WO2017045415A1 - 内容投放方法及装置 - Google Patents

内容投放方法及装置 Download PDF

Info

Publication number
WO2017045415A1
WO2017045415A1 PCT/CN2016/082502 CN2016082502W WO2017045415A1 WO 2017045415 A1 WO2017045415 A1 WO 2017045415A1 CN 2016082502 W CN2016082502 W CN 2016082502W WO 2017045415 A1 WO2017045415 A1 WO 2017045415A1
Authority
WO
WIPO (PCT)
Prior art keywords
content
user account
evaluation value
user
account
Prior art date
Application number
PCT/CN2016/082502
Other languages
English (en)
French (fr)
Inventor
卢铮
Original Assignee
腾讯科技（深圳）有限公司
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 腾讯科技（深圳）有限公司 filed Critical 腾讯科技（深圳）有限公司
Priority to JP2018514335A priority Critical patent/JP6661754B2/ja
Priority to EP16845529.3A priority patent/EP3352121A1/en
Publication of WO2017045415A1 publication Critical patent/WO2017045415A1/zh
Priority to US15/842,275 priority patent/US10621516B2/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/55—Push-based network services
    • 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
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • 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
    • 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
    • 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
    • G06Q30/0278—Product appraisal
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/06—Generation of reports
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/52—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail for supporting social networking services
    • 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
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking

Definitions

  • the present invention relates to the field of Internet technologies, and in particular, to a content delivery method and a content delivery device.
  • Content delivery is a way to promote specific multimedia content such as advertising content, news content, notification content, audio and video to a specific group of people.
  • an existing social network-based application usually publishes an advertisement content through a public number or a social network application account with a high degree of interest, and the other user account that pays attention to the social network application account can receive the advertisement, thereby realizing the advertisement content. Delivery.
  • the content delivery system in the conventional technology generally estimates the degree of interest of the user to the content to be delivered, and then pushes the content to the estimated user account with a higher degree of interest.
  • a content usage method with a higher usage rate is to select a user account to serve according to the click rate and conversion rate of the delivered content (in the advertisement delivery system, that is, the probability that the user purchases the corresponding product after clicking the advertisement).
  • the inventors have found through research that the content delivery system in the conventional technology has at least the following problems: in order to allow more users to receive the content delivery system, that is, to increase the breadth of content delivery, the content delivery system needs to screen a large number of The user account is delivered, and the content is pushed to more user accounts, which greatly reduces the accuracy of content delivery.
  • a method of content delivery including:
  • a content delivery device comprising:
  • An initialization module configured to traverse a user account, and generate an initial content propagation evaluation value of the first user account traversed
  • a friend search module configured to traverse the user account, and search for the friend account of the second user account that is traversed
  • a content propagation evaluation value update module configured to calculate a weighted sum of content propagation evaluation values of the friend accounts of the second user account, and update the content according to the sum of the weighted sum and the first forwarding probability of the second user account Content dissemination evaluation value of the second user account;
  • a content data pushing module configured to: when the content propagation evaluation value of each of the second user accounts converges, jump out of an iteration, select a user account to push content data according to the content propagation evaluation value; and call the friend to search when not converging
  • the module and content propagation evaluation value update module iterates.
  • the content delivery method and device utilize the association and forwarding operations between user accounts in the social network application, and only need to push the content data to a small portion of the user account, and then deliver the content data to the small portion of the user by sequentially forwarding the content data.
  • the content data is further spread from the small portion of the user account to a wider range.
  • the degree of association between users also ensures that the user clicks or views the received content data forwarded by the user concerned by the user. The rate, so as to improve the accuracy and breadth of the content data delivery.
  • FIG. 1 is a flow chart of a content delivery method in an embodiment
  • FIG. 2 is a schematic diagram of a social relationship chain of user accounts in an embodiment
  • FIG. 3 is a flow chart showing a process of calculating a weight coefficient in one embodiment
  • FIG. 4 is a flow chart showing a process of correcting content propagation evaluation values in an embodiment
  • FIG. 5 is a flowchart of selecting a user account for content delivery according to a click rate in an embodiment
  • FIG. 6 is a schematic diagram of a content delivery device in an embodiment
  • FIG. 7 is a schematic structural diagram of a computer device running the foregoing content delivery method in an embodiment.
  • a new type is proposed.
  • a content delivery method the method is based on a mapping relationship between a plurality of user accounts including a large number of user accounts and a user relationship between the user accounts (ie, a friend relationship established by a limited number of user accounts), for example, If users A and B are friends, B and C are friends, and C and D are friends, then the social network application of ABCD exists between A and D), which utilizes the user accounts in the social network application.
  • Association and forwarding operations just inside The data is pushed to a small portion of the user account, and then the content data is passed to other user accounts on the social relationship chain on the small portion of the user account by forwarding in turn.
  • the content data is further spread from the small portion of the user account to a wider range.
  • the degree of association between the users also ensures that the user clicks or views the probability of receiving the content data forwarded by the user concerned by the user, thereby reducing the initial need under the premise of ensuring the accuracy and breadth of the delivery of the content data.
  • the number of user accounts that push content data reduces the cost of delivery.
  • the implementation of the method may depend on a computer program, which may be a social networking chain application such as a social network application, an instant messaging application, a web game application, etc., and the computer program may run on a von Neumann system.
  • a computer program which may be a social networking chain application such as a social network application, an instant messaging application, a web game application, etc.
  • the computer program may run on a von Neumann system.
  • the computer system may be a server device of the social networking chain application, such as the social network application, the instant messaging application, the web game application, and the like.
  • the content delivery method includes:
  • Step S102 Acquire a traversal user account, and generate an initial content propagation evaluation value of the traversed first user account.
  • the content data is the data that needs to be delivered to the user account.
  • the content data may be advertisement content data, news content data, multimedia content data, and the like.
  • the content data that needs to be delivered to the user account is an advertisement.
  • content data that needs to be delivered to a user account is news information.
  • the content data can typically contain a link and a summary description of the complete content corresponding to the link.
  • the link may jump to the purchase page, the read page of the complete news content, or the play page of the multimedia file.
  • the content data to be served may include an advertisement image, text information of the advertisement, and a link corresponding to the e-commerce purchase page. If the content data is pushed to a certain user account, the user who logs in using the user account can receive the advertisement content data on the terminal, and by clicking the link in the advertisement, the user can jump to the corresponding e-commerce purchase page. .
  • the present invention uses the "content propagation evaluation value" metric to evaluate the ability of a user account to disseminate content data.
  • the larger the content propagation evaluation value of a user account the more the number of users that may be transmitted when the user account receives the content data; the more the content propagation evaluation value of a user account Small means that when the user account receives the content data, the number of users that may be transmitted is small.
  • the first forwarding probability of each user account in the social network application may be pre-calculated, and the first forwarding probability is used as the initial content propagation evaluation value of each user account.
  • the first forwarding probability of the user account forwarding the received content data may be calculated by acquiring the number N1 of the posted content data, obtaining the number N2 of the content data forwarded by the user account A, and forwarding the user account A by calculating N2/N1. The first forwarding probability of the content data.
  • a page of an app of a news category periodically posts news with a share button based on a social networking application.
  • the number N1 of posting news on the page may be counted, and the number N2 of forwarding the news using the user account A is calculated, and then the N2/N1 is calculated to obtain the first forwarding probability that the user account A forwards the page to release.
  • the input content data may be acquired in advance, and a keyword or a tag matching the received content data is extracted, and the number N1 of publications of the published content data matching the keyword or the tag is obtained, and the user account A is obtained. Forwarding the number N2 of the content data that also matches the keyword or tag, and then calculating N2/N1 to obtain the first forwarding probability that the user account A forwards the page to release the gratification.
  • a specific first forwarding probability is calculated as an initial content propagation evaluation value of each user account, so that content communication capabilities corresponding to each user account can be determined for the received different content data, so that the content Data can spread more widely.
  • the vocabulary that defines the user identifiers in the first, second, etc. used herein is a process for distinguishing multiple traversals in the execution process of the method, and the first user account is the initial calculation.
  • the second user account is a stage of iteratively updating the content distribution evaluation value of the user account in the social network application, traversing the user account in the social network application;
  • the third user account and the fourth user account are respectively the stages of correcting the content propagation evaluation value of the user account in the social network application, and traversing the user account in the social network application.
  • the user accounts referred to by the first, second, etc. user accounts are still user accounts in the social network application, and are not described below.
  • the iterative process may be entered.
  • the iterative process one iteration will traverse all user accounts, and the above traversal process is distinguished, and the user account traversed in the iterative process is marked as the second user account.
  • Iterated The process is a process of updating the content propagation evaluation value of the second user account traversed according to the content of the other user account, until the content propagation evaluation value of all the user accounts converges to not change (or change) with the evolution of the iterative process. The amount is less than the threshold).
  • Step S104 Traversing the user account and searching for the friend account of the second user account traversed.
  • Step S106 Calculate a weighted sum of the content propagation evaluation values of the friend accounts of the second user account, and update the content of the second user account according to the sum of the weighted sum and the first forwarding probability of the second user account. Spread the evaluation value.
  • FIG. 2 illustrates a social relationship chain between user accounts of social network applications on which the method relies. If the user accounts in the social network application are A, B, C, D, E, and F, where A and B, C, and D are friends, B and A, and F are friends, C and A, D, E is a friend relationship, D and A, C, and E are friends, E and C, D, and F are friends, and F and B and E are friends.
  • First traverse from user account A, traversing to A, can be executed:
  • A's friend account namely B, C, D
  • ⁇ b, ⁇ c, ⁇ d is the weighted sum of the contents of the evaluation calculation propagation time B, C, D value MPA b, MPA e, MPA d respective weight coefficients.
  • the first forwarding probability ⁇ a of A can be obtained, and the content of the first forwarding probability of A and the sum of the foregoing weighted sums are obtained:
  • MPA a ⁇ a + ⁇ b ⁇ MPA b + ⁇ c ⁇ MPA c + ⁇ d ⁇ MPA d
  • the iterative process is the first iteration, when traversing to A, the content propagation evaluation value of A is updated by the initial first forwarding probability ⁇ a to:
  • MPA a ⁇ a + ⁇ b ⁇ MPA b + ⁇ c ⁇ MPA c + ⁇ d ⁇ MPA d
  • MPA b ⁇ b + ⁇ a ⁇ MPA a + ⁇ f ⁇ MPA f
  • the iterative process is the first iteration, when traversing to B, the content propagation evaluation value of B is updated by the initial first forwarding probability B a to:
  • MPA b ⁇ b + ⁇ a ⁇ MPA a + ⁇ f ⁇ MPA f
  • MPA a ⁇ a + ⁇ b ⁇ MPA b + ⁇ c ⁇ MPA c + ⁇ d ⁇ MPA d
  • A, B, C, D, E, and F may be sequentially traversed in the above manner to calculate respective content propagation evaluation values. Then, the content propagation evaluation values of A, B, C, D, E, and F can be gradually corrected as the number of iterations increases until convergence converges without change.
  • the content propagation evaluation value MPA ti can be according to the formula:
  • MPA tj is the j-th friend of Ni in the i-th user account traversed when calculating MPA ti
  • the content propagation evaluation value of the account, ⁇ j is the weight coefficient of the content propagation evaluation value of the jth friend account in Ni of the i-th user account traversed.
  • the weighting coefficient of the content propagation evaluation value of each friend account may be set according to the content forwarding capability of the second user account of each friend account of the second user account traversed, specifically, as shown in the figure As shown in 3, it can be executed:
  • Step S202 Traversing the friend account of the second user account, and calculating a second forwarding probability of the traversed friend account to the second user account.
  • Step S204 Set the second forwarding probability to a weighting coefficient of the content propagation evaluation value of the friend account that traverses the second user account.
  • Step S206 Calculate a weighted sum of the content propagation evaluation values of the friend accounts of the second user account according to the weight coefficient of the content distribution evaluation value of the friend account of the second user account.
  • the second forwarding probability of the second user account of the second user account is the ratio of the friend account to the content of the second user account, and the specific executable:
  • the friend account of A is found as B, C, and D. If A posts the content (for example, posting a microblog, posting a circle of friends, publishing a blog post, etc.), the number of times is 100. B forwarded 20 times, friend C forwarded 30 times, and friend D forwarded 40 times, then friend B's second forwarding probability for user A was 0.2, and friend C's second forwarding probability for user A was 0.3, and friend D for User A's second forwarding probability is 0.4, so the foregoing content propagation evaluation value of A can be calculated according to the formula:
  • MPA a ⁇ a +0.2 ⁇ MPA b +0.3 ⁇ MPA c +0.4 ⁇ MPA d
  • each user account may be traversed in each iteration, as in the previous example, after traversing A, B, C, D, E, and F in each iteration, It can be determined whether the content propagation evaluation value of each user account is converged, and if yes, step S108 is performed: selecting the user account to push the content data according to the content propagation evaluation value. If the content propagation evaluation value of the second user account does not converge, return to the iteration, and step S104 is performed to traverse the user account again.
  • a change value when updating the content propagation evaluation value of each of the second user accounts may be acquired, and when each of the change values is less than a threshold, determining content propagation of each of the second user accounts The evaluation value converges.
  • step S104 and step S106 are performed k times in an iterative manner, the change value of MPA ka to MPA k-1a is smaller than the threshold value, and the change value of MPA kb to MPA k-1b is smaller than the threshold value. If the change value of MPA kf to MPA k-1f is less than the threshold value, it can be determined that the content propagation evaluation value of each of the second user accounts converges to end the iteration, and step S108 is performed.
  • the number of traversals of the second user account that performs the traversing of the calculated forwarding probability may be obtained, and when the number of traversal times is greater than or equal to the threshold, determining each of the second user accounts The content propagation evaluation value converges.
  • Step S104 and step S106 are performed 1000 times, and the content propagation evaluation values of each of A, B, C, D, E, and F are obtained.
  • Steps S108 are performed for the MPA 1000a , the MPA 1000b , and the MPA 1000f .
  • the content propagation evaluation value of each user account converges after the iteration of k times, the content propagation evaluation value of each user account is obtained as MPA ki , and then the size of the MPA ki can be sorted, and then According to the sorting, the user account with a larger MPA ki is selected as the target of the content data push.
  • the user account is selected as the target of content data push according to the page order of A, B, C, D, E, and F.
  • the preset number of serving users can be obtained; the user accounts are sorted according to the content propagation evaluation value, and the user account pushing the content data in the forefront of the sorted user accounts is selected.
  • a and B can be selected as targets for content data push.
  • the content data will have a high probability of being forwarded by A and B to be received by C, D, and E, so that when the content data is pushed to a smaller user account, the content data can still be delivered to the user. More user accounts.
  • the content propagation evaluation value of each user account obtained after the iteration may be corrected, and the correction manner may include multiple methods, and the following manner is adopted. Two embodiments are set forth.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • Step S302 Traverse the user account, obtain the content propagation evaluation value of the traversed third user account, and find a collection of user accounts whose content propagation evaluation value is greater than the content distribution evaluation value of the third user account.
  • the third user account is the user account traversed when traversing the user account in the modification process of the embodiment.
  • MPA ka >MPA kd >MPA kf >MPA kb >MPA ke >MPA kc is obtained , in this traversal, if traversing to F, the content propagation evaluation value of the found content is greater than F.
  • the user accounts for the value are A and D.
  • Step S304 Search for a minimum value of the number of user accounts that are separated from the searched user account by the third user account in the social relationship chain.
  • F and A are not friends, but F and B are friends, and B and A are friends. Therefore, in the social relationship chain of F to A, the number of user accounts is 2 (ie, B and A).
  • F and D are not friends, but F and E are friends, and E and D are friends. Therefore, in the social relationship chain of F to D, the number of user accounts is 2 (ie, E and D). Therefore, the minimum number of user accounts that are separated from A by A in the social relationship chain and the number of user accounts that are separated from D is 2.
  • Step S306 Calculate the correction coefficient by using the minimum value as an independent variable according to a preset increment function.
  • Step S308 Update the content propagation evaluation value of the third user account according to the correction coefficient.
