WO2020088050A1

WO2020088050A1 - Information generation method and device

Info

Publication number: WO2020088050A1
Application number: PCT/CN2019/101727
Authority: WO
Inventors: 洪春晓; 谷长胜
Original assignee: 北京字节跳动网络技术有限公司
Priority date: 2018-10-31
Filing date: 2019-08-21
Publication date: 2020-05-07
Also published as: CN111131356A; CN111131356B

Abstract

Embodiments of the present application disclose an information generation method and a device. An example embodiment of the method comprises: acquiring probability information corresponding to each user identifier in a user identifier set, wherein the probability information comprises a first probability being the probability of a user achieving a target conversion goal associated with target push information; acquiring a historical average conversion rate, wherein the historical average conversion rate is associated with the target push information and the target conversion goal; determining a ratio of the first probability corresponding to each user identifier in the user identifier set to the historical average conversion rate; and selecting, on the basis of the ratio, user identifiers from the user identifier set as target user identifiers, and generating a target user identifier set.

Description

Method and device for generating information

This application requires the priority of the Chinese patent application filed on October 31, 2018, with the application number 201811289912.X. The entire content of this application is incorporated by reference in this application.

Technical field

The embodiments of the present application relate to the field of computer technology, such as a method and an apparatus for generating information.

Background technique

Information push, also known as "webcast", is a technology that reduces information overload by pushing certain information on the Internet through certain technical standards or protocols. Information push technology can reduce the time users spend searching on the Internet by actively pushing information to users.

There is interactive push information in the push information in the related art, and the user can select (for example, click) such push information to enter a specific website or open a specific window. In addition, such push information is generally associated with conversion goals, such as activation, payment, phone calls, and so on. In the information push method in related technologies, when target users are selected for the push information, usually the stored user data related to the push information and the conversion target are directly used to predict the probability of the user in the user group under the conversion target , And then select users from the user group as the target users based on the probability. This target user selection method is generally not conducive to the improvement of the push effect of the push information.

Summary of the invention

The embodiments of the present application provide a method and a device for generating information.

In a first aspect, an embodiment of the present application provides a method for generating information. The method includes: obtaining probability information corresponding to each user ID in a user ID set, where the probability information includes a first probability, and the The first probability is the probability that the user achieves the target conversion goal associated with the target push information; obtain the historical average conversion probability, wherein the historical average conversion probability is associated with the target push information and the target conversion goal; determine A ratio of the first probability corresponding to each user ID in the user ID set and the historical average conversion probability; based on the ratio, a user ID is selected from the user ID set as the target user ID to generate the target user ID set.

In a second aspect, an embodiment of the present application provides an apparatus for generating information. The apparatus includes: a first acquiring unit configured to acquire probability information corresponding to each user ID in a user ID set, wherein the probability The information includes a first probability, the first probability is the probability that the user achieves the target conversion goal associated with the target push information; the second acquisition unit is configured to obtain a historical average conversion probability, wherein the historical average conversion probability is The target push information is associated with the target conversion target; a determining unit is configured to determine a ratio of the first probability corresponding to each user identification in the user identification set and the historical average conversion probability; the generating unit Is configured to select a user ID from the set of user IDs as the target user ID based on the ratio, and generate a target user ID set.

In a third aspect, an embodiment of the present application provides an electronic device including: at least one processor; a storage device on which at least one program is stored; when the at least one program is executed by the at least one processor So that the at least one processor implements the method described in any one of the implementation manners of the first aspect.

According to a fourth aspect, an embodiment of the present application provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, a method as described in any implementation manner of the first aspect is implemented.

BRIEF DESCRIPTION

FIG. 1 is an exemplary system architecture diagram to which an embodiment of the present application can be applied;

2 is a flowchart of an embodiment of a method for generating information according to the present application;

3 is a schematic diagram of an application scenario according to the method for generating information according to the present application;

4 is a flowchart of another embodiment of a method for generating information according to the present application;

5 is a schematic structural diagram of an embodiment of an apparatus for generating information according to the present application;

6 is a schematic structural diagram of a computer system suitable for implementing an electronic device according to an embodiment of the present application.

detailed description

The application will be described in detail below with reference to the drawings and embodiments. It can be understood that the example embodiments described herein are only used to explain the related invention, but not to limit the invention. It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings.

It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other if there is no conflict. The present application will be described in detail below with reference to the drawings and in conjunction with the embodiments.

FIG. 1 shows an exemplary system architecture 100 of an embodiment of an information generation method or an information generation apparatus to which the present application can be applied.

As shown in FIG. 1, the system architecture 100 may include

terminal devices

101, 102, 103, a network 104, and a server 105. The network 104 is a medium used to provide a communication link between the

terminal devices

101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, and so on.

