CN111368050A - Document page pushing method and device - Google Patents

Document page pushing method and device Download PDF

Info

Publication number
CN111368050A
CN111368050A CN202010123950.9A CN202010123950A CN111368050A CN 111368050 A CN111368050 A CN 111368050A CN 202010123950 A CN202010123950 A CN 202010123950A CN 111368050 A CN111368050 A CN 111368050A
Authority
CN
China
Prior art keywords
document page
user
document
pages
association
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
CN202010123950.9A
Other languages
Chinese (zh)
Other versions
CN111368050B (en
Inventor
毛铁峥
李子健
赵子元
颜强
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.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen 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 Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN202010123950.9A priority Critical patent/CN111368050B/en
Publication of CN111368050A publication Critical patent/CN111368050A/en
Application granted granted Critical
Publication of CN111368050B publication Critical patent/CN111368050B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/332—Query formulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/332—Query formulation
    • G06F16/3325—Reformulation based on results of preceding query
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/335—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0613—Third-party assisted
    • 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
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Machine Translation (AREA)

Abstract

The embodiment of the application provides a method and a device for pushing a document page. The method comprises the following steps: receiving a first set of document pages provided by a third party platform according to a user's intent; determining the relevance between the document pages in the first document page set and the user intention through a user behavior characteristic model, wherein the relevance is used for representing the degree of the document pages in the first document page set conforming to the user intention; filtering the document pages with the relevance lower than a preset threshold in the first document page set to obtain a second document page set; and pushing the document pages in the second document page set to a user. According to the technical scheme of the embodiment of the application, the accuracy of pushing the document page can be improved.

