CN111626805B - Information display method and device - Google Patents

Information display method and device Download PDF

Info

Publication number
CN111626805B
CN111626805B CN201910151726.8A CN201910151726A CN111626805B CN 111626805 B CN111626805 B CN 111626805B CN 201910151726 A CN201910151726 A CN 201910151726A CN 111626805 B CN111626805 B CN 111626805B
Authority
CN
China
Prior art keywords
information
target item
item information
sample
click rate
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.)
Active
Application number
CN201910151726.8A
Other languages
Chinese (zh)
Other versions
CN111626805A (en
Inventor
刘孟涛
赵夕炜
徐夙龙
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 Wodong Tianjun Information Technology Co Ltd
Original Assignee
Beijing Wodong Tianjun Information 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 Wodong Tianjun Information Technology Co Ltd filed Critical Beijing Wodong Tianjun Information Technology Co Ltd
Priority to CN201910151726.8A priority Critical patent/CN111626805B/en
Publication of CN111626805A publication Critical patent/CN111626805A/en
Application granted granted Critical
Publication of CN111626805B publication Critical patent/CN111626805B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0641—Shopping interfaces

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

The embodiment of the application discloses an information display method and device. One embodiment of the method comprises the following steps: for target item information in at least one target item information, acquiring a feature set related to the target item information, and inputting features included in the feature set related to the target item information into a pre-trained click rate prediction model to obtain click rate information of the target item information; determining the to-be-presented rank of at least one target item information according to the obtained at least one click rate information; generating a display message according to the to-be-displayed rank of at least one target article information; and sending the display information to the terminal equipment in communication connection so that the terminal equipment displays at least one piece of target article information according to the display information. In the embodiment, when the click rate of the target item information is generated, the historical display rank of the target item information is taken into consideration, and then the click rate of the target item information in each display rank is obtained.

