CN111125574A - Method and apparatus for generating information - Google Patents

Method and apparatus for generating information Download PDF

Info

Publication number
CN111125574A
CN111125574A CN201811290026.9A CN201811290026A CN111125574A CN 111125574 A CN111125574 A CN 111125574A CN 201811290026 A CN201811290026 A CN 201811290026A CN 111125574 A CN111125574 A CN 111125574A
Authority
CN
China
Prior art keywords
information
type
displayed
cost value
compensation
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.)
Granted
Application number
CN201811290026.9A
Other languages
Chinese (zh)
Other versions
CN111125574B (en
Inventor
谷长胜
杨希旺
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.)
Beijing ByteDance Network Technology Co Ltd
Original Assignee
Beijing ByteDance Network Technology 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 Beijing ByteDance Network Technology Co Ltd filed Critical Beijing ByteDance Network Technology Co Ltd
Priority to CN201811290026.9A priority Critical patent/CN111125574B/en
Priority to PCT/CN2019/102176 priority patent/WO2020088058A1/en
Publication of CN111125574A publication Critical patent/CN111125574A/en
Application granted granted Critical
Publication of CN111125574B publication Critical patent/CN111125574B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/957—Browsing optimisation, e.g. caching or content distillation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Electrically Operated Instructional Devices (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

The embodiment of the application discloses a method and a device for generating information. One embodiment of the method comprises: acquiring at least one first type of information to be displayed and user information of a target user, which are to be pushed to a terminal of the target user to be displayed on the terminal; for first type information to be displayed in the obtained at least one first type information to be displayed, obtaining characteristic information of the first type information to be displayed; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user; and generating a first type of display cost value of the first type of information to be displayed based on the obtained compensation display cost value and the obtained concerned scores. The method and the device improve the accuracy of generating the first-type display cost value and are beneficial to improving the pertinence of displaying different types of information to be displayed to users.

