CN113763005A - Picture advertisement pushing method, electronic equipment and computer readable storage medium - Google Patents

Picture advertisement pushing method, electronic equipment and computer readable storage medium Download PDF

Info

Publication number
CN113763005A
CN113763005A CN202011012322.XA CN202011012322A CN113763005A CN 113763005 A CN113763005 A CN 113763005A CN 202011012322 A CN202011012322 A CN 202011012322A CN 113763005 A CN113763005 A CN 113763005A
Authority
CN
China
Prior art keywords
keyword
keywords
picture advertisement
objects
keyword set
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202011012322.XA
Other languages
Chinese (zh)
Other versions
CN113763005B (en
Inventor
梁舒飞
王喆
陈召新
赫南
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 CN202011012322.XA priority Critical patent/CN113763005B/en
Publication of CN113763005A publication Critical patent/CN113763005A/en
Application granted granted Critical
Publication of CN113763005B publication Critical patent/CN113763005B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0253—During e-commerce, i.e. online transactions
    • 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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • 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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0269—Targeted advertisements based on user profile or attribute
    • G06Q30/0271—Personalized advertisement
    • 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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0277—Online advertisement

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present disclosure provides a picture advertisement push method, including: acquiring a search keyword of a user; matching the search keywords of the user with a keyword set of the picture advertisement; under the condition that the keyword set of the picture advertisement is determined to have the keyword matched with the search keyword of the user, pushing a first picture advertisement corresponding to the matched keyword to the user; the keyword set of the picture advertisement is set in the following way: determining a first object set related to the picture advertisement according to the information of the picture advertisement; expanding the first set of objects using a graph embedding model to obtain a second set of objects; determining a first set of keywords related to the second set of objects; determining a second keyword set related to the picture advertisement according to the first keyword set; and expanding the second keyword set to obtain the keyword set of the picture advertisement.