Description

Information display method and device
Technical Field
The embodiment of the application relates to the technical field of computers, in particular to an information display method and device.
Background
Various item information is usually displayed in each e-commerce platform, and the click rate of the displayed item information often needs to be predicted. In practice, when predicting the click rate of the item information, the historical display rank of the item information is often not taken into consideration, and thus the click rate of the item information at each display rank cannot be obtained.
Disclosure of Invention
The embodiment of the application provides an information display method and device.
In a first aspect, an embodiment of the present application provides an information display method, including: for target item information in at least one target item information, acquiring a feature set related to the target item information, and inputting features included in the feature set related to the target item information into a pre-trained click rate prediction model to obtain click rate information of the target item information, wherein the feature set comprises display level information representing historical display levels of the target item information, the click rate information is used for representing click rates of the target item information in preset numbers of display levels, and the click rate prediction model is used for representing a corresponding relation between the feature set related to the item information and the click rate information of the item information; determining the to-be-presented rank of at least one target item information according to the obtained at least one click rate information; generating a display message according to the to-be-displayed rank of the at least one target item information, wherein the display message is used for identifying the to-be-displayed rank of the at least one target item information; and sending the display information to the terminal equipment in communication connection so that the terminal equipment displays at least one piece of target article information according to the display information.
In some embodiments, the click rate prediction model is trained by: acquiring a sample set, wherein the sample comprises a sample feature set and sample click rate information, the sample feature set is related to sample article information, the sample feature set comprises display level information for representing historical display levels of the sample article information, and the sample click rate information is used for representing click rates of the sample article information in a preset number of display levels; selecting at least one sample from the set of samples and performing the training steps of: respectively inputting sample features included in the sample feature set of the selected at least one sample into an initial model to obtain click rate information of sample article information related to the sample feature set of each sample in the at least one sample; comparing click rate information of sample item information related to a sample feature set of each of the at least one sample with corresponding sample click rate information; determining whether the initial model is trained based on the comparison result; in response to determining that the training is complete, an initial model of the training completion is determined as a click rate prediction model.
In some embodiments, the step of training the click rate prediction model further comprises: in response to determining that the untraining is complete, adjusting relevant parameters of the initial model, and selecting at least one unused sample from the set of samples, continuing to perform the training step using the adjusted initial model as the initial model.
In some embodiments, at least one of the following is also included in the feature set: user characteristics associated with the target item information, item characteristics of the item indicated by the target item information, search information for the target item information, click information characterizing whether the target item information is clicked, and the like.
In some embodiments, before determining the to-be-presented ranking of the at least one target item information according to the obtained at least one click rate information, the method further includes: and generating a target article information set, wherein the target article information set comprises at least one piece of target article information.
In some embodiments, determining the to-be-presented ranking of the at least one target item information according to the obtained at least one click rate information includes: selecting a presentation rank from a preset number of presentation ranks, and executing the following determining steps: determining the product of the click rate of the target item information in the target item information set in the display order and the multiplication value corresponding to the target item information based on at least one click rate information; determining the target article information indicated by the maximum product as target article information to be displayed by the display rank; the method further comprises the following steps: deleting target item information indicated by the maximum product from the set of target item information in response to determining that there is an unselected presentation rank; using the target item information set after deleting the target item information as a target item information set; selecting unselected presentation orders from the preset number of presentation orders, and continuing to execute the determining step.
In some embodiments, the generating the display information according to the to-be-displayed ranking of the at least one target item information includes: and sequencing the at least one piece of target article information according to the sequence from the small sequence to the large sequence of the sequence number of the to-be-displayed bit of the at least one piece of target article information, and generating sequencing information as display information.
In a second aspect, an embodiment of the present application provides an information display apparatus, including: the acquisition unit is configured to acquire a feature set related to the target item information for the target item information in at least one target item information, and input features included in the feature set related to the target item information into a pre-trained click rate prediction model to obtain click rate information of the target item information, wherein the feature set comprises display level information representing historical display levels of the target item information, the click rate information is used for representing click rates of the target item information in a preset number of display levels, and the click rate prediction model is used for representing a corresponding relation between the feature set related to the item information and the click rate information of the item information; a determining unit configured to determine a rank to be presented of the at least one target item information according to the obtained at least one click rate information; the first generation unit is configured to generate a display message according to the to-be-displayed rank of the at least one target item information, wherein the display information is used for identifying the to-be-displayed rank of the at least one target item information; and the sending unit is configured to send the display information to the terminal equipment in communication connection so that the terminal equipment displays at least one piece of target article information according to the display information.
In some embodiments, the apparatus further comprises a training unit, wherein the model training unit comprises: the acquisition module is configured to acquire a sample set, wherein the sample comprises a sample feature set related to sample article information and sample click rate information, the sample feature set comprises display bit information representing historical display bit of the sample article information, and the sample click rate information is used for representing click rate of the sample article information in a preset number of display bit; a training module configured to select at least one sample from a set of samples, and perform the training steps of: respectively inputting sample features included in the sample feature set of the selected at least one sample into an initial model to obtain click rate information of sample article information related to the sample feature set of each sample in the at least one sample; comparing click rate information of sample item information related to a sample feature set of each of the at least one sample with corresponding sample click rate information; determining whether the initial model is trained based on the comparison result; in response to determining that the training is complete, an initial model of the training completion is determined as a click rate prediction model.
In some embodiments, the model training unit further comprises: and the adjusting module is configured to respond to the determination of the non-training completion, adjust relevant parameters of the initial model, select at least one unused sample from the sample set, and continue to execute the training step by using the adjusted initial model as the initial model.
In some embodiments, at least one of the following is also included in the feature set: user characteristics associated with the target item information, item characteristics of the item indicated by the target item information, search information for the target item information, click information characterizing whether the target item information is clicked, and the like.
In some embodiments, the apparatus further comprises: and a second generation unit configured to generate a set of target item information, wherein the set of target item information includes at least one piece of target item information.
In some embodiments, the determining unit includes: a determining module configured to select a presentation rank from a preset number of presentation ranks, and perform the following determining steps: determining the product of the click rate of the target item information in the target item information set in the display order and the multiplication value corresponding to the target item information based on at least one click rate information; determining the target article information indicated by the maximum product as target article information to be displayed by the display rank; an execution module configured to delete target item information indicated by the maximum product from the set of target item information in response to determining that there is an unselected presentation rank; using the target item information set after deleting the target item information as a target item information set; selecting unselected presentation orders from the preset number of presentation orders, and continuing to execute the determining step.
In some embodiments, the generating unit is further configured to: and sequencing the at least one piece of target article information according to the sequence from the small sequence to the large sequence of the sequence number of the to-be-displayed bit of the at least one piece of target article information, and generating sequencing information as display information.
In a third aspect, an embodiment of the present application provides a server, including: one or more processors; a storage device having one or more programs stored thereon; the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method as described in any of the implementations of the first aspect.
In a fourth aspect, embodiments of the present application provide a computer readable medium having stored thereon a computer program which, when executed by a processor, implements a method as described in any of the implementations of the first aspect.
According to the information display method and device provided by the embodiment of the application, firstly, for each piece of target item information in at least one piece of target item information, a feature set related to the target item information can be acquired, and the feature set related to the target item information is input into a pre-trained click rate prediction model to obtain the click rate information of the target item information. Then, the to-be-presented rank of the at least one target item information can be determined according to the obtained at least one click rate information. And then, generating a display message according to the determined to-be-displayed rank of the at least one target item information. Therefore, the generated display information can be sent to the terminal equipment in communication connection, so that the terminal equipment displays the at least one piece of target object information according to the received display information. Therefore, when the click rate of the target item information is generated, the historical display rank of the target item information can be taken into consideration, and the click rate of the target item information in each display rank can be obtained.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the accompanying drawings in which:
FIG. 1 is an exemplary system architecture diagram in which an embodiment of the present application may be applied;
FIG. 2 is a flow chart of one embodiment of an information presentation method according to the present application;
fig. 3 is a schematic diagram of an application scenario of an information presentation method according to an embodiment of the present application;
FIG. 4 is a flow chart of yet another embodiment of an information presentation method according to the present application;
FIG. 5 is a schematic diagram of an embodiment of an information presentation apparatus according to the present application;
FIG. 6 is a schematic diagram of a computer system suitable for use with a server implementing an embodiment of the application.
Detailed Description
The application is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting of the application. It should be noted that, for convenience of description, only the portions related to the present application are shown in the drawings.
It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The application will be described in detail below with reference to the drawings in connection with embodiments.
Fig. 1 shows an exemplary architecture 100 in which the information presentation method or information presentation apparatus of the present application may be applied.
As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, a network 103, and a server 104. The network 103 is the medium used to provide communication links between the terminal devices 101, 102 and the server 104. The network 103 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.
The terminal devices 101, 102 interact with the server 104 through the network 103 to receive or send messages or the like. Various communication client Applications (APP) may be installed on the terminal devices 101, 102, such as shopping applications, search applications, browser applications, web browser applications, and the like.
The terminal devices 101 and 102 may be hardware or software. When the terminal devices 101, 102 are hardware, they may be various electronic devices having a display screen and supporting web browsing functions, including but not limited to smartphones, tablet computers, electronic book readers, laptop and desktop computers, and the like. When the terminal devices 101, 102 are software, they can be installed in the above-listed electronic devices. Which may be implemented as multiple software or software modules (e.g., to provide distributed services), or as a single software or software module. The present application is not particularly limited herein.
The server 104 may be a server providing various services. For example, it may be a background server of a shopping class application on the terminal device 101, 102. As an example, the background server may send the generated display information to the terminal device, so that the terminal device displays at least one target item information according to the received display information.
The server 104 may be hardware or software. When the server 104 is hardware, it may be implemented as a distributed server cluster formed by a plurality of servers, or as a single server. When server 104 is software, it may be implemented as multiple software or software modules (e.g., to provide distributed services), or as a single software or software module. The present application is not particularly limited herein.
It should be noted that, the information display method provided in the embodiment of the present application is generally executed by the server 104, and accordingly, the information display device is generally disposed in the server 104.
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 an information presentation method according to the present application is shown. The information display method comprises the following steps:
step 201, for at least one piece of target item information, acquiring a feature set related to the target item information, and inputting features included in the feature set related to the target item information into a pre-trained click rate prediction model to obtain click rate information of the target item information.
In this embodiment, the execution subject of the information presentation method (such as the server 104 shown in fig. 1) may acquire, from a local or communicatively connected database server, a feature set associated with each of the at least one target item information.
In practice, the item information is typically information used in an e-commerce platform to introduce various items, and may include, but is not limited to, at least one of the following: pictures, text, video. For example, the item is a mobile phone of some type. Then, the item information may be composed of at least one of: the picture of the mobile phone introduces the text of the mobile phone and relates to the video of the mobile phone. The target article information may also be article information specified in advance according to actual demands. For example, item information may be retrieved from the same term specified in item information presented in the e-commerce platform. For example, the pre-specified item information is displayed under the same category (such as a mobile phone category) in the e-commerce platform. The items indicated by the at least one target item information generally belong to the same category. For example, the indicated items are all cell phones.
The feature set may include presentation rank information that characterizes a historical presentation rank of the target item information. The historical display rank may be a rank of the target article information displayed in the e-commerce platform in a past period of time. It should be noted that the presentation bit information may be represented by at least one of the following: numbers, letters, words. Take the example of representing the presentation bit order information by a number. The historical display orders of the target object information in the e-commerce platform comprise the 2 nd bit, the 3 rd bit and the 6 th bit. In practice, the presentation bit information may be a sequence. As an example, in some application scenarios, each component in the sequence of presentation rank information characterizing the historical presentation ranks of the target item information may be used to characterize a sequence number of the historical presentation ranks at which the target item information was presented. For example, the sequence of the presentation rank information for characterizing the historical presentation ranks of the target item information is "2,3,6", and then it can be understood that the target item information has been presented at the ranks of the 2 nd, 3 rd, and 6 th bits. Or in other application scenes, each component in the sequence of the display level information for representing the historical display level of the target article information can be used for representing whether the historical display level corresponding to the position of the component in the sequence is displayed or not. For example, "1" may be used to indicate that the target item information was presented at the historical presentation rank corresponding to the position of the component in the sequence, and "0" may be used to indicate that the target item was not presented at the historical presentation rank corresponding to the position of the component in the sequence. At this time, the sequence of the presentation rank information for characterizing the history presentation rank of the target item information is "0,1,1,0,0,1", and then the first component "0" in the sequence indicates that the target item information was not presented at the 1 st presentation rank, and the second component "1" in the sequence indicates that the target item information was presented at the 2 nd presentation rank. Similarly, it can be concluded that the target item information was once presented at the 2 nd, 3 rd and 6 th presentation orders, but not at the 1 st, 4 th and 5 th presentation orders.
In this embodiment, the execution body may further input the features included in the feature set related to each piece of target item information into a pre-trained click rate prediction model, to obtain click rate information of each piece of target item information.
The click rate information is used for representing the click rate of the target article information in a preset number of display orders. In practice, the click rate information may be a sequence of a preset number of click rates, a first component in the sequence may be the click rate of the target item information at the 1 st display level, a second component in the sequence may be the click rate of the target item information at the 2 nd display level, and so on. It should be noted that the preset number may be determined according to actual needs, and is not specifically limited herein.
The click rate prediction model may be used to characterize a correspondence between a feature set associated with item information and click rate information of the item information. For example, the click rate prediction model may be a correspondence table in which a technician counts the obtained correspondence between presentation bit information and click rate information according to a large amount of history data. As another example, the click rate prediction model may also be a machine learning model trained according to the methods described in steps 401 and 402.
As an example, the execution body may input the presentation bit information included in the feature set related to each target item information into the correspondence table, and then determine click rate information corresponding to the presentation bit information matched with the input presentation bit information as the click rate information of the target item information. Here, the matching may be the same, or the similarity may be equal to or greater than a predetermined threshold.
In some optional implementations of this embodiment, at least one of the following may be further included in the feature set: user characteristics associated with the target item information, item characteristics of the item indicated by the target item information, search information for the target item information, click information characterizing whether the target item information is clicked, and the like.
The user characteristics may be characteristics of a user who browses the target item information within a preset time, including but not limited to at least one of the following: gender, age, region, purchasing power, brand preferences, account numbers, membership ratings, etc. The item characteristics may be characteristics of the item indicated by the target item information, including but not limited to at least one of: brands, prices, models, sales, user scores, etc. The above-mentioned search information may be information related to the user searching for the target article information within a preset time, including but not limited to at least one of the following: search terms, time when the user searches for the target item information, weather conditions when the user searches for the target item information, and the like. The click information may be information for indicating whether the target item information is clicked within a preset time, and may be indicated by any one of the following: numbers, letters, words, etc. For example, the number "1" may indicate that the target item information is clicked, and the number "0" may indicate that the target item information is not clicked.
In these implementations, the executing entity may obtain click rate information of each target item information through the following steps. First, the features included in the feature set are encoded. Then, for each of a preset number of display orders, designating a weight value of the feature included in the feature set in the display order; multiplying the encoded features by the weight value at the presentation bit number, and adding the multiplied result values. And then, the obtained preset number of result values can be converted into a range of 0 to 1 to be used as click rate information of the target article information, namely the click rate of the target article information in the preset number of display orders. In practice, the features may be encoded by various methods, such as one-hot encoding (one-bit efficient encoding), TF-IDF (Term Frequency-inverse text Frequency index) algorithm, and the like. The predetermined number of resulting values obtained may be converted into a range of 0 to 1 using various methods, such as Sigmoid function, dispersion normalization (min-max normalization), Z-score normalization, etc.
Step 202, determining the to-be-presented rank of at least one target item information according to the obtained at least one click rate information.
In this embodiment, after obtaining the click rate information of the at least one piece of target item information, the click rate of each piece of target item information in the preset number of display levels may be determined. Then, the executing body may further determine the to-be-presented rank of each target item information.