Description

Method and apparatus for generating information
Technical Field
The embodiment of the application relates to the technical field of computers, in particular to a method and a device for generating information.
Background
The existing method for displaying information is to display various types of information in corresponding display bits.
With the development of internet technology, users browse information more and more frequently through terminals, and in order to display different types of display information on user terminals more specifically, different types of display information are usually displayed in different display areas of terminal equipment respectively. In order to push the presentation information to the user as pointedly as possible, the presentation cost values of the presentation information (i.e., numerical values representing the cost of the provider of the presentation information for pushing the presentation information to the user) are generally obtained, so that the ranking of the presentation information is determined according to the presentation cost values.
Disclosure of Invention
The embodiment of the application provides a method and a device for generating information.
In a first aspect, an embodiment of the present application provides a method for generating information, where the method includes: acquiring at least one first type of information to be displayed and user information of a target user, which are to be pushed to a terminal of the target user to be displayed on the terminal; for first type information to be displayed in the obtained at least one first type information to be displayed, obtaining characteristic information of the first type information to be displayed; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user; and generating a first type of display cost value of the first type of information to be displayed based on the obtained compensation display cost value and the obtained concerned scores.
In some embodiments, the method further comprises: acquiring at least one preset second type of information to be displayed and a preset second type of display cost value corresponding to the second type of information to be displayed; combining the obtained at least one second type display cost value and the obtained at least one first type display cost value into a display cost value set, and combining the at least one first type information to be displayed and the at least one second type information to be displayed into an information set to be displayed; sorting the display cost values in the display cost value set based on the display cost values in the display cost value set; and selecting a display cost value from the sorted display cost value set, and pushing the information to be displayed, which corresponds to the selected display cost value and is contained in the information set to be displayed, to the terminal.
In some embodiments, the scoring model of interest includes at least two scoring sub-models and a result generation sub-model; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user, wherein the concerned score comprises the following steps: for a scoring submodel of at least two scoring submodels, inputting the acquired characteristic information and user information into the scoring submodel to obtain a score corresponding to the scoring submodel; and generating a submodel according to the obtained grading input result to obtain a concerned grade representing the degree of the first type of information to be displayed concerned by the target user, wherein the result generation submodel is used for weighting and summing the obtained grades based on preset weight to obtain the concerned grade.
In some embodiments, for a scoring sub-model of the at least two scoring sub-models, the scoring sub-model is trained in advance as follows: acquiring a training sample set corresponding to the grading sub-model, wherein the training sample comprises preset characteristic information of first-class information to be displayed of the sample, sample user information of a sample user who browses the first-class information to be displayed of the sample, and pre-labeled labeling information used for representing whether the sample user performs target operation on the first-class information to be displayed of the sample; and by utilizing a machine learning method, taking the characteristic information and the sample user information included in the training samples in the obtained training sample set as input, taking the input characteristic information and the marking information corresponding to the sample user information as expected output, and training to obtain the scoring sub-model.
In some embodiments, the compensation presentation cost value used to generate the first type of presentation cost value is obtained in advance by:
acquiring a preset proportion set, wherein the proportion is the proportion of the number of first type information to be displayed and the number of second type information to be displayed, which are pushed to a terminal used by a user in the preset user set within a preset time period; determining the number of lost users corresponding to the proportion in the proportion set, wherein the lost users are users who push the first type of information to be displayed and the second type of information to be displayed according to the proportion within a preset time period, and do not use a terminal to receive the pushed first type of information to be displayed and the pushed second type of information to be displayed after the preset time period; determining a supplementary cost value corresponding to the proportion based on the determined number of the lost users, wherein the supplementary cost value is used for representing the cost paid by the users which are added by the same amount as the lost users; determining the number of users who use the terminal to browse the first type of information to be displayed pushed according to the proportion in a preset time period, dividing the determined supplementary cost value by the determined number, and determining the obtained quotient as the compensation display cost value corresponding to the proportion; selecting the compensation presentation cost value from the determined compensation presentation cost values as a compensation presentation cost value used for generating the first type of presentation cost value.
In some embodiments, selecting a compensation presentation cost value from the determined compensation presentation cost values as the compensation presentation cost value used to generate the first type of presentation cost value comprises: for the compensation display cost values in the determined compensation display cost values, determining the sum of the second type of display cost values of the second type of information to be displayed, which is pushed according to the proportion corresponding to the compensation display cost values, as a second type of total display cost values corresponding to the compensation display cost values; determining a difference value between the second type total display cost value and a supplementary cost value corresponding to the compensation display cost value; and selecting a maximum value from the determined difference values, and taking the compensation display cost value corresponding to the maximum value as the compensation display cost value used for generating the first type of display cost value.
In a second aspect, an embodiment of the present application provides an apparatus for generating information, where the apparatus includes: the terminal comprises a first acquisition unit, a second acquisition unit and a display unit, wherein the first acquisition unit is configured to acquire at least one first type of information to be displayed and user information of a target user, which are to be pushed to the terminal of the target user to be displayed on the terminal; the generating unit is configured to acquire feature information of the first type of information to be displayed for the first type of information to be displayed in the acquired at least one first type of information to be displayed; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user; and generating a first type of display cost value of the first type of information to be displayed based on the obtained compensation display cost value and the obtained concerned scores.
In some embodiments, the apparatus further comprises: the second acquisition unit is configured to acquire at least one preset second type of information to be displayed and a second type of display cost value corresponding to the preset second type of information to be displayed; a merging unit configured to merge the obtained at least one second-type presentation cost value and the obtained at least one first-type presentation cost value into a presentation cost value set, and merge the at least one first-type information to be presented and the at least one second-type information to be presented into an information set to be presented; the sorting unit is configured to sort the display cost values in the display cost value set based on the display cost values in the display cost value set; and the pushing unit is configured to select the display cost value from the sorted display cost value set and push the information to be displayed, which is contained in the information set to be displayed and corresponds to the selected display cost value, to the terminal.
In some embodiments, the scoring model of interest includes at least two scoring sub-models and a result generation sub-model; and the generating unit includes: the first generation module is configured to input the acquired feature information and the user information into the scoring submodel of the at least two scoring submodels to obtain a score corresponding to the scoring submodel; and the second generation module is configured to generate a sub-model according to the obtained grading input result, and obtain a concerned grade representing the degree of the first type of information to be displayed, which is concerned by the target user, wherein the result generation sub-model is used for weighting and summing the obtained grades based on preset weight, so as to obtain the concerned grade.
In some embodiments, for a scoring sub-model of the at least two scoring sub-models, the scoring sub-model is trained in advance as follows: acquiring a training sample set corresponding to the grading sub-model, wherein the training sample comprises preset characteristic information of first-class information to be displayed of the sample, sample user information of a sample user who browses the first-class information to be displayed of the sample, and pre-labeled labeling information used for representing whether the sample user performs target operation on the first-class information to be displayed of the sample; and by utilizing a machine learning method, taking the characteristic information and the sample user information included in the training samples in the obtained training sample set as input, taking the input characteristic information and the marking information corresponding to the sample user information as expected output, and training to obtain the scoring sub-model.
In some embodiments, the compensation presentation cost value used to generate the first type of presentation cost value is obtained in advance by: acquiring a preset proportion set, wherein the proportion is the proportion of the number of first type information to be displayed and the number of second type information to be displayed, which are pushed to a terminal used by a user in the preset user set within a preset time period; determining the number of lost users corresponding to the proportion in the proportion set, wherein the lost users are users who push the first type of information to be displayed and the second type of information to be displayed according to the proportion within a preset time period, and do not use a terminal to receive the pushed first type of information to be displayed and the pushed second type of information to be displayed after the preset time period; determining a supplementary cost value corresponding to the proportion based on the determined number of the lost users, wherein the supplementary cost value is used for representing the cost paid by the users which are added by the same amount as the lost users; determining the number of users who use the terminal to browse the first type of information to be displayed pushed according to the proportion in a preset time period, dividing the determined supplementary cost value by the determined number, and determining the obtained quotient as the compensation display cost value corresponding to the proportion; selecting the compensation presentation cost value from the determined compensation presentation cost values as a compensation presentation cost value used for generating the first type of presentation cost value.