The client can use the

terminal devices

101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications may be installed on the

terminal devices

101, 102, and 103, such as web browser applications, applications for customers to configure information for target push information, and so on.

The

terminal devices

101, 102, and 103 may be hardware or software. When the

terminal devices

101, 102, and 103 are hardware, they may be various electronic devices with display screens, including but not limited to smart phones, tablet computers, laptop portable computers, and desktop computers. When the

terminal devices

101, 102, and 103 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (for example to provide distributed services), or as a single software or software module. There is no specific limit here.

The server 105 may be a server that provides various services. For example, the server 105 can support the client to set conversion targets for the target push information through the

terminal devices

101, 102, and 103. The server 105 may also perform corresponding processing operations based on the conversion target.

It should be noted that the method for generating information provided by the embodiments of the present application is generally executed by the server 105. Accordingly, the device for generating information is generally provided in the server 105.

It should be pointed out that the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, to provide distributed services), or as a single software or software module. There is no specific limit here.

It should be understood that the numbers of terminal devices, networks, and servers in FIG. 1 are only schematic. According to the implementation needs, there can be any number of terminal devices, networks and servers.

Referring to FIG. 2, a flow 200 of an embodiment of a method for generating information according to the present application is shown. The process 200 of the method for generating information includes steps 201 to 204.

In step 201, probability information corresponding to each user ID in the user ID set is obtained.

In this embodiment, the execution subject of the method for generating information (for example, the server 105 shown in FIG. 1) may acquire probability information corresponding to each user ID in the user ID set in response to receiving the corresponding information generation instruction. The probability information may include the first probability. The first probability may be the probability that the user achieves the target conversion goal corresponding to the target push information.

The above target conversion target may be a conversion target set by the customer for the above target push information. Conversion goals may include, but are not limited to, download completion, installation completion, activation, payment, form submission, phone dialing, etc., for example. For example, the above-mentioned execution subject can provide a conversion target setting interface to the client, and the client can access the interface through the client (such as the

terminal devices

101, 102, and 103 shown in FIG. 1) according to actual needs, and set the corresponding conversion target for the target push information .

The above target push information may be interactive push information. In an embodiment, the above-mentioned target push information may be push information that is in a cold start phase and is interactive. It should be understood that, for any piece of push information, if the push information has not been pushed to the user group or has only been pushed a limited number of times, and the push information is associated with less user data, the push information may be said to be in a cold start phase . The above target push information can be used to introduce customers' products or services to the user group, for example. The target push information may be push information in various forms (for example, pictures, text links, etc.), and this embodiment does not limit the form of the target push information.

The above information generation instruction may be sent by the client through the client, or may be sent by a server that is connected to the above-mentioned executive body in remote communication, which is not specifically limited herein. In addition, the information generation instruction may include, for example, the information identifier of the target push information and the conversion target identifier of the target conversion target.

The user identification set may be obtained by the execution subject from the connected server in advance, or may be generated by the execution subject in advance, and is not specifically limited herein.

In this embodiment, the execution subject may use various methods to obtain the first probability corresponding to each user ID in the user ID set. As an example, the execution subject may locally run a thread for real-time obtaining the probability of the user indicated by the user identification in the user identification set under the target conversion target. For each user ID in the user ID set, the execution subject may determine the probability of the user indicated by the user ID acquired by the thread most recently under the target conversion target as the first probability corresponding to the user ID.

For another example, the execution subject may also determine the first probability corresponding to each user identification in the user identification set by performing the following probability determination step: acquiring an operation information set, where the operation information set may include the user identification set Operation information corresponding to the user ID in and related to the push information in the category to which the above target push information belongs; parsing the operation information set to determine that the user indicated by each user ID in the user ID set achieves the above target conversion The probability of the target is determined as the first probability corresponding to the user identification.

It should be noted that the category to which the above target push information belongs may be a coarse-grained category or a fine-grained category, which is not specifically limited herein. For example, assuming that the target push information is associated with game software, the category to which the target push information belongs may be software or game software.

For the operation information corresponding to any user ID, the operation information may include the information ID of the push information in the above category that the user indicated by the user ID has clicked, and a series of operations performed after clicking the push information Operation ID. Taking the push information related to game software as an example, a series of operations performed by the user after clicking the push information may include, but not limited to, download, installation, activation, payment, and so on.

It should be understood that the above-mentioned execution subject may use various methods to obtain the above-mentioned operation information set. As an example, the above-mentioned execution subject may obtain a preset first set of user information associated with the category to which the above-mentioned target push information belongs, from a server locally or remotely connected thereto. Wherein, the first user information may include a user ID and operation information corresponding to the user ID and associated with the push information under the category. For each user ID in the user ID set, the above-mentioned execution subject may find the first user information including the user ID in the first user information set, and extract operation information from the first user information. Then, the above-mentioned executive body can compose the extracted pieces of operation information into an operation information set.