Description

Document page pushing method and device
Technical Field
The application relates to the technical field of computers and the Internet, in particular to a method and a device for pushing a document page.
Background
In a scenario of pushing a document page, for example, a scenario of pushing a document page for a user through an application platform, generally, the application platform requests a third party platform to provide the document page pushed to the user for the application platform by allocating the identified user intention and user intention requirement to the third party platform. However, in the prior art, a situation that some wrong document pages are pushed to a user often occurs, and therefore, how to improve the accuracy of pushing the document pages is a technical problem to be solved urgently.
Disclosure of Invention
The embodiment of the application provides a method and a device for pushing a document page, a computer readable medium and electronic equipment, so that the pushing accuracy of the document page can be improved at least to a certain extent.
Other features and advantages of the present application will be apparent from the following detailed description, or may be learned by practice of the application.
According to an aspect of an embodiment of the present application, there is provided a method for pushing a document page, including: receiving a first set of document pages provided by a third party platform according to a user's intent; determining the relevance between the document pages in the first document page set and the user intention through a user behavior characteristic model, wherein the relevance is used for representing the degree of the document pages in the first document page set conforming to the user intention; filtering the document pages with the relevance lower than a preset threshold in the first document page set to obtain a second document page set; and pushing the document pages in the second document page set to a user.
According to an aspect of the embodiments of the present application, there is provided a document page pushing apparatus, including: a receiving unit, configured to receive a first set of document pages provided by a third-party platform according to a user intention; a first determining unit, configured to determine, through a user behavior feature model, a degree of association between a document page in the first document page set and the user intention, where the degree of association is used to characterize a degree to which the document page in the first document page set meets the user intention; the filtering unit is used for filtering the document pages with the relevance lower than a preset threshold value in the first document page set to obtain a second document page set; and the pushing unit is used for pushing the document pages in the second document page set to the user.
In some embodiments of the present application, based on the foregoing solution, the apparatus further includes a first obtaining unit, configured to obtain a keyword, which is input by a user and used for searching a document page, before receiving a first document page set provided by a third-party platform according to a user intention, so as to determine the user intention of the user; a sending unit, configured to send the user intention to a third party platform to request the third party platform to provide the first set of document pages.
In some embodiments of the present application, based on the foregoing solution, the apparatus further includes a second obtaining unit, configured to obtain a behavior data set of a full number of users within a historical predetermined time interval; the construction unit is used for constructing a heteromorphic graph network according to the behavior data set; and the training unit is used for training a machine learning model through the heterogeneous graph network to obtain the user behavior characteristic model.
In some embodiments of the present application, based on the foregoing scheme, the second obtaining unit is configured to: behavior data sets of a full number of users at the time of inputting a keyword, at the time of clicking on a document page, and at the time of operating inside the document page within a predetermined time interval in history are acquired.
In some embodiments of the present application, based on the foregoing solution, the construction unit is configured to: determining the operation behaviors of the keywords, the document pages and the full amount of users in the document pages in the behavior data set as the network nodes of the heteromorphic graph; determining the association among the keywords in the behavior data set, the association between the keywords and the document page and the association between the document page and the operation behaviors of the full amount of users in the document page as an abnormal composition network connection edge; and constructing the heterogeneous graph network based on the heterogeneous graph network nodes and the heterogeneous graph network connection edges.
In some embodiments of the present application, based on the foregoing solution, the apparatus further includes a second determining unit, configured to determine, before the heterogeneous graph network is constructed based on the heterogeneous graph network nodes and the heterogeneous graph network connection edges, feature vectors of the respective heterogeneous graph network nodes based on content of keywords in the behavior data set, content of document pages, and operation behavior features of a full amount of users inside the document pages; and determining the length of each heterogeneous graph network connection edge based on the association times among the keywords in the behavior data set, the association times among the keywords and the document pages and the association times among the document pages and the operation behaviors of the whole users in the document pages, wherein the length of the connection edge is used for measuring the closeness degree of the association among different heterogeneous graph network nodes.
In some embodiments of the present application, based on the foregoing solution, the apparatus further includes a feedback unit, configured to, after filtering document pages in the first document page set whose association degree is lower than a predetermined threshold, feed back the document pages whose association degree is lower than the predetermined threshold to a corresponding third-party platform, so as to request the third-party platform to change the provided document pages.
In some embodiments of the present application, based on the foregoing scheme, the second obtaining unit is configured to: after the document pages in the second document page set are pushed to the user, acquiring feedback behavior data of the user aiming at the pushed document pages; the training unit is configured to: and training the user behavior feature model based on the feedback behavior data of the user.
In some embodiments of the present application, based on the foregoing scheme, the document page comprises an applet page.
According to an aspect of the embodiments of the present application, there is provided a computer readable medium, on which a computer program is stored, which when executed by a processor implements the method for pushing document pages as described in the above embodiments.
According to an aspect of an embodiment of the present application, there is provided an electronic device including: one or more processors; a storage device for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the method of pushing document pages as described in the above embodiments.
In the technical scheme provided by some embodiments of the application, firstly, a user behavior feature model is used, then, the association degree between the document pages in a first document page set and the user intention is determined in the first document page set provided by a third-party platform according to the user intention through the user behavior feature model, the document pages with the association degree lower than a preset threshold value in the first document page set are filtered to obtain a second document page set, and finally, the document pages in the second document page set are pushed to a user. Because the relevance is used for representing the degree that the document pages in the first document page set meet the intention of the user, the document pages with the relevance lower than a preset threshold are filtered, so that the document pages pushed to the user meet the intention and requirements of the user better, and the accuracy of pushing the document pages can be improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application. It is obvious that the drawings in the following description are only some embodiments of the application, and that for a person skilled in the art, other drawings can be derived from them without inventive effort. In the drawings:
FIG. 1 shows a schematic diagram of an exemplary system architecture to which aspects of embodiments of the present application may be applied;
FIG. 2 is a diagram illustrating an application scenario of a document page pushing method according to an embodiment of the present application;
FIG. 3 shows a flow diagram of a method of pushing a document page according to one embodiment of the present application;
FIG. 4 illustrates a flowchart of a method prior to receiving a first set of document pages provided by a third party platform according to a user's intent, according to one embodiment of the present application;
FIG. 5 illustrates a flow diagram of a method of obtaining a model of a user behavior feature according to one embodiment of the present application;
FIG. 6 illustrates a detailed flow diagram for constructing a heterogeneous graph network according to one embodiment of the present application;
FIG. 7 illustrates a flow diagram of a method prior to building a heterogeneous graph network, according to one embodiment of the present application;
FIG. 8 shows a schematic structural diagram of a heteromorphic graphic network according to one embodiment of the present application;
FIG. 9 illustrates a flowchart of a method after pushing document pages in a second set of document pages to a user, according to one embodiment of the present application;
FIG. 10 illustrates a flow diagram of a method for pushing an applet page according to one embodiment of the present application;
FIG. 11 shows a block diagram of a pushing device of document pages according to an embodiment of the present application;
FIG. 12 illustrates a schematic structural diagram of a computer system suitable for use in implementing the electronic device of an embodiment of the present application.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art.
Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the subject matter of the present application can be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known methods, devices, implementations, or operations have not been shown or described in detail to avoid obscuring aspects of the application.
The block diagrams shown in the figures are functional entities only and do not necessarily correspond to physically separate entities. I.e. these functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor means and/or microcontroller means.
The flow charts shown in the drawings are merely illustrative and do not necessarily include all of the contents and operations/steps, nor do they necessarily have to be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
Fig. 1 shows a schematic diagram of an exemplary system architecture to which the technical solution of the embodiments of the present application can be applied.
As shown in fig. 1, the system architecture may include a terminal device (e.g., one or more of a smartphone 101, a tablet computer 102, and a portable computer 103 shown in fig. 1, but may also be a desktop computer, etc.), a network 104, and a server 105. The network 104 serves as a medium for providing communication links between terminal devices and the server 105. Network 104 may include various connection types, such as wired communication links, wireless communication links, and so forth.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation. For example, server 105 may be a server cluster comprised of multiple servers, or the like.
It is noted that the terms first, second and the like in the description and claims of the present application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the objects so used are interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in other sequences than those illustrated or described herein.
In an embodiment of the present application, the server 105 may receive a first document page set provided by a third-party platform according to a user intention, determine a relevance between a document page in the first document page set and the user intention in the first document page set through a user behavior feature model, filter a document page in the first document page set, where the relevance is lower than a predetermined threshold, obtain a second document page set, and finally send a document page in the second document page set to a terminal device, so as to push the document page to the user.
In another embodiment of the present application, the server 105 may also be configured to directly receive, by the terminal device, a first document page set provided by the third-party platform according to the user intention, determine, in the first document page set, the association degree between a document page in the first document page set and the user intention through the user behavior feature model, filter a document page in the first document page set, where the association degree is lower than a predetermined threshold value, obtain a second document page set, and finally directly push, by the terminal device, a document page in the second document page set to the user.
It should be noted that the pushing method of the document page provided in the embodiment of the present application is generally executed by the server 105, and accordingly, a pushing device of the document page is generally disposed in the server 105. However, in other embodiments of the present application, the terminal device may also have a similar function as the server, so as to execute the pushing scheme of the document page provided by the embodiments of the present application.
In addition, in the present application, the type of the document page may be an applet page, a public number page, or a web page. It should be noted that the types of document pages listed above are merely exemplary, and those skilled in the art will appreciate that other types of document pages may also be included in practical applications.
Fig. 2 is a schematic view of an application scenario of a document page pushing method according to an embodiment of the present application, and in particular, the application scenario shown in the figure is a scenario in which the document page pushing method proposed in the present application is applied to pushing an applet page for a user.
In the application scenario shown in the figure, a user may express his intention and demand by inputting "ABC marine park" in an APP application search bar, the background recognizes the intention of the user to order ABC marine park attraction tickets, a third party platform operated by ticketing (for example, a third party platform corresponding to "123 ticketing applet") provides applet pages (document pages) for ordering ABC marine park attraction tickets to the APP application background according to the intention of the user, then, after filtering some applet pages which do not conform to the intention of the user, the APP application background pushes and shows the applet pages which conform to the intention of the user to the user through an APP interface (for example, an interface shown in 201 in fig. 2), and the user enters the applet pages which order ABC marine park attraction tickets (for example, a page shown in 202 in fig. 2) by clicking one of the applet pages, the order ticket is then confirmed through a series of operations.
The implementation details of the technical solution of the embodiment of the present application are set forth in detail below:
according to a first aspect of the present disclosure, a method for pushing a document page is provided.
Referring to fig. 3, a flowchart of a document page pushing method according to an embodiment of the present application is shown, which may be performed by a device having a calculation processing function, such as the server 105 shown in fig. 1, or a terminal device as shown in fig. 1. As shown in fig. 3, the method for pushing document pages at least includes steps 320 to 380:
in step 320, a first set of document pages provided by a third-party platform according to a user intent is received.
In the application, the third-party platform generally establishes a corresponding document page in the background according to the needs and intentions of the user, and displays the corresponding document page to the user at the front end according to the service needs and intentions of the user.
However, it should be noted that in actual operation, due to the error of the third party platform, a problem of document page error may occur, for example, a problem that the document page provided by the third party platform does not conform to the service requirement of the user. Taking the application scenario shown in FIG. 2 where the document page is an applet page as an example, when the user's service requirement is to order ABC Marine park attraction tickets, the third party platform may provide an applet page ordering other Marine park attraction tickets by mistake (an applet page ordering other Marine park attraction tickets is established for the intent of "ordering ABC Marine park attraction tickets").
In one embodiment of the present application, the method as shown in FIG. 4 may also be implemented prior to receiving the first set of document pages provided by the third party platform according to the user's intent.
Referring to fig. 4, which shows a flowchart of a method before receiving a first document page set provided by a third party platform according to a user intention according to an embodiment of the present application, specifically, the method may include steps 311 to 312:
step 311, obtaining a keyword for searching a document page input by a user to determine a user intention of the user.
In the present application, the user's intention may be determined by obtaining keywords input by the user, for example, in the application scenario shown in fig. 2 where the document page is an applet page, the keyword "ABC marine park" input by the user may be determined by keywords, which may be implemented based on the existing semantic recognition technology.
Step 312, sending the user intention to a third party platform to request the third party platform to provide the first document page set.
In the application, the sending of the identified user intention to the third-party platform may be simultaneously sent to a plurality of third-party platforms, or may be separately sent to one third-party platform, and the purpose of the sending is to provide a first document page set for requesting one or more third-party platforms. It should be noted that the document page set may include a plurality of document pages, or may include only one document page.
With continued reference to FIG. 3, in step 340, a degree of association between the document pages in the first document page set and the user intention is determined through a user behavior feature model, and the degree of association is used for characterizing a degree of conformity of the document pages in the first document page set with the user intention.
In one embodiment of the present application, the user behavior feature model may be obtained by a method as shown in fig. 5.
Referring to fig. 5, a flowchart of a method for obtaining a user behavior feature model according to an embodiment of the present application is shown, which may specifically include steps 341 to 343:
step 341, acquiring a behavior data set of the total number of users in a historical preset time interval.
In a specific implementation of one embodiment, the behavior data set of the full-user within the predetermined time interval of the history may be the behavior data set of the full-user when the full-user inputs a keyword within the predetermined time interval of the history, when a document page is clicked, and when an operation is performed inside the document page.
Specifically, the predetermined time interval may be one week, one month, or one year. It is to be noted that the predetermined time interval may be arbitrary and is not limited to those exemplified previously.
It should be further noted that the behavior data of the user when inputting the keywords may refer to the content of the keywords input by the user, such as inputting "ABC marine park", or re-inputting another keyword after inputting one keyword, such as re-inputting "ABC marine park", or selecting the association keywords corresponding to the keywords input by the user, such as selecting the corresponding association keywords "ABC marine park" and "ABC marine park" skipped by the search bar after the user inputs "ABC": "ABC ocean park".
Behavior data of a user when clicking on a document page may refer to a user's selection behavior of the document page in a pushed document page, e.g., in FIG. 2, after the user enters "ABC ocean park" in the search bar, the first document page is selected in the two pushed document pages.
The behavior data of the user when operating inside the document page may refer to the specific operation of the user inside the document page after clicking the document page into the document page, for example, in fig. 2, after the user clicks the first document page, the operation behavior of "confirm reservation" is clicked inside the document page.
And 342, constructing a heteromorphic graph network according to the behavior data set.
In a specific implementation of an embodiment, the building a heterogeneous graph network according to the behavior data set may be implemented by the steps shown in fig. 6.
Referring to fig. 6, a detailed flowchart for constructing an heteromorphic graph network according to an embodiment of the present application is shown, which may specifically include steps 3421 to 3423:
step 3421, determining the keywords in the behavior data set, the document page, and the operation behaviors of the total users in the document page as the nodes of the heteromorphic network.
Step 3422, determining the association between the keywords in the behavior data set, the association between the keywords and the document page, and the association between the document page and the operation behavior of the full amount of users in the document page as the heterogeneous graph network connection edge.
It should be noted that the association between the keywords refers to an association between two keywords that are continuously input by the user, for example, after the user inputs one keyword "ABC ocean park" in the search bar, another keyword "ABC ocean amusement park" is input or switched in the search bar again, and then an association exists between the keyword "ABC ocean park" and the keyword "ABC ocean amusement park"; the association between the keywords and the document page means that after a keyword is input by a user, a document page is selected and clicked in a pushed document page in a gathering manner, and then, an association is generated between the input keyword and the document page selected and clicked; the association between the document page and the operation behaviors of the total users in the document page refers to that the user performs a series of operation behaviors in the document page selected and clicked, and then the association is generated between the document page selected and clicked by the user and the operation behaviors performed in the document page selected and clicked by the user.
Step 3423, building the heterogeneous graph network based on the heterogeneous graph network nodes and the heterogeneous graph network connection edges.
In a specific implementation of an embodiment, before constructing a heterogeneous graph network based on the heterogeneous graph network nodes and the heterogeneous graph network connection edges, a method as shown in fig. 7 may be further implemented.
Referring to fig. 7, a flowchart of a method before constructing a heterogeneous graph network according to an embodiment of the present application is shown, which may specifically include steps 331 to 332:
step 331, determining feature vectors of each heterogeneous graph network node based on the content of the keywords in the behavior data set, the content of the document page, and the operation behavior features of the full amount of users in the document page.
Step 332, determining the length of each heterogeneous graph network connection edge based on the association times among the keywords in the behavior data set, the association times among the keywords and the document page, and the association times among the document page and the operation behaviors of the full number of users in the document page, wherein the length of the connection edge is used for measuring the closeness degree of the association among different heterogeneous graph network nodes.
In order to enable those skilled in the art to understand the heterogeneous graph network as described above, the following explanation will be made with reference to the structure of the heterogeneous graph network shown in fig. 8.
Referring to fig. 8, a schematic structural diagram of a heteromorphic graphic network according to one embodiment of the present application is shown.
As shown in the figure, the structure of the heteromorphic graph network is composed of 3 types of nodes (keyword nodes, document page nodes and operation behavior nodes) and 3 types (edge 1, edge 2 and edge 3) of connecting edges.
For the keyword node, the same keyword historically input or selected by the full-scale user may be used as a keyword node, wherein feature vectors of the respective corresponding keyword nodes may be determined according to the content of the respective keyword, for example, a word vector corresponding to the word "ABC marine park" may be used as a feature vector corresponding to the keyword "ABC marine park" node, and it should be noted that each keyword corresponds to a unique feature vector.
For the document page node, the document page is provided by the third-party platform according to the user intention, each document page provided by the third-party platform can be used as a document page node, wherein the feature vector of each corresponding document page node can be determined according to the content of each document page, for example, page data such as a title and a document corresponding to the document page can be converted into a vector to be used as the feature vector of the document page node.
For the operation behavior node, the operation behavior of the user inside the document page may be taken as the operation behavior node, wherein the specific operation behavior capable of being inside the document page may be numbered, for example, 1 forward number is "1", 1 second staying on the page is numbered as "2", the confirmation number is "3", the exit page number is "4", and the like. Further, the specific operation behavior of the user in the document page can obtain a combination of numbers, so that the combination of numbers can be used as a feature vector for characterizing the operation behavior of the user in the document page.
It should be noted that the operation behavior of the user in the document page can be determined through the feature vector corresponding to the operation behavior node, and further the recognition degree of the user on the pushed document page is determined, for example, when the operation behavior after the user clicks to enter a document page is to exit the page immediately, it indicates that the user does not recognize the document page.
For the dotted-dashed edges shown in fig. 8, the dotted-dashed edges are connecting edges between the keyword nodes, and are a bidirectional edge for performing association between the keywords, wherein the length of the dotted-dashed edges is determined by the number of times of association between the keywords, and is used for representing the degree of closeness of association between the keywords. For example, the longer the edge 1 is as shown in the figure, the more the association number between the keyword 1 and the keyword 2 is represented, that is, the more the number of times of switching between the keyword 1 and the keyword 2 is represented.
For the long-dashed edge shown in fig. 8, the long-dashed edge is a connecting edge between the keyword node and the document page node, is a one-way edge, and is pointed to the document page node by the keyword node, and the length of the long-dashed edge is determined by the number of associations between the keyword node and the document page node, and is used for representing the closeness of the association between the keyword node and the document page node. For example, the longer the edge 2 is as shown in the figure, the more times the document page corresponding to the document page node 1 is selected and clicked by the total users in the document pages recommended by the third-party platform according to the corresponding intention of the keyword 1.
For the solid line edge shown in fig. 8, the solid line edge is a connecting edge between the document page node and the operation behavior node of the user in the document page, and is a one-way edge, and the document page node points to the operation behavior node, and the length of the solid line edge is determined by the number of times of association between the document page node and the operation behavior node, and is used for representing the degree of closeness of association between the document page node and the operation behavior node. For example, the longer the edge 3 is as shown in the figure, the more times the operation behavior corresponding to the node 1 is performed in the document page corresponding to the node 1 representing the full amount of users.
And 343, training a machine learning model through the heterogeneous graph network to obtain the user behavior feature model.
Because the constructed heterogeneous graph network contains behavior information from the input of a full amount of users from a keyword to the operation in a document page, the machine learning model is trained through the heterogeneous graph network, so that the machine learning model learns the behavior information from the input of the keyword to the selection of the clicking of the document page and the operation in the document page of the full amount of users in the heterogeneous graph network, and the user behavior feature model is obtained.
In the present application, the determination of the relevance between the document page in the first document page set and the user intention through the user behavior feature model may be understood according to the following embodiments:
in one embodiment of the present application, for example, the association degree of the document page a with the user intention corresponding to the keyword a may be calculated by the following procedure:
firstly, determining the incidence relation between the nodes corresponding to the keywords a and the nodes corresponding to the document page A through a user behavior characteristic model, then determining the incidence relation between the nodes corresponding to the document page A and the nodes corresponding to the operation behaviors of the full-scale user in the document page A, and finally calculating the incidence degree between the nodes corresponding to the keywords a and the nodes corresponding to the document page A through the length of the connecting edge between the nodes corresponding to the document page A and each operation behavior node, thereby obtaining the incidence degree of the user intention corresponding to the documents page A and the keywords a.
For example, historically, 1000 users have operation behaviors in the document page a corresponding to the keyword a, where the operation behaviors of 750 users represent that the document page a can be pushed, and the operation behaviors of 250 users represent that the document page a cannot be pushed, and then the association degree between the document page a and the user intention corresponding to the keyword a may be determined as: 750/1000 is 75%.
For those skilled in the art, it should be understood that the degree of association can be used to characterize the degree to which the document pages in the first set of document pages meet the user's intent.
Continuing to refer to FIG. 3, in step 360, the document pages in the first document page set whose association degree is lower than the predetermined threshold are filtered to obtain a second document page set.
In a specific implementation of step 360, for example, the document page set provided by the third-party platform includes A, B, C, D4 document pages, where the association degrees of the 4 document pages with the user intention are a (90%), B (60%), C (80%), and D (40%), respectively, and if a value with an association degree of 70% is taken as a predetermined threshold, a second document page set including document page a (90%) and document page C (80%) is obtained after filtering the 4 document pages.
In an embodiment of the present application, after filtering the document pages in the first document page set whose association degree is lower than the predetermined threshold, the document pages whose association degree is lower than the predetermined threshold may be further fed back to the corresponding third-party platform, so as to request the third-party platform to change the provided document pages.
Continuing with FIG. 3, in step 380, the document pages in the second set of document pages are pushed to the user.
In an embodiment of the present application, after pushing the document pages in the second document page set to the user, the method as shown in fig. 9 may also be implemented.
Referring to fig. 9, which is a flowchart illustrating a method after pushing document pages in the second document page set to a user according to an embodiment of the present application, specifically, the method may include steps 391 to 392:
step 391, obtaining feedback behavior data of the user aiming at the pushed document page.
Specifically, in the present application, the feedback behavior data of the user for the pushed document page may refer to behavior data of the user when clicking on the pushed document page and when operating inside the clicked document page.
Step 392, training the user behavior feature model based on the feedback behavior data of the user.
In this embodiment, the advantage of further training the user behavior feature model through the behavior data fed back by the user is that the behavior information of the user in the user behavior feature model can be enriched, so that the determined association degree between the document page and the user intention is more accurate.
The following takes a pushing scenario of an applet page as an example with reference to fig. 10, and further explains a specific application of the technical solution in the embodiment of the present application:
referring to FIG. 10, a flow diagram of a method for pushing applet pages is shown, according to one embodiment of the present application. Specifically, the method comprises steps 1001 to 1004:
step 1001, identifying a user intention through a keyword input by a user and used for searching the applet page, and sending the user intention to a third-party platform.
Step 1002, the third party platform provides a first applet page set including at least one applet page to a user according to the user intent.
Step 1003, training based on user behavior data obtained by a total amount of historical users during searching of the small program pages to obtain a user behavior feature model, and filtering the small program pages which do not accord with the user intention in the first small program page set to obtain a second small program page set.
It should be noted that, regarding the user behavior feature model, the user behavior data of historically a full number of users searching a small program page may be first constructed as a heterogeneous graph network, and then the user behavior feature model is obtained through training of the heterogeneous graph network.
And step 1004, pushing the applet pages in the second applet page set to the user.
In the scenario of pushing the applet page, some erroneous documents may appear in the document corresponding to the applet page established by the third-party platform, which may cause that some applet pages that are actually not compliant with the user's intention or even deviate from the user's intention may exist in the first applet page set provided by the third-party platform for the user according to the user's intention. Therefore, the applet pages in the first applet page set which do not meet the user intention can be filtered through the user behavior feature model provided by the application.
In this application, the principle that the applet pages in the first applet page set that do not meet the user intention can be filtered through the user behavior feature model is as follows: the user behavior feature model is obtained by training based on historical user behavior data of a full amount of users when searching the small program page, and for small program pages which are provided by some third-party platforms and do not accord with the user intention in the historical user behavior data of the full amount of users when searching the small program page, historical user pairs can generate some non-approved behaviors, such as behavior of closing the small program page. Therefore, the user behavior feature model trained based on the historical user behavior data of the whole user in searching the applet pages has the capability of identifying (for example, identified by the technical means of determining the association degree between the user application and the document page proposed in the above embodiment of the present application) the applet pages which do not meet the user intention, so that the applet pages which do not meet the user intention in the first applet page set can be filtered through the user behavior feature model.
Furthermore, user behavior data of a whole amount of historical users during searching of the small program pages are constructed into the heterogeneous graph network, so that the user behavior feature model can learn the historical user behavior data more comprehensively, and the small program pages which do not accord with the user intention can be identified by the user behavior feature model more strongly.
For those skilled in the art, it should be understood that the filtering of the applet pages in the first set of applet pages that do not meet the user's intent by the user behavior feature model has the benefits of: the applet pages pushed to the user can be made more accurate.
In summary, in the technical solution provided in the foregoing embodiment of the present application, a user behavior feature model is first used, then a relevance between a document page in a first document page set and a user intention is determined in the first document page set provided by a third party platform according to the user intention through the user behavior feature model, a document page in the first document page set, whose relevance is lower than a predetermined threshold, is filtered to obtain a second document page set, and finally, a document page in the second document page set is pushed to a user. Because the relevance is used for representing the degree that the document pages in the first document page set meet the intention of the user, the document pages with the relevance lower than a preset threshold are filtered, so that the document pages pushed to the user meet the intention and requirements of the user better, and the accuracy of pushing the document pages can be improved.
The following describes embodiments of an apparatus of the present application, which may be used to execute a method for pushing document pages in the above embodiments of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, please refer to the embodiments of the document page pushing method described above in the present application.
FIG. 11 shows a block diagram of a pushing device of document pages according to one embodiment of the present application.
Referring to fig. 11, a document page pushing device 1100 according to an embodiment of the present application includes: a receiving unit 1101, a first determining unit 1102, a filtering unit 1103 and a pushing unit 1104.
Wherein, the receiving unit 1101 is used for receiving a first document page set provided by a third party platform according to a user intention; a first determining unit 1102, configured to determine, through a user behavior feature model, a degree of association between a document page in the first document page set and the user intention, where the degree of association is used to characterize a degree to which the document page in the first document page set meets the user intention; a filtering unit 1103, configured to filter document pages in the first document page set, where the association degree is lower than a predetermined threshold, to obtain a second document page set; a pushing unit 1104, configured to push the document pages in the second document page set to the user.
In some embodiments of the present application, based on the foregoing solution, the apparatus further includes a first obtaining unit, configured to obtain a keyword, which is input by a user and used for searching a document page, before receiving a first document page set provided by a third-party platform according to a user intention, so as to determine the user intention of the user; a sending unit, configured to send the user intention to a third party platform to request the third party platform to provide the first set of document pages.
In some embodiments of the present application, based on the foregoing solution, the apparatus further includes a second obtaining unit, configured to obtain a behavior data set of a full number of users within a historical predetermined time interval; the construction unit is used for constructing a heteromorphic graph network according to the behavior data set; and the training unit is used for training a machine learning model through the heterogeneous graph network to obtain the user behavior characteristic model.
In some embodiments of the present application, based on the foregoing scheme, the second obtaining unit is configured to: behavior data sets of a full number of users at the time of inputting a keyword, at the time of clicking on a document page, and at the time of operating inside the document page within a predetermined time interval in history are acquired.
In some embodiments of the present application, based on the foregoing solution, the construction unit is configured to: determining the operation behaviors of the keywords, the document pages and the full amount of users in the document pages in the behavior data set as the network nodes of the heteromorphic graph; determining the association among the keywords in the behavior data set, the association between the keywords and the document page and the association between the document page and the operation behaviors of the full amount of users in the document page as an abnormal composition network connection edge; and constructing the heterogeneous graph network based on the heterogeneous graph network nodes and the heterogeneous graph network connection edges.
In some embodiments of the present application, based on the foregoing solution, the apparatus further includes a second determining unit, configured to determine, before the heterogeneous graph network is constructed based on the heterogeneous graph network nodes and the heterogeneous graph network connection edges, feature vectors of the respective heterogeneous graph network nodes based on content of keywords in the behavior data set, content of document pages, and operation behavior features of a full amount of users inside the document pages; and determining the length of each heterogeneous graph network connection edge based on the association times among the keywords in the behavior data set, the association times among the keywords and the document pages and the association times among the document pages and the operation behaviors of the whole users in the document pages, wherein the length of the connection edge is used for measuring the closeness degree of the association among different heterogeneous graph network nodes.
In some embodiments of the present application, based on the foregoing solution, the apparatus further includes a feedback unit, configured to, after filtering document pages in the first document page set whose association degree is lower than a predetermined threshold, feed back the document pages whose association degree is lower than the predetermined threshold to a corresponding third-party platform, so as to request the third-party platform to change the provided document pages.
In some embodiments of the present application, based on the foregoing scheme, the second obtaining unit is configured to: after the document pages in the second document page set are pushed to the user, acquiring feedback behavior data of the user aiming at the pushed document pages; the training unit is configured to: and training the user behavior feature model based on the feedback behavior data of the user.
In some embodiments of the present application, based on the foregoing scheme, the document page comprises an applet page.
FIG. 12 illustrates a schematic structural diagram of a computer system suitable for use in implementing the electronic device of an embodiment of the present application.
It should be noted that the computer system 1200 of the electronic device shown in fig. 12 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.
As shown in fig. 12, the computer system 1200 includes a Central Processing Unit (CPU)1201, which can perform various appropriate actions and processes, such as performing the methods described in the above embodiments, according to a program stored in a Read-Only Memory (ROM) 1202 or a program loaded from a storage section 1208 into a Random Access Memory (RAM) 1203. In the RAM 1203, various programs and data necessary for system operation are also stored. The CPU 1201, ROM 1202, and RAM 1203 are connected to each other by a bus 1204. An Input/Output (I/O) interface 1205 is also connected to bus 1204.
The following components are connected to the I/O interface 1205: an input section 1206 including a keyboard, a mouse, and the like; an output section 1207 including a Display device such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and a speaker; a storage section 1208 including a hard disk and the like; and a communication section 1209 including a network interface card such as a LAN (Local area network) card, a modem, or the like. The communication section 1209 performs communication processing via a network such as the internet. A driver 1210 is also connected to the I/O interface 1205 as needed. A removable medium 1211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is mounted on the drive 1210 as necessary, so that a computer program read out therefrom is mounted into the storage section 1208 as necessary.
In particular, according to embodiments of the application, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, embodiments of the present application 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 by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 1209, and/or installed from the removable medium 1211. The computer program executes various functions defined in the system of the present application when executed by a Central Processing Unit (CPU) 1201.
It should be noted that the computer readable medium shown in the embodiments of the present application 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), a flash Memory, an optical fiber, a portable Compact Disc Read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable 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 this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. 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.
The units described in the embodiments of the present application may be implemented by software, or may be implemented by hardware, and the described units may also be disposed in a processor. Wherein the names of the elements do not in some way constitute a limitation on the elements themselves.
As another aspect, the present application also provides a computer-readable medium, which may be contained in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by an electronic device, cause the electronic device to implement the method described in the above embodiments.
It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the application. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present application.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains.
It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims (10)

1. A method for pushing a document page, the method comprising:
receiving a first set of document pages provided by a third party platform according to a user's intent;
determining the relevance between the document pages in the first document page set and the user intention through a user behavior characteristic model, wherein the relevance is used for representing the degree of the document pages in the first document page set conforming to the user intention;
filtering the document pages with the relevance lower than a preset threshold in the first document page set to obtain a second document page set;
and pushing the document pages in the second document page set to a user.
2. The method of claim 1, wherein prior to receiving the first set of document pages provided by the third-party platform according to the user's intent, the method further comprises:
acquiring a keyword which is input by a user and used for searching a document page so as to determine the user intention of the user;
sending the user intent to a third party platform to request the third party platform to provide the first set of document pages.
3. The method of claim 1, wherein the user behavior feature model is obtained by:
acquiring a behavior data set of a full amount of users in a historical preset time interval;
constructing a heteromorphic graph network according to the behavior data set;
and training a machine learning model through the heterogeneous graph network to obtain the user behavior characteristic model.
4. The method of claim 3, wherein the obtaining a set of behavior data of a full number of users within a predetermined time interval comprises:
behavior data sets of a full number of users at the time of inputting a keyword, at the time of clicking on a document page, and at the time of operating inside the document page within a predetermined time interval in history are acquired.
5. The method of claim 4, wherein constructing a heterogeneous graph network from the set of behavior data comprises:
determining the operation behaviors of the keywords, the document pages and the full amount of users in the document pages in the behavior data set as the network nodes of the heteromorphic graph;
determining the association among the keywords in the behavior data set, the association between the keywords and the document page and the association between the document page and the operation behaviors of the full amount of users in the document page as an abnormal composition network connection edge;
and constructing the heterogeneous graph network based on the heterogeneous graph network nodes and the heterogeneous graph network connection edges.
6. The method of claim 5, wherein prior to constructing the heterogeneous graph network based on the heterogeneous graph network nodes and the heterogeneous graph network connection edges, the method further comprises:
determining the feature vector of each heterogeneous graph network node based on the content of the keywords in the behavior data set, the content of the document page and the operation behavior features of the full amount of users in the document page;
and determining the length of each heterogeneous graph network connection edge based on the association times among the keywords in the behavior data set, the association times among the keywords and the document pages and the association times among the document pages and the operation behaviors of the whole users in the document pages, wherein the length of the connection edge is used for measuring the closeness degree of the association among different heterogeneous graph network nodes.
7. The method according to claim 1, wherein after filtering document pages in the first set of document pages for which the degree of association is below a predetermined threshold, the method further comprises:
and feeding back the document page with the relevance lower than a preset threshold value to a corresponding third-party platform to request the third-party platform to change the provided document page.
8. The method of claim 1, wherein after pushing document pages in the second set of document pages to a user, the method further comprises:
acquiring feedback behavior data of a user aiming at a pushed document page;
and training the user behavior feature model based on the feedback behavior data of the user.
9. The method of any of claims 1-8, wherein the document page comprises an applet page.
10. A device for pushing pages of a document, comprising:
a receiving unit, configured to receive a first set of document pages provided by a third-party platform according to a user intention;
a first determining unit, configured to determine, through a user behavior feature model, a degree of association between a document page in the first document page set and the user intention, where the degree of association is used to characterize a degree to which the document page in the first document page set meets the user intention;
the filtering unit is used for filtering the document pages with the relevance lower than a preset threshold value in the first document page set to obtain a second document page set;
and the pushing unit is used for pushing the document pages in the second document page set to the user.
CN202010123950.9A 2020-02-27 2020-02-27 Method and device for pushing document pages Active CN111368050B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010123950.9A CN111368050B (en) 2020-02-27 2020-02-27 Method and device for pushing document pages

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010123950.9A CN111368050B (en) 2020-02-27 2020-02-27 Method and device for pushing document pages

Publications (2)

Publication Number Publication Date
CN111368050A true CN111368050A (en) 2020-07-03
CN111368050B CN111368050B (en) 2023-07-21

Family

ID=71206508

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010123950.9A Active CN111368050B (en) 2020-02-27 2020-02-27 Method and device for pushing document pages

Country Status (1)

Country Link
CN (1) CN111368050B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112860626A (en) * 2021-02-04 2021-05-28 北京百度网讯科技有限公司 Document sorting method and device and electronic equipment
CN113032549A (en) * 2021-05-31 2021-06-25 北京明略昭辉科技有限公司 Document sorting method and device, electronic equipment and storage medium

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105335391A (en) * 2014-07-09 2016-02-17 阿里巴巴集团控股有限公司 Processing method and device of search request on the basis of search engine
CN105659225A (en) * 2013-07-26 2016-06-08 微软技术许可有限责任公司 Query expansion and query-document matching using path-constrained random walks
CN105701155A (en) * 2015-12-30 2016-06-22 百度在线网络技术（北京）有限公司 Information push method and the device
CN107302566A (en) * 2017-05-27 2017-10-27 冯小平 The method and apparatus of pushed information
CN107463704A (en) * 2017-08-16 2017-12-12 北京百度网讯科技有限公司 Searching method and device based on artificial intelligence
CN107832414A (en) * 2017-11-07 2018-03-23 百度在线网络技术（北京）有限公司 Method and apparatus for pushed information
CN108540508A (en) * 2017-03-02 2018-09-14 百度在线网络技术（北京）有限公司 Method, apparatus and equipment for pushed information
US20190258721A1 (en) * 2018-02-19 2019-08-22 Microsoft Technology Licensing, Llc Standardized entity representation learning for smart suggestions
CN110569437A (en) * 2019-09-05 2019-12-13 腾讯科技（深圳）有限公司 click probability prediction and page content recommendation methods and devices

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105659225A (en) * 2013-07-26 2016-06-08 微软技术许可有限责任公司 Query expansion and query-document matching using path-constrained random walks
CN105335391A (en) * 2014-07-09 2016-02-17 阿里巴巴集团控股有限公司 Processing method and device of search request on the basis of search engine
CN105701155A (en) * 2015-12-30 2016-06-22 百度在线网络技术（北京）有限公司 Information push method and the device
CN108540508A (en) * 2017-03-02 2018-09-14 百度在线网络技术（北京）有限公司 Method, apparatus and equipment for pushed information
CN107302566A (en) * 2017-05-27 2017-10-27 冯小平 The method and apparatus of pushed information
CN107463704A (en) * 2017-08-16 2017-12-12 北京百度网讯科技有限公司 Searching method and device based on artificial intelligence
CN107832414A (en) * 2017-11-07 2018-03-23 百度在线网络技术（北京）有限公司 Method and apparatus for pushed information
US20190258721A1 (en) * 2018-02-19 2019-08-22 Microsoft Technology Licensing, Llc Standardized entity representation learning for smart suggestions
CN110569437A (en) * 2019-09-05 2019-12-13 腾讯科技（深圳）有限公司 click probability prediction and page content recommendation methods and devices

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112860626A (en) * 2021-02-04 2021-05-28 北京百度网讯科技有限公司 Document sorting method and device and electronic equipment
CN112860626B (en) * 2021-02-04 2023-07-28 北京百度网讯科技有限公司 Document ordering method and device and electronic equipment
CN113032549A (en) * 2021-05-31 2021-06-25 北京明略昭辉科技有限公司 Document sorting method and device, electronic equipment and storage medium

Also Published As

Publication number Publication date
CN111368050B (en) 2023-07-21

Similar Documents

Publication Publication Date Title
US9020862B2 (en) Method and system for computer question-answering
US20120290509A1 (en) Training Statistical Dialog Managers in Spoken Dialog Systems With Web Data
US20190179966A1 (en) Method and apparatus for identifying demand
CN109871311B (en) Method and device for recommending test cases
CN103207899A (en) Method and system for recommending text files
US11651015B2 (en) Method and apparatus for presenting information
CN111552880A (en) Data processing method, device and medium based on knowledge graph and electronic equipment
CN114385780B (en) Program interface information recommendation method and device, electronic equipment and readable medium
WO2012158571A2 (en) Training statistical dialog managers in spoken dialog systems with web data
CN111274490B (en) Method and device for processing consultation information
CN106471497A (en) Auxiliary using context browses
CN112966081A (en) Method, device, equipment and storage medium for processing question and answer information
CN111368050A (en) Document page pushing method and device
CN113971243A (en) Data processing method, system, equipment and storage medium applied to questionnaire survey
CN114528391A (en) Method, device and equipment for training question-answer pair scoring model and storage medium
CN113590756A (en) Information sequence generation method and device, terminal equipment and computer readable medium
CN113870998A (en) Interrogation method, device, electronic equipment and storage medium
CN108509442B (en) Search method and apparatus, server, and computer-readable storage medium
CN110705308B (en) Voice information domain identification method and device, storage medium and electronic equipment
CN116525093A (en) Prediction method, device, equipment and storage medium for session ending
CN109471969A (en) Application search method, device and equipment
CN110795424B (en) Characteristic engineering variable data request processing method and device and electronic equipment
CN114297380A (en) Data processing method, device, equipment and storage medium
CN115238165A (en) Information pushing method and device based on machine learning, storage medium and terminal
CN109408716B (en) Method and device for pushing information

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