In some embodiments, selecting a compensation presentation cost value from the determined compensation presentation cost values as the compensation presentation cost value used to generate the first type of presentation cost value comprises: for the compensation display cost values in the determined compensation display cost values, determining the sum of the second type of display cost values of the second type of information to be displayed, which is pushed according to the proportion corresponding to the compensation display cost values, as a second type of total display cost values corresponding to the compensation display cost values; determining a difference value between the second type total display cost value and a supplementary cost value corresponding to the compensation display cost value; and selecting a maximum value from the determined difference values, and taking the compensation display cost value corresponding to the maximum value as the compensation display cost value used for generating the first type of display cost value.
In a third aspect, an embodiment of the present application provides a server, where the server includes: one or more processors; a storage device having one or more programs stored thereon; when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the method as described in any implementation of the first aspect.
In a fourth aspect, the present application provides a computer-readable medium, on which a computer program is stored, which, when executed by a processor, implements the method as described in any implementation manner of the first aspect.
According to the method and the device for generating information, the first type of information to be displayed and the user information of the target user, which are pushed to the terminal of the target user to be displayed on the terminal, are obtained by the aid of the pre-trained focused score model, focused scores corresponding to the first type of information to be displayed are obtained, compensation display cost values corresponding to the first type of information to be displayed are obtained, and the first type of display cost values of the first type of information to be displayed are obtained on the basis of the focused scores and the compensation display cost values.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is an exemplary system architecture diagram in which one embodiment of the present application may be applied;
FIG. 2 is a flow diagram of one embodiment of a method for generating information, according to an embodiment of the present application;
FIG. 3 is a schematic diagram of an application scenario of a method for generating information according to an embodiment of the present application;
FIG. 4 is a flow diagram of yet another embodiment of a method for generating information according to an embodiment of the present application;
FIG. 5 is a block diagram of one embodiment of an apparatus for generating information according to an embodiment of the present application;
FIG. 6 is a schematic block diagram of a computer system suitable for use in implementing a server according to embodiments of the present application.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
Fig. 1 illustrates an exemplary system architecture 100 to which a method for generating information or an apparatus for generating information of embodiments of the present application may 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 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 may have various communication client applications installed thereon, such as a web browser application, a shopping application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like.
The terminal apparatuses 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices with display screens, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like. When the terminal apparatuses 101, 102, 103 are software, they can be installed in the electronic apparatuses listed above. It may be implemented as multiple pieces of software or software modules (e.g., software or software modules used to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.
The server 105 may be a server that provides various services, such as a background information processing server that provides support for information presented on the terminal devices 101, 102, 103. The background information processing may process the acquired first type of information to be displayed and user information of a user who is to use the terminal device to browse the first type of information to be displayed, and generate a first type of display cost value of the first type of information to be displayed.
It should be noted that the method for generating information provided in the embodiment of the present application is generally performed by the server 105, and accordingly, the apparatus for generating information is generally disposed in the server 105.
The server may be hardware or software. When the server is hardware, it may be implemented as a distributed server cluster formed by multiple servers, or may be implemented as a single server. When the server is software, it may be implemented as multiple pieces of software or software modules (e.g., software or software modules used to provide distributed services), or as a single piece of software or software module. And is not particularly limited herein.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
With continued reference to FIG. 2, a flow 200 of one embodiment of a method for generating information in accordance with the present application is shown. The method for generating information comprises the following steps:
step 201, at least one first type of information to be displayed, which is to be pushed to a terminal of a target user to be displayed on the terminal, and user information of the target user are obtained.
In this embodiment, an executing body (for example, a server shown in fig. 1) of the method for generating information may obtain at least one piece of information to be presented, which is to be pushed to a terminal of a target user, of a first type to be presented and user information of the target user from a remote place or a local place through a wired connection manner or a wireless connection manner. The target user may be a user who is to browse the information for executing the main body push by using a terminal (for example, a terminal device shown in fig. 1) used by the target user. The user information of the target user may be used to characterize characteristics of the target user, including but not limited to at least one of: gender, age, interests, etc. of the target user.
The first type of information to be displayed may be information belonging to a preset category, such as article information, video information, and the like. The first type of information to be presented may include, but is not limited to, at least one of: pictures, text, audio, video, link addresses, etc.
Step 202, for a first type of information to be displayed in the acquired at least one first type of information to be displayed, acquiring characteristic information of the first type of information to be displayed; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user; and generating a first type of display cost value of the first type of information to be displayed based on the obtained compensation display cost value and the obtained concerned scores.
In this embodiment, for the first type of information to be displayed in the at least one first type of information to be displayed acquired in step 201, the executing body may execute the following steps for the first type of information to be displayed:
step 2021, acquiring the feature information of the first type of information to be displayed.
Specifically, the executing body may obtain the characteristic information of the first type of information to be presented from a remote location or a local location. The characteristic information is used for representing the characteristics of the first type of information to be displayed. The first type of characteristics of the information to be presented may include, but is not limited to, at least one of: the title of the first type of information to be displayed, the type of the first type of information to be displayed, the link address included in the first type of information to be displayed and the like. The characteristic information may be information included in the first type of information to be displayed (for example, a title of the first type of information to be displayed), or may be information related to the first type of information to be displayed (for example, type information representing a type to which the first type of information to be displayed belongs, generation time of the first type of information to be displayed, and the like).
Step 2022, inputting the acquired feature information and the user information into a pre-trained focused score model, and obtaining a focused score for representing the degree of focusing on the first type of information to be displayed by the target user.
The focused score can be used for representing the degree of focus of the first type of information to be displayed by the target user. In general, the attention score may be a numerical value of 0 or more and 1 or less, and the larger the attention score is, the greater the degree of attention of the first type of information to be presented by the target user is.
The concerned score model is used for representing the corresponding relation between the characteristic information of the display information, the user information of the user browsing the display information and the concerned score for representing the concerned degree of the display information concerned by the user.
Specifically, the attention score model may be a correspondence table in which a plurality of pieces of feature information, user information, and attention scores are stored, which is previously prepared by a technician based on statistics of a large number of pieces of feature information, user information, and attention scores. The score to be paid attention in the correspondence table may be statistics of scores of the first type of information to be displayed, which are performed in advance by a technician for a user group to which a user represented by the user information belongs (for example, which may be represented by a preset user information set), and a score to be paid attention obtained by calculating (for example, averaging the scores of the users). The execution body may search the corresponding score of interest from the correspondence table according to the input feature information and the user information. The score model of interest may be a model obtained by training an initial model (e.g., a neural network) by a machine learning method based on a preset training sample. By using the concerned scoring model, different concerned scores can be obtained according to the characteristic information and the user information of different first-class information to be displayed, and the concerned scoring model is helpful for improving the accuracy of generating the concerned scores.
In some optional implementations of the present embodiment, the scoring model of interest may include at least two scoring sub-models and a result generation sub-model. Wherein, the submodel in at least two submodels is respectively connected with the result generation model. The score of interest may be obtained as follows:
firstly, for a scoring sub-model in at least two scoring sub-models, inputting the acquired feature information and user information into the scoring sub-model to obtain a score corresponding to the scoring sub-model. The scoring submodel is used for representing the corresponding relation between the characteristic information, the user information and the score.
And then, generating a sub-model by using the obtained scoring input result to obtain a focused score for representing the focused degree of the first type of information to be displayed by the target user. And the result generation submodel is used for weighting and summing the obtained scores based on preset weights to obtain the concerned scores.
In particular, the scoring sub-model may correspond to a preset scoring category. For example, the scoring categories may include, but are not limited to: the score is used for representing the click probability of the user for the first type of information to be displayed corresponding to the characteristic information, the score is used for representing the forwarding probability of the user for the first type of information to be displayed corresponding to the characteristic information, and the score is used for representing the comment probability of the user for the first type of information to be displayed corresponding to the characteristic information. The click probability, the forwarding probability and the comment probability may be probabilities obtained by counting the behaviors of the user in advance. For example, the click probability may be a ratio of the number of the display information clicked by the user in a preset time period to the number of the browsed information, the forwarding probability may be a ratio of the number of the display information forwarded by the user in the preset time period to the number of the browsed information, and the comment probability may be a ratio of the number of the display information commented by the user in the preset time period to the number of the browsed information.