For another example, the above-mentioned execution subject may obtain a preset second user information set from a server locally or remotely connected to the server. The second user information may include a user ID and various operation information corresponding to the user ID. For each user ID in the user ID set, the above-mentioned executive body can find out the second user information including the user ID in the second user information set, and extract the attribute belonging to the target push information from the second user information Operation information related to the push information under the category. Then, the above-mentioned execution subject may compose each piece of extracted operation information into the above-mentioned operation information set.

It should be noted that the above-mentioned execution subject can use the target prediction model to parse the operation information set to determine the probability of the user indicated by each user ID in the user ID set under the target conversion target. As an example, the above-mentioned executive body may combine each user identifier in the above-mentioned user identifier set with the information identifier of the above target push information to form an identifier pair. Then, the above-mentioned executive body may input the formed multiple identification pairs and the operation information corresponding to the user identification included in each identification pair of the multiple identification pairs into the target prediction model to obtain a prediction result. The prediction result may include the probability that the user indicated by each user ID in the user ID set is under the target conversion target.

It should be noted that the target prediction model may be a model that is learned in advance using a multi-task learning (MTL) method and is used to make a probabilistic prediction related to the above target conversion target. The input of the target prediction model may include an identification pair and operation information corresponding to the user identification in the identification pair. The identification pair may include a user identification and an information identification for pushing information. The output of the target prediction model may include the probability that the user indicated by the user identification in the input identification pair is under the target conversion target.

The multi-task learning method is a machine learning method based on shared representation, where multiple related tasks are learned together. Multi-task learning involves the simultaneous parallel learning of multiple related tasks and the simultaneous back propagation of gradients. Multiple tasks help each other learn through the underlying shared representation and improve the generalization effect. Since the multi-task learning method is a well-known technology widely researched and applied at present, it will not be repeated here.

In step 202, the historical average conversion probability is obtained.

In this embodiment, the above-mentioned execution subject can also obtain the historical average conversion probability. Wherein, the historical average conversion probability may be the probability associated with the target push information and the target conversion target.

As an example, the above-mentioned execution subject may obtain historical predicted conversion data related to the above-mentioned target push information and the above-mentioned target conversion target recorded in a set historical time period, and calculate the historical average conversion probability. The historical predicted conversion data may include, for example, a user identification group and the probability of the user indicated by each user identification in the user identification group under the above target conversion target. Wherein, each user ID in the user ID group may be included in the above user ID set. The above-mentioned executive body can count the number of user IDs in the user ID group. The above-mentioned execution subject can also calculate the sum of the probabilities corresponding to the multiple user IDs in the user ID group in the historical estimated conversion data. The above-mentioned executive body may determine the ratio of the sum to the number as the historical average conversion rate.

In step 203, the ratio of the first probability corresponding to each user ID in the user ID set and the historical average conversion probability is determined.

In this embodiment, the execution subject may calculate the ratio of the first probability corresponding to each user ID in the user ID set and the historical average conversion probability.

In step 204, based on the ratio, the user ID is selected from the user ID set as the target user ID to generate the target user ID set.

In this embodiment, based on the ratio determined in step 203, the execution subject may select a user ID from the user ID set as the target user ID to generate the target user ID set. For example, the above-mentioned execution subject may combine each selected target user ID into a set of target user IDs.

Here, the above-mentioned executive body may adopt various methods to select the target user identification. As an example, for each user ID in the user ID set, the above-mentioned executive body may determine whether the ratio of the first probability corresponding to the user ID and the historical average conversion probability is lower than a ratio threshold (for example, 1), if not, Then, the above-mentioned execution subject may select the user ID as the target user ID.

For another example, for each user ID in the user ID set, the above-mentioned execution subject may determine whether the ratio of the first probability corresponding to the user ID and the historical average conversion probability is within a preset value interval. If it is within a preset value interval, the above-mentioned execution subject may select the user ID as the target user ID.

It should be understood that the ratio threshold and the preset value interval can be set according to actual needs, and are not specifically limited here.

It should be noted that the target user indicated by the target user ID in the target user ID set may be referred to as a premium user. High-quality users generally have specific information browsing habits or operating habits. In the subsequent information push, by pushing the above target push information to the user terminal of the target user indicated by the target user ID in the target user ID set, the push effect of the above target push information can be improved.