For example, for each of the preset number of display ranks, the execution body may first determine a click rate of each target item information at the display rank, and then may determine target item information having the largest click rate and not determined the display rank as target item information to be displayed at the display rank. After the target item information to be displayed for each display rank is determined, it means that the rank to be displayed for each target item information is determined. If at least two pieces of target item information with the same click rate exist at a certain display level, the execution body may further determine a click bid (the item information is clicked once, and the advertiser's bid) of the at least two pieces of target item information, and further determine the target item information with the highest click bid as the target item information to be displayed at the display level.
In some optional implementations of this embodiment, the executing entity may generate the set of target item information before determining the to-be-presented ranking of the at least one target item information. The target article information set may include at least one piece of target article information described above. Specifically, the executing body may directly determine the set of at least one piece of target item information as the target item information set.
In some optional implementations of this embodiment, after generating the set of target item information, the executing entity may further determine a to-be-presented ranking of each of the at least one target item information by.
In the first step, the execution body may select a presentation rank from the preset number of presentation ranks, and then execute the determining steps described in the following steps S1 and S2.
And step S1, determining the product of the click rate of the target item information in the target item information set in the display order and the multiplication value corresponding to the target item information based on at least one click rate information. Where the multiplied value is typically a click bid for the target item information.
First, for each piece of target item information in the target item information set, the execution body may determine a click rate of the target item information at a selected display level from click rate information of the target item information. Then, for each target item information in the target item information set, the executing entity may multiply a click rate of the target item information at the selected presentation level by a click bid of the target item information. It should be noted that, the execution body may select the display order from the preset number of display orders according to the order of the sequence numbers of the display orders from small to large, or may select the display order from the preset number of display orders according to the actual requirement.
As an example, the target item information ad is included in the target item information set 1 Target article information ad 2 … and target article information ad n . Wherein the target article information ad 1 The click rate information of (2) is "p 11 ，p 12 ，…，p 1k … ", target article information ad 2 The click rate information of (2) is "p 21 ，p 22 ，…，p 2k … ", target article information ad n The click rate information of (2) is "p n1 ，p n2 ，…，p nk …). Target article information ad 1 Target article information ad 2 And target article information ad n The click bids for (a), b, and c, respectively. Where n represents the number of the target article information, k represents the number of the display bit, and p nk Information ad representing target article n Click rate at the kth display level. Taking the selected display level as the kth display level as an example, the execution body may multiply the click rate of each target item information in the target item information set at the kth display level with the click bid of the target item information to obtain a product (a×p) 1k )，(b*p 2k )，…，(c*p nk )。
It will be appreciated that, at the selected presentation level, a corresponding product may be obtained for each of the target item information in the set of target item information.
And S2, determining the target item information indicated by the maximum product as the target item information to be displayed by the display rank. Where the multiplied value is typically a click bid for the item information.
After the selected display order obtains the product corresponding to each piece of the target item information in the target item information set, the execution body may determine the piece of target item information with the largest product as the piece of target item information to be displayed in the selected display order.
Continuing with the above example, if the product (a p 1k ) Maximum, the executing entity may then execute the target item information ad 1 And determining the target article information to be displayed for the kth display level.
Second, deleting target item information indicated by the maximum product from the target item information set in response to determining that the unselected presentation bit exists; using the target item information set after deleting the target item information as a target item information set; selecting an unselected presentation rank from the preset number of presentation ranks, and continuing to execute the determining step.
Still another example, the executing entity may delete the target item information ad from the target item information set in response to determining that there are not selected ones of the preset number of presentation orders 1 . At this time, the target article information ad is included in the target article information set 2 …, target article information ad n . Then, the execution subject may use the target item information set after deleting the target item information as the target item information set for executing the determining step next time. Then, the executing body may select the unselected presentation bit, and continue to execute the determining step.
Therefore, the execution main body can determine the target article information to be displayed in each display level in the preset number of display levels according to the generated at least one click rate information, namely, determine the level to be displayed of each target article information.
And 203, generating a display message according to the to-be-displayed rank of the at least one target object information.
In this embodiment, after determining the to-be-presented ranking of each of the at least one target item information, the execution body may further generate presentation information. The display information is used for identifying the to-be-displayed rank of the at least one target article information. In practice, the presentation information may take various forms, which may include, for example, but not limited to, at least one of the following: numbers, letters, words, pictures, etc.
As an example, if the target article information ad 1 Target article information ad 2 And target article information ad 3 The display information can be various information ad capable of identifying the target object 1 Target article information ad 2 And target article information ad 3 To be presented with the information of the rank. For example, the presentation information may be "ad 1:2 ，ad 2:1 ,ad 3:3 ”。
In some optional implementations of this embodiment, the generating display information according to the to-be-displayed ranking of the at least one target item information may further include: and sequencing the at least one piece of target article information according to the sequence from the small sequence to the large sequence of the sequence number of the to-be-displayed bit of the at least one piece of target article information, and generating sequencing information as display information. Wherein the ranking information may be information for identifying a ranking result of the at least one target item information. In practice, the ranking information may take various forms, which may include, for example, but not limited to, at least one of the following: numbers, letters, words, pictures, etc.
Also as an example, the target article information ad is displayed in the order of the sequence number of the rank to be displayed from small to large 1 Target article information ad 2 And target article information ad 3 Ranking the articles to obtain ranking information which can be various information ad capable of identifying the target articles 1 Target article information ad 2 And target article information ad 3 Is provided for the ranking result. For example, the ordering information may be "1:ad 2 ，2:da 1 ，3:da 3 ”。
And step 204, the display information is sent to the terminal equipment in communication connection, so that the terminal equipment displays at least one piece of target article information according to the display information.
In this embodiment, after the display information is generated, the execution body may send the display information to a terminal device connected in communication, and then the terminal device may update the target article information displayed by the preset number of display orders according to the received display information.
For example, for the display level in the preset number of display levels, the terminal device may determine whether the target item information currently displayed by the display level is consistent with the target item information to be displayed by the display level indicated by the display information, and if not, the terminal device may update the target item information currently displayed by the display level to the target item information to be displayed by the display level indicated by the display information.
As yet another example, for each of the at least one target item information, the terminal device may determine whether a current display rank of the target item information is consistent with a to-be-displayed rank of the target item information indicated by the display information, and if not, the terminal device may display the target item information on the to-be-displayed rank of the target item information indicated by the display information.
With continued reference to fig. 3, fig. 3 is a schematic diagram of an application scenario of the information presentation method according to the present embodiment. In the application scenario of fig. 3, the server 301 is a background server of a shopping class application installed in the smart phone 302, and the target item information a, the target item information B, and the target item information C are item information under a certain category in the shopping class application installed in the smart phone 302.
First, the server 301 may locally acquire the feature set 303 of the target item information a, the feature set 304 of the target item information B, and the feature set 305 of the target item information C. The feature set 303 includes the display rank information of the target item information a, the feature set 304 includes the display rank information of the target item information B, and the feature set 305 includes the display rank information of the target item information C.
The click rate prediction model 306 is taken as a correspondence table as an example. And the corresponding relation table displays the bit information and click rate information for associated storage. Then, the server 301 may input the display rank information in the feature set 303, the display rank information in the feature set 304, and the display rank information in the feature set 305 into the correspondence table to perform matching, so as to obtain click rate information 307 of the target item information a, click rate information 308 of the target item information B, and click rate information 309 of the target item information C.
Then, for each of the 1 st to 3 rd display orders, the server 301 may determine the click rates of the target item information a, the target item information B, and the target item information C at the display order, and determine the target item information having the largest click rate and not determined the display order as the target item information to be displayed at the display order. It will be appreciated that after determining the target item information to be presented for each presentation rank, the target item information a, target item information B, and target item information C may be determined for each presentation rank. Assuming that the to-be-presented rank of the target article information A is the 2 nd to-be-presented rank, the to-be-presented rank of the target article information B is the 1 st to-be-presented rank, and the to-be-presented rank of the target article information C is the 3 rd to-be-presented rank. Further, the server 301 may generate the display information 310, as shown in the figure, "target item information a:2, target item information B:1, and target item information C:3".
Further, the server 301 may send the presentation information 310 to the terminal device 302. Upon receiving the display information 310, the terminal device 302 may determine whether the current display ranks of the target item information a, the target item information B, and the target item information C are consistent with the to-be-displayed ranks indicated by the display information 310, respectively. If not, the target item information A, the target item information B and the target item information C are displayed on the display order indicated by the display information 310.
In the method provided by the embodiment of the application, first, for each piece of target item information in at least one piece of target item information, a feature set related to the target item information may be acquired, and the feature set related to the target item information is input into a pre-trained click rate prediction model to obtain click rate information of the target item information. Then, the to-be-presented rank of the at least one target item information can be determined according to the obtained at least one click rate information. And then, generating a display message according to the determined to-be-displayed rank of the at least one target item information. Therefore, the generated display information can be sent to the terminal equipment in communication connection, so that the terminal equipment displays the at least one piece of target object information according to the received display information. Thus, when the click rate of the target item information is generated, the click rate of the target item information at each display level can be obtained by taking the historical display level of the target item information into consideration.
With further reference to fig. 4, a flow 400 of yet another embodiment of an information presentation method is shown. The flow 400 of the information presentation method includes the following steps:
in step 401, a sample set is obtained.
In this embodiment, the samples in the sample set may include sample feature sets and sample click rate information related to sample item information. The sample article information can be article information selected by a technician from article information displayed by the e-commerce platform. The sample click rate information may be used to characterize the click rate of the sample item information at a preset number of display levels. A sample feature set is typically composed of features related to sample item information. For example, presentation rank information that characterizes historical presentation ranks of sample item information may be included in the sample feature set.
In some optional implementations of this embodiment, at least one of the following may be included in the sample feature set: user characteristics associated with the sample item information, item characteristics of the item indicated by the sample item information, search information for the sample item information, click information characterizing whether the sample item information is clicked, and the like.
In this embodiment, the execution subject (such as the server 104 shown in fig. 1) that trains the click rate prediction model may acquire a sample set by various methods. For example, if the sample set is stored locally, the executing body that trains the click rate prediction model may directly acquire the sample set from the local. For example, if the sample set is stored in a database server of the communication connection, the execution subject that trains the click rate prediction model may obtain the sample set from the database server of the communication connection.
Step 402, at least one sample is selected from the set of samples, and the following training steps are performed.
In this embodiment, first, an execution subject that trains the click rate prediction model may select at least one sample from a sample set. Then, the click rate prediction model may be obtained through training in steps 4021 to 4024.
In step 4021, sample features included in the sample feature set of the selected at least one sample are respectively input into an initial model, so as to obtain click rate information of sample article information related to the sample feature set of each sample in the at least one sample.
In this embodiment, for each sample of the at least one sample selected, the execution body that trains the click rate prediction model may encode the features included in the sample feature set for that sample. At least one encoded sample may then be obtained. Further, for each sample in the at least one sample after encoding, the features included in the sample feature set of the sample may be input into the initial model to obtain click rate information of sample item information related to the sample feature set of the sample. The above-mentioned manner of encoding the sample feature may be various manners, such as one-hot encoding, TF-IDF algorithm, and the like. The initial model may be a machine learning model built using an artificial neural network (Artificial Neural Network, ANN) and may include an input layer, a pooling layer, a fully connected layer, and an output layer. The pooling layer can reduce the dimension of the coded features input by the input layer. The full connection layer can integrate the characteristics after dimension reduction. The output layer is preset with a preset number of outputs, wherein the preset number can be determined according to actual requirements. The number of layers of the pooling layer and the full-connection layer included in the initial model may be arbitrary, and is not particularly limited herein.
Specifically, first, the execution body for training the click rate prediction model may select a sample from the at least one encoded sample. Then, steps S1 to S6 shown below are performed on the selected encoded samples.
Step S1, through an input layer, sample characteristics included in a sample characteristic set of the selected coded samples are input to an initial model.
And S2, reducing the dimension of the input sample features through a pooling layer.
And S3, integrating the sample characteristics after dimension reduction through the full connection layer to obtain the integrated characteristics. In practice, at each display level, the above-mentioned integrated features are preset with corresponding weight values.
And S4, in each display level in the preset number of display levels, the execution body of the training click rate prediction model can multiply the synthesized characteristic with the corresponding weight value and add the multiplied result.
Step S5, the obtained preset number of added results may be converted into a range from 0 to 1, which is used as click rate information of sample item information related to the sample feature set to which the input sample feature belongs.
In practice, various methods may be used to convert the preset number of resulting values into a range of 0 to 1, such as Sigmoid function, dispersion normalization, Z-score normalization, and the like.
And step S6, outputting the obtained click rate information through an output layer.
It should be understood that the execution body for training the click rate prediction model may perform the above steps S1 to S6 for each sample after the above at least one code, respectively.
Step 4022, comparing the click rate information of the sample item information related to the sample feature set of each of the at least one sample with the corresponding sample click rate information.
In this embodiment, for each sample in the at least one sample, the execution body for training the click rate prediction model may compare the obtained click rate information with the sample click rate information of the sample, so as to determine a difference between the obtained click rate information and the sample click rate information of the sample. Thus, the execution subject training the click rate prediction model may add the obtained at least one difference to obtain a total difference. In practice, the total difference may be obtained by various loss functions. Such as a logarithmic loss function, a square loss function, etc.
In some optional implementations of this embodiment, if at least one of the following is further included in the sample feature set: user characteristics associated with the sample item information, item characteristics of the item indicated by the sample item information, search information for the sample item information, click information characterizing whether the sample item information is clicked, and the like. At this time, the execution subject who trains the click rate prediction model may also determine the total difference through the following loss function.
Where N represents the total number of sample item information related to the sample feature set of the encoded sample, and i represents the serial number of the sample item information. M represents the total number of the display bit times, and j represents the serial number of the display bit times. y is i Indicating whether the i-th sample item information is clicked, e.g., "0" indicates not clicked, and "1" indicates clicked. loss represents a cross entropy loss function, p j And the click rate of the sample article information at the j-th display level is represented. I represents an indication function, S i Representing the presentation rank of the i-th sample item information.
The definition of the cross entropy loss function is as follows:
loss(y i ，p j )＝-[y i lnp j +(1-y i )ln(1-p j )]
the definition of the indication function is as follows:
it can be seen that when S i When equal to j, the value of the indicator function is 1, when S i And when the value is not equal to j, the value of the indication function is 0.
Step 4023, determining whether the initial model is trained based on the comparison result.
In this embodiment, the execution subject that trains the click rate prediction model may determine whether the initial model is trained based on the comparison result. Specifically, if the obtained total difference is smaller than or equal to the preset difference, it may be determined that the initial model training is completed.
In step 4024, in response to determining that the training is complete, the initial model after the training is determined to be the click rate prediction model.
In this embodiment, in response to determining that the above initial model training is completed, the execution subject that trains the click rate prediction model may determine the initial model for which training is completed as the click rate prediction model.
In some optional implementations of this embodiment, in response to determining that the initial model is not trained, the executing entity that trains the click rate prediction model may adjust relevant parameters of the initial model. Then, at least one unused sample is selected from the sample set, and the adjusted initial model is used as the initial model, and the steps 4021 to 4024 are continuously executed until a preset training ending condition is met, so that training is ended. Here, the preset training end conditions include, but are not limited to, at least one of: the training times exceed the preset times; the training time exceeds the preset duration; the resulting total difference is less than or equal to the preset difference. In practice, the initial model may be adjusted by various algorithms, such as BP (Back Propagation) algorithm, SGD (Stochastic Gradient Descent, random gradient descent) algorithm.
It should be noted that, the execution subject of the training click rate prediction model may be the same as or different from the execution subject of the information display method. If the structure information and the parameter value of the trained click rate prediction model are the same, after the click rate prediction model is obtained through training, the execution main body for training the click rate prediction model can store the structure information and the parameter value of the trained click rate prediction model locally. If the structure information and the parameter value of the trained click rate prediction model are different, after the click rate prediction model is obtained through training, an execution main body for training the click rate prediction model can send the structure information and the parameter value of the trained click rate prediction model to the execution main body of the information display method.
Step 403, for the target item information in the at least one target item information, acquiring a feature set related to the target item information, and inputting features included in the feature set related to the target item information into a pre-trained click rate prediction model to obtain click rate information of the target item information.
Step 404, determining the to-be-presented rank of at least one target item information according to the obtained at least one click rate information.
Step 405, generating a display message according to the to-be-displayed rank of at least one target item information.
And step 406, the display information is sent to the terminal equipment in communication connection, so that the terminal equipment displays at least one piece of target article information according to the display information.
The steps 403, 404, 405 and 406 are identical to the steps 201, 202, 203 and 204 in the foregoing embodiments, and the descriptions of the steps 201, 202, 203 and 204 are also applicable to the steps 403, 404, 405 and 406, which are not repeated herein.
As can be seen from fig. 4, compared with the embodiment corresponding to fig. 2, the flow 400 of the information presentation method in this embodiment represents the step of training the click rate prediction model by using the machine learning method. Therefore, the scheme described in the embodiment can take the historical display level of the sample article information into consideration, and train to obtain the click rate prediction model capable of predicting the click rate of the target article information in each display level. Therefore, the click rate of the target object information in each display level can be obtained through the trained click rate prediction model. The prior method for training the click rate prediction model generally does not consider the characteristic of historical display level of the article information, the predicted click rate is always a general click rate of the article information, the click rate of the article information in each display level is not the click rate, and the prediction accuracy is lower. In order to improve the prediction accuracy, the click rate obtained by prediction may be corrected by a method of rank correction, but the prediction efficiency may be lowered. In contrast, in the scheme of the embodiment, the prediction accuracy can be further improved on the premise of not affecting the prediction efficiency.
With further reference to fig. 5, as an implementation of the method shown in the foregoing figures, the present application provides an embodiment of an information display apparatus, 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 information presentation apparatus 500 provided in the present embodiment includes an acquisition unit 501, a determination unit 502, a first generation unit 503, and a transmission unit 504. Wherein the obtaining unit 501 may be configured to: for target item information in at least one target item information, acquiring a feature set related to the target item information, and inputting features included in the feature set related to the target item information into a pre-trained click rate prediction model to obtain click rate information of the target item information, wherein the feature set comprises display level information representing historical display levels of the target item information, the click rate information is used for representing click rates of the target item information in preset number of display levels, and the click rate prediction model is used for representing a corresponding relation between the feature set related to the item information and the click rate information of the item information. The determining unit 502 may be configured to: and determining the to-be-presented rank of at least one target item information according to the obtained at least one click rate information. The first generation unit 503 may be configured to: and generating a display message according to the to-be-displayed rank of the at least one target item information, wherein the display message is used for identifying the to-be-displayed rank of the at least one target item information. The transmitting unit 504 may be configured to: and sending the display information to the terminal equipment in communication connection so that the terminal equipment displays at least one piece of target article information according to the display information.
In the present embodiment, in the information display apparatus 500: the specific processes of the obtaining unit 501, the determining unit 502, the first generating unit 503 and the transmitting unit 504 and the technical effects thereof may refer to the descriptions related to step 201, step 202, step 203 and step 204 in the corresponding embodiment of fig. 2, and are not repeated here.
In some optional implementations of this embodiment, the apparatus 500 may further include a training unit (not shown in the figures). The training unit may include an acquisition module (not shown in the figure) and a training module (not shown in the figure). The acquisition module may be configured to: the method comprises the steps of obtaining a sample set, wherein the sample comprises a sample feature set and sample click rate information, the sample feature set is related to sample article information, the sample feature set comprises display level information representing historical display levels of the sample article information, and the sample click rate information is used for representing click rates of the sample article information in a preset number of display levels. The training module may be configured to: selecting at least one sample from the set of samples and performing the training steps of: respectively inputting sample features included in the sample feature set of the selected at least one sample into an initial model to obtain click rate information of sample article information related to the sample feature set of each sample in the at least one sample; comparing click rate information of sample item information related to a sample feature set of each of the at least one sample with corresponding sample click rate information; determining whether the initial model is trained based on the comparison result; in response to determining that the training is complete, an initial model of the training completion is determined as a click rate prediction model.
In some optional implementations of this embodiment, the model training unit may further include an adjustment module (not shown in the figure). Wherein the adjustment module may be configured to: in response to determining that the untraining is complete, adjusting relevant parameters of the initial model, and selecting at least one unused sample from the set of samples, continuing to perform the training step using the adjusted initial model as the initial model.
In some optional implementations of this embodiment, at least one of the following may be further included in the feature set: user characteristics associated with the target item information, item characteristics of the item indicated by the target item information, search information for the target item information, click information characterizing whether the target item information is clicked, and the like.
In some optional implementations of this embodiment, the apparatus 500 may further include a second generating unit (not shown in the figure). Wherein the second generation unit may be configured to: a set of target item information is generated. The set of target item information may include at least one target item information therein.
In some optional implementations of this embodiment, the determining unit 502 may include: a determination module (not shown) and an execution module (not shown). Wherein the determination module may be configured to: selecting a presentation rank from a preset number of presentation ranks, and executing the following determining steps: determining the product of the click rate of the target item information in the target item information set in the display order and the multiplication value corresponding to the target item information based on at least one click rate information; and determining the target article information indicated by the maximum product as the target article information to be displayed by the display rank. The execution module may be configured to: deleting target item information indicated by the maximum product from the set of target item information in response to determining that there is an unselected presentation rank; using the target item information set after deleting the target item information as a target item information set; selecting unselected presentation orders from the preset number of presentation orders, and continuing to execute the determining step.
In some optional implementations of this embodiment, the first generating unit 503 may be further configured to: and sequencing the at least one piece of target article information according to the sequence from the small sequence to the large sequence of the sequence number of the to-be-displayed bit of the at least one piece of target article information, and generating sequencing information as display information.
The apparatus 500 provided in the foregoing embodiment of the present application may first obtain click rate information of each target item information by acquiring a feature set related to each target item information in at least one target item information through the acquiring unit 501, and inputting features included in the feature set related to each item information into a pre-trained click rate prediction model. Then, the number of bits to be presented of the at least one target item information may be determined by the determining unit 502 according to the obtained at least one click rate information. Then, the first generating unit 503 may generate a display message according to the to-be-displayed rank of the at least one target item information. Further, the display information may be transmitted to the communicatively connected terminal device through the transmitting unit 504, so that the terminal device displays at least one target item information according to the display information. Therefore, when the click rate of the target item information is generated, the historical display rank of the target item information can be taken into consideration, and the click rate of the target item information in each display rank can be obtained.
Referring now to FIG. 6, there is illustrated a schematic diagram of a computer system 600 suitable for use with a server embodying embodiments of the present application. The server illustrated in fig. 6 is merely an example, and should not be construed as limiting the functionality and scope of use of embodiments of the present application.
As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU) 601, which 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 required for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other through a bus 604. An input/output (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, mouse, etc.; an output portion 607 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), or the like; 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 drive 610 is also connected to the I/O interface 605 as needed. Removable media 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is installed as needed on drive 610 so that a computer program read therefrom is installed as needed into storage section 608.
In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network through the communication portion 609, and/or installed from the removable medium 611. The above-described functions defined in the method of the present application are performed when the computer program is executed by a Central Processing Unit (CPU) 601.
The computer readable medium of the present application may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present application, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations 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 ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present 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 involved in the embodiments of the present application may be implemented in software or in hardware. The described units may also be provided in a processor, for example, described as: a processor includes an acquisition unit, a determination unit, a first generation unit, and a transmission unit. The names of these units do not constitute a limitation on the unit itself in some cases, and for example, the determining unit may also be described as "a unit that determines the rank to be presented of at least one target item information based on the obtained at least one click rate information".
As another aspect, the present application also provides a computer-readable medium that may be contained in the server described in the above embodiment; or may exist alone without being assembled into the server. The computer readable medium carries one or more programs which, when executed by the server, cause the server to: for target item information in at least one target item information, acquiring a feature set related to the target item information, and inputting features included in the feature set related to the target item information into a pre-trained click rate prediction model to obtain click rate information of the target item information, wherein the feature set comprises display level information representing historical display levels of the target item information, the click rate information is used for representing click rates of the target item information in preset numbers of display levels, and the click rate prediction model is used for representing a corresponding relation between the feature set related to the item information and the click rate information of the item information; determining the to-be-presented rank of at least one target item information according to the obtained at least one click rate information; generating a display message according to the to-be-displayed rank of the at least one target item information, wherein the display message is used for identifying the to-be-displayed rank of the at least one target item information; and sending the display information to the terminal equipment in communication connection so that the terminal equipment displays at least one piece of target article information according to the display information.
The above description is only illustrative of the preferred embodiments of the present application and of the principles of the technology employed. It will be appreciated by persons skilled in the art that the scope of the application referred to in the present application is not limited to the specific combinations of the technical features described above, but also covers other technical features formed by any combination of the technical features described above or their equivalents without departing from the inventive concept described above. Such as the above-mentioned features and the technical features disclosed in the present application (but not limited to) having similar functions are replaced with each other.

Claims (10)

1. An information display method, comprising:
for target item information in at least one target item information, acquiring a feature set related to the target item information, and inputting features included in the feature set related to the target item information into a pre-trained click rate prediction model to obtain click rate information of the target item information, wherein the feature set comprises display level information representing historical display levels of the target item information, the click rate information is used for representing click rates of the target item information in at least two display levels, and the click rate prediction model is used for representing a corresponding relation between the feature set related to the item information and the click rate information of the item information;
Determining the to-be-presented rank of the at least one target item information according to the obtained at least one click rate information;
generating display information according to the to-be-displayed rank of the at least one target item information, wherein the display information is used for identifying the to-be-displayed rank of the at least one target item information;
and sending the display information to a terminal device in communication connection so that the terminal device displays the at least one piece of target article information according to the display information.
2. The method of claim 1, wherein the click rate prediction model is trained by:
acquiring a sample set, wherein the sample comprises a sample feature set and sample click rate information, the sample feature set is related to sample article information, the sample feature set comprises display level information for representing historical display levels of the sample article information, and the sample click rate information is used for representing click rates of the sample article information in at least two display levels;
selecting at least one sample from the set of samples, and performing the training steps of: respectively inputting sample features included in the sample feature set of the selected at least one sample into an initial model to obtain click rate information of sample article information related to the sample feature set of each sample in the at least one sample; comparing click rate information of sample item information related to a sample feature set of each of the at least one sample with corresponding sample click rate information; determining whether the initial model is trained based on the comparison result; in response to determining that the training is complete, an initial model of the training completion is determined as a click rate prediction model.
3. The method of claim 2, wherein training the click rate prediction model further comprises:
and in response to determining that the training is not completed, adjusting relevant parameters of the initial model, selecting at least one unused sample from the sample set, and continuing to execute the training step by using the adjusted initial model as the initial model.
4. The method of claim 1, wherein the feature set further comprises at least one of: the user characteristics related to the target article information, the article characteristics of the article indicated by the target article information, and the search information of the target article information, which characterizes whether the target article information is clicked or not, are searched by a user.
5. The method according to any one of claims 1-4, wherein prior to said determining a to-be-presented ranking of the at least one target item information from the resulting at least one click-through rate information, the method further comprises:
generating a target article information set, wherein the target article information set comprises the at least one piece of target article information.
6. The method of claim 5, wherein the determining the to-be-presented ranking of the at least one target item information according to the obtained at least one click rate information comprises:
Selecting a presentation rank from the at least two presentation ranks, and performing the determining step of: determining the product of the click rate of the target item information in the target item information set in the display level and the multiplication value corresponding to the target item information based on the at least one click rate information; determining the target article information indicated by the maximum product as target article information to be displayed by the display rank;
the method further comprises the steps of: deleting target item information indicated by the maximum product from the set of target item information in response to determining that there is an unselected presentation rank; using the target item information set after deleting the target item information as a target item information set; selecting an unselected presentation rank from the at least two presentation ranks, and continuing to perform the determining step.
7. The method of any of claims 1-4, wherein the generating presentation information according to the to-be-presented ranking of the at least one target item information comprises:
and sequencing the at least one piece of target article information according to the sequence of the sequence number of the to-be-displayed rank of the at least one piece of target article information from small to large, and generating sequencing information as the display information.
8. An information presentation apparatus comprising:
the acquisition unit is configured to acquire a feature set related to the target item information for the target item information in at least one target item information, and input features included in the feature set related to the target item information into a pre-trained click rate prediction model to obtain click rate information of the target item information, wherein the feature set comprises display level information representing historical display levels of the target item information, the click rate information is used for representing click rates of the target item information in at least two display levels, and the click rate prediction model is used for representing a corresponding relation between the feature set related to the item information and the click rate information of the item information;
a determining unit configured to determine a rank to be presented of the at least one target item information according to the obtained at least one click rate information;
a first generation unit configured to generate display information according to a to-be-displayed rank of the at least one target item information, wherein the display information is used for identifying the to-be-displayed rank of the at least one target item information;
and the sending unit is configured to send the display information to the terminal equipment in communication connection so that the terminal equipment displays the at least one piece of target article information according to the display information.
9. 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, causes the one or more processors to implement the method of any of claims 1-7.
10. A computer readable medium having stored thereon a computer program, wherein the program when executed by a processor implements the method of any of claims 1-7.
CN201910151726.8A 2019-02-28 2019-02-28 Information display method and device Active CN111626805B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910151726.8A CN111626805B (en) 2019-02-28 2019-02-28 Information display method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910151726.8A CN111626805B (en) 2019-02-28 2019-02-28 Information display method and device

Publications (2)

Publication Number Publication Date
CN111626805A CN111626805A (en) 2020-09-04
CN111626805B true CN111626805B (en) 2023-11-03

Family

ID=72258722

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910151726.8A Active CN111626805B (en) 2019-02-28 2019-02-28 Information display method and device

Country Status (1)

Country Link
CN (1) CN111626805B (en)

Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000163477A (en) * 1998-11-27 2000-06-16 Nec Corp Display advertisement selecting method
CN103793388A (en) * 2012-10-29 2014-05-14 阿里巴巴集团控股有限公司 Method and device for search result sorting
CN104992348A (en) * 2015-06-24 2015-10-21 深圳市腾讯计算机系统有限公司 Method and device for displaying information
CN105160548A (en) * 2015-08-20 2015-12-16 北京奇虎科技有限公司 Method and apparatus for predicting advertisement click-through rate
CN105183772A (en) * 2015-08-07 2015-12-23 百度在线网络技术（北京）有限公司 Release information click rate estimation method and apparatus
CN105447045A (en) * 2014-09-02 2016-03-30 阿里巴巴集团控股有限公司 Information ordering method and device and information providing method and system
CN105653545A (en) * 2014-11-10 2016-06-08 阿里巴巴集团控股有限公司 Method and device for providing service object information in page
CN106033583A (en) * 2016-05-18 2016-10-19 杭州算子科技有限公司 Advertisement paid listing method and system applied to electronic commerce
CN107105031A (en) * 2017-04-20 2017-08-29 北京京东尚科信息技术有限公司 Information-pushing method and device
CN107590689A (en) * 2017-08-25 2018-01-16 北京奇艺世纪科技有限公司 A kind of ad data recommends method and system
CN107767174A (en) * 2017-10-19 2018-03-06 厦门美柚信息科技有限公司 The Forecasting Methodology and device of a kind of ad click rate
CN108108998A (en) * 2017-12-14 2018-06-01 百度在线网络技术（北京）有限公司 Showing advertisement method and apparatus, server, storage medium
CN108230101A (en) * 2017-12-29 2018-06-29 百度在线网络技术（北京）有限公司 Information recommendation method and device
CN108805305A (en) * 2017-04-28 2018-11-13 北京嘀嘀无限科技发展有限公司 The method, apparatus and storage medium of the director information for going to one place are provided

Patent Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000163477A (en) * 1998-11-27 2000-06-16 Nec Corp Display advertisement selecting method
CN103793388A (en) * 2012-10-29 2014-05-14 阿里巴巴集团控股有限公司 Method and device for search result sorting
CN105447045A (en) * 2014-09-02 2016-03-30 阿里巴巴集团控股有限公司 Information ordering method and device and information providing method and system
CN105653545A (en) * 2014-11-10 2016-06-08 阿里巴巴集团控股有限公司 Method and device for providing service object information in page
CN104992348A (en) * 2015-06-24 2015-10-21 深圳市腾讯计算机系统有限公司 Method and device for displaying information
CN105183772A (en) * 2015-08-07 2015-12-23 百度在线网络技术（北京）有限公司 Release information click rate estimation method and apparatus
CN105160548A (en) * 2015-08-20 2015-12-16 北京奇虎科技有限公司 Method and apparatus for predicting advertisement click-through rate
CN106033583A (en) * 2016-05-18 2016-10-19 杭州算子科技有限公司 Advertisement paid listing method and system applied to electronic commerce
CN107105031A (en) * 2017-04-20 2017-08-29 北京京东尚科信息技术有限公司 Information-pushing method and device
CN108805305A (en) * 2017-04-28 2018-11-13 北京嘀嘀无限科技发展有限公司 The method, apparatus and storage medium of the director information for going to one place are provided
CN107590689A (en) * 2017-08-25 2018-01-16 北京奇艺世纪科技有限公司 A kind of ad data recommends method and system
CN107767174A (en) * 2017-10-19 2018-03-06 厦门美柚信息科技有限公司 The Forecasting Methodology and device of a kind of ad click rate
CN108108998A (en) * 2017-12-14 2018-06-01 百度在线网络技术（北京）有限公司 Showing advertisement method and apparatus, server, storage medium
CN108230101A (en) * 2017-12-29 2018-06-29 百度在线网络技术（北京）有限公司 Information recommendation method and device

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
基于细粒度标签的在线视频广告投放机制研究;陆枫;王子锐;廖小飞;金海;;计算机研究与发展(第12期);全文 *
金紫嫣 ; 张娟 ; 李向军 ; 温海平 ; 张华薇 ; .一种带标签的协同过滤广告推荐算法.计算机工程.2018,(第04期),全文. *

Also Published As

Publication number Publication date
CN111626805A (en) 2020-09-04

Similar Documents

Publication Publication Date Title
CN109460513B (en) Method and apparatus for generating click rate prediction model
CN109522483B (en) Method and device for pushing information
KR102155261B1 (en) Wide and deep machine learning models
CN107330715B (en) Method and device for selecting picture advertisement material
US10354184B1 (en) Joint modeling of user behavior
CN111581926B (en) Document generation method, device, equipment and computer readable storage medium
CN110866040B (en) User portrait generation method, device and system
CN111695041B (en) Method and device for recommending information
CN109377284B (en) Method and electronic equipment for pushing information
CN114119123A (en) Information pushing method and device
CN113779186A (en) Text generation method and device
CN117216393A (en) Information recommendation method, training method and device of information recommendation model and equipment
CN112989174A (en) Information recommendation method and device, medium and equipment
CN111626805B (en) Information display method and device
CN115293818A (en) Advertisement putting and selecting method and device, equipment and medium thereof
US10699127B1 (en) Method and apparatus for adjusting parameter
CN113327151A (en) Commodity object recommendation method and device, computer equipment and storage medium
CN111125501B (en) Method and device for processing information
CN111506643B (en) Method, device and system for generating information
CN112308942A (en) Method and apparatus for generating image
CN111666481B (en) Data mining method, device, computer readable medium and electronic equipment
CN113327147A (en) Method and device for displaying article information
CN111859220A (en) Method and device for displaying information
CN111784377A (en) Method and apparatus for generating information
CN116911304B (en) Text recommendation method and device

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