As an example, the score submodel may be a correspondence table in which a plurality of feature information, user information, and scores are stored, which is previously prepared by a technician based on statistics of a large amount of feature information, user information, and scores belonging to score types corresponding to the score submodel. The score in the correspondence table may be obtained by calculating (for example, calculating a ratio of the number of the display information clicked by the user in a preset time period to the number of the browsed information) through statistics performed by a technician on operations (for example, clicking) performed on the first type of information to be displayed by the user represented by the user information in advance. The execution body may look up a corresponding score from the correspondence table according to the input feature information and the user information. The scoring submodel may be a model obtained by training an initial model (for example, a neural network) by a machine learning method based on a preset training sample.
Optionally, for a scoring sub-model of the at least two scoring sub-models, the scoring sub-model may be trained in advance according to the following steps:
firstly, a training sample set corresponding to the scoring sub-model is obtained. The training sample comprises preset characteristic information of the first type of information to be displayed of the sample, sample user information of a sample user who browses the first type of information to be displayed of the sample, and pre-labeled labeling information used for representing whether the sample user performs target operation on the first type of information to be displayed of the sample. For example, the annotation information may be a number, such as "0" indicating that the sample user has not performed the target operation, and "1" indicating that the sample user has performed the target operation. Wherein the target operation may include, but is not limited to, at least one of: click operation, comment operation, forward operation, and the like.
And then, using a machine learning method to input the feature information and the sample user information included in the training samples in the obtained training sample set, using the input feature information and the labeling information corresponding to the sample user information as expected output, and training to obtain the scoring sub-model. It should be noted that the score output by the trained score submodel may be a numerical value between 0 and 1, which represents the probability that the sample user performs the target operation on the first type of information to be displayed of the sample.
In particular, the scoring sub-model may be a model obtained by training an initial model. The initial model may include, but is not limited to, at least one of: FM (Factorization Machine) models, neural network models, and the like. The initial model may be provided with initial parameters, which may be continuously adjusted during the training process. The execution subject training the conversion rate prediction model may calculate a loss value based on a preset loss function, and determine whether the initial model is trained according to the loss value. Here, it should be noted that the loss value can be used to characterize the difference between the actual output and the expected output. In practice, various loss functions preset can be used to calculate the loss value of the actual output relative to the labeled output. For example, the loss value may be calculated using a logarithmic loss function, a cross-entropy loss function, or the like.
Step 2023, obtaining a preset compensation presentation cost value, and generating a first type presentation cost value of the first type of information to be presented based on the obtained compensation presentation cost value and the obtained score to be paid attention to.
Specifically, as an example, the execution subject may multiply the obtained compensation presentation cost value and the obtained score of interest to obtain a first type presentation cost value. Alternatively, the executing entity may multiply the obtained compensation presentation cost value and the obtained score of interest, and then perform other processing (for example, add or subtract the product obtained after the multiplication with a preset value) to obtain the first type presentation cost value. Therefore, technicians can adjust the first type of display cost values, and accordingly, when the information to be displayed in multiple types exists, the sequencing of the information to be displayed in the multiple types can be pertinently adjusted by adjusting the first type of display cost values.
The compensation presentation cost value can be used for representing the cost of a presenter who pushes a certain first type of information set to be presented in order to enable a user to browse the first type of information to be presented in the first type of information set to be presented. The presenter may be a unit or an individual who uses the execution main body to present the first type of information to be presented to the user. For example, assuming that a user receives a first type of information to be presented using a target application installed on a terminal used by the user, the owner of the target application is a pusher. In practice, the pusher pays a certain cost to obtain a new user, and a compensation display cost value can be calculated based on the paid cost. For example, the compensation presentation cost value is obtained by dividing the cost value representing the cost paid by the cost value by the number of users browsing the first type of information to be presented in the first type of information to be presented set.
Generally, the higher the score of interest is, the higher the interest degree of the target user on the first type of information to be displayed is, the higher the obtained first type of display cost value is, the higher the cost for representing that the presenter pushes the first type of information to be displayed to the target user is, and the higher the value for representing the first type of information to be displayed is.
In some optional implementations of this embodiment, the compensation presentation cost value used for generating the first type presentation cost value may be obtained in advance by the execution subject or other electronic device through the following steps:
step one, acquiring a preset proportion set. The ratio is the ratio of the number of the first type of information to be displayed and the number of the second type of information to be displayed, which are pushed to terminals used by users in a preset user group within a preset time period. The user in the user set may be a user who browses information pushed by the execution main body by using a terminal used by the user. The second type of information to be presented may be presentation information of a different category than the first type of presentation information. For example, the first type of display information is article type information, and the second type of display information is product promotion type information.
Step two, for the proportion in the proportion set, the following steps are executed:
first, the number of lost users corresponding to the ratio is determined. The lost user is a user who pushes the first type of information to be displayed and the second type of information to be displayed according to the proportion in a preset time period, and does not use a terminal to receive the pushed first type of information to be displayed and the pushed second type of information to be displayed after the preset time period. As an example, assuming that the preset time period is one month, the first type of information to be displayed and the second type of information to be displayed are pushed to the terminal used by the user in the user set according to the ratio, after one month, some users in the user set do not use the terminal to receive the first type of information to be displayed and the second type of information to be displayed pushed by the execution main body any more, and then the some users lose users.
Then, based on the determined number of lost users, a supplemental cost value corresponding to the ratio is determined. Wherein the supplemental cost value is used to characterize the cost of adding an equal amount of users as lost users. Generally, the information presenter can increase the number of users at a cost. For example, the execution main body pushes the first type of information to be displayed and the second type of information to be displayed to the terminals in which the target applications (i.e., the applications for receiving the first type of information to be displayed and the second type of information to be displayed) are installed in advance, and in general, in order to install the target applications in the terminals in advance, an information pusher needs to pay a certain cost. Generally, a preset per-person supplementary cost value for representing the cost paid by adding one user can be obtained, and the product of the per-person supplementary cost value and the number of lost users is determined as the supplementary cost value corresponding to the proportion. Alternatively, the execution body may store a preset correspondence table representing a correspondence between the number of lost users and the supplemental cost value, and search for the supplemental cost value corresponding to the determined number of lost users from the correspondence table.
And finally, determining the number of users who use the terminal to browse the first type of information to be displayed pushed according to the proportion in a preset time period, dividing the determined supplementary cost value by the determined number, and determining the obtained quotient as the compensation display cost value corresponding to the proportion.
And step three, selecting the compensation display cost value from the determined compensation display cost values as the compensation display cost value used for generating the first type of display cost value.
And after the second step is executed, the compensation display cost value corresponding to the proportion in the proportion set can be obtained. The execution subject may select the compensation presentation cost value from the compensation presentation cost values according to various methods as the compensation presentation cost value used for generating the first type of presentation cost value. As an example, a minimum value or a median number, etc. may be selected from the respective compensation presentation cost values as the compensation presentation cost values used to generate the first type of presentation cost values.
In some optional implementations of this embodiment, the executing entity or other electronic device may determine the compensation presentation cost value used for generating the first type of presentation cost value according to the following steps:
firstly, for the compensation presentation cost values in the determined compensation presentation cost values, determining the sum of the second type presentation cost values of the second type information to be presented, which is pushed according to the proportion corresponding to the compensation presentation cost values, as the second type total presentation cost value corresponding to the compensation presentation cost values. And determining the difference value between the second type total display cost value and the supplementary cost value corresponding to the compensation display cost value.
Generally, a provider of a second type of information to be displayed (for example, the second type of information to be displayed is used to represent a certain product, and then an owner of the product is a provider of the second type of information to be displayed) needs to pay a certain cost for a user to browse the second type of information to be displayed, each piece of second type of information to be displayed corresponds to a second type of display cost value, and the second type of display cost value is used to represent a cost paid by the provider of the second type of information to be displayed for pushing the second type of information to be displayed to the user. For a certain compensation presentation cost value, adding the second type presentation cost values corresponding to the compensation presentation cost value to obtain a second type total presentation cost value used for representing the total cost paid by the provider of each second type information to be presented.
And then, selecting a maximum value from the determined difference values, and taking a compensation display cost value corresponding to the maximum value as a compensation display cost value used for generating the first type of display cost value. In practice, each provider of the second type of information to be displayed usually pays a cost to the above-mentioned pusher, and the total cost paid by the maximum value representing each provider of the second type of information to be displayed is the largest difference from the total cost paid by the above-mentioned pusher.