Referring to FIG. 3, FIG. 3 is a schematic diagram of an application scenario of the method for generating information according to this embodiment. In the application scenario of FIG. 3, the ratio threshold is 1. Customer C has set a paid conversion target for target push message B. The target push information B may be push information used to introduce the game software of customer C to the user group. The user identification set may include a user identification A1 and a user identification A2. The server (as indicated by reference number 301) may obtain the first probabilities corresponding to the user identification A1 and the user identification A2, respectively, in response to receiving the information generation instruction. The information generation instruction may include, for example, the information identification of the target push information B and the conversion target identification of the paid conversion target. The first probability may be the probability that the user uses customer C's game software to pay. Here, it is assumed that the first probability corresponding to the user identification A1 acquired by the server is 0.4 (as indicated by reference number 302), and the first probability corresponding to the user identification A2 is 0.6 (as indicated by reference number 303).

Then, the above-mentioned server may also calculate the historical average conversion probability based on the historical estimated conversion data associated with the target push information B and the paid conversion target acquired before the above-mentioned process 200 is executed this time. The historical estimated conversion data may include, for example, the user identification group and the probability that the user indicated by the user identification in the user identification group uses the above-mentioned game software of customer C to pay. Each user ID in the user ID group may be included in the user ID set. Here, it is assumed that the historical average conversion probability calculated by the server is 0.5 (as indicated by reference number 304).

Then, the above server can also calculate a ratio 0.8 of the first probability 0.4 corresponding to the user identification A1 and the historical average conversion probability 0.5 (as indicated by reference numeral 305). The above server can also calculate a ratio of the first probability 0.6 corresponding to the user identification A2 and the historical average conversion probability 0.5 of 1.2 (as indicated by reference number 306).

Finally, since the ratio 1.2 is not lower than the ratio threshold 1, the server may select the user ID A2 from the user ID set as the target user ID, and generate a target user ID set including the user ID A2 (as indicated by reference number 307).

The method provided in the above embodiment of the present application obtains a first probability corresponding to each user ID in the user ID set, where the first probability is the probability that the user currently achieves the target conversion target corresponding to the target push information; Then obtain the historical average conversion probability, where the historical average conversion probability is associated with the user identification set, the target push information, and the target conversion target; then determine the first probability and the historical average conversion probability corresponding to each user identification in the user identification set Based on the ratio, in order to select the user ID from the user ID set as the target user ID and generate the target user ID set. The generation of the target user identification set with higher effectiveness is realized, which is beneficial to the promotion of the push effect of the target push information.

Referring to FIG. 4, it shows a flow 400 of yet another embodiment of a method of generating information. The process 400 of the method for generating information includes steps 401 to 405.

In step 401, a first probability and a second probability corresponding to each user ID in the user ID set are obtained.

In this embodiment, the execution subject of the method for generating information (for example, the server 105 shown in FIG. 1) may obtain the first probability corresponding to each user ID in the user ID set in response to receiving the corresponding information generation instruction And second probability. The first probability may be the probability that the user achieves the target conversion goal corresponding to the target push information. The second probability may be the probability that the user clicks on the target to push the information.

The above target conversion target may be a conversion target set by the customer for the above target push information. Conversion goals may include, but are not limited to, download completion, installation completion, activation, payment, form submission, phone dialing, etc., for example. The above target conversion target can correspond to the expected cost. The unit of expected cost may be yuan. The expected cost can be set by the customer for the above target conversion goal. The expected cost may be the cost that the customer is willing to pay for a conversion. For example, the above-mentioned execution subject can provide a conversion target setting interface to the client, and the client can access the interface through the client (such as the

terminal devices

101, 102, and 103 shown in FIG. 1) to set the corresponding conversion target for the target push information and The conversion goal sets the corresponding expected cost.

The above target push information may be interactive push information. In an embodiment, the target push information may be push information that is in a cold start phase and is interactive. It should be understood that, for any piece of push information, if the push information has not been pushed to the user group or has only been pushed a limited number of times, and the push information is associated with less user data, the push information may be said to be in a cold start phase . The above target push information can be used to introduce customers' products or services to the user group, for example. The target push information may be push information in various forms (for example, pictures, text links, etc.), and this embodiment does not limit the form of the target push information.

In this embodiment, the above-mentioned execution subject may use various methods to obtain the first probability and the second probability corresponding to each user ID in the above-mentioned user ID set. Here, for the method of acquiring the first probability, reference may be made to the relevant description of step 201 in the embodiment corresponding to FIG. 2, and details are not described herein again. In addition, the thread in the embodiment corresponding to FIG. 2 may also be used to obtain in real time the probability that the user indicated by the user ID in the user ID set clicks on the target push information. For each user ID in the user ID set, the execution subject may determine the probability that the user indicated by the user ID acquired by the thread most recently clicks on the target push information as the second probability corresponding to the user ID.