  • MPA f(n) ⁇ MPA, and f(n) is an increasing function
  • n is the corresponding minimum value of the user account.
  • f(n) can be preset to:
  • the content propagation evaluation value greater than MPA kd is MPA ka
  • the minimum value of the number of user accounts separated by A and D in the social relationship chain is 1, and the content propagation evaluation value of the corrected D is MPA kd. That is:
  • the content propagation evaluation value greater than MPA kf is MPA ka and MPA kd , and the minimum value of the number of user accounts separated by F between A and D in the social relationship chain is 2, and the content propagation evaluation value of the modified F is corrected.
  • MPA kf is :
  • the MPA kd is larger than the MPA kf
  • D and A are relatively close in the social relationship chain
  • the probability that the user accounts sequentially transmitted after A and D are forwarded may be the same batch of user accounts.
  • F and A are far away in the social relationship chain
  • the distribution points are more dispersed and can be spread to a wider range of transmission through forwarding.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • Step S402 Traversing the user account, and acquiring the click probability of the traversed fourth user account on the content data.
  • Step S404 Calculate a weighted sum of the content propagation evaluation value of the fourth user account and the click probability of the fourth user account on the content data.
  • Step S406 Push the content data according to the weighted sum selection user account.
  • the fourth user account is a user account traversed when the user account is traversed during the modification process of the embodiment.
  • the user account is selected as the target of the content data push, not only the content communication capability of the user account but also the probability of the user account clicking or converting the content data is needed.
  • the MPA kf is small relative to the MPA kd , if F has a high probability of clicking the content data for viewing browsing or conversion, F is still selected as the target of the content data push.
  • F is still selected as the target of the content data push.
  • both the content dissemination capability of the user account and the degree of adaptation of the content data and the user account are referenced, thereby improving the accuracy of the content data delivery.
  • a new content delivery device includes an initialization module 102, a friend search module 104, a content propagation evaluation value update module 106, and a content data push module 108, wherein:
  • the initialization module 102 is configured to traverse the user account and generate an initial content propagation evaluation value of the first user account traversed.
  • the buddy search module 104 is configured to traverse the user account and search for the buddy account of the traversed second user account.
  • the content propagation evaluation value update module 106 is configured to calculate a weighted sum of the content propagation evaluation values of the friend accounts of the second user account, and update the location according to the sum of the weighted sum and the first forwarding probability of the second user account.
  • the content propagation evaluation value of the second user account is described.
  • the content data pushing module 108 is configured to: when the content propagation evaluation value of each of the second user accounts converges, jump out of the iteration, select a user account to push content data according to the content propagation evaluation value; and call the friend when not converging
  • the lookup module 104 and the content propagation evaluation value update module 106 iterate.
  • the initialization module 102 is further configured to calculate a first forwarding probability of the traversed first user account, and set a first forwarding probability to an initial content propagation rating of the traversed first user account. value.
  • the content propagation evaluation value update module 106 is further configured to traverse the friend account of the second user account, and calculate a second forwarding probability of the traversed friend account to the second user account;
  • the second forwarding probability is set as a weighting coefficient of the content propagation evaluation value of the friend account traversing the second user account; and calculating the weight coefficient according to the content propagation evaluation value of the friend account of the second user account.
  • the content propagation evaluation value update module 106 is further configured to obtain the number of times the second user account is published, and obtain the number of times the traversed friend account forwards the content of the second user account. And dividing the number of forwardings by the number of publications to calculate a second forwarding probability of the traversed friend account to the second user account.
  • the apparatus further includes a first convergence determining module 110, configured to acquire a change value when updating a content propagation evaluation value of each of the second user accounts, where each of the change values When the threshold value is less than the threshold, it is determined that the content propagation evaluation value of each of the second user accounts converges.
  • a first convergence determining module 110 configured to acquire a change value when updating a content propagation evaluation value of each of the second user accounts, where each of the change values When the threshold value is less than the threshold, it is determined that the content propagation evaluation value of each of the second user accounts converges.
  • the apparatus further includes a second convergence determining module 112, configured to acquire a number of traversal times of performing the traversing the calculated second forwarding user account, in the When the number of traversal times is greater than or equal to the threshold, it is determined that the content propagation evaluation value of each of the second user accounts converges.
  • a second convergence determining module 112 configured to acquire a number of traversal times of performing the traversing the calculated second forwarding user account, in the When the number of traversal times is greater than or equal to the threshold, it is determined that the content propagation evaluation value of each of the second user accounts converges.
  • the apparatus further includes a content propagation evaluation value correction module 114, And traversing a user account, obtaining a content propagation evaluation value of the traversed third user account, and searching for a collection of user accounts whose content propagation evaluation value is greater than a content distribution evaluation value of the third user account; searching for the third a minimum value of the number of user accounts separated from the searched user account by the user account; the correction coefficient is calculated according to the preset increment function with the minimum value as an independent variable; The content propagation evaluation value of the third user account is updated.
  • the content data pushing module 108 is further configured to traverse the user account, obtain the click probability of the traversed fourth user account to the content data, and calculate a content propagation evaluation value of the fourth user account. a weighted sum of the click probability of the content data by the fourth user account; and pushing the content data according to the weighted sum selection user account.
  • the content data pushing module 108 is further configured to obtain a preset number of serving users; sort the user accounts according to the content propagation evaluation value, and select a user who is in the forefront of the sorted user accounts.
  • the account push content data.
  • the content delivery method according to FIGS. 1 and 3 to 5 may be performed by respective units in the content delivery apparatus shown in FIG. 6.
  • steps S102, S104, S106, and S108 shown in FIG. 1 may be performed by the initialization module 102, the buddy lookup module 104, the content propagation evaluation value update module 106, and the data content push module 108 shown in FIG. 6, respectively;
  • the steps S202, S204, and S106 shown in FIG. 6 can be performed by the content propagation evaluation value update module 106 shown in FIG. 6;
  • the steps S302, S304, S306, and S308 shown in FIG. 4 can be propagated by the content evaluation value shown in FIG.
  • the modification module 114 is executed; the steps S402, S404, and S406 shown in FIG. 5 can be performed by the content data push module 108 shown in FIG. 6.
  • each unit in the content delivery apparatus shown in FIG. 6 may be separately or completely combined into one or several additional units, or one of the units may be further removed. It is divided into a plurality of units that are functionally smaller, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present invention.
  • the above units are divided based on logical functions. In practical applications, the functions of one unit may also be implemented by multiple units, or the functions of multiple units may be implemented by one unit. In other embodiments of the invention, the terminal device may also include other modules. However, in practical applications, these functions can also be implemented by other units, and can be implemented by multiple units.
  • FIGS. 1 and 3 to 5 A computer program capable of executing the content delivery method as shown in FIGS. 1 and 3 to 5 is executed on a general-purpose computing device such as a computer that accesses a processing element such as a memory access memory (RAM) or a read only memory (ROM) and a storage element (for example, a computer)
  • the program code is included to construct a content delivery device as shown in FIG. 6, and to implement a content delivery method according to an embodiment of the present invention.
  • the computer program can be recorded, for example, on a computer readable recording medium, and loaded in and run in the above-described computing device by a computer readable recording medium.
  • the content delivery method and device utilize the association and forwarding operations between user accounts in the social network application, and only need to push the content data to a small portion of the user account, and then deliver the content data to the small portion of the user by sequentially forwarding the content data.
  • the content data is further spread from the small portion of the user account to a wider range.
  • the degree of association between the users also ensures the probability that the user clicks or views the received content data forwarded by the user concerned by the user, thereby ensuring the accuracy and breadth of the delivery of the content data.
  • FIG. 7 illustrates a server device of a von Neumann system-based computer system 10 that operates the first or second content delivery method described above.
  • the computer system may be a server, a server cluster, or a server cluster device that provides a virtual machine as a server program execution environment to execute the above content delivery method.
  • the network interface 1001, the processor 1002, and the memory 1003 connected through a system bus may be included.
  • the memory 1003 may include an external memory 10032 (eg, a hard disk, an optical disk, or a floppy disk, etc.) and an internal memory 10034.
  • the operation of the method is based on a computer program, the program file of which is stored in the external memory 10032 of the aforementioned von Neumann system-based computer system 10, and is loaded into the internal memory 10034 at runtime, It is then compiled into a machine code and passed to the processor 1002 for execution, so that the logical initialization module 102, the friend lookup module 104, the content propagation evaluation value update module 106, and the content are formed in the von Neumann system-based computer system 10.
  • Data push module 108 During the execution of the content delivery method, the input parameters are received by the network interface 1001, and are transferred to the cache in the memory 1003, and then input to the processor 1002 for processing, and the processed result data is cached in the memory 1003 for subsequent processing. It is processed or passed to the network interface 1001 for output.
  • the memory 1003 stores a set of program codes, and the processor 1002 calls the program code stored in the memory 1003 to perform the following operations:
  • the processor 1002 invokes the program code stored in the memory 1003 to perform the operation of generating the initial content propagation evaluation value of the first user traversed, including:
  • the processor 1002 invokes the program code stored in the memory 1003 to perform the operation of calculating the weighted sum of the content propagation evaluation values of the friend account of the second user account, and may further include:
  • the processor 1002 invokes the program code stored in the memory 1003 to perform the operation of calculating the second forwarding probability of the buddy account that is traversed to the second user account, and may also include:
  • the processor 1002 calls the program code stored in the memory 1003, Used to do the following:
  • the processor 1002 invokes program code stored in the memory 1003 and is also used to perform the following operations:
  • the processor 1002 may further include:
  • the processor 1002 invokes the program code stored in the memory 1003 to perform the operation of selecting the user account to push the content data according to the content propagation evaluation value, and may further include:
  • the processor 1002 invokes the program code stored in the memory 1003 to perform the operation of selecting the user account to push the content data according to the content propagation evaluation value, and may further include:
  • the user accounts are sorted according to the content propagation evaluation value, and the user account pushing the content data in the forefront of the sorted user accounts is selected.
  • a "computer readable medium” can be any apparatus that can contain, store, communicate, propagate, or transport a program for use in an instruction execution system, apparatus, or device, or in conjunction with such an instruction execution system, apparatus, or device.
  • computer readable media include the following: electrical connections (electronic devices) having one or more wires, portable computer disk cartridges (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM).
  • the computer readable medium may even be a paper or other suitable medium on which the program can be printed, as it may be optically scanned, for example by paper or other medium, followed by editing, interpretation or, if appropriate, other suitable The method is processed to obtain the program electronically and then stored in computer memory.