With continued reference to fig. 3, fig. 3 is a schematic diagram of an application scenario of the method for generating information according to the present embodiment. In the application scenario of fig. 3, the server 301 first obtains locally three pieces of first information to be presented 3021, 3022, and 3023 to be presented to the terminal of the target user for presentation on the terminal and user information 304 (e.g., gender information, geographical location information, age information, etc.) of the target user. The first type of information to be displayed is news information, including news titles, images and the like. Then, the server 301 obtains feature information (for example, a title, a type name, and the like of the first type information to be displayed) of each first type information to be displayed, which are 3031, 3032, and 3033, respectively. Each feature information and user information 304 is input into the pre-trained attention score model 305 in turn, resulting in attention scores 3061 (e.g., "0.9"), 3062 (e.g., "0.5"), 3063 (e.g., "0.6") for characterizing the degree to which the first type of information to be presented is attended to by the target user. Then, the server 301 obtains a preset compensation presentation cost value 307 (e.g., "0.5"), and multiplies the compensation presentation cost value 307 by each of the scores to be watched, so as to obtain a first-type presentation cost value 3081 (e.g., "0.45"), 3082 (e.g., "0.25"), 3083 (e.g., "0.3") of each first-type information to be presented.
According to the method provided by the embodiment of the application, the first-class display cost value of the first-class information to be displayed is obtained by using the concerned scoring model and the preset compensation display cost value, so that different first-class information to be displayed and different user information can be analyzed, the accuracy of generating the first-class display cost value is improved, and the method is beneficial to improving the pertinence of displaying various information to be displayed to a user under the condition that various information to be displayed needs to be displayed to the user.
With further reference to fig. 4, a flow 400 of yet another embodiment of a method for generating information is shown. The flow 400 of the method for generating information comprises the steps of:
step 401, at least one first type of information to be displayed pushed to a terminal of a target user to be displayed on the terminal and user information of the target user are obtained.
In this embodiment, step 401 is substantially the same as step 201 in the corresponding embodiment of fig. 2, and is not described here again.
Step 402, for a first type of information to be displayed in the obtained at least one first type of information to be displayed, obtaining characteristic information of the first type of information to be displayed; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user; and generating a first type of display cost value of the first type of information to be displayed based on the obtained compensation display cost value and the obtained concerned scores.
In this embodiment, step 402 is substantially the same as step 202 in the corresponding embodiment of fig. 2, and is not described herein again.
Step 403, obtaining at least one preset second type of information to be displayed and a second type of display cost value corresponding to the preset second type of information to be displayed.
In this embodiment, an execution subject (for example, a server shown in fig. 1) of the method for generating information may obtain at least one preset second type of information to be displayed and a second type of display cost value corresponding to the preset second type of information to be displayed from a remote place or a local place through a wired connection manner or a wireless connection manner.
The second type of information to be displayed may be display information of a different category from the first type of display information. For example, the first type of display information is article type information, and the second type of display information is product promotion type information. The second type of display cost value is used for representing the cost (such as price, integral and the like) paid by the provider of the second type of information to be displayed for pushing the second type of information to be displayed to the user, and the larger the second type of display cost value is, the larger the cost paid by the provider of the second type of information to be displayed for pushing the second type of information to be displayed to the user is, and the larger the value of the second type of information to be displayed can be represented.
Step 404, merging the obtained at least one second-type presentation cost value and the obtained at least one first-type presentation cost value into a presentation cost value set, and merging the at least one first-type information to be presented and the at least one second-type information to be presented into an information set to be presented.
In this embodiment, the executing entity may combine the obtained at least one second-type presentation cost value and the obtained at least one first-type presentation cost value into a presentation cost value set, and combine the at least one first-type information to be presented and the at least one second-type information to be presented into an information set to be presented.
And 405, sorting the display cost values in the display cost value set based on the display cost values in the display cost value set.
In this embodiment, the execution subject may sort the presentation cost values in the presentation cost value set based on the size of the presentation cost values in the presentation cost value set.
Specifically, as an example, the execution subjects may be sorted in the order of the presentation cost values from large to small. The first-type display cost values obtained according to the steps 401 to 402 can represent the value of the first-type information to be displayed, so the display cost values in the display cost value set can represent the values of different types of information to be displayed. Therefore, the basis for sequencing different types of information to be displayed can be enriched.
And 406, selecting a display cost value from the sorted display cost value sets, and pushing the information to be displayed, which is contained in the information set to be displayed and corresponds to the selected display cost value, to the terminal.
In this embodiment, the execution subject may first select a presentation cost value from the sorted presentation cost value sets according to various methods. For example, all or part of the presentation cost values may be selected, and when part of the presentation cost values is selected, the part of the presentation cost values arranged at the front may be selected from the sorted presentation cost value sets according to a preset selection number or a preset selection ratio.
Then, the execution subject may push the information to be presented, which is included in the information set to be presented and corresponds to the selected presentation cost value, to the terminal. Specifically, as an example, the execution main body may sort, according to the descending order of the selected presentation cost values, the information to be presented corresponding to each presentation cost value, and sequentially push, according to the sort, each information to be presented to the terminal.
As can be seen from fig. 4, compared with the embodiment corresponding to fig. 2, the flow 400 of the method for generating information in this embodiment highlights a step of selecting and pushing information to be presented from an information set to be presented, which is composed of at least one first type of information to be presented and at least one second type of information to be presented. Therefore, the scheme described in the embodiment can mix and sort different types of information to be displayed, thereby improving the pertinence of information push and being beneficial to displaying various information to be displayed on the user terminal in a targeted and sequential manner.
With further reference 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 which is particularly applicable to various electronic devices.
As shown in fig. 5, the apparatus 500 for generating information of the present embodiment includes: a first obtaining unit 501, configured to obtain at least one first type of information to be presented, which is to be pushed to a terminal of a target user to be presented on the terminal, and user information of the target user; a generating unit 502 configured to acquire, for a first type of information to be displayed in the acquired at least one first type of information to be displayed, feature information of the first type of information to be displayed; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user; and generating a first type of display cost value of the first type of information to be displayed based on the obtained compensation display cost value and the obtained concerned scores.
In this embodiment, the first obtaining unit 501 may obtain, remotely or locally through a wired connection manner or a wireless connection manner, at least one piece of information to be presented to be pushed to a terminal of a target user to be presented on the terminal and user information of the target user. The target user may be a user who is to browse information pushed by the apparatus 500 using a terminal (e.g., a terminal device shown in fig. 1) used by the target user. The user information of the target user may be used to characterize characteristics of the target user, including but not limited to at least one of: gender, age, interests, etc. of the target user.
The first type of information to be displayed may be information belonging to a preset category, such as article information, video information, and the like. The first type of information to be presented may include, but is not limited to, at least one of: pictures, text, audio, video, link addresses, etc.
In this embodiment, for a first type of information to be shown in at least one first type of information to be shown acquired by the first acquiring unit 501, the generating unit 502 may perform the following steps for the first type of information to be shown:
step 5021, obtaining the characteristic information of the first type of information to be displayed.
Specifically, the first obtaining unit 501 may obtain the feature information of the first type of information to be presented from a remote location or a local location. The characteristic information is used for representing the characteristics of the first type of information to be displayed. The first type of characteristics of the information to be presented may include, but is not limited to, at least one of: the title of the first type of information to be displayed, the type of the first type of information to be displayed, the link address included in the first type of information to be displayed and the like. The feature information may be information included in the first type of information to be displayed (for example, a title of the first type of information to be displayed), or may be preset information related to the first type of information to be displayed (for example, type information representing a type to which the first type of information to be displayed belongs, generation time of the first type of information to be displayed, and the like).
Step 5022, inputting the acquired feature information and the user information into a pre-trained concerned score model, and obtaining a concerned score for representing the degree of the first type of information to be displayed concerned by the target user.
The focused score can be used for representing the degree of focus of the first type of information to be displayed by the target user. In general, the attention score may be a numerical value of 0 or more and 1 or less, and the larger the attention score is, the greater the degree of attention of the first type of information to be presented by the target user is.
The concerned score model is used for representing the corresponding relation between the characteristic information of the display information, the user information of the user browsing the display information and the concerned score for representing the concerned degree of the display information concerned by the user. Specifically, the attention score model may be a correspondence table in which a plurality of pieces of feature information, user information, and attention scores are stored, which is previously prepared by a technician based on statistics of a large number of pieces of feature information, user information, and attention scores. The score to be paid attention in the correspondence table may be statistics of scores of the first type of information to be displayed, which are performed in advance by a technician for a user group to which a user represented by the user information belongs (for example, which may be represented by a preset user information set), and a score to be paid attention obtained by calculating (for example, averaging the scores of the users). The generating unit 502 may search the corresponding attention score from the correspondence table according to the input feature information and the user information. The score model of interest may be a model obtained by training an initial model (e.g., a neural network) by a machine learning method based on a preset training sample.
Step 5023, acquiring a preset compensation display cost value, and generating a first type display cost value of the first type of information to be displayed based on the acquired compensation display cost value and the acquired concerned scores.
Specifically, as an example, the generation unit 502 may multiply the obtained compensation presentation cost value and the obtained score of interest to obtain a first-type presentation cost value. Alternatively, the generating unit 502 may multiply the obtained compensation presentation cost value and the obtained attention score, and then perform other processing (for example, add or subtract a preset value to or from the product obtained after the multiplication) to obtain the first type presentation cost value.
The compensation presentation cost value can be used for representing the cost of a presenter who pushes a certain first type of information set to be presented in order to enable a user to browse the first type of information to be presented in the first type of information set to be presented. The presenter may be a unit or an individual who uses the apparatus 500 to present the first type of information to be presented to the user. For example, assuming that a user receives a first type of information to be presented using a target application installed on a terminal used by the user, the owner of the target application is a pusher. In practice, the pusher pays a certain cost to obtain a new user, and a compensation display cost value can be calculated based on the paid cost. For example, the compensation presentation cost value is obtained by dividing the cost value representing the cost paid by the cost value by the number of users browsing the first type of information to be presented in the first type of information to be presented set.
Generally, the higher the score of interest is, the higher the interest degree of the target user on the first type of information to be displayed is, the higher the obtained first type of display cost value is, the higher the cost for representing that the presenter pushes the first type of information to be displayed to the target user is, and the higher the value for representing the first type of information to be displayed is.
In some optional implementations of this embodiment, the apparatus 500 may further include: a second obtaining unit (not shown in the figures) configured to obtain at least one preset second type of information to be displayed and a second type of display cost value corresponding to the preset second type of information to be displayed; a merging unit (not shown in the figure), configured to merge the obtained at least one second-type presentation cost value and the obtained at least one first-type presentation cost value into a presentation cost value set, and merge the at least one first-type information to be presented and the at least one second-type information to be presented into an information set to be presented; an ordering unit (not shown in the figure) configured to order the presentation cost values in the presentation cost value set based on the size of the presentation cost values in the presentation cost value set; and a pushing unit (not shown in the figure) configured to select a presentation cost value from the sorted presentation cost value sets, and push the information to be presented, which is included in the information set to be presented and corresponds to the selected presentation cost value, to the terminal.
In some optional implementations of this embodiment, the scoring model of interest may include at least two scoring sub-models and a result generation sub-model; and the generating unit includes: a first generating module (not shown in the figure), configured to, for a scoring sub-model of at least two scoring sub-models, input the obtained feature information and user information into the scoring sub-model, so as to obtain a score corresponding to the scoring sub-model; and a second generating module (not shown in the figure) configured to input the obtained score into a sub-model, and obtain a score to be attended, which represents a degree that the first type of information to be presented is attended by the target user, wherein the result generating sub-model is used for performing weighted summation on the obtained scores based on preset weights to obtain a score to be attended.
In some optional implementations of this embodiment, for a scoring sub-model of the at least two scoring sub-models, the scoring sub-model may be trained in advance according to the following steps: acquiring a training sample set corresponding to the grading sub-model, wherein the training sample comprises preset characteristic information of first-class information to be displayed of the sample, sample user information of a sample user who browses the first-class information to be displayed of the sample, and pre-labeled labeling information used for representing whether the sample user performs target operation on the first-class information to be displayed of the sample; and by utilizing a machine learning method, taking the characteristic information and the sample user information included in the training samples in the obtained training sample set as input, taking the input characteristic information and the marking information corresponding to the sample user information as expected output, and training to obtain the scoring sub-model.
In some optional implementations of this embodiment, the compensation presentation cost value used for generating the first type of presentation cost value may be obtained in advance through the following steps: acquiring a preset proportion set, wherein the proportion is the proportion of the number of first type information to be displayed and the number of second type information to be displayed, which are pushed to a terminal used by a user in the preset user set within a preset time period; determining the number of lost users corresponding to the proportion in the proportion set, wherein the lost users are users who push the first type of information to be displayed and the second type of information to be displayed according to the proportion within a preset time period, and do not use a terminal to receive the pushed first type of information to be displayed and the pushed second type of information to be displayed after the preset time period; determining a supplementary cost value corresponding to the proportion based on the determined number of the lost users, wherein the supplementary cost value is used for representing the cost paid by the users which are added by the same amount as the lost users; determining the number of users who use the terminal to browse the first type of information to be displayed pushed according to the proportion in a preset time period, dividing the determined supplementary cost value by the determined number, and determining the obtained quotient as the compensation display cost value corresponding to the proportion; selecting the compensation presentation cost value from the determined compensation presentation cost values as a compensation presentation cost value used for generating the first type of presentation cost value.
In some optional implementations of this embodiment, selecting, from the determined compensation presentation cost values, a compensation presentation cost value as a compensation presentation cost value used to generate the first type of presentation cost value includes: for the compensation display cost values in the determined compensation display cost values, determining the sum of the second type of display cost values of the second type of information to be displayed, which is pushed according to the proportion corresponding to the compensation display cost values, as a second type of total display cost values corresponding to the compensation display cost values; determining a difference value between the second type total display cost value and a supplementary cost value corresponding to the compensation display cost value; and selecting a maximum value from the determined difference values, and taking the compensation display cost value corresponding to the maximum value as the compensation display cost value used for generating the first type of display cost value.
The device provided by the above embodiment of the application obtains the first kind of display cost value of the first kind of information to be displayed by using the concerned scoring model and the preset compensation display cost value, so that the different first kind of information to be displayed and the different user information can be analyzed, the accuracy of generating the first kind of display cost value is improved, and the device is helpful for improving the pertinence of displaying various information to be displayed to a user under the condition that the various information to be displayed needs to be displayed to the user.
Referring now to FIG. 6, shown is a block diagram of a computer system 600 suitable for use in implementing a server according to embodiments of the present application. The server shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.
As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU)601 that can 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 section 608 into a Random Access Memory (RAM) 603. 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 via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, and the like; an output portion 607 including a display such as a Liquid Crystal Display (LCD) and a speaker; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. 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 mounted on the drive 610 as necessary, so that a computer program read out therefrom is mounted in the storage section 608 as necessary.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the 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 computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 609, and/or installed from the removable medium 611. The computer program performs the above-described functions defined in the method of the present application when executed by a Central Processing Unit (CPU) 601.
It should be noted that the computer readable medium described herein can be a computer readable signal medium or a computer readable medium or any combination of the two. A computer readable medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable 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 present application, a computer readable 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 this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable 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: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like 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 type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart 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 application. 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 units described in the embodiments of the present application may be implemented by software or hardware. The described units may also be provided in a processor, and may be described as: a processor includes a first acquisition unit and a generation unit. The names of the units do not form a limitation on the units themselves in some cases, for example, the first obtaining unit may also be described as a "unit for obtaining at least one of first type information to be presented to be pushed to a terminal of a target user for presentation on the terminal and user information of the target user".
As another aspect, the present application also provides a computer-readable medium, which may be contained in the server described in the above embodiments; or may exist separately and not be assembled into the server. The computer readable medium carries one or more programs which, when executed by the server, cause the server to: acquiring at least one first type of information to be displayed and user information of a target user, which are to be pushed to a terminal of the target user to be displayed on the terminal; for first type information to be displayed in the obtained at least one first type information to be displayed, obtaining characteristic information of the first type information to be displayed; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user; and generating a first type of display cost value of the first type of information to be displayed based on the obtained compensation display cost value and the obtained concerned scores.
The above description is only a preferred embodiment of the application and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention herein disclosed is not limited to the particular combination of features described above, but also encompasses other arrangements formed by any combination of the above features or their equivalents without departing from the spirit of the invention. For example, the above features may be replaced with (but not limited to) features having similar functions disclosed in the present application.

Claims (14)

1. A method for generating information, comprising:
the method comprises the steps of obtaining at least one first type of information to be displayed and user information of a target user, wherein the first type of information to be displayed is pushed to a terminal of the target user so as to be displayed on the terminal;
for first type information to be displayed in the obtained at least one first type information to be displayed, obtaining characteristic information of the first type information to be displayed; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user; and generating a first type of display cost value of the first type of information to be displayed based on the obtained compensation display cost value and the obtained concerned scores.
2. The method of claim 1, wherein the method further comprises:
acquiring at least one preset second type of information to be displayed and a preset second type of display cost value corresponding to the second type of information to be displayed;
merging the obtained at least one second type display cost value and the obtained at least one first type display cost value into a display cost value set, and merging the at least one first type information to be displayed and the at least one second type information to be displayed into an information set to be displayed;
sorting the display cost values in the display cost value set based on the display cost values in the display cost value set;
and selecting a display cost value from the sorted display cost value set, and pushing the information to be displayed, which corresponds to the selected display cost value and is contained in the information set to be displayed, to the terminal.
3. The method of claim 1, wherein the scoring model of interest includes at least two scoring sub-models and a result generation sub-model; and
inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user, wherein the concerned score comprises:
for a scoring sub-model in the at least two scoring sub-models, inputting the acquired feature information and the user information into the scoring sub-model to obtain a score corresponding to the scoring sub-model;
and inputting the obtained scores into the result generation submodel to obtain the concerned scores representing the degree of the first type of information to be displayed concerned by the target user, wherein the result generation submodel is used for weighting and summing the obtained scores based on preset weight to obtain the concerned scores.
4. The method of claim 3, wherein, for a scoring sub-model of the at least two scoring sub-models, the scoring sub-model is trained in advance as follows:
acquiring a training sample set corresponding to the grading sub-model, wherein the training sample comprises preset characteristic information of first-class information to be displayed of the sample, sample user information of a sample user who browses the first-class information to be displayed of the sample, and pre-labeled labeling information used for representing whether the sample user performs target operation on the first-class information to be displayed of the sample;
and by utilizing a machine learning method, taking the characteristic information and the sample user information included in the training samples in the obtained training sample set as input, taking the input characteristic information and the marking information corresponding to the sample user information as expected output, and training to obtain the scoring sub-model.
5. The method according to one of claims 1 to 4, wherein the compensation presentation cost value used for generating the first type of presentation cost value is obtained in advance by:
acquiring a preset proportion set, wherein the proportion is the proportion of the number of first type information to be displayed and the number of second type information to be displayed, which are pushed to a terminal used by a user in the preset user set within a preset time period;
determining the number of lost users corresponding to the proportion in the proportion set, wherein the lost users are users who push the first type of information to be displayed and the second type of information to be displayed according to the proportion in a preset time period, and do not use a terminal to receive the pushed first type of information to be displayed and the pushed second type of information to be displayed after the preset time period; determining a supplementary cost value corresponding to the proportion based on the determined number of the lost users, wherein the supplementary cost value is used for representing the cost paid by the users which are added by the same amount as the lost users; determining the number of users who use the terminal to browse the first type of information to be displayed pushed according to the proportion in the preset time period, dividing the determined supplementary cost value by the determined number, and determining the obtained quotient as the compensation display cost value corresponding to the proportion;
selecting the compensation presentation cost value from the determined compensation presentation cost values as a compensation presentation cost value used for generating the first type of presentation cost value.
6. The method of claim 5, wherein said selecting a compensation presentation cost value from the determined compensation presentation cost values as the compensation presentation cost value used to generate the first type of presentation cost value comprises:
for the compensation display cost values in the determined compensation display cost values, determining the sum of the second type of display cost values of the second type of information to be displayed, which is pushed according to the proportion corresponding to the compensation display cost values, as a second type of total display cost values corresponding to the compensation display cost values; determining a difference value between the second type total display cost value and a supplementary cost value corresponding to the compensation display cost value;
and selecting a maximum value from the determined difference values, and taking the compensation display cost value corresponding to the maximum value as the compensation display cost value used for generating the first type of display cost value.
7. An apparatus for generating information, comprising:
the terminal comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is configured to acquire at least one first type of information to be displayed and user information of a target user, which are to be pushed to the terminal of the target user to be displayed on the terminal;
the generating unit is configured to acquire feature information of the first type of information to be displayed for the first type of information to be displayed in the acquired at least one first type of information to be displayed; inputting the acquired feature information and the user information into a pre-trained concerned score model to obtain a concerned score for representing the degree of the first type of information to be displayed concerned by the target user; and generating a first type of display cost value of the first type of information to be displayed based on the obtained compensation display cost value and the obtained concerned scores.
8. The apparatus of claim 7, wherein the apparatus further comprises:
the second acquisition unit is configured to acquire at least one preset second type of information to be displayed and a second type of display cost value corresponding to the preset second type of information to be displayed;
a merging unit configured to merge the obtained at least one second-type presentation cost value and the obtained at least one first-type presentation cost value into a presentation cost value set, and merge the at least one first-type information to be presented and the at least one second-type information to be presented into an information set to be presented;
the sorting unit is configured to sort the display cost values in the display cost value set based on the size of the display cost values in the display cost value set;
and the pushing unit is configured to select a display cost value from the sorted display cost value set and push the information to be displayed, which corresponds to the selected display cost value and is contained in the information set to be displayed, to the terminal.
9. The apparatus of claim 7, wherein the scoring model of interest comprises at least two scoring sub-models and a result generation sub-model; and
the generation unit includes:
the first generation module is configured to input the acquired feature information and the user information into a scoring sub-model of the at least two scoring sub-models to obtain a score corresponding to the scoring sub-model;
and the second generation module is configured to input the obtained scores into the result generation submodel to obtain the concerned scores representing the degree of the first type of information to be displayed, which is concerned by the target user, wherein the result generation submodel is used for weighting and summing the obtained scores based on preset weights to obtain the concerned scores.
10. The apparatus of claim 9, wherein for a scoring sub-model of the at least two scoring sub-models, the scoring sub-model is trained in advance as follows:
acquiring a training sample set corresponding to the grading sub-model, wherein the training sample comprises preset characteristic information of first-class information to be displayed of the sample, sample user information of a sample user who browses the first-class information to be displayed of the sample, and pre-labeled labeling information used for representing whether the sample user performs target operation on the first-class information to be displayed of the sample;
and by utilizing a machine learning method, taking the characteristic information and the sample user information included in the training samples in the obtained training sample set as input, taking the input characteristic information and the marking information corresponding to the sample user information as expected output, and training to obtain the scoring sub-model.
11. The apparatus according to one of claims 7 to 10, wherein the compensation presentation cost value used for generating the first type of presentation cost value is obtained in advance by:
acquiring a preset proportion set, wherein the proportion is the proportion of the number of first type information to be displayed and the number of second type information to be displayed, which are pushed to a terminal used by a user in the preset user set within a preset time period;
determining the number of lost users corresponding to the proportion in the proportion set, wherein the lost users are users who push the first type of information to be displayed and the second type of information to be displayed according to the proportion in a preset time period, and do not use a terminal to receive the pushed first type of information to be displayed and the pushed second type of information to be displayed after the preset time period; determining a supplementary cost value corresponding to the proportion based on the determined number of the lost users, wherein the supplementary cost value is used for representing the cost paid by the users which are added by the same amount as the lost users; determining the number of users who use the terminal to browse the first type of information to be displayed pushed according to the proportion in the preset time period, dividing the determined supplementary cost value by the determined number, and determining the obtained quotient as the compensation display cost value corresponding to the proportion;
selecting the compensation presentation cost value from the determined compensation presentation cost values as a compensation presentation cost value used for generating the first type of presentation cost value.
12. The apparatus of claim 11, wherein said selecting a compensation presentation cost value from the determined compensation presentation cost values as the compensation presentation cost value used to generate the first type of presentation cost value comprises:
for the compensation display cost values in the determined compensation display cost values, determining the sum of the second type of display cost values of the second type of information to be displayed, which is pushed according to the proportion corresponding to the compensation display cost values, as a second type of total display cost values corresponding to the compensation display cost values; determining a difference value between the second type total display cost value and a supplementary cost value corresponding to the compensation display cost value;
and selecting a maximum value from the determined difference values, and taking the compensation display cost value corresponding to the maximum value as the compensation display cost value used for generating the first type of display cost value.
13. A server, comprising:
one or more processors;
a storage device having one or more programs stored thereon,
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-6.
14. A computer-readable medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-6.
CN201811290026.9A 2018-10-31 2018-10-31 Method and device for generating information Active CN111125574B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201811290026.9A CN111125574B (en) 2018-10-31 2018-10-31 Method and device for generating information
PCT/CN2019/102176 WO2020088058A1 (en) 2018-10-31 2019-08-23 Information generating method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811290026.9A CN111125574B (en) 2018-10-31 2018-10-31 Method and device for generating information