In an embodiment, the target prediction model in the embodiment corresponding to FIG. 2 may be associated with the click target while linking the above target conversion target. The probability under the click target can be referred to as the probability that the user clicks the corresponding push information. Therefore, the output of the target prediction model includes the probability that the user indicated by the user identification in the input identification pair is under the target conversion target, and may also include the probability of the user under the click target. Therefore, while using the target prediction model to obtain the first probability corresponding to each user ID in the user ID set, the execution subject can also obtain the second probability corresponding to the user ID.

In step 402, the historical average conversion probability is obtained.

In step 403, the ratio of the first probability corresponding to each user ID in the user ID set and the historical average conversion probability is determined.

In step 404, based on the ratio, the user identification is selected from the user identification set as the target user identification to generate the target user identification set.

In this embodiment, for the explanation of step 402, step 403, and step 404, reference may be made to the relevant descriptions of step 202, step 203, and step 204 in the embodiment corresponding to FIG. 2, and details are not described herein again.

In step 405, for each target user ID in the set of target user IDs, a generating step is performed for each target user ID to generate value information corresponding to each target user ID.

In this embodiment, for each target user ID in the set of target user IDs, the execution subject may perform a generating step for each target user ID to generate value information corresponding to each target user ID.

Here, the above-mentioned execution subject may use each target user ID as the user ID to be processed, and perform the following generation steps:

In the first step, the above-mentioned execution subject may perform sampling in a normal distribution based on the first probability corresponding to the user identifier to be processed to obtain a sample value. Here, the first probability corresponding to the user identifier to be processed is represented as P. The normal distribution followed by the sampled value can be expressed as:

Among them, N can represent a normal distribution (Normal distribution). Click can represent the number of times the target user indicated by the to-be-processed user ID clicks on each piece of preset information. The preset information may be information set in advance according to the purpose of the aforementioned target push information. Assuming that the above target push information is push information used to introduce customers' shopping applications to the user group, the preset information may be, for example, item information. The user can open the corresponding item information interface in the shopping application by clicking the item information, for example. It should be understood that the preset information can be set according to actual needs, and is not specifically limited here.

Step 2: The execution subject may determine the value of the target user indicated by the user ID to be processed based on the expected cost, the sample value, and the first probability and the second probability corresponding to the user ID to be processed.

As an example, the execution subject may determine the first value corresponding to the user ID to be processed based on the expected cost and the first probability and the second probability corresponding to the user ID to be processed. The execution subject may also determine the second value corresponding to the user identifier to be processed based on the expected cost, the sample value, and the second probability corresponding to the user identifier to be processed. Then the above-mentioned executive body may determine the sum of the first value and the second value as the value of the target user indicated by the user ID to be processed.

Here, the execution subject may determine the product of the first cost and the first probability and the second probability corresponding to the user identifier to be processed as the first value. For example, the execution subject may determine the product of the expected cost, the sampled value, and the second probability corresponding to the user ID to be processed as the second value.

It should be noted that the above sampling value may be a value for measuring the exploration value. The above-mentioned first value may be understood as the actual cost value of the target user indicated by the to-be-processed user identifier. The actual cost value may be the fee that the service provider of the information push service used by the customer (for example, the above-mentioned executive body) should subsequently charge the customer for the target user. The above second value can be understood as the exploration value of the target user. The above service provider generally does not need to charge the customer an amount of the exploration value for the target user.

In an embodiment, in order to prevent the customer from cheating, the above-mentioned executive body may determine the above-mentioned first value and the above-mentioned second value using the following determination steps: determine the product of the above-mentioned sample value and the first preset value (for example, 0.5), based on The product of the value and the first preset value, the expected cost and the first probability and the second probability corresponding to the user ID to be processed determine the first value; determine the second preset value (for example, 1) and the first The difference of the preset values; determine the product of the sampled value and the difference, and determine the second value based on the product of the sampled value and the difference, the expected cost, and the second probability corresponding to the user ID to be processed . Among them, the first preset value can be used to control the exploration range and is a hyperparameter. The first preset value and the second preset value can be set according to actual needs, and are not specifically limited herein.

In an embodiment, for the first value, the execution subject may first calculate the product of the sampled value and the first preset value, and then calculate the product and the first probability corresponding to the user ID to be processed. Sum, and then calculate the product of the sum, the second probability corresponding to the user ID to be processed and the expected cost, and determine the product as the first value.

For the above-mentioned second value, the executive body may first calculate the difference between the above-mentioned second preset value and the above-mentioned first preset value, and then calculate the product of the difference and the above-mentioned sample value, and then calculate the product, the above The product of the expected cost and the second probability corresponding to the to-be-processed user ID is determined as the second value.

It should be noted that, when determining the first value, by using the sampled value and the first preset value to participate in the calculation, the calculated first value can include a certain exploration value, which can enable the customer to carry out the cost in the future When paying, pay a certain premium user exploration fee.