Description

Picture advertisement pushing method, electronic equipment and computer readable storage medium
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a method for pushing a picture advertisement, an electronic device, and a computer-readable storage medium.
Background
At present, the electronic commerce platforms are spreading fiercely in various technical fields. The application of computing advertisements enables a platform and an advertiser to depict user requirements and accurately locate effective audience groups, so that the platform and the advertiser also become one of key research directions of an e-commerce platform. How to process and apply massive user and commodity data is a key for designing and realizing e-commerce advertisement products.
The main actions of the user on the E-commerce website include browsing, purchasing, shopping cart adding, paying attention, searching and the like. The search content of the user can accurately reflect the recent interest of the user and cover a more reasonable advertisement material range. Therefore, extracting keywords from search content as user tags becomes one of the effective methods for the user interest mining process, and advertisement products related to keyword targeting are produced accordingly. However, the existing scheme of the keyword-oriented advertisement product still cannot meet the user requirements in the aspects of intellectualization, accuracy, stability, compatibility and the like, and the user experience is influenced.
Disclosure of Invention
In view of this, the present disclosure provides a picture advertisement pushing method, which mines search keywords related to a picture advertisement from multiple dimensions according to complete information of the picture advertisement, and also expands search keywords from similar product expansion and similar/identical search keywords, so as to expand a targeting range, improve accuracy, stability and compatibility, and improve user experience on the premise of ensuring relevance.
One aspect of the present disclosure provides a picture advertisement pushing method, including: acquiring a search keyword of a user; matching the search keywords of the user with a keyword set of the picture advertisement; under the condition that the keyword set of the picture advertisement is determined to have the keyword matched with the search keyword of the user, pushing a first picture advertisement corresponding to the matched keyword to the user; the keyword set of the picture advertisement is set in the following way: determining a first object set related to the picture advertisement according to the information of the picture advertisement; expanding the first set of objects using a graph embedding model to obtain a second set of objects; determining a first set of keywords related to the second set of objects; determining a second keyword set related to the picture advertisement according to the first keyword set; and expanding the second keyword set to obtain the keyword set of the picture advertisement.
According to an embodiment of the present disclosure, the expanding the first set of objects to obtain the second set of objects by using the graph embedding model includes: constructing an object graph according to the user behavior sequence, wherein the object graph comprises a plurality of objects; processing the object graph by using the graph embedding model to obtain an object vector; determining a similar object for each object of the plurality of objects from the object vector; and expanding the first object set by using the similar objects to obtain a second object set.
According to an embodiment of the present disclosure, the constructing the object graph according to the user behavior sequence includes: constructing a user behavior sequence according to at least one of clicking, adding and submitting confirmation behaviors of a user on an object; determining a set of objects that interact with behaviors in the sequence of user behaviors; and constructing the object graph by using the object sets interacted with the behaviors in the user behavior sequence.
According to an embodiment of the present disclosure, the expanding the second keyword set to obtain the keyword set of the picture advertisement includes: sequencing the historical search keywords of the user according to the sequence of character strings to obtain a first similar keyword set: performing word segmentation on historical search keywords of the user, and obtaining a second similar keyword set by using a word vector model based on the keywords after word segmentation; constructing a keyword graph by using keywords having searching and clicking relations with the second object set, and obtaining a third similar keyword set by using the graph embedding model based on the keyword graph; and expanding the second keyword set according to the first similar keyword set, the second similar keyword set and the third similar keyword set to obtain the keyword set of the picture advertisement.
According to an embodiment of the present disclosure, the expanding the second keyword set according to the first similar keyword set, the second similar keyword set, and the third similar keyword set to obtain the keyword set of the picture advertisement includes: determining similar keywords of which the edit distances between the keywords in the first similar keyword set, the second similar keyword set and the third similar keyword set and the keywords in the second keyword set are smaller than or equal to a threshold value; and expanding the second keyword set by using similar keywords of which the editing distance with the keywords in the second keyword set is less than or equal to a threshold value to obtain the keyword set of the picture advertisement.
According to an embodiment of the present disclosure, the determining the first set of keywords related to the second set of objects includes: acquiring keywords having searching and clicking relations with the second object set; and determining a first keyword set related to the second object set according to the keywords having searching and clicking relations with the second object set.
According to an embodiment of the present disclosure, the determining a second keyword set related to a picture advertisement according to the first keyword set includes: determining the relevancy of each keyword in the first keyword set and the picture advertisement; and determining a second keyword set related to the picture advertisement according to the relevance of each keyword in the first keyword set and the picture advertisement.
According to an embodiment of the present disclosure, the determining, according to the picture advertisement information, a first object set related to the picture advertisement includes: determining shop ID information, advertisement activity URL information, order following object ID information and picture URL information according to the picture advertisement information; according to the shop ID information and the advertisement activity URL information, objects under shops and under activity landing pages are obtained, a documentary object is obtained according to the documentary object ID information, and the objects under shops and under activity landing pages and the documentary object form a candidate object set; acquiring category words and brand words according to the picture URL information; and filtering the objects in the candidate object set according to the category words and the brand words so as to obtain the objects under the category and the brand, thereby determining a first object set related to the picture advertisement.
Another aspect of the present disclosure provides an electronic device including: one or more processors; a storage device for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the picture advertisement push method.
Another aspect of the present disclosure provides a computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the picture advertisement push method.
According to the embodiment of the disclosure, the technical problems of insufficient information utilization, keyword information loss, limited orientation range and poor stability and compatibility in the related technology can be at least partially solved. According to the method and the device, the search keywords related to the picture advertisements are mined from multiple dimensions through the complete information of the picture advertisements, and the search keywords are expanded from similar commodity expansion and similar/same search keywords, so that the orientation range is expanded, the accuracy, the stability and the compatibility are improved, and the user experience is improved on the premise of ensuring the relevance.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to limit the disclosure. In the drawings:
fig. 1 schematically illustrates an application scenario in which a picture advertisement push method may be applied according to an embodiment of the present disclosure.
Fig. 2 schematically shows a flowchart of a picture advertisement push method according to an embodiment of the present disclosure.
Fig. 3 schematically illustrates the arrangement of keyword sets of a picture advertisement according to an embodiment of the present disclosure.
FIG. 4 schematically illustrates a flow chart for determining a first set of objects related to a photo advertisement based on information of the photo advertisement according to an embodiment of the present disclosure.
FIG. 5 schematically illustrates a flow diagram for expanding the first set of objects to obtain a second set of objects using a graph embedding model, according to an embodiment of the disclosure.
Fig. 6 schematically shows a flow chart for determining a first set of keywords related to the second set of objects according to an embodiment of the present disclosure.
Fig. 7 schematically shows a flow chart for determining a second keyword set related to a picture advertisement according to the first keyword set according to an embodiment of the present disclosure.
Fig. 8 schematically shows a flowchart of expanding the second keyword set to obtain the keyword set of the photo advertisement according to an embodiment of the present disclosure.
Fig. 9 schematically shows a flowchart of expanding the second keyword set according to the first similar keyword set, the second similar keyword set and the third similar keyword set to obtain the keyword set of the picture advertisement according to the embodiment of the present disclosure.
Fig. 10 schematically shows a block diagram of an electronic device adapted to implement the above described method according to an embodiment of the present disclosure.
Detailed Description
Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is illustrative only and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present disclosure.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.
All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.
Where a convention analogous to "at least one of A, B and C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B and C" would include but not be limited to systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.). Where a convention analogous to "A, B or at least one of C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B or C" would include but not be limited to systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).
In the process of implementing the present disclosure, the inventor finds that the picture advertisement push modes are mainly divided into the following two types:
(1) advertiser specified targeting keywords
The method comprises the steps that an advertiser selects a plurality of entries for advertisement materials delivered by the advertiser to serve as keyword labels of the advertisement materials, inverted indexes from the keyword labels to the advertisement materials are generated, after a user request arrives at an online advertisement system, the advertisement system obtains search content of the user in a period of time, and the search content of the user in the period of time is divided into a plurality of keywords by using a word cutting dictionary to serve as the keyword labels of the user. Thereafter, the advertisement system retrieves the inverted index using the keyword tags of the user, recalling the advertisement material matching the user.
(2) System mining advertisement material, search content respectively relevant product word and brand word
And the advertisement platform excavates related product words, brand words and the combination of the related product words and the brand words according to the advertisement material information to serve as keyword labels of the advertisement material, and generates inverted indexes from the keyword labels to the advertisement material. After a user request reaches an online advertising system, the advertising system acquires search contents of the user in a period of time, and mines related product words, brand words and a combination of the related product words and the brand words as keyword tags of the user. Thereafter, the advertisement system retrieves the inverted index using the user's keyword tag and recalls advertisement material matching the user.
The above two picture advertisement pushing manners still have the following challenges:
(1) insufficient availability of information and loss of valuable information
Due to the fact that the information utilization degree of the picture advertisement material is insufficient, certain keyword information loss is caused, and therefore the advertiser cannot target more high-quality flow, and the method gradually becomes the bottleneck of further growth of keyword targeting services.
(2) Lack of correlation spread and limited orientation range
Because only advertisement materials and user search information are used, relevance expansion is not carried out, the number of the extracted keyword labels is insufficient, and the directional range and the triggering capability are very limited.
(3) Single digging method, poor stability and compatibility
The user search content, the commodity and the advertisement material have strong diversity, the interest of the user on the commodity and the advertisement also can obviously change along with the time migration, and a single mining mode is adopted, so that the method has a good effect on some users and the advertisement material only in some time periods.
In view of this, embodiments of the present disclosure provide a picture advertisement pushing method, an electronic device, and a storage medium. The method comprises the following steps: acquiring a search keyword of a user; matching the search keywords of the user with a keyword set of the picture advertisement; and under the condition that the keywords matched with the search keywords of the user exist in the keyword set of the picture advertisement, pushing a first picture advertisement corresponding to the matched keywords to the user. The keyword set of the picture advertisement is set in the following way: determining a first object set related to the picture advertisement according to the information of the picture advertisement; expanding the first set of objects using a graph embedding model to obtain a second set of objects; determining a first set of keywords related to the second set of objects; determining a second keyword set related to the picture advertisement according to the first keyword set; and expanding the second keyword set to obtain the keyword set of the picture advertisement.
Fig. 1 schematically illustrates an application scenario in which a picture advertisement push method may be applied according to an embodiment of the present disclosure. It should be noted that fig. 1 is only an example of a system architecture to which the embodiments of the present disclosure may be applied to help those skilled in the art understand the technical content of the present disclosure, and does not mean that the embodiments of the present disclosure may not be applied to other devices, systems, environments or scenarios.
As shown in fig. 1, the application scenario includes a client 100, and various communication client applications, such as a shopping application, a web browser application, a search application, an instant messaging tool, a mailbox client, and/or social platform software, etc. (for example only), may be installed on the client 100.
According to an embodiment of the present disclosure, the client 100 may be, for example, various electronic devices having a display screen and supporting APP, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
According to the embodiment of the disclosure, the terminal device 100 can intelligently push the matched picture advertisement to the user in response to the operation of the user to access the shopping application, for example.
According to an embodiment of the present disclosure, a picture advertisement push method performed by a client includes: acquiring a search keyword of a user; matching the search keywords of the user with a keyword set of the picture advertisement; and under the condition that the keywords matched with the search keywords of the user exist in the keyword set of the picture advertisement, pushing a first picture advertisement corresponding to the matched keywords to the user.
Fig. 2 schematically shows a flowchart of a picture advertisement push method according to an embodiment of the present disclosure.
As shown in fig. 2, the method includes operations S201 to S203.
In operation S201, a search keyword of a user is acquired.
According to the embodiment of the disclosure, the search keyword of the user may be a historical search keyword of the user or a current search keyword of the user. The historical search keywords of the user may be search keywords of the user over one or more historical time periods. For example, the search keyword of the user may be determined by a search record of the user, and accordingly, the historical search keyword of the user may be determined by a historical search record of the user, and the current search keyword of the user may be determined by a current search record of the user.
In operation S202, the search keyword of the user is matched with a keyword set of a picture advertisement.
According to an embodiment of the present disclosure, the keyword set of the picture advertisement is a keyword set corresponding to the picture advertisement, for example, a keyword set related to the picture advertisement, or a similar keyword set obtained by expanding the related keywords. The keyword set of the picture advertisement in operation S202 may be implemented by the following procedure described in fig. 3, and will not be described in detail here.
In operation S203, in a case where it is determined that a keyword matching the search keyword of the user exists in the keyword set of the picture advertisement, a first picture advertisement corresponding to the matching keyword is pushed to the user.
According to the embodiment of the disclosure, the search keyword of the user in the last 7 days can be taken, the search keyword is strictly matched with the related search keywords after all the picture advertisements are expanded, and if the matching is successful, the picture advertisement is regarded as the picture advertisement and is targeted to the user. Of course, 7 days is merely illustrative and may be adjusted as desired.
Illustratively, the search keyword of the user is a large-screen mobile phone, the keyword set of the picture advertisement includes a large-screen mobile phone, a curved-screen television, a mobile phone support, and the like, and the keyword set of the picture advertisement has picture advertisements corresponding to the large-screen mobile phone, the curved-screen television, and the mobile phone support, respectively. The search keyword "large-screen cell phone" of the user is matched with the "large-screen cell phone" in which the keywords of the picture advertisement are concentrated, and in this case, the picture advertisement corresponding to the large-screen cell phone is pushed to the user.
According to the method and the device, the intelligent orientation of the picture advertisement is realized according to the behavior of the search keywords of the user, which can intuitively express the interest and the intention of the user, and the keyword set of the picture advertisement obtained by mining and expanding.
Fig. 3 schematically illustrates the arrangement of keyword sets of a picture advertisement according to an embodiment of the present disclosure.
As shown in fig. 3, the setting of the keyword set of the picture advertisement of this embodiment may include operations S301 to S305, for example.
In operation S301, a first set of objects related to a picture advertisement is determined according to information of the picture advertisement.
According to the embodiment of the present disclosure, the first object set is a set of objects related to the picture advertisement determined according to the picture advertisement information, that is, a set of one or more objects related to the picture advertisement. Illustratively, the object is a commodity. This operation S301 may be implemented by a flow described in the following fig. 4, and will not be described in detail here.
In operation S302, the first set of objects is expanded using a graph embedding model to obtain a second set of objects.
According to the embodiment of the present disclosure, the second set of objects is a set of objects obtained by expanding the first set of objects using a graph embedding model, that is, the second set of objects includes both the objects in the first set of objects and the objects expanded on the basis of the first set of objects. In other words the first set of objects is a subset of the second set of objects. This operation S302 may be implemented by a flow described later in fig. 5, and will not be described in detail here.
In operation S303, a first set of keywords related to the second set of objects is determined.
According to an embodiment of the present disclosure, the first keyword set is a keyword set related to the second object set, that is, a set formed by keywords related to each object in the second object set, which is referred to as an object-related keyword set for short, and in a case that the object is a commodity, the keyword is related to the commodity. This operation S303 may be implemented by a flow described later in fig. 6, and will not be described in detail here.
In operation S304, a second keyword set related to the picture advertisement is determined according to the first keyword set.
According to the embodiment of the disclosure, the first keyword set is a keyword set related to an object, and the second keyword set is a keyword set related to a picture advertisement. This operation S304 may be implemented by a flow described later in fig. 7, and will not be described in detail here.
In operation S305, the second keyword set is expanded to obtain a keyword set of the picture advertisement.
According to the embodiment of the disclosure, the first keyword set is a keyword set related to an object, the second keyword set is a keyword set related to a picture advertisement, and the keyword set of the picture advertisement is a keyword set obtained after the second keyword set is expanded, that is, the keyword set of the picture advertisement includes both keywords in the second keyword set (keyword set related to the picture advertisement) and keywords in a keyword set (keyword set similar to the picture advertisement) expanded on the basis of the second keyword set. In other words, the second keyword set is a subset of the keyword set of the photo advertisement. This operation S305 may be implemented by a flow described later in fig. 8, and is not described in detail here.
FIG. 4 schematically illustrates a flow chart for determining a first set of objects related to a photo advertisement based on information of the photo advertisement according to an embodiment of the present disclosure.
As shown in fig. 4, determining the first set of objects related to the picture advertisement according to the information of the picture advertisement of this embodiment may include, for example, operations S401 to S404.
In operation S401, store ID information, advertisement campaign URL information, merchandising object ID information, and picture URL information are determined from picture advertisement information.
Illustratively, store ID information, inventory object ID information, campaign URL information, and picture URL information are parsed from advertiser impression information. Furthermore, advertiser ID information and the like can also be analyzed.
In operation S402, objects under a shop and under a campaign landing page are acquired from the shop ID information and the advertisement campaign URL information, and a documentary object is acquired from the documentary object ID information, and the objects under the shop and under the campaign landing page and the documentary object constitute a candidate object set.
In operation S403, category words and brand words are obtained according to the picture URL information.
According to the embodiment of the disclosure, the picture information is obtained according to the picture URL information, and category words and brand words in characters contained in the picture are extracted through ocr technology and named entity recognition technology.
In operation S404, filtering the objects in the candidate object set according to the category words and the brand words, so as to obtain the objects under the category and the brand, thereby determining a first object set related to the picture advertisement.
According to the embodiment of the disclosure, the category word and the brand word extracted in operation S403 are used to filter the object set (i.e., the candidate object set) generated in operation S402, and only the objects under the category and the brand in the candidate object set are reserved. Illustratively, the filtered set of objects may be saved on the HDFS for subsequent reading.
According to the embodiment of the disclosure, the object information related to the picture advertisement is mined in various ways, including but not limited to object information such as objects under advertisers stores, order following objects and objects under advertisement activities, and character information in the picture, category words and brand words in the picture are extracted, objects irrelevant to the category and the brand are filtered out according to the category words and the brand words, and the relevance of the picture advertisement and the objects is guaranteed.
FIG. 5 schematically illustrates a flow diagram for expanding the first set of objects to obtain a second set of objects using a graph embedding model, according to an embodiment of the disclosure.
As shown in fig. 5, the expanding the first set of objects to obtain the second set of objects by using the graph embedding model of this embodiment may include, for example, operations S501 to S504.
In operation S501, an object graph is constructed according to a user behavior sequence.
According to an embodiment of the present disclosure, the object map includes a plurality of objects. Illustratively, constructing the object graph according to the user behavior sequence includes: constructing a user behavior sequence according to at least one of clicking, adding and submitting confirmation behaviors of a user on an object; determining a set of objects that interact with behaviors in the sequence of user behaviors; and constructing the object graph by using the object sets interacted with the behaviors in the user behavior sequence.
According to an embodiment of the present disclosure, in a case where the object is a commodity, the addition is, for example, a shopping cart, and the confirmation of submission is, for example, a purchase.
In operation S502, the object graph is processed by using the graph embedding model to obtain an object vector.
According to an embodiment of the present disclosure, the graph embedding model is a trained graph embedding model. And obtaining a vector representation (called an object vector for short) of each node in the object graph, namely the object by using the trained graph embedding model. According to an embodiment of the present disclosure, the graph embedding model includes, but is not limited to, Node2Vec, ProNE, and the like.
In operation S503, a similar object for each object of the plurality of objects is determined according to the object vector.
According to the embodiment of the present disclosure, the cosine similarity between each object in the object diagram and other objects in the object diagram is calculated according to the object vector obtained in operation S502, so as to obtain n objects with the most similar similarity to each object. Illustratively, n objects with the largest cosine similarity with each object are obtained as similar objects of the object.
In operation S504, the first set of objects is expanded with the similar objects to obtain a second set of objects.
According to the embodiment of the present disclosure, the first object set obtained in operation S301 is expanded to obtain the second object set according to the similar object obtained in operation S503, and the number of the expanded objects may be set according to a requirement. Illustratively, the second set of objects may be saved on the HDFS for subsequent reading.
Fig. 6 schematically shows a flow chart for determining a first set of keywords related to the second set of objects according to an embodiment of the present disclosure.
As shown in fig. 6, determining the first set of keywords related to the second set of objects in this embodiment may include, for example, operations S601 to S602.
In operation S601, a keyword having a search and click relationship with the second object set is obtained.
According to the embodiment of the disclosure, the keywords having search and click relations with the second object set in one or more historical time periods can be obtained. Illustratively, keywords having a search and click relationship with the second set of objects within the last 3 days are obtained. Of course, the 3 days are merely illustrative and may be adjusted as desired.
In operation S602, a first keyword set related to the second object set is determined according to keywords having a search and click relationship with the second object set.
According to the embodiment of the disclosure, the obtained keywords having a search and click relationship with the second object set may be screened according to a preset rule, so as to determine the first keyword set related to the second object set, where the preset rule is, for example, that the occurrence frequency exceeds a frequency threshold, and the click rate exceeds a click rate threshold. Correspondingly, for example, the keywords with the frequency exceeding a frequency threshold and the click rate exceeding a click rate threshold are screened out from the acquired keywords with the searching and click relation with the second object set, so as to form the first keyword set related to the second object set. Exemplarily, the keywords which have searching and clicking relations with the second object set in the last 3 days are obtained, and are arranged according to the reverse order of occurrence frequency, the first 100 keywords are taken from each object, and then 10 keywords with the highest clicking rate are taken from the first 100 keywords to form the search keywords related to the object, wherein 100 and 10 keywords are only exemplary illustrations and can be adjusted as required.
Fig. 7 schematically shows a flow chart for determining a second keyword set related to a picture advertisement according to the first keyword set according to an embodiment of the present disclosure.
As shown in fig. 7, the example of determining the second keyword set related to the picture advertisement according to the first keyword set in this embodiment may include operations S701 to S702, for example.
In operation S701, a relevance of each keyword in the first keyword set to the picture advertisement is determined.
According to the embodiment of the present disclosure, the first keyword set obtained in operation S602 is ranked and aggregated onto the picture advertisement, and the ranking considers the occurrence frequency of the search keyword, the click rate, and the relevancy score between the object and the picture advertisement.
According to the embodiment of the disclosure, the relevancy of each keyword in the first keyword set to the picture advertisement satisfies the following relational expression:
Figure BDA0002697217280000121
wherein score represents a relevance score of the search keyword and the picture advertisement, fre represents an occurrence frequency of the search keyword, ctr represents a click rate, and m represents a relevance score of the object and the picture advertisement.
In operation S702, a second keyword set related to the picture advertisement is determined according to the relevance of each keyword in the first keyword set to the picture advertisement.
According to the embodiment of the disclosure, a plurality of keywords can be screened out according to the relevancy between each keyword in the first keyword set and the picture advertisement and a preset rule, so as to form a keyword set related to the picture advertisement. The preset rule is, for example, k keywords with the highest relevancy exceeding a relevancy threshold, and the like. Illustratively, 20 keywords with the highest relevancy are taken to form a keyword set related to the picture advertisement, namely a second keyword set after being sorted according to relevancy. Of these, 20 are merely exemplary and may be adjusted as needed.
According to the method and the device, timeliness, hot degree, click rate and correlation with the object of the search keyword are comprehensively considered, and stability and compatibility of the mining effect are guaranteed.
Fig. 8 schematically shows a flowchart of expanding the second keyword set to obtain the keyword set of the photo advertisement according to an embodiment of the present disclosure.
As shown in fig. 8, the expanding the second keyword set to obtain the keyword set of the picture advertisement according to this embodiment may include operations S801 to S804, for example.
In operation S801, the user history search keywords are sorted in the character string order, resulting in a first set of similar keywords.
According to the embodiment of the disclosure, the search keywords input by the user in the last month can be selected and preprocessed, invalid characters such as punctuations, special symbols and web page links are removed, only Chinese, English and numbers are reserved, the processed keywords are sorted according to the sequence of character strings, the keywords with the same sorting result are similar keywords with the sequence normalization, and the similarity value is 1. Wherein the last month is merely an exemplary illustration and can be adjusted as desired.
In operation S802, word segmentation is performed on the user history search keyword, and a second similar keyword set is obtained by using a word vector model based on the word segmented keyword.
According to the embodiment of the disclosure, phrases such as Chinese and English brand words, category words, product words and the like related to all objects can be obtained from an object information table, an object related dictionary is built, then word segmentation is performed on historical search keywords of a user, the object related word table is loaded during word segmentation to obtain a more accurate word segmentation result, the search keyword phrases after word segmentation are input into a word vector model, word vectors corresponding to all the phrases are output, cosine similarity between every two keywords is calculated according to the word vectors, q keywords which are most similar to each other and are generated by the word vector model are obtained, and the similarity is a cosine similarity value. Illustratively, the word segmentation tools include, but are not limited to, common Chinese word segmentation tools such as jieba, LTP, etc., and the word vector models include, but are not limited to, word2vec, fastText, etc.
In operation S803, a keyword graph is constructed using keywords having a search and click relationship with the second object set, and a third similar keyword set is obtained using the graph embedding model based on the keyword graph.
According to the embodiment of the disclosure, each node in the object graph constructed in operation S501 is replaced by a search keyword that is searched for with the object and has the highest frequency of co-occurrence when clicked, so as to form a new graph, that is, a keyword graph, the new graph is input into the graph embedding model trained in operation S502, so as to obtain vector representation, referred to as keyword vector, of each node of the new graph, and cosine similarity between nodes is calculated, where the similarity is a cosine similarity, so as to obtain y keywords that are most similar to each keyword produced by the graph embedding model.
In operation S804, the second keyword set is expanded according to the first similar keyword set, the second similar keyword set, and the third similar keyword set to obtain the keyword set of the picture advertisement. This operation S804 can be implemented by a flow described in the following fig. 9, and will not be described in detail here.
According to the embodiment of the present disclosure, the similar keywords obtained in the operations S801, S802, and S803 may be fused, for example, the result generated in the three ways is de-duplicated, each similar keyword is matched with the original keyword, during matching, the english object words in the keywords are normalized into chinese object words according to the chinese-english object dictionary, then the editing distance insensitive to the sequence between the similar keyword phrases and the original keyword phrases is calculated, the similar keywords having the editing distance greater than an editing distance threshold value with the original keyword are removed, and the similar keywords meeting the condition are retained. Wherein the edit distance threshold is a configurable value.
Because the search behavior diversity of the user is strong, especially, the same semantic has various expression modes in the e-commerce search and presents long-tail distribution, the keywords with the same/similar semantics in the long-tail distribution need to be fully mined on the premise of ensuring the correlation among similar search keywords. The embodiment of the disclosure uses various methods to calculate similar search keywords, and performs fusion based on rules to calculate similarity of the search keywords from various angles.
Fig. 9 schematically shows a flowchart of expanding the second keyword set according to the first similar keyword set, the second similar keyword set and the third similar keyword set to obtain the keyword set of the picture advertisement according to the embodiment of the present disclosure.
As shown in fig. 9, the expanding the second keyword set according to the first similar keyword set, the second similar keyword set, and the third similar keyword set to obtain the keyword set of the picture advertisement according to this embodiment may include, for example, operations S901 to S902.
In operation S901, determining similar keywords of which edit distances between the keywords in the first similar keyword set, the second similar keyword set, and the third similar keyword set and the keywords in the second keyword set are less than or equal to a threshold;
in operation S902, the second keyword set is expanded by using similar keywords whose edit distances to the keywords in the second keyword set are less than or equal to a threshold value, so as to obtain a keyword set of the picture advertisement.
The method and the device for expanding the similarity of the search keywords relevant to the excavated picture advertisement perform similarity expansion, the similarity expansion mode comprises phrase sequence normalization, similar keywords are calculated based on a word vector model, similar keywords are calculated based on a picture embedding model, and the obtained similar keywords are fused, so that the purpose of expanding the directional range is achieved.
Fig. 10 schematically shows a block diagram of an electronic device adapted to implement the above described method according to an embodiment of the present disclosure. The electronic device shown in fig. 10 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. 10, an electronic device 1000 according to an embodiment of the present disclosure includes a processor 1001 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)1002 or a program loaded from a storage section 1008 into a Random Access Memory (RAM) 1003. Processor 1001 may include, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor and/or associated chipset, and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), among others. The processor 1001 may also include onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of a method flow according to embodiments of the present disclosure.
In the RAM 1003, various programs and data necessary for the operation of the electronic apparatus 1000 are stored. The processor 1001, ROM 1002, and RAM 1003 are connected to each other by a bus 1004. The processor 1001 performs various operations of the method flow according to the embodiments of the present disclosure by executing programs in the ROM 1002 and/or the RAM 1003. Note that the programs may also be stored in one or more memories other than the ROM 1002 and the RAM 1003. The processor 1001 may also perform various operations of the method flows according to embodiments of the present disclosure by executing programs stored in the one or more memories.
Electronic device 1000 may also include an input/output (I/O) interface 1005, the input/output (I/O) interface 1005 also being connected to bus 1004, according to an embodiment of the present disclosure. Electronic device 1000 may also include one or more of the following components connected to I/O interface 1005: an input section 1006 including a keyboard, a mouse, and the like; an output section 1007 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage portion 1008 including a hard disk and the like; and a communication section 1009 including a network interface card such as a LAN card, a modem, or the like. The communication section 1009 performs communication processing via a network such as the internet. The driver 1010 is also connected to the I/O interface 1005 as necessary. A removable medium 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 1010 as necessary, so that a computer program read out therefrom is mounted into the storage section 1008 as necessary.
According to embodiments of the present disclosure, method flows according to embodiments of the present disclosure 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 storage medium, the computer program containing program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication part 1009 and/or installed from the removable medium 1011. The computer program performs the above-described functions defined in the system of the embodiment of the present disclosure when executed by the processor 1001. The systems, devices, apparatuses, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the present disclosure.
The present disclosure also provides a computer-readable storage medium, which may be contained in the apparatus/device/system described in the above embodiments; or may exist separately and not be assembled into the device/apparatus/system. The computer-readable storage medium carries one or more programs which, when executed, implement the method according to an embodiment of the disclosure.
According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium. Examples may include, but are not limited to: 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), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present 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.
For example, according to embodiments of the present disclosure, a computer-readable storage medium may include the ROM 1002 and/or the RAM 1003 described above and/or one or more memories other than the ROM 1002 and the RAM 1003.
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 or flowchart illustration, and combinations of blocks in the block diagrams 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. Those skilled in the art will appreciate that various combinations and/or combinations of features recited in the various embodiments and/or claims of the present disclosure can be made, even if such combinations or combinations are not expressly recited in the present disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments and/or claims of the present disclosure may be made without departing from the spirit or teaching of the present disclosure. All such combinations and/or associations are within the scope of the present disclosure.
The embodiments of the present disclosure have been described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described separately above, this does not mean that the measures in the embodiments cannot be used in advantageous combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be devised by those skilled in the art without departing from the scope of the present disclosure, and such alternatives and modifications are intended to be within the scope of the present disclosure.

Claims (10)

1. A picture advertisement pushing method comprises the following steps:
acquiring a search keyword of a user;
matching the search keywords of the user with a keyword set of the picture advertisement; and
under the condition that the keyword set of the picture advertisement is determined to have the keyword matched with the search keyword of the user, pushing a first picture advertisement corresponding to the matched keyword to the user;
the keyword set of the picture advertisement is set in the following way:
determining a first object set related to the picture advertisement according to the information of the picture advertisement;
expanding the first set of objects using a graph embedding model to obtain a second set of objects;
determining a first set of keywords related to the second set of objects;
determining a second keyword set related to the picture advertisement according to the first keyword set; and
and expanding the second keyword set to obtain a keyword set of the picture advertisement.
2. The method of claim 1, wherein expanding the first set of objects to obtain a second set of objects using a graph embedding model comprises:
constructing an object graph according to the user behavior sequence, wherein the object graph comprises a plurality of objects;
processing the object graph by using the graph embedding model to obtain an object vector;
determining a similar object for each object of the plurality of objects from the object vector;
and expanding the first object set by using the similar objects to obtain a second object set.
3. The method of claim 2, wherein constructing the object graph from the sequence of user behaviors comprises:
constructing a user behavior sequence according to at least one of clicking, adding and submitting confirmation behaviors of a user on an object;
determining a set of objects that interact with behaviors in the sequence of user behaviors;
and constructing the object graph by using the object sets interacted with the behaviors in the user behavior sequence.
4. The method of claim 1, wherein expanding the second set of keywords to obtain a set of keywords for the photo advertisement comprises:
sequencing the historical search keywords of the user according to the sequence of character strings to obtain a first similar keyword set:
performing word segmentation on historical search keywords of the user, and obtaining a second similar keyword set by using a word vector model based on the keywords after word segmentation;
constructing a keyword graph by using keywords having searching and clicking relations with the second object set, and obtaining a third similar keyword set by using the graph embedding model based on the keyword graph;
and expanding the second keyword set according to the first similar keyword set, the second similar keyword set and the third similar keyword set to obtain the keyword set of the picture advertisement.
5. The method of claim 4, wherein expanding the second keyword set according to the first, second, and third similar keyword sets to obtain a keyword set of the picture advertisement comprises:
determining similar keywords of which the edit distances between the keywords in the first similar keyword set, the second similar keyword set and the third similar keyword set and the keywords in the second keyword set are smaller than or equal to a threshold value;
and expanding the second keyword set by using similar keywords of which the editing distance with the keywords in the second keyword set is less than or equal to a threshold value to obtain the keyword set of the picture advertisement.
6. The method of claim 1, wherein determining a first set of keywords related to the second set of objects comprises:
acquiring keywords having searching and clicking relations with the second object set;
and determining a first keyword set related to the second object set according to the keywords having searching and clicking relations with the second object set.
7. The method of claim 1, wherein determining a second set of keywords related to a picture advertisement from the first set of keywords comprises:
determining the relevancy of each keyword in the first keyword set and the picture advertisement;
and determining a second keyword set related to the picture advertisement according to the relevance of each keyword in the first keyword set and the picture advertisement.
8. The method of claim 1, wherein determining a first set of objects related to the picture advertisement from picture advertisement information comprises:
determining shop ID information, advertisement activity URL information, order following object ID information and picture URL information according to the picture advertisement information;
according to the shop ID information and the advertisement activity URL information, objects under shops and under activity landing pages are obtained, a documentary object is obtained according to the documentary object ID information, and the objects under shops and under activity landing pages and the documentary object form a candidate object set;
acquiring category words and brand words according to the picture URL information; and
and filtering the objects in the candidate object set according to the category words and the brand words so as to obtain the objects under the category and the brand, thereby determining a first object set related to the picture advertisement.
9. An electronic device, comprising:
one or more processors;
a storage device for storing one or more programs,
wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-8.
10. A computer readable storage medium having stored thereon executable instructions which, when executed by a processor, cause the processor to perform the method of any one of claims 1 to 8.
CN202011012322.XA 2020-09-23 2020-09-23 Picture advertisement pushing method, electronic equipment and computer readable storage medium Active CN113763005B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011012322.XA CN113763005B (en) 2020-09-23 2020-09-23 Picture advertisement pushing method, electronic equipment and computer readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011012322.XA CN113763005B (en) 2020-09-23 2020-09-23 Picture advertisement pushing method, electronic equipment and computer readable storage medium

Publications (2)

Publication Number Publication Date
CN113763005A true CN113763005A (en) 2021-12-07
CN113763005B CN113763005B (en) 2024-06-18

Family

ID=78785848

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011012322.XA Active CN113763005B (en) 2020-09-23 2020-09-23 Picture advertisement pushing method, electronic equipment and computer readable storage medium

Country Status (1)

Country Link
CN (1) CN113763005B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115618035A (en) * 2022-10-14 2023-01-17 北京京航计算通讯研究所 Picture recommendation method based on picture exposure conversion rate prediction model

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20130082539A (en) * 2011-12-08 2013-07-22 애드티브이노베이션(주) Internet advertisement system and method for displaying search keyword
WO2017000513A1 (en) * 2015-06-30 2017-01-05 百度在线网络技术（北京）有限公司 Information pushing method and apparatus based on user search behavior, storage medium, and device
WO2017181663A1 (en) * 2016-04-21 2017-10-26 北京搜狗科技发展有限公司 Method and device for matching image to search information
CN107451161A (en) * 2016-06-01 2017-12-08 阿里巴巴集团控股有限公司 Show method for pushing, device and the platform of object
KR101870353B1 (en) * 2017-01-26 2018-06-22 강원오픈마켓 주식회사 Proposing system for search advertising using issue keyword and associated keyword
CN108415961A (en) * 2018-02-06 2018-08-17 厦门集微科技有限公司 A kind of advertising pictures recommendation method and device
CN109446402A (en) * 2017-08-29 2019-03-08 阿里巴巴集团控股有限公司 A kind of searching method and device
CN109993570A (en) * 2019-01-14 2019-07-09 深圳市东信时代信息技术有限公司 A kind of orientation launches the method and system of moving advertising
CN110287440A (en) * 2019-06-27 2019-09-27 北京金山安全软件有限公司 Search engine optimization method and device, computer equipment and computer-readable storage medium
CN110851637A (en) * 2019-09-26 2020-02-28 华为技术有限公司 Picture searching method and device
CN111368171A (en) * 2020-02-27 2020-07-03 腾讯科技（深圳）有限公司 Keyword recommendation method, related device and storage medium
CN111401974A (en) * 2020-06-03 2020-07-10 北京每日优鲜电子商务有限公司 Information sending method, information sending device, electronic equipment and computer readable medium
CN111699503A (en) * 2019-06-04 2020-09-22 深圳海付移通科技有限公司 Advertisement recommendation method and device and electronic equipment