Publications (2)

Publication Number Publication Date
CN111125574A true CN111125574A (en) 2020-05-08
CN111125574B CN111125574B (en) 2023-04-28

Family

ID=70464575

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811290026.9A Active CN111125574B (en) 2018-10-31 2018-10-31 Method and device for generating information

Country Status (2)

Country Link
CN (1) CN111125574B (en)
WO (1) WO2020088058A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113179430A (en) * 2021-04-19 2021-07-27 广州欢网科技有限责任公司 Smart television information loading method and system for different users
CN113722624A (en) * 2020-05-25 2021-11-30 北京沃东天骏信息技术有限公司 Method and device for displaying service icon
CN114461081A (en) * 2020-11-10 2022-05-10 北京搜狗科技发展有限公司 Information display method and device and electronic equipment
CN116384473A (en) * 2023-06-02 2023-07-04 支付宝(杭州)信息技术有限公司 Calculation graph improvement and information pushing method and device

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113627454B (en) * 2020-05-09 2024-06-18 北京沃东天骏信息技术有限公司 Article information clustering method, pushing method and device
CN112182374B (en) * 2020-09-25 2024-02-13 宇文道静 Inventory control method, apparatus, electronic device, and computer-readable medium
CN112348356A (en) * 2020-11-05 2021-02-09 北京字节跳动网络技术有限公司 Work quality determination method and device, computer equipment and readable storage medium
CN113781168A (en) * 2021-09-07 2021-12-10 深圳壹账通智能科技有限公司 Part display method, device, equipment and storage medium
CN114003188B (en) * 2021-10-09 2024-02-20 北京用友薪畴数字科技有限公司 Information encryption display method, device, electronic equipment and computer readable medium
CN116011505B (en) * 2023-03-15 2024-05-14 图灵人工智能研究院（南京）有限公司 Multi-module dynamic model training method and device based on feature comparison
CN117238451B (en) * 2023-11-16 2024-02-13 北京无疆脑智科技有限公司 Training scheme determining method, device, electronic equipment and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060248035A1 (en) * 2005-04-27 2006-11-02 Sam Gendler System and method for search advertising
CN105809558A (en) * 2016-03-15 2016-07-27 平安科技（深圳）有限公司 Social network based recommendation method and apparatus
CN105912686A (en) * 2016-04-18 2016-08-31 上海珍岛信息技术有限公司 Search engine marketing bid method and system based on machine learning
CN107341187A (en) * 2017-06-07 2017-11-10 努比亚技术有限公司 Search processing method, device, equipment and computer-readable storage medium

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105868237A (en) * 2015-12-09 2016-08-17 乐视网信息技术（北京）股份有限公司 Multimedia data recommendation method and server
CN105677881B (en) * 2016-01-12 2020-07-03 腾讯科技（深圳）有限公司 Information recommendation method and device and server
CN106021577B (en) * 2016-06-01 2019-07-23 北京金山安全软件有限公司 Information pushing method and device and electronic equipment
CN106339502A (en) * 2016-09-18 2017-01-18 电子科技大学 Modeling recommendation method based on user behavior data fragmentation cluster
CN106886918A (en) * 2017-02-06 2017-06-23 中国联合网络通信集团有限公司 A kind of determination method of targeted customer, apparatus and system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060248035A1 (en) * 2005-04-27 2006-11-02 Sam Gendler System and method for search advertising
CN105809558A (en) * 2016-03-15 2016-07-27 平安科技（深圳）有限公司 Social network based recommendation method and apparatus
CN105912686A (en) * 2016-04-18 2016-08-31 上海珍岛信息技术有限公司 Search engine marketing bid method and system based on machine learning
CN107341187A (en) * 2017-06-07 2017-11-10 努比亚技术有限公司 Search processing method, device, equipment and computer-readable storage medium

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113722624A (en) * 2020-05-25 2021-11-30 北京沃东天骏信息技术有限公司 Method and device for displaying service icon
CN114461081A (en) * 2020-11-10 2022-05-10 北京搜狗科技发展有限公司 Information display method and device and electronic equipment
CN113179430A (en) * 2021-04-19 2021-07-27 广州欢网科技有限责任公司 Smart television information loading method and system for different users
CN116384473A (en) * 2023-06-02 2023-07-04 支付宝(杭州)信息技术有限公司 Calculation graph improvement and information pushing method and device
CN116384473B (en) * 2023-06-02 2023-08-08 支付宝(杭州)信息技术有限公司 Calculation graph improvement and information pushing method and device

Also Published As

Publication number Publication date
WO2020088058A1 (en) 2020-05-07
CN111125574B (en) 2023-04-28

Similar Documents

Publication Publication Date Title
CN109460513B (en) Method and apparatus for generating click rate prediction model
CN111125574B (en) Method and device for generating information
CN109460514B (en) Method and device for pushing information
CN109522483B (en) Method and device for pushing information
CN109976997B (en) Test method and device
CN110298716B (en) Information pushing method and device
CN108536867B (en) Method and apparatus for generating information
CN110413872B (en) Method and device for displaying information
CN107908662B (en) Method and device for realizing search system
CN111695041B (en) Method and device for recommending information
CN116541610B (en) Training method and device for recommendation model
CN111311294A (en) Data processing method, device, medium and electronic equipment
CN111104590A (en) Information recommendation method, device, medium and electronic equipment
CN111126649B (en) Method and device for generating information
CN111125502B (en) Method and device for generating information
CN113822734A (en) Method and apparatus for generating information
CN111768218B (en) Method and device for processing user interaction information
CN109472028B (en) Method and device for generating information
CN111723274B (en) Method and device for processing information
CN111125501B (en) Method and device for processing information
CN111125514B (en) Method, device, electronic equipment and storage medium for analyzing user behaviors
CN111784091B (en) Method and device for processing information
CN109584012B (en) Method and device for generating item push information
CN109727072B (en) Method and apparatus for processing information
CN112348594A (en) Method, device, computing equipment and medium for processing article demands

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
GR01 Patent grant
GR01 Patent grant