Step 3: The execution subject may generate value information corresponding to the user ID to be processed based on the value of the target user indicated by the user ID to be processed. The value information can be used to indicate the value of the target user. The value information may include, for example, the user identification to be processed and the value of the target user.

In some implementations of this embodiment, the execution subject may determine the sum of the first values corresponding to the target user IDs in the target user ID set as a total bid, generate feedback information based on the total bid, and convert This feedback information is sent to the client's client. The total bid may be the total cost that the customer should pay for the target conversion goal. The feedback information can be used to feed back to the customer that the total cost that should be paid for the target conversion goal is the total bid. In addition, the feedback information may include, for example, the information identifier of the target push information, the conversion target name of the target conversion target, and the total bid.

As can be seen from FIG. 4, compared with the embodiment corresponding to FIG. 2, the process 400 of the method for generating information in this embodiment highlights the determination of the target user indicated by each target user ID in the set of target user IDs Value, and the step of generating value information corresponding to the target user identification based on the value. Therefore, the solution described in this embodiment can realize the exploration of high-quality users and the mining of the value of high-quality users.

In an implementation of the method for generating information provided by the embodiments of the present application, the execution subject of the method (for example, the server 105 shown in FIG. 1) may also push the target push information to the target in the target user identification set The user terminal of the target user indicated by the user identification (for example, each target user identification) to improve the push effect of the target push information.

Referring to FIG. 5, as an implementation of the method shown in the above figures, the present application provides an embodiment of an apparatus for generating information, which corresponds to the method embodiment shown in FIG. 2, and the apparatus can be applied Used in various electronic devices.

As shown in FIG. 5, the information generating apparatus 500 of this embodiment includes: the first obtaining unit 501 is configured to obtain probability information corresponding to each user ID in the user ID set, where the probability information may include the first probability , The first probability is the probability that the user achieves the target conversion goal associated with the target push information; the second obtaining unit 502 is configured to obtain a historical average conversion probability, where the historical average conversion probability may be related to the target push information and the target conversion goal The determination unit 503 is configured to determine the ratio of the first probability corresponding to each user ID in the user ID set and the historical average conversion probability; the generating unit 504 is configured to select the user ID from the user ID set based on the ratio Target user ID, to generate a set of target user IDs.

In this embodiment, in the apparatus 500 for generating information: the processing of the first acquiring unit 501, the second acquiring unit 502, the determining unit 503, and the generating unit 504 can refer to

steps

201 and 202 in the corresponding embodiment of FIG. The relevant descriptions of step 203 and step 204 will not be repeated here.

In some implementations of this embodiment, the target conversion target may correspond to the expected cost in advance, the probability information may further include a second probability, and the second probability may be the probability that the user clicks on the target push information; and the above apparatus 500 may further include: A generating unit (not shown in the figure) is configured to, for each target user ID in the set of target user IDs, take each target user ID as the user ID to be processed, and perform the following generation step: based on the user ID Corresponding first probability, normal sampling is performed to obtain the sampled value; based on the expected cost, the sampled value and the first probability and the second probability corresponding to the user ID to be processed, the target user ’s target indicated by the user ID to be processed is determined Value; generate value information corresponding to the user identification to be processed based on the value.

In some implementations of this embodiment, the first generating unit may include: a first determining subunit (not shown in the figure) configured to be based on the first cost and the second probability corresponding to the user ID to be processed based on the expected cost Probability, to determine the first value corresponding to the user ID to be processed; the second determination subunit (not shown in the figure) is configured to determine the value based on the expected cost, the sample value, and the second probability corresponding to the user ID to be processed The second value corresponding to the to-be-processed user identifier; the third determining subunit (not shown in the figure) is configured to determine the sum of the first value and the second value as the value of the target user.

In some implementations of this embodiment, the first determining subunit may be configured to: determine the product of the sample value and the first preset value, based on the product of the sample value and the first preset value, the expected cost, and the user to be processed Identify the first probability and the second probability corresponding to it, and determine the first value.

In some implementations of this embodiment, the second determining subunit may be configured to: determine the difference between the second preset value and the first preset value; determine the product of the sample value and the difference value, based on the sample value and the difference The product of the value, the expected cost, and the second probability corresponding to the user ID to be processed determine the second value.

In some implementations of this embodiment, the foregoing apparatus 500 further includes: a second generation unit (not shown in the figure) configured to assign a first value corresponding to each of the target user IDs in the target user ID set The sum of is determined as the total bid, and feedback information is generated based on the total bid; the sending unit (not shown in the figure) is configured to send the feedback information to the client's client associated with the target push information.

