing the second history push information by adopting the preset channel aiming at each piece of second history push information in M pieces of recently pushed second history push information;
and determining the median of the maximum K historical result index gains in the M historical result index gains as a current gain threshold, wherein K is smaller than M.
9. A push channel selection device, comprising:
the first determining module is used for determining a result index gain of pushing information to be pushed by adopting a preset channel based on a causal inference model in response to receiving a pushing request, wherein the pushing request comprises the information to be pushed;
and the second determining module is used for determining whether to adopt the preset channel to push the information to be pushed or not according to the result index gain.
10. A computer readable medium on which a computer program is stored, characterized in that the program, when being executed by a processing device, carries out the steps of the method according to any one of claims 1-8.
11. An electronic device, comprising:
a storage device having a computer program stored thereon;
processing means for executing said computer program in said storage means to carry out the steps of the method according to any one of claims 1-8.
CN202311311241.3A 2023-10-10 2023-10-10 Push channel selection method, push channel selection device, push channel selection medium and electronic equipment Pending CN117278617A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311311241.3A CN117278617A (en) 2023-10-10 2023-10-10 Push channel selection method, push channel selection device, push channel selection medium and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311311241.3A CN117278617A (en) 2023-10-10 2023-10-10 Push channel selection method, push channel selection device, push channel selection medium and electronic equipment

Publications (1)

Publication Number Publication Date
CN117278617A true CN117278617A (en) 2023-12-22

Family

ID=89212162

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311311241.3A Pending CN117278617A (en) 2023-10-10 2023-10-10 Push channel selection method, push channel selection device, push channel selection medium and electronic equipment

Country Status (1)

Country Link
CN (1) CN117278617A (en)

Similar Documents

Publication Publication Date Title
CN110658961B (en) Information display method and device and electronic equipment
CN110781373B (en) List updating method and device, readable medium and electronic equipment
CN111246228B (en) Method, device, medium and electronic equipment for updating gift resources of live broadcast room
CN117241092A (en) Video processing method and device, storage medium and electronic equipment
CN110737846B (en) Social interface recommendation method and device, electronic equipment and storage medium
CN117035842A (en) Model training method, traffic prediction method, device, equipment and medium
CN117278617A (en) Push channel selection method, push channel selection device, push channel selection medium and electronic equipment
CN113660699A (en) Intelligent cluster networking method and device and electronic equipment
CN116820539B (en) System software operation maintenance system and method based on Internet
CN114979757B (en) Video publishing method, device, equipment and storage medium
CN112015625B (en) Alarm device control method, device, electronic device and computer readable medium
CN113010784B (en) Method, apparatus, electronic device and medium for generating prediction information
CN117200942A (en) Data processing method, device, medium and electronic equipment
CN117235535B (en) Abnormal supply end power-off method and device, electronic equipment and medium
CN110532475B (en) Information estimation method and device, electronic equipment and storage medium
WO2024012306A1 (en) Method and apparatus for determining neural network model structure, device, medium, and product
CN117082273A (en) Video playing duration prediction method and device, electronic equipment, medium and product
CN116340632A (en) Object recommendation method, device, medium and electronic equipment
CN116010867A (en) Negative sample determination method, device, medium and electronic equipment
CN117156147A (en) Video transcoding method, device, equipment and storage medium
CN116319932A (en) Training method, device, equipment and storage medium of content push model
CN117473107A (en) Training method, device, equipment and storage medium for media content click model
CN112256307A (en) Gray scale distribution method and device, electronic equipment and computer readable medium
CN116758458A (en) Video analysis method, device, equipment and storage medium
CN116204722A (en) Content recommendation method, device, equipment and medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination