CN115795176A - Material sequence generation method, apparatus, device, medium, and program product - Google Patents

Material sequence generation method, apparatus, device, medium, and program product Download PDF

Info

Publication number
CN115795176A
CN115795176A CN202211248320.XA CN202211248320A CN115795176A CN 115795176 A CN115795176 A CN 115795176A CN 202211248320 A CN202211248320 A CN 202211248320A CN 115795176 A CN115795176 A CN 115795176A
Authority
CN
China
Prior art keywords
recommended
creative
target
creative material
historical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211248320.XA
Other languages
Chinese (zh)
Inventor
刘银星
阮涛
张政
吕晶晶
庞新强
王维珍
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 Jingdong Century Trading Co Ltd
Beijing Wodong Tianjun Information Technology Co Ltd
Original Assignee
Beijing Jingdong Century Trading Co Ltd
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 Jingdong Century Trading Co Ltd, Beijing Wodong Tianjun Information Technology Co Ltd filed Critical Beijing Jingdong Century Trading Co Ltd
Priority to CN202211248320.XA priority Critical patent/CN115795176A/en
Publication of CN115795176A publication Critical patent/CN115795176A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Abstract

Embodiments of the present disclosure disclose material sequence generation methods, apparatus, devices, media, and program products. One embodiment of the method comprises: in response to the fact that the target creative material to be recommended exists in the creative material set to be recommended, determining a target creative material set to be recommended; for each target creative material to be recommended, screening a preset number of target historical recommendation creative materials from the historical recommendation creative materials in a centralized manner; generating an estimated click rate of each target creative material to be recommended according to the click rate of each target historical recommendation creative material; determining the click rate of each multi-time recommended creative material; and sequencing all the creative materials to be recommended to obtain a creative material sequence to be recommended. The embodiment is related to artificial intelligence, and a more accurate material recommendation sequence aiming at the creative material set to be recommended can be generated by determining the click rate corresponding to the creative material to be recommended of the target, so that the material recommendation effect is improved.