The device provided by the above embodiment of the present application obtains probability information corresponding to each user ID in the user ID set, wherein the probability information includes a first probability, and the first probability is that the user achieves the target associated with the target push information Probability of conversion target; then obtain the historical average conversion probability, where the historical average conversion probability is associated with the target push information and the target conversion target; and then determine the first probability and historical average conversion corresponding to each user ID in the user ID set Probability ratio, so that based on the ratio, the user ID is selected from the user ID set as the target user ID to generate the target user ID set. The generation of the target user identification set with higher effectiveness is realized, which is beneficial to the promotion of the push effect of the target push information.

6, which shows a schematic structural diagram of a computer system 600 suitable for implementing an electronic device (for example, the server 105 shown in FIG. 1) of the embodiment of the present application. The electronic device shown in FIG. 6 is only an example, and should not bring any limitation to the functions and use scope of the embodiments of the present application.

As shown in FIG. 6, the computer system 600 includes a central processing unit (Central Processing Unit, CPU) 601, which can be loaded into a random unit according to a program stored in a read-only memory (Read-Only Memory, ROM) 602 or from the storage section 608 The program in the memory (Random Access Memory, RAM) 603 is accessed to perform various appropriate actions and processes. In the RAM 603, various programs and data necessary for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other through a bus 604. An input / output (Input / Output, I / O) interface 605 is also connected to the bus 604.

The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc .; an output section 607 including a cathode ray tube (Cathode Ray Tube, CRT), liquid crystal display (Liquid Crystal Display, LCD), etc. A storage section 608 including a hard disk, etc .; and a communication section 609 including a network interface card such as a local area network (Local Area Network, LAN) card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. The driver 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is installed on the drive 610 as necessary, so that the computer program read out therefrom is installed into the storage section 608 as needed.

In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product that includes a computer program carried on a computer-readable medium, the computer program containing program code for performing the method shown in the flowchart. In such an embodiment, the computer program may be downloaded and installed from the network through the communication section 609, or installed from the removable medium 611 in at least one way. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present application are executed.

It should be noted that the computer-readable medium shown in this application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with at least one wire, portable computer disk, hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable only Read memory ((Erasable Programmable Read-Only Memory, EPROM) or flash memory, optical fiber, portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), optical storage device, magnetic storage device, or any suitable of the above In this application, the computer-readable storage medium can be any tangible medium that contains or stores a program, which can be used by or in combination with the instruction execution system, apparatus, or device. In this application, the computer-readable storage medium The signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal may take many forms, including but not limited to electromagnetic signals, optical signals, or the above Any suitable combination. The computer-readable signal medium can also Any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program contained on the computer-readable medium The code may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical cable, radio frequency (Radio Frequency, RF), etc., or any suitable combination of the above.

The computer program code for performing the operations of the present application may be written in one or more programming languages or a combination thereof, the programming languages including object-oriented programming languages such as Java, Smalltalk, C ++, as well as conventional Procedural programming language-such as "C" language or similar programming language. The program code may be executed entirely on the user's computer, partly on the user's computer, as an independent software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In situations involving remote computers, 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 an Internet service provider Internet connection).

The flowchart and block diagrams in the drawings illustrate the possible implementation architecture, functions, and operations of the system, method, and computer program product according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains at least one Execute instructions. 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 represented in succession may actually be executed in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and a combination of blocks in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be used It is realized by a combination of dedicated hardware and computer instructions.

The units described in the embodiments of the present application may be implemented in software or hardware. The described unit may also be provided in the processor. For example, it may be described as: a processor includes a first acquisition unit, a second acquisition unit, a determination unit, and a generation unit. Among them, the names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the first obtaining unit may also be described as a unit that obtains probability information corresponding to each user ID in the user ID set. ".

As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the foregoing embodiment; or may exist alone without being assembled into the electronic device in. The computer-readable medium carries one or more programs. When the one or more programs are executed by one of the electronic devices, the electronic device causes the electronic device to acquire probability information corresponding to each user ID in the user ID set, where , The probability information may include a first probability, and the first probability may be the probability that the user achieves the target conversion goal associated with the target push information; obtain the historical average conversion probability, where the historical average conversion probability may be the same as the target push information and the target conversion goal Correlation; determine the ratio of the first probability corresponding to each user ID in the user ID set and the historical average conversion probability; based on the ratio, select the user ID from the user ID set as the target user ID to generate the target user ID set.