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20130082539A (en) * 2011-12-08 2013-07-22 애드티브이노베이션(주) Internet advertisement system and method for displaying search keyword
WO2017000513A1 (en) * 2015-06-30 2017-01-05 百度在线网络技术（北京）有限公司 Information pushing method and apparatus based on user search behavior, storage medium, and device
WO2017181663A1 (en) * 2016-04-21 2017-10-26 北京搜狗科技发展有限公司 Method and device for matching image to search information
CN107451161A (en) * 2016-06-01 2017-12-08 阿里巴巴集团控股有限公司 Show method for pushing, device and the platform of object
KR101870353B1 (en) * 2017-01-26 2018-06-22 강원오픈마켓 주식회사 Proposing system for search advertising using issue keyword and associated keyword
CN109446402A (en) * 2017-08-29 2019-03-08 阿里巴巴集团控股有限公司 A kind of searching method and device
CN108415961A (en) * 2018-02-06 2018-08-17 厦门集微科技有限公司 A kind of advertising pictures recommendation method and device
CN109993570A (en) * 2019-01-14 2019-07-09 深圳市东信时代信息技术有限公司 A kind of orientation launches the method and system of moving advertising
CN111699503A (en) * 2019-06-04 2020-09-22 深圳海付移通科技有限公司 Advertisement recommendation method and device and electronic equipment
CN110287440A (en) * 2019-06-27 2019-09-27 北京金山安全软件有限公司 Search engine optimization method and device, computer equipment and computer-readable storage medium
CN110851637A (en) * 2019-09-26 2020-02-28 华为技术有限公司 Picture searching method and device
CN111368171A (en) * 2020-02-27 2020-07-03 腾讯科技（深圳）有限公司 Keyword recommendation method, related device and storage medium
CN111401974A (en) * 2020-06-03 2020-07-10 北京每日优鲜电子商务有限公司 Information sending method, information sending device, electronic equipment and computer readable medium

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
WEI-HAO HWAN, YEONG-SHENG CHEN, TSANG-MING JIANG: "Personalized Internet Advertisement Recommendation Service Based on Keyword Similarity", 2015 IEEE 39TH ANNUAL COMPUTER SOFTWARE AND APPLICATIONS CONFERENCE, 24 September 2015 (2015-09-24) *
XIAO YANG, TAO DENG, ZHI GUO, ZONGYAO DING: "Advertising Keyword Recommendation based on Supervised Link Prediction in Multi-Relational Network", WWW \'17 COMPANION: PROCEEDINGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB COMPANION, 3 April 2017 (2017-04-03) *
卢军;李哲;黄一杰;焦利涛;: "一种基于协同过滤与语义分析的个性化网络广告投放方法研究", 现代电子技术, no. 19, 1 June 2016 (2016-06-01) *
沈军彩;: "用户行为数据分析下的信息推送系统的设计", 现代电子技术, no. 17, 1 September 2017 (2017-09-01) *
钱程;阳小兰;: "基于语义扩展的网页标签推荐系统", 计算机工程与科学, no. 10, 15 October 2012 (2012-10-15) *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115618035A (en) * 2022-10-14 2023-01-17 北京京航计算通讯研究所 Picture recommendation method based on picture exposure conversion rate prediction model

Also Published As

Publication number Publication date
CN113763005B (en) 2024-06-18

Similar Documents

Publication Publication Date Title
US20210374196A1 (en) Keyword and business tag extraction
US10210243B2 (en) Method and system for enhanced query term suggestion
JP6022056B2 (en) Generate search results
CN107679119B (en) Method and device for generating brand derivative words
JP2018106658A (en) Method and apparatus for pushing information
US20080319974A1 (en) Mining geographic knowledge using a location aware topic model
US20160103758A1 (en) Online product testing using bucket tests
US20110077998A1 (en) Categorizing online user behavior data
CN104899322A (en) Search engine and implementation method thereof
CN110674620A (en) Target file generation method, device, medium and electronic equipment
CN112136127B (en) Action indicator for search operation output element
CN108153909B (en) Keyword putting word-expanding method and device, electronic equipment and storage medium
CN106681598B (en) Information input method and device
CN107609192A (en) The supplement searching method and device of a kind of search engine
CN104142990A (en) Search method and device
CN112668320A (en) Model training method and device based on word embedding, electronic equipment and storage medium
CN103365876B (en) Method and equipment for generating network operation auxiliary information based on relational graph
CN104199952A (en) Method and device for acquiring information of interest points
CN111475725A (en) Method, apparatus, device, and computer-readable storage medium for searching for content
CN109952571A (en) Image search result based on context
CN115423555A (en) Commodity recommendation method and device, electronic equipment and storage medium
CN111538830A (en) French retrieval method, French retrieval device, computer equipment and storage medium
CN113763005B (en) Picture advertisement pushing method, electronic equipment and computer readable storage medium
CN107665442B (en) Method and device for acquiring target user
JP5559750B2 (en) Advertisement processing apparatus, information processing system, and advertisement processing method

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