Description

Material sequence generation method, apparatus, device, medium, and program product
Technical Field
Embodiments of the present disclosure relate to the field of computer technologies, and in particular, to a method, an apparatus, a device, a medium, and a program product for generating a material sequence.
Background
Currently, the material recommending end often displays the item characteristics of each item by pushing the material (for example, the item creative image) of each item. For pushing newly generated materials, the commonly adopted mode is as follows: and setting a specific threshold value aiming at the newly generated material by using a greedy algorithm so that the subsequent material recommending end gives sufficient exposure to the material recommending end.
However, the inventor finds that when the newly generated material is pushed in the above manner, the following technical problems often exist:
whether the newly generated material is a high-quality creative material with high quality and high potential cannot be determined, and the newly generated material is directly exposed sufficiently, so that the recommendation effect of the actual material is possibly poor.
The above information disclosed in this background section is only for enhancement of understanding of the background of the inventive concept and, therefore, it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art.
Disclosure of Invention
This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.
Some embodiments of the present disclosure propose a material sequence generation method, apparatus, device, medium, and program product to solve the technical problems mentioned in the background section above.
In a first aspect, some embodiments of the present disclosure provide a method for generating a material sequence, including: in response to the fact that the target creative material to be recommended exists in the acquired creative material set to be recommended, determining a target creative material set to be recommended, wherein the target creative material to be recommended is a creative material meeting target material determination conditions; screening a preset number of target historical recommendation creative materials from the obtained historical recommendation creative material sets to obtain a target historical recommendation creative material set for each target creative material to be recommended in the target creative material set to be recommended, wherein the material association degree between the target historical recommendation creative material and the target creative material to be recommended is larger than a preset threshold value; generating an estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each target history recommended creative material in the obtained target history recommended creative material set group; determining the click rate of each multi-time recommended creative material in a multi-time recommended creative material set, wherein the multi-time recommended creative material set is a material set of the creative material set to be recommended except the target creative material set to be recommended; and sequencing all creative materials to be recommended included in the creative material set to be recommended according to the estimated click rate set corresponding to the creative material set to be recommended and the click rate set corresponding to the creative material set recommended for multiple times to obtain a creative material sequence to be recommended.
Optionally, the preset material conditions include: a plurality of preset material sub-conditions, the plurality of preset material sub-conditions comprising: the time interval between the material generation time corresponding to the creative material to be recommended and the current time is less than the preset time, and the exposure corresponding to the creative material to be recommended is less than the preset exposure value; and the above-mentioned creative material set to be recommended of determining the target includes: determining material generation time and exposure corresponding to each creative material to be recommended in the creative material set to be recommended; and determining the creative materials to be recommended, which meet the plurality of preset material sub-conditions in a centralized manner, as target creative materials to be recommended.
Optionally, the filtering a predetermined number of target history recommended creative materials from the obtained history recommended creative material sets to obtain a target history recommended creative material set includes: inputting each historical recommended creative material in the historical recommended creative material set to a pre-trained material coding model to generate a historical material coding vector to obtain a historical material coding vector set; inputting the target creative material to be recommended to the material coding model to generate a coding vector of the material to be recommended; and according to the historical material coding vector set and the material coding vector to be recommended, screening out historical recommendation creative materials with vector distances meeting a preset distance condition from the historical recommendation creative material set, and taking the historical recommendation creative materials as target historical recommendation creative materials to obtain a target historical recommendation creative material set, wherein the vector distance is the vector distance between the historical material coding vector in the historical material coding vector set and the material coding vector to be recommended.
Optionally, the generating of the estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each target history recommended creative material in the obtained target history recommended creative material set group includes: and calculating the click rate of each target historical recommendation creative material in the target historical recommendation creative material set corresponding to the target creative material to be recommended to obtain a calculated value serving as the estimated click rate corresponding to the target creative material to be recommended.
Optionally, the material coding model is trained by the following steps: obtaining a training sample set, wherein the training samples comprise: training creative materials and article product word labels; training an initial article product word classification model by using the training sample set to obtain a trained article product word classification model, wherein the trained article product word classification model comprises: and (5) a material coding model.
Optionally, the method further includes: responding to the situation that no target creative material to be recommended exists in the creative material set to be recommended, inputting each creative material to be recommended in the creative material set to be recommended to a material click rate generation model, and outputting a material click rate to obtain a material click rate set; and sequencing all the creative materials to be recommended included in the creative material set to be recommended according to the material click rate set to obtain a creative material sequence to be recommended.
Optionally, the method further includes: and sending at least one creative material to be recommended, which has a click rate meeting a preset click rate condition, in the creative material sequence to be recommended to a material recommending end.
In a second aspect, some embodiments of the present disclosure provide a material sequence generating apparatus, including: the device comprises a first determining unit, a second determining unit and a judging unit, wherein the first determining unit is configured to respond to the fact that a target creative material to be recommended exists in the acquired creative material set to be recommended, and the target creative material to be recommended is a creative material meeting target material determining conditions; the screening unit is configured to screen a preset number of target historical recommendation creative materials from the acquired historical recommendation creative materials for each target to-be-recommended creative material in the target to-be-recommended creative material set to obtain a target historical recommendation creative material set, wherein the material association degree between the target historical recommendation creative material and the target to-be-recommended creative material is greater than a preset threshold value; the generating unit is configured to generate the estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each target history recommended creative material in the obtained target history recommended creative material set group; the second determining unit is configured to determine the click rate of each multi-time recommendation creative material in a multi-time recommendation creative material set, wherein the multi-time recommendation creative material set is a material set of the to-be-recommended creative material set except the target to-be-recommended creative material set; and the sequencing unit is configured to sequence all the creative materials to be recommended included in the creative material set to be recommended according to the estimated click rate set corresponding to the target creative material set to be recommended and the click rate sets corresponding to the multiple recommended creative material sets to obtain a creative material sequence to be recommended.
Optionally, the preset material conditions include: a plurality of preset material sub-conditions, the plurality of preset material sub-conditions comprising: the time interval between the material generation time corresponding to the creative material to be recommended and the current time is less than the preset time, and the exposure corresponding to the creative material to be recommended is less than the preset exposure value; and the first determination unit may be configured to: determining material generation time and exposure corresponding to each creative material to be recommended in the creative material set to be recommended; and determining the creative materials to be recommended, which meet the plurality of preset material sub-conditions in a centralized manner, as target creative materials to be recommended.
Optionally, the screening unit may be configured to: inputting each historical recommended creative material in the historical recommended creative material set to a pre-trained material coding model to generate a historical material coding vector to obtain a historical material coding vector set; inputting the target creative material to be recommended to the material coding model to generate a coding vector of the material to be recommended; and according to the historical material coding vector set and the material coding vector to be recommended, screening historical recommended creative materials with vector distances meeting preset distance conditions from the historical recommended creative material set, and taking the historical recommended creative materials as target historical recommended creative materials to obtain a target historical recommended creative material set, wherein the vector distances are the vector distances between the historical material coding vectors in the historical material coding vector set and the material coding vectors to be recommended.
Optionally, the generating unit may be configured to: and calculating the click rate of each target historical recommendation creative material in the target historical recommendation creative material set corresponding to the target creative material to be recommended to obtain a calculated value serving as the estimated click rate corresponding to the target creative material to be recommended.
Optionally, the material coding model is trained by the following steps: obtaining a training sample set, wherein the training samples comprise: training creative materials and article product word labels; training an initial article product word classification model by using the training sample set to obtain a trained article product word classification model, wherein the trained article product word classification model comprises: and (5) material coding model.
Optionally, the apparatus further comprises: responding to the situation that no target creative material to be recommended exists in the creative material set to be recommended, inputting each creative material to be recommended in the creative material set to be recommended to a material click rate generation model, and outputting a material click rate to obtain a material click rate set; and sequencing all the creative materials to be recommended included in the creative material set to be recommended according to the material click rate set to obtain a creative material sequence to be recommended.
Optionally, the apparatus further comprises: and sending at least one creative material to be recommended, which has a click rate meeting a preset click rate condition, in the creative material sequence to be recommended to a material recommending end.
In a third aspect, some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, which when executed by 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, some embodiments of the disclosure provide a computer readable medium having a computer program stored thereon, where the program when executed by a processor implements a method as described in any of the implementations of the first aspect.
In a fifth aspect, some embodiments of the present disclosure provide a computer program product comprising a computer program that, when executed by a processor, implements the method described in any of the implementations of the first aspect above.
The above embodiments of the present disclosure have the following beneficial effects: according to the material sequence generation method of some embodiments of the disclosure, a more accurate material recommendation sequence aiming at the creative material set to be recommended can be generated by determining the click rate corresponding to the target creative material to be recommended, so that the material recommendation effect is improved. Specifically, the reason why the material recommendation effect is not good is that: whether newly generated materials are high-quality creative materials with high quality and high potential cannot be determined, and sufficient exposure is directly given to the newly generated materials, so that the actual material recommendation effect is possibly poor. Based on this, in the material sequence generation method of some embodiments of the present disclosure, first, in response to that a target creative material to be recommended exists in an obtained creative material set to be recommended, a target creative material set to be recommended is determined from the creative material set to be recommended, so as to subsequently generate a click rate corresponding to each target creative material to be recommended. Then, for each target creative material to be recommended in the target creative material set to be recommended, screening a preset number of target historical recommendation creative materials from the acquired historical recommendation creative material sets to obtain a target historical recommendation creative material set. Here, since the target creative material to be recommended does not have a real click rate, it is not possible to determine whether the target creative material to be recommended is a high-quality creative material with high quality and high potential. Therefore, the preset number of target historical recommendation creative materials with the material association degree with the target recommendation creative material are selected from the historical recommendation creative materials, and the click rate of the target recommendation creative materials can be conveniently estimated subsequently. Then, according to the click rate of each target historical recommendation creative material in the obtained target historical recommendation creative material set group, the estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set can be estimated more accurately. And then, determining the click rate of each multi-time recommended creative material in the multi-time recommended creative material set so as to be used for sequencing the materials of the subsequent creative material set to be recommended. And finally, sequencing all creative materials to be recommended included in the creative material set to be recommended according to the estimated click rate set corresponding to the creative material set to be recommended and the click rate set corresponding to the creative material set recommended for multiple times to obtain a creative material sequence to be recommended. Here, the material quality of the creative material is measured by using the estimated click rate of the target recommended creative material as a standard. And measuring the material quality of the multi-time recommended creative materials through the click rate of the multi-time recommended creative materials. Therefore, the recommendation creative material sets are sorted based on the click rates corresponding to the various creative materials to be recommended in the creative material sets to be recommended, and the effect of sorting the materials according to the material quality corresponding to the materials can be achieved. The material recommendation effect can be greatly improved by recommending each material based on the creative material sequence to be recommended.
Drawings
The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and elements are not necessarily drawn to scale.
1-2 are schematic diagrams of one application scenario of a material sequence generation method according to some embodiments of the present disclosure;
FIG. 3 is a flow diagram of some embodiments of a method of material sequence generation according to the present disclosure;
FIG. 4 is a flow diagram of further embodiments of a method of generating a sequence of material according to the present disclosure;
FIG. 5 is a block diagram of some embodiments of a material sequence generating apparatus according to the present disclosure;
FIG. 6 is a schematic structural diagram of an electronic device suitable for use in implementing some embodiments of the present disclosure.
Detailed Description
Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is to be understood that the disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.
It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings. The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.
It should be noted that the terms "first", "second", and the like in the present disclosure are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence of the functions performed by the devices, modules or units.
It is noted that references to "a" or "an" in this disclosure are intended to be illustrative rather than limiting, and that those skilled in the art will appreciate that references to "one or more" are intended to be exemplary and not limiting unless the context clearly indicates otherwise.
The names of messages or information exchanged between devices in the embodiments of the present disclosure are for illustrative purposes only, and are not intended to limit the scope of the messages or information.
The operations of collection, storage, use, etc. of creative materials (e.g., item creative images, item creative videos) involved in the present disclosure go to the extent that relevant organizations or individuals have obligations including developing material security impact assessment, fulfilling notification obligations to material creating entities, soliciting authorization and consent of material creating entities in advance, etc., before performing the respective operations.
The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
Fig. 1-2 are schematic diagrams of an application scenario of a material sequence generation method according to some embodiments of the present disclosure.
In the application scenarios of fig. 1-2, first, the electronic device 101 can determine a set of target creative materials to be recommended in response to the presence of the target creative materials to be recommended in the set of acquired creative materials to be recommended 101. The target creative materials to be recommended are creative materials meeting target material determination conditions. In the application scenario, the creative material set 101 to be recommended may include: creative material to be recommended 1011, creative material to be recommended 1012, creative material to be recommended 1013, and creative material to be recommended 1014. The target to be recommended creative material can be the to be recommended creative material 1013. The target material determination condition can be that the material content corresponding to the target creative material to be recommended is a triangle. Then, for each of the target creative materials to be recommended in the target creative material set to be recommended, the electronic device 101 may screen a predetermined number of target historical recommended creative materials from the acquired historical recommended creative material set 102 to obtain a target historical recommended creative material set. And the material association degree between the target historical recommendation creative material and the target to-be-recommended creative material is greater than a preset threshold value. In this application scenario, the historical recommended creative material set 102 includes: historical recommended creative material 1021, historical recommended creative material 1022, historical recommended creative material 1023, and historical recommended creative material 1024. The predetermined number may be 2 for the target to be recommended creative material being to be recommended creative material 1013. The corresponding target history recommended creative material set comprises: historical recommended creative assets 1022 and historical recommended creative assets 1024. Then, the electronic device 101 may generate the estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each target history recommended creative material in the obtained target history recommended creative material set. In this application scenario, the creative material to be recommended for the target is creative material to be recommended 1013, and the corresponding creative material set of the target history recommendation includes: historical recommended creative material 1022 and historical recommended creative material 1024. The historically recommended creative assets 1022 may be "0.3" corresponding to the click through rate 106. The historically recommended creative material 1024 may be "0.5" for click rate 107. And averaging the click rate 106 and the click rate 107 to obtain an average click rate which is used as an estimated click rate 108 corresponding to the creative material 1013 to be recommended. Wherein, the estimated click rate 108 may be "0.4". In turn, the electronic device 101 can determine the click through rate of each of the multiple recommended creative materials in the set of multiple recommended creative materials. The creative material sets recommended for multiple times are material sets with the creative material sets to be recommended removed from the target creative material sets to be recommended. In the application scenario, the multiple recommendation of the creative material set includes: a to-be-recommended creative material 1011, a to-be-recommended creative material 1012, and a to-be-recommended creative material 1014. The click through rate 103 of the creative material 1011 to be recommended may be "0.45". The click through rate 104 for the creative material 1012 to be recommended may be "0.6". The click through rate 105 for the creative material 1014 to be recommended may be "0.1". Finally, the electronic device 101 may sort the creative materials to be recommended included in the creative material set 101 to be recommended according to the estimated click rate set corresponding to the target creative material set to be recommended and the click rate sets corresponding to the multiple recommended creative material sets, so as to obtain a creative material sequence 109 to be recommended. In the application scenario, the creative material sets 101 to be recommended are sorted according to the order of the click rates of the creative materials to be recommended from small to large, and a creative material sequence 109 to be recommended is obtained. The to-be-recommended creative material sequence 109 may be [ to-be-recommended creative material 1014, to-be-recommended creative material 1013, to-be-recommended creative material 1011, to-be-recommended creative material 1012].
The electronic device 101 may be hardware or software. When the electronic device is hardware, the electronic device may be implemented as a distributed cluster formed by a plurality of servers or terminal devices, or may be implemented as a single server or a single terminal device. When the electronic device is embodied as software, it may be installed in the above-listed hardware devices. It may be implemented, for example, as multiple software or software modules to provide distributed services, or as a single software or software module. And is not particularly limited herein.
It should be understood that the number of electronic devices in fig. 1-2 is merely illustrative. There may be any number of electronic devices, as desired for an implementation.
With continuing reference to FIG. 3, a flow 300 of some embodiments of a method of generating a sequence of material according to the present disclosure is shown. The material sequence generation method comprises the following steps:
step 301, in response to the fact that the target creative material to be recommended exists in the acquired creative material set to be recommended, determining a target creative material set to be recommended.
In some embodiments, an executing agent of the above-described material sequence generation method (e.g., electronic device 101 of figure 1) can determine a set of target creative materials to be recommended in response to the presence of the target creative material to be recommended in the set of acquired creative materials to be recommended. The target creative materials to be recommended are creative materials meeting target material determination conditions. The target material determination condition may be a condition for determining whether the creative material to be recommended is a new material (i.e., a newly generated material). The target creative material to be recommended is a newly generated creative material to be recommended. In practice, the target material determination conditions are correspondingly changed according to different judgment bases of newly generated creative materials to be recommended. For example, the criterion for the newly generated creative material to be recommended is the recommendation frequency, and the target material condition may be that the recommendation frequency of the target creative material to be recommended is less than a predetermined frequency. For another example, the criterion of the newly generated creative material to be recommended is the material delivery duration, and the target material condition may be that the material delivery duration of the target creative material to be recommended is less than a predetermined duration. The creative material set to be recommended can be at least one creative material to be recommended. In the e-market scenario, the creative material may be a creative item material designed for the item. The creative material can be a material in a picture form and can also be a material in a video form.
It should be noted that when the target creative material to be recommended appears, the material recommending end is in a cold start period to determine how to recommend the target creative material to be recommended.
As an example, for the target material recommendation condition that the recommendation frequency of the target creative material to be recommended is less than the predetermined frequency, the execution main body may determine whether the target creative material to be recommended exists in the acquired creative material set to be recommended by the following steps:
the method comprises the steps of firstly, determining material recommendation times corresponding to each creative material to be recommended in a creative material set to be recommended, and obtaining a material recommendation time set.
And secondly, in response to the fact that the material recommendation times are smaller than 1 in the material recommendation time set, determining that the target creative material to be recommended exists in the creative material set to be recommended.
In some optional implementations of some embodiments, the preset material condition includes: a plurality of preset material sub-conditions. Wherein, the plurality of preset material sub-conditions comprise: the time interval between the material generation time corresponding to the creative material to be recommended and the current time is less than the preset time, and the exposure corresponding to the creative material to be recommended is less than the preset exposure value. The method for determining the target creative material set to be recommended comprises the following steps:
the method comprises the steps of firstly, determining material generation time and exposure corresponding to each creative material to be recommended in the creative material set to be recommended.
For example, the set of creative materials to be recommended includes: the creative material recommendation system comprises a first creative material to be recommended, a second creative material to be recommended and a third creative material to be recommended. The material generation time of the first creative material to be recommended is 2021 year, 8 month and 1 day. The material generation time of the second creative material to be recommended is 2021 year, 9 month and 1 day. The material generation time of the creative material to be recommended is 2022, 4 month and 15 days. The exposure of the first creative material to be recommended is 1000 times. The exposure of the second creative material to be recommended is 500 times. The exposure of the creative material to be recommended for the third time is 3000 times.
And secondly, determining the creative materials to be recommended, which meet the plurality of preset material sub-conditions in a centralized manner, as target creative materials to be recommended.
For example, the predetermined period is 1 month. The predetermined exposure value is 2500. The current time is 2022 years, 5 months and 1 day. The time interval corresponding to the first creative material to be recommended is 8 months. The time interval corresponding to the second creative material to be recommended is 7 months. The time interval corresponding to the third creative material to be recommended is half a month. Because the exposure of the creative material to be recommended for the third time is 3000 times, the creative material to be recommended for the third time meets a plurality of preset material sub-conditions. Namely, the third creative material to be recommended is determined as the target creative material to be recommended.
Step 302, for each target to-be-recommended creative material in the target to-be-recommended creative material set, screening a predetermined number of target historical recommended creative materials from the acquired historical recommended creative material set to obtain a target historical recommended creative material set.
In some embodiments, the execution subject may filter a predetermined number of target historical recommended creative materials from the obtained historical recommended creative material sets for each of the target creative materials to be recommended in the target creative material sets to obtain target historical recommended creative material sets. And the material association degree between the target historical recommendation creative material and the target to-be-recommended creative material is greater than a preset threshold value. The material association degree characterizes the material similarity between the target historical recommendation creative material and the target creative material to be recommended. For example, the material similarity may be one of: material content similarity and material style similarity. The historical recommendation creative material can be creative material which is historically recommended multiple times by the material recommending end. The material recommending end can be a terminal for recommending materials. In practice, the historically recommended creative material set may be a plurality of creative materials that the material recommender recommended within a predetermined historical time period. For example, the current time is 2022 years, 5 months, 1 day. The predetermined historical period of time may be from 1/month 1/2021 to 3/month 1/year.
As an example, the execution agent may first input each of the historical recommended creative material in the set of historical recommended creative materials to a creative material style determination model to generate a historical creative material style. Wherein the creative material style determination model may be a model that determines the style of the material. For example, the creative material style determination model may be a multi-layer Convolutional Neural Networks (CNN) model. And then, inputting the target creative material to be recommended to the creative material style determination model to generate a creative material style to be recommended. And finally, screening out historical recommended creative materials with the same creative material style as the creative materials to be recommended from the historical recommended creative material set, and taking the historical recommended creative materials as target historical recommended creative materials to obtain a target historical recommended creative material set.
In some alternative implementations of some embodiments, the filtering of a predetermined number of target historical recommended creative materials from the set of retrieved historical recommended creative materials to obtain a set of target historical recommended creative materials may include:
firstly, inputting each historical recommended creative material in the historical recommended creative material set to a pre-trained material coding model to generate a historical material coding vector to obtain a historical material coding vector set.
The material coding model may be a model in which the material is vector-coded. The historical material coding vectors can represent material characteristic information of the historical recommended creative materials.
Here, the execution agent uses the history material coding vector as an index, and stores the corresponding history recommended creative material as a value in a faiss index library.
Optionally, the material coding model includes: a residual network model and a plurality of serial full link layers; and inputting each historical recommended creative material in the historical recommended creative material set into a pre-trained material coding model to generate a historical material coding vector, wherein the method comprises the following steps:
and step 1, inputting the historical recommended creative materials into the residual error network model to obtain a model output result.
And 2, inputting the output result of the model to the plurality of serial full-connection layers to obtain the historical material coding vector.
And secondly, inputting the target creative material to be recommended to the material coding model to generate a coding vector of the material to be recommended.
The coding vector of the material to be recommended can represent material characteristic information of the target creative material to be recommended.
Here, the execution main body takes the coding vector of the material to be recommended as an index, and takes the corresponding target creative material to be recommended as a value, and stores the value in the faiss index library.
And thirdly, according to the historical material coding vector set and the material coding vector to be recommended, screening out historical recommended creative materials with vector distances meeting preset distance conditions from the historical recommended creative material set, using the historical recommended creative materials as target historical recommended creative materials, and obtaining a target historical recommended creative material set.
And the vector distance is the vector distance between the historical material coding vector in the historical material coding vector set and the material coding vector to be recommended. The predetermined distance condition may be that the vector distance between the historical recommended creative material and the coding vector of the material to be recommended is greater than a predetermined distance value. For example, the predetermined distance value may be 0.5.
As an example, first, the execution subject may determine, as a vector distance, a cosine distance between each of the history material encoding vectors in the history material encoding vector set and the material encoding vector to be recommended. And then, history recommendation creative materials with vector distances meeting preset distance conditions are screened out from the history recommendation creative material sets and serve as target history recommendation creative materials, and a target history recommendation creative material set is obtained.
Optionally, the material coding model is trained by the following steps:
in the first step, a training sample set is obtained.
Wherein, training the sample includes: training creative materials and article product word labels.
And secondly, training the initial article product word classification model by using the training sample set to obtain a trained article product word classification model.
Wherein, the above-mentioned article product word classification model after training includes: and (5) a material coding model. The item product word classification model can be a model for determining the item product words of the items corresponding to the creative materials. The article product word classification model may include: a material coding model and a plurality of fully connected layers connected in series.
And 303, generating the estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each target history recommended creative material in the obtained target history recommended creative material set group.
In some embodiments, the execution agent may generate a pre-estimated click rate for each of the target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each of the obtained target historical recommended creative material sets in the target historical recommended creative material set. And each target historical recommendation creative material has a corresponding click rate.
As an example, first, the execution body may determine a maximum click through rate and a minimum click through rate among the various target historical recommended creative materials in the set of target historical recommended creative materials. And then, removing the maximum click rate and the minimum click rate from the click rate set corresponding to the target historical recommendation creative material set group to obtain a removed click rate set. And finally, determining the average value of all click rates in the removed click rate set as the click rate corresponding to the target historical recommendation creative material.
Step 304, determining the click rate of each multi-time recommended creative material in the multi-time recommended creative material set.
In some embodiments, the execution subject can determine click through rates for individual ones of the plurality of recommended creative materials in the plurality of recommended creative material sets. The creative material sets to be recommended for multiple times are material sets, wherein the creative material sets to be recommended for multiple times are material sets, and the creative material sets to be recommended for the multiple times are material sets, from which the target creative material sets to be recommended are removed.
As an example, the execution subject can query the database for click-through rates for each of the multiple recommended creative materials in the multiple recommended creative material set.
And 305, sequencing all creative materials to be recommended included in the creative material set to be recommended according to the estimated click rate set corresponding to the target creative material set to be recommended and the click rate set corresponding to the creative material set recommended for multiple times to obtain a creative material sequence to be recommended.
In some embodiments, the execution main body may sort, according to the pre-estimated click rate set corresponding to the target creative material set to be recommended and the click rate sets corresponding to the multiple-time recommended creative material sets, each creative material to be recommended included in the creative material set to be recommended, so as to obtain a creative material sequence to be recommended.
As an example, the execution main body may sort, in order from small click rate to large click rate, each creative material to be recommended included in the creative material set to be recommended according to the pre-estimated click rate set corresponding to the target creative material set to be recommended and the click rate sets corresponding to the creative material sets recommended multiple times, so as to obtain a creative material sequence to be recommended.
In some optional implementations of some embodiments, after step 305, the following step is further included:
the method comprises the steps of responding to the situation that no target creative material to be recommended exists in the creative material set to be recommended, inputting each creative material to be recommended in the creative material set to be recommended to a material click rate generation model, outputting a material click rate, and obtaining a material click rate set.
The material click rate generation model can be a model for generating the material click rate aiming at the target moment based on the historical creative material set of the article corresponding to the creative material to be recommended. For example, the material click rate generation model may be a recurrent neural network model.
And secondly, sequencing all the creative materials to be recommended included in the creative material set to be recommended according to the material click rate set to obtain a creative material sequence to be recommended.
As an example, the execution main body may sort the creative materials to be recommended included in the creative material set to be recommended according to a sequence of the click rates of the materials from small to large, so as to obtain a creative material sequence to be recommended.
Optionally, the steps further comprise:
and sending at least one creative material to be recommended, which has a click rate meeting a preset click rate condition, in the creative material sequence to be recommended to a material recommending end. The material recommending terminal can be a terminal for recommending materials (namely, putting the materials). The preset click through rate condition may be at least one creative material to be recommended in the sequence of creative materials to be recommended with a click through rate at a top predetermined number.
The above embodiments of the present disclosure have the following advantages: according to the material sequence generation method of some embodiments of the disclosure, a more accurate material recommendation sequence aiming at the creative material set to be recommended can be generated by determining the click rate corresponding to the target creative material to be recommended, so that the material recommendation effect is improved. Specifically, the reason why the material recommendation effect is not good is that: whether the newly generated material is a high-quality creative material with high quality and high potential cannot be determined, and the newly generated material is directly exposed sufficiently, so that the recommendation effect of the actual material is possibly poor. Based on this, in the material sequence generation method of some embodiments of the disclosure, first, in response to the fact that the target creative material to be recommended exists in the acquired creative material set to be recommended, a target creative material set to be recommended is determined from the creative material set to be recommended, so as to generate a click rate corresponding to each target creative material to be recommended subsequently. Then, for each target creative material to be recommended in the target creative material set to be recommended, screening a preset number of target historical recommendation creative materials from the acquired historical recommendation creative material sets to obtain a target historical recommendation creative material set. Here, since the target creative material to be recommended does not have a real click rate, it is not possible to determine whether the target creative material to be recommended is a high-quality creative material with high quality and high potential. Therefore, the preset number of target historical recommendation creative materials with the material association degree with the target recommendation creative material are selected from the historical recommendation creative materials, and the click rate of the target recommendation creative materials can be conveniently estimated subsequently. Then, according to the click rate of each target historical recommendation creative material in the obtained target historical recommendation creative material set group, the estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set can be accurately estimated. And then, determining the click rate of each multi-time recommended creative material in the multi-time recommended creative material set so as to be used for sequencing the materials of the subsequent creative material set to be recommended. And finally, sequencing all creative materials to be recommended included in the creative material set to be recommended according to the estimated click rate set corresponding to the creative material set to be recommended and the click rate set corresponding to the creative material set recommended for multiple times to obtain a creative material sequence to be recommended. Here, the material quality of the creative material is measured by using the estimated click rate of the target recommended creative material as a standard. And measuring the material quality of the repeatedly recommended creative materials through the click rate of the repeatedly recommended creative materials. Therefore, the recommendation creative material sets are sorted based on the click rates corresponding to the creative materials to be recommended in the recommendation creative material sets, and the effect of sorting the materials according to the material quality corresponding to the materials can be achieved. The material recommendation effect can be greatly improved by recommending each material based on the creative material sequence to be recommended.
With further reference to fig. 4, a flow 400 of further embodiments of a method of generating a sequence of material according to the present disclosure is shown. The material sequence generation method comprises the following steps:
step 401, in response to the fact that the target creative material to be recommended exists in the acquired creative material set to be recommended, determining a target creative material set to be recommended.
Step 402, for each target creative material to be recommended in the target creative material set to be recommended, screening a predetermined number of target historical recommendation creative materials from the acquired historical recommendation creative material sets to obtain a target historical recommendation creative material set.
In some embodiments, specific implementations of steps 401 to 402 and technical effects thereof may refer to steps 301 to 302 in the embodiment corresponding to fig. 3, and are not described herein again.
Step 403, for each of the creative materials to be recommended in the set of creative materials to be recommended, calculating the click rate of each of the creative materials to be recommended in the set of creative materials to be recommended corresponding to the target creative material to be recommended, and obtaining a calculated value as the estimated click rate corresponding to the creative material to be recommended.
In some embodiments, an execution agent (e.g., the electronic device 101 shown in fig. 1) may perform, for each of the set of target creative materials to be recommended, a calculation process on a click rate of each of the set of target historical recommendation creative materials corresponding to the set of target creative materials to be recommended, and obtain a calculated value as a pre-estimated click rate corresponding to the set of target creative materials to be recommended.
For example, for a target to-be-recommended creative material, a corresponding target historical recommended creative material set includes: the first target history recommended creative material, the second target history recommended creative material and the third target history recommended creative material. The click rate corresponding to the first target history recommended creative material is 2%. The click rate corresponding to the second target history recommended creative material is 4%. The click rate corresponding to the third target history recommended creative material is 6%. The click rate corresponding to the first target history recommended creative material, the click rate corresponding to the second target history recommended creative material and the click rate corresponding to the third target history recommended creative material are averaged to obtain an average click rate. The average click rate was 4%. The click rate corresponding to the creative material to be recommended by the target is 4%.
Step 404, determining the click rate of each of the multiple recommended creative materials in the multiple recommended creative material set.
And 405, sequencing all the creative materials to be recommended included in the creative material set to be recommended according to the estimated click rate set corresponding to the target creative material set to be recommended and the click rate set corresponding to the creative material set recommended for multiple times to obtain a creative material sequence to be recommended.
In some embodiments, specific implementations of steps 404 to 405 and technical effects thereof may refer to steps 304 to 305 in the embodiment corresponding to fig. 3, and are not described herein again.
As can be seen from fig. 4, compared with the description of some embodiments corresponding to fig. 3, in the process 400 of the material sequence generation method in some embodiments corresponding to fig. 4, the click through rate of each target historical recommended creative material in the target historical recommended creative material set corresponding to each target creative material to be recommended is calculated, so that the click through rate of each target creative material to be recommended can be represented more accurately. More accurate creative material sequences to be recommended can be generated conveniently in a follow-up mode.
With further reference to fig. 5, as an implementation of the methods illustrated in the above figures, the present disclosure provides some embodiments of a material sequence generating apparatus, which correspond to those method embodiments illustrated in fig. 3, and which may be particularly applicable in various electronic devices.
As shown in fig. 5, a material sequence generating apparatus 500 includes: a first determination unit 501, a screening unit 502, a generation unit 503, a second determination unit 504, and a sorting unit 505. The first determining unit 501 is configured to determine a target creative material set to be recommended in response to the presence of a target creative material to be recommended in the acquired creative material set to be recommended, wherein the target creative material to be recommended is a creative material meeting a target material determination condition; a screening unit 502 configured to screen a predetermined number of target history recommended creative materials from the acquired history recommended creative material sets for each target to-be-recommended creative material in the target to-be-recommended creative material sets to obtain target history recommended creative material sets, wherein a material association degree between the target history recommended creative material and the target to-be-recommended creative material is greater than a predetermined threshold value; a generating unit 503 configured to generate an estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each target history recommended creative material in the obtained target history recommended creative material set group; a second determining unit 504 configured to determine a click rate of each of a plurality of times of recommended creative materials in a plurality of times of recommended creative material sets, wherein the plurality of times of recommended creative material sets are material sets from which the target creative material set to be recommended is removed from the creative material set to be recommended; and the sorting unit 505 is configured to sort each creative material to be recommended included in the creative material set to be recommended according to the pre-estimated click rate set corresponding to the target creative material set to be recommended and the click rate set corresponding to the multiple-time recommended creative material sets, so as to obtain a creative material sequence to be recommended.
In some optional implementations of some embodiments, the preset material condition includes: a plurality of preset material sub-conditions, the plurality of preset material sub-conditions comprising: the time interval between the material generation time corresponding to the creative material to be recommended and the current time is smaller than the preset time, and the exposure corresponding to the creative material to be recommended is smaller than the preset exposure value; and the first determining unit 501 in the apparatus 500 described above may be further configured to: determining material generation time and exposure corresponding to each creative material to be recommended in the creative material set to be recommended; and determining the creative materials to be recommended, which meet the plurality of preset material sub-conditions in the creative material to be recommended in a centralized manner, as target creative materials to be recommended.
In some optional implementations of some embodiments, the screening unit 502 in the apparatus 500 described above may be further configured to: inputting each historical recommended creative material in the historical recommended creative material set to a pre-trained material coding model to generate a historical material coding vector to obtain a historical material coding vector set; inputting the target creative material to be recommended to the material coding model to generate a coding vector of the material to be recommended; and according to the historical material coding vector set and the material coding vector to be recommended, screening out historical recommendation creative materials with vector distances meeting a preset distance condition from the historical recommendation creative material set, and taking the historical recommendation creative materials as target historical recommendation creative materials to obtain a target historical recommendation creative material set, wherein the vector distance is the vector distance between the historical material coding vector in the historical material coding vector set and the material coding vector to be recommended.
In some optional implementations of some embodiments, the generating unit 503 in the apparatus 500 may be further configured to: and calculating the click rate of each target historical recommendation creative material in the target historical recommendation creative material set corresponding to the target creative material to be recommended to obtain a calculated value serving as the estimated click rate corresponding to the target creative material to be recommended.
In some optional implementations of some embodiments, the material coding model is trained by: obtaining a training sample set, wherein the training samples comprise: training creative materials and article product word labels; training an initial article product word classification model by using the training sample set to obtain a trained article product word classification model, wherein the trained article product word classification model comprises: and (5) material coding model.
In some optional implementations of some embodiments, the apparatus 500 further includes: an input unit and a material sorting unit (not shown). Wherein the input unit may be configured to: and responding to the situation that no target creative material to be recommended exists in the creative material set to be recommended, inputting each creative material to be recommended in the creative material set to be recommended to a material click rate generation model, outputting a material click rate, and obtaining a material click rate set. The material ordering unit may be configured to: and sequencing all creative materials to be recommended included in the creative material set to be recommended according to the material click rate set to obtain a creative material sequence to be recommended.
In some optional implementations of some embodiments, the apparatus 500 further includes: a transmitting unit (not shown). Wherein the transmitting unit may be configured to: and sending at least one creative material to be recommended, which has a click rate meeting a preset click rate condition, in the creative material sequence to be recommended to a material recommending end.
It will be understood that the elements described in the apparatus 500 correspond to various steps in the method described with reference to fig. 3. Thus, the operations, features and resulting advantages described above with respect to the method are also applicable to the apparatus 500 and the units included therein, and are not described herein again.
Referring now to FIG. 6, a block diagram of an electronic device (e.g., electronic device 101 of FIG. 1) 600 suitable for use in implementing some embodiments of the present disclosure is shown. The electronic device shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 6, the electronic device 600 may include a processing means (e.g., central processing unit, graphics processor, etc.) 601 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 602 or a program loaded from a storage means 608 into a Random Access Memory (RAM) 603. In the RAM603, various programs and data necessary for the operation of the electronic apparatus 600 are also stored. The processing device 601, the ROM 602, and the RAM603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
Generally, the following devices may be connected to the I/O interface 605: input devices 606 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 607 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 608 including, for example, tape, hard disk, etc.; and a communication device 609. The communication means 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. While fig. 6 illustrates an electronic device 600 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may be alternatively implemented or provided. Each block shown in fig. 6 may represent one device or may represent multiple devices as desired.
In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In some such embodiments, the computer program may be downloaded and installed from a network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. The computer program, when executed by the processing device 601, performs the above-described functions defined in the methods of some embodiments of the present disclosure.
It should be noted that the computer readable medium described above in some embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable 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 some embodiments of the disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (Hyper Text Transfer Protocol), and may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: in response to the fact that the target creative material to be recommended exists in the acquired creative material set to be recommended, determining a target creative material set to be recommended, wherein the target creative material to be recommended is a creative material meeting the target material determination condition; screening a preset number of target historical recommendation creative materials from the obtained historical recommendation creative material sets to obtain a target historical recommendation creative material set for each target creative material to be recommended in the target creative material set to be recommended, wherein the material association degree between the target historical recommendation creative material and the target creative material to be recommended is larger than a preset threshold value; generating an estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each target history recommended creative material in the obtained target history recommended creative material set group; determining the click rate of each multi-time recommendation creative material in a multi-time recommendation creative material set, wherein the multi-time recommendation creative material set is a material set of the to-be-recommended creative material set except the target to-be-recommended creative material set; and sequencing all creative materials to be recommended included in the creative material set to be recommended according to the estimated click rate set corresponding to the creative material set to be recommended and the click rate set corresponding to the creative material set recommended for multiple times to obtain a creative material sequence to be recommended.
Computer program code for carrying out operations for embodiments of the present disclosure 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 type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in some embodiments of the present disclosure may be implemented by software or hardware. The described units may also be provided in a processor, and may be described as: a processor includes a first determining unit, a screening unit, a generating unit, a second determining unit, and a sorting unit. Where the names of these units do not constitute a limitation on the unit itself under certain circumstances, for example, the first determining unit may also be described as a "unit that determines a set of target creative materials to be recommended in response to the presence of target creative materials to be recommended in the set of acquired creative materials to be recommended".
The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems on a chip (SOCs), complex Programmable Logic Devices (CPLDs), and the like.
Some embodiments of the present disclosure also provide a computer program product comprising a computer program that, when executed by a processor, implements any of the material sequence generation methods described above.
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure are mutually replaced to form the technical solution.

Claims (11)

1. A method of generating a sequence of material, comprising:
in response to the fact that the target creative material to be recommended exists in the acquired creative material set to be recommended, determining a target creative material set to be recommended, wherein the target creative material to be recommended is a creative material meeting the target material determination condition;
screening a preset number of target historical recommendation creative materials from the acquired historical recommendation creative material sets for each target to-be-recommended creative material in the target to-be-recommended creative material set to obtain a target historical recommendation creative material set, wherein the material association degree between the target historical recommendation creative material and the target to-be-recommended creative material is greater than a preset threshold value;
generating an estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each target history recommended creative material in the obtained target history recommended creative material set group;
determining the click rate of each multi-time recommendation creative material in a multi-time recommendation creative material set, wherein the multi-time recommendation creative material set is a material set of the to-be-recommended creative material set except the target to-be-recommended creative material set;
and sequencing all creative materials to be recommended included in the creative material set to be recommended according to the estimated click rate set corresponding to the creative material set to be recommended and the click rate set corresponding to the creative material set recommended for multiple times to obtain a creative material sequence to be recommended.
2. The method of claim 1, wherein the preset material conditions comprise: a plurality of preset material sub-conditions, the plurality of preset material sub-conditions comprising: the time interval between the material generation time corresponding to the creative material to be recommended and the current time is less than the preset time, and the exposure corresponding to the creative material to be recommended is less than the preset exposure value; and
the method for determining the target creative material set to be recommended comprises the following steps:
determining material generation time and exposure corresponding to each creative material to be recommended in the creative material set to be recommended;
and determining the creative materials to be recommended, which meet the plurality of preset material sub-conditions in the creative material to be recommended in a centralized manner, as target creative materials to be recommended.
3. The method of claim 1, wherein the filtering of a predetermined number of target historical recommended creative materials from the acquired set of historical recommended creative materials to obtain a set of target historical recommended creative materials comprises:
inputting each historical recommended creative material in the historical recommended creative material set to a pre-trained material coding model to generate a historical material coding vector to obtain a historical material coding vector set;
inputting the target creative material to be recommended to the material coding model to generate a coding vector of the material to be recommended;
and according to the historical material coding vector set and the material coding vector to be recommended, screening historical recommended creative materials with vector distances meeting a preset distance condition from the historical recommended creative material set, using the historical recommended creative materials as target historical recommended creative materials, and obtaining a target historical recommended creative material set, wherein the vector distance is the vector distance between the historical material coding vector in the historical material coding vector set and the material coding vector to be recommended.
4. The method of claim 1, wherein generating a predictive click rate for each of the set of target to-be-recommended creative materials based on the click rate of each of the set of derived target historical recommended creative materials comprises:
and calculating the click rate of each target historical recommendation creative material in the target historical recommendation creative material set corresponding to the target creative material to be recommended to obtain a calculated numerical value serving as the estimated click rate corresponding to the target creative material to be recommended.
5. A method according to claim 3, wherein the material coding model is trained by:
obtaining a training sample set, wherein the training samples comprise: training creative materials and article product word labels;
training an initial article product word classification model by using the training sample set to obtain a trained article product word classification model, wherein the trained article product word classification model comprises: and (5) material coding model.
6. The method of claim 1, wherein the method further comprises:
responding to the situation that no target creative material to be recommended exists in the creative material set to be recommended, inputting each creative material to be recommended in the creative material set to be recommended to a material click rate generation model, outputting a material click rate, and obtaining a material click rate set;
and sequencing all the creative materials to be recommended included in the creative material set to be recommended according to the material click rate set to obtain a creative material sequence to be recommended.
7. The method of claim 1 or 6, wherein the method further comprises:
and sending at least one creative material to be recommended, which is in the creative material sequence to be recommended and has a click rate meeting a preset click rate condition, to a material recommending end.
8. A material sequence generating apparatus comprising:
the device comprises a first determining unit, a second determining unit and a judging unit, wherein the first determining unit is configured to respond to the fact that a target creative material to be recommended exists in the acquired creative material set to be recommended, and the target creative material to be recommended is a creative material meeting target material determining conditions;
the screening unit is configured to screen a preset number of target historical recommendation creative materials from the acquired historical recommendation creative material sets for each target to-be-recommended creative material in the target to-be-recommended creative material sets to obtain target historical recommendation creative material sets, wherein the material association degree between the target historical recommendation creative materials and the target to-be-recommended creative materials is larger than a preset threshold value;
the generating unit is configured to generate the estimated click rate of each target to-be-recommended creative material in the target to-be-recommended creative material set according to the click rate of each target history recommended creative material in the obtained target history recommended creative material set group;
a second determination unit configured to determine a click rate for each of a plurality of times of recommended creative material sets, wherein the plurality of times of recommended creative material sets are material sets from which the target creative material set to be recommended is removed from the creative material set to be recommended;
and the sequencing unit is configured to sequence all the creative materials to be recommended included in the creative material set to be recommended according to the estimated click rate set corresponding to the target creative material set to be recommended and the click rate sets corresponding to the multiple recommended creative material sets to obtain a creative material sequence to be recommended.
9. An electronic device, comprising:
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 recited in any of claims 1-7.
10. A computer-readable medium, on which a computer program is stored, wherein the program, when executed by a processor, implements the method of any one of claims 1-7.
11. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-7.
CN202211248320.XA 2022-10-12 2022-10-12 Material sequence generation method, apparatus, device, medium, and program product Pending CN115795176A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211248320.XA CN115795176A (en) 2022-10-12 2022-10-12 Material sequence generation method, apparatus, device, medium, and program product

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211248320.XA CN115795176A (en) 2022-10-12 2022-10-12 Material sequence generation method, apparatus, device, medium, and program product

Publications (1)

Publication Number Publication Date
CN115795176A true CN115795176A (en) 2023-03-14

Family

ID=85432840

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211248320.XA Pending CN115795176A (en) 2022-10-12 2022-10-12 Material sequence generation method, apparatus, device, medium, and program product

Country Status (1)

Country Link
CN (1) CN115795176A (en)

Similar Documents

Publication Publication Date Title
CN110688528B (en) Method, apparatus, electronic device, and medium for generating classification information of video
CN110929799B (en) Method, electronic device, and computer-readable medium for detecting abnormal user
CN114049072B (en) Index determination method and device, electronic equipment and computer readable medium
CN113836406A (en) Information flow recommendation method and device
CN112836128A (en) Information recommendation method, device, equipment and storage medium
CN114119123A (en) Information pushing method and device
CN115062119B (en) Government affair event handling recommendation method and device
CN115757933A (en) Recommendation information generation method, device, equipment, medium and program product
CN115983902A (en) Information pushing method and system based on user real-time event
WO2022218068A1 (en) Material delivery method and apparatus, and device and medium
CN115795176A (en) Material sequence generation method, apparatus, device, medium, and program product
CN113592607A (en) Product recommendation method and device, storage medium and electronic equipment
CN114970444A (en) Information processing method, device, terminal and storage medium
CN113762581A (en) Method and device for generating activity population prediction model, electronic equipment and medium
CN115169321B (en) Logistics content text checking method and device, electronic equipment and computer medium
CN116800834B (en) Virtual gift merging method, device, electronic equipment and computer readable medium
CN112860999B (en) Information recommendation method, device, equipment and storage medium
CN116596585B (en) User satisfaction obtaining method and device, electronic equipment and storage medium
CN116881097B (en) User terminal alarm method, device, electronic equipment and computer readable medium
CN116843800B (en) Animation information transmission method, device, electronic equipment and computer readable medium
CN115223113B (en) Training sample set cleaning method and device
CN113077352B (en) Insurance service article recommending method based on user information and insurance related information
CN116823407B (en) Product information pushing method, device, electronic equipment and computer readable medium
CN117235535B (en) Abnormal supply end power-off method and device, electronic equipment and medium
CN115169852A (en) Information sending method, information sending device, electronic equipment, medium and computer program product

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