Claims

A method of generating information, including:

Acquiring probability information corresponding to each user ID in the user ID set, wherein the probability information includes a first probability, and the first probability is a probability that the user reaches a target conversion target associated with the target push information;

Obtaining a historical average conversion probability, wherein the historical average conversion probability is associated with the target push information and the target conversion target;

Determining the ratio of the first probability corresponding to each user ID in the user ID set and the historical average conversion probability;

Based on the ratio, a user ID is selected as the target user ID from the user ID set to generate a target user ID set.
The method according to claim 1, wherein the target conversion target corresponds to an expected cost in advance, and the probability information further includes a second probability, and the second probability is a probability that the user clicks on the target push information;

The method also includes:

For each target user ID in the set of target user IDs, taking each target user ID as the user ID to be processed, the following generation process is performed:

Based on the first probability corresponding to the to-be-processed user ID, sampling in a normal distribution to obtain a sample value;

Determine the value of the target user indicated by the user ID to be processed based on the expected cost, the sampled value and the first probability and the second probability corresponding to the user ID to be processed;

Generate value information corresponding to the user ID to be processed based on the value.
The method according to claim 2, wherein the determination of the to-be-processed user identifier indicates based on the expected cost, the sampled value, and the first probability and the second probability corresponding to the to-be-processed user identity The value of target users includes:

Determine the first value corresponding to the user ID to be processed based on the expected cost and the first probability and the second probability corresponding to the user ID to be processed;

Determine a second value corresponding to the user ID to be processed based on the expected cost, the sampled value and the second probability corresponding to the user ID to be processed;

The sum of the first value and the second value is determined as the value of the target user.
The method according to claim 3, wherein the first value and the second probability corresponding to the user ID to be processed are determined based on the expected cost and the first probability corresponding to the user ID to be processed, include:

Determine the product of the sampled value and the first preset value, based on the product of the sampled value and the first preset value, the expected cost, and the first probability and the second probability corresponding to the user identification to be processed, Determine the first value.
The method according to claim 4, wherein the second value corresponding to the user ID to be processed is determined based on the expected cost, the sampled value, and the second probability corresponding to the user ID to be processed ,include:

Determine the difference between the second preset value and the first preset value;

Determine the product of the sampled value and the difference, and determine the second value based on the product of the sampled value and the difference, the expected cost, and the second probability corresponding to the user ID to be processed .
The method according to any one of claims 3-5, further comprising:

Determining the sum of the first values corresponding to the plurality of target user identifiers in the target user identifier set as a total bid, and generating feedback information based on the total bid;

Send the feedback information to the client's client associated with the target push information.
A device for generating information, including:

The first obtaining unit is configured to obtain probability information corresponding to each user ID in the user ID set, wherein the probability information includes a first probability, and the first probability is that the user achieves the association with the target push information Probability of goal conversion goal;

The second obtaining unit is configured to obtain a historical average conversion probability, wherein the historical average conversion probability is associated with the target push information and the target conversion target;

The determining unit is configured to determine the ratio of the first probability corresponding to each user ID in the user ID set and the historical average conversion probability;

The generating unit is configured to select a user ID from the user ID set as the target user ID based on the ratio, and generate a target user ID set.
The apparatus according to claim 7, wherein the target conversion target corresponds to an expected cost in advance, and the probability information further includes a second probability, and the second probability is a probability that the user clicks on the target push information;

The device also includes:

The first generation unit is configured to perform the following generation process for each target user ID in the set of target user IDs, using each target user ID as the user ID to be processed:

Based on the first probability corresponding to the to-be-processed user ID, sampling in a normal distribution to obtain a sample value;

Determine the value of the target user indicated by the user ID to be processed based on the expected cost, the sampled value and the first probability and the second probability corresponding to the user ID to be processed;

Generate value information corresponding to the user ID to be processed based on the value.
The apparatus according to claim 8, wherein the first generating unit comprises:

The first determining subunit is configured to determine a first value corresponding to the user ID to be processed based on the expected cost and the first probability and the second probability corresponding to the user ID to be processed;

The second determining subunit is configured to determine a second value corresponding to the user ID to be processed based on the expected cost, the sample value, and the second probability corresponding to the user ID to be processed;

The third determining subunit is configured to determine the sum of the first value and the second value as the value of the target user.
The apparatus according to claim 9, wherein the first determining subunit is configured to:

Determine the product of the sampled value and the first preset value, based on the product of the sampled value and the first preset value, the expected cost, and the first probability and the second probability corresponding to the user identification to be processed, Determine the first value.
The apparatus according to claim 10, wherein the second determining subunit is configured to:

Determine the difference between the second preset value and the first preset value;

Determine the product of the sampled value and the difference, and determine the second value based on the product of the sampled value and the difference, the expected cost, and the second probability corresponding to the user ID to be processed .
The device according to any one of claims 9-11, further comprising:

The second generating unit is configured to determine the sum of the first values corresponding to the plurality of target user identifiers in the target user identifier set as a total bid, and generate feedback information based on the total bid;

The sending unit is configured to send the feedback information to the client of the client associated with the target push information.
An electronic device, including:

A processor

A storage device on which a program is stored,

When the one program is executed by the one processor, the one processor implements the method according to any one of claims 1-6.
A computer-readable medium having a computer program stored thereon, wherein the program according to any one of claims 1-6 is implemented when the program is executed by a processor.