  • portions of the invention may be implemented in hardware, software, firmware or a combination thereof.
  • multiple steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system.
  • a suitable instruction execution system For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques well known in the art: having logic gates for implementing logic functions on data signals. Discrete logic circuits, application specific integrated circuits with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.
  • the above mentioned storage medium may be a read only memory, a magnetic disk or an optical disk or the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Human Resources & Organizations (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Data Mining & Analysis (AREA)
  • Operations Research (AREA)
  • General Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Probability & Statistics with Applications (AREA)
  • Algebra (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Computing Systems (AREA)
  • Information Transfer Between Computers (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本发明实施例公开了一种内容投放方法，包括：遍历用户账号，生成遍历到的第一用户的初始的内容传播评价值；迭代执行：遍历用户账号，查找所述遍历到的第二用户账号的好友账号；计算所述第二用户账号的好友账号的内容传播评价值的加权和，根据所述加权和与所述第二用户账号的第一转发概率的和更新所述第二用户账号的内容传播评价值；在各个所述第二用户账号的内容传播评价值收敛时，跳出迭代，根据所述内容传播评价值选择用户账号推送内容数据。本发明还公开了另一种内容投放装置。本发明中的内容投放方法及装置可以降低投放成本。

Description

内容投放方法及装置
本专利申请要求2015年09月17日提交的中国专利申请号为201510593480.1，发明名称为“内容投放方法及装置”的优先权，该申请的全文以引用的方式并入本申请中。
技术领域
本发明涉及互联网技术领域，尤其涉及一种内容投放方法及一种内容投放装置。
背景技术
内容投放即为将特定的广告内容、新闻内容、通知内容、音视频等多媒体资源内容推广给特定人群的一种方式。例如，现有的基于社交网络应用通常通过公众号或受关注度较高的社交网络应用账号发布广告内容，关注该社交网络应用账号的其他用户账号即可接收到该广告，从而实现广告内容的投放。
传统技术中的内容投放系统为了提高投放的精度，通常先预估用户对待投放内容的兴趣程度，然后将内容推送给估算得到的兴趣程度较高的用户账号。例如，使用率较高的内容投放方法为，根据投放内容的点击率和转化率(在广告投放系统中，即为用户点击广告后购买相应产品的概率)选择用户账号进行投放。
然而，发明人经研究发现，传统技术中的内容投放系统至少存在如下问题：传统技术中的内容投放系统为了让更多的用户接收到内容投放系统，即提高内容投放的广度，需要筛选出大量的用户账户进行投放，将投放内容推送给较多的用户账号，从而大大使得内容投放的精度较低。
发明内容
基于此，为了解决前述传统技术中的内容投放方式在选择推送对象时，需要对大量的用户账号推送内容，从而造成了内容投放的精度较低的问题，特提出了一种新的内容投放方法。
一种内容投放方法，包括：
获取内容数据，遍历用户账号，生成遍历到的第一用户账号转发所述内容 数据的初始的内容传播评价值；
迭代执行：
遍历用户账号，查找所述遍历到的第二用户账号的好友账号；
计算所述第二用户账号的好友账号的内容传播评价值的加权和，根据所述加权和与所述第二用户账号的第一转发概率的和更新所述第二用户账号的内容传播评价值；
在各个所述第二用户账号的内容传播评价值收敛时，跳出迭代，根据所述内容传播评价值选择用户账号推送内容数据。
此外，为了解决前述传统技术中的内容投放方式在选择推送对象时，需要对大量的用户账号推送内容，从而造成了内容投放的精度较低的问题，特提出了一种新的内容投放装置。
一种内容投放装置，包括：
初始化模块，用于遍历用户账号，生成遍历到的第一用户账号的初始的内容传播评价值；
迭代执行：
好友查找模块，用于遍历用户账号，查找所述遍历到的第二用户账号的好友账号；
内容传播评价值更新模块，用于计算所述第二用户账号的好友账号的内容传播评价值的加权和，根据所述加权和与所述第二用户账号的第一转发概率的和更新所述第二用户账号的内容传播评价值；
内容数据推送模块，用于在各个所述第二用户账号的内容传播评价值收敛时，跳出迭代，根据所述内容传播评价值选择用户账号推送内容数据；在不收敛时，调用所述好友查找模块和内容传播评价值更新模块进行迭代。
实施本发明实施例，将具有如下有益效果：
上述内容投放方法及装置利用了社交网络应用中的用户账号之间的关联与转发操作，只需将内容数据推送给小部分的用户账号，然后通过依次转发将内容数据传递给该小部分的用户账号上的社交关系链上的其他用户账号。将内容数据再由该小部分的用户账号扩散至较广的范围。同时，用户之间的关联程度也保证了用户点击或查看接收到的该用户关注的用户转发的内容数据的概 率，从而在提高了内容数据的投放精度的和广度的前提下。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
其中：
图1为一个实施例中一种内容投放方法的流程图；
图2为一个实施例中用户账号的社交关系链的示意图；
图3为一个实施例中计算权重系数的过程的流程图；
图4为一个实施例中对内容传播评价值进行修正过程的流程图；
图5为一个实施例中结合点击率选择用户账号进行内容投放的流程图；
图6为一个实施例中一种内容投放装置的示意图；及
图7为一个实施例中运行前述内容投放方法的计算机设备的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
为了解决前述传统技术中的内容投放方式在选择推送对象时，需要对大量的用户账号推送内容，从而造成了内容投放的精度较低的问题，在一个实施例中，特提出了一种新的内容投放方法，该方法基于包含有大量用户账号且用户账号之间存在社交关系链(即通过有限个数的用户账户两两之间的好友关系建立的任意两个用户账号的映射关系，例如，若用户A与B为好友，B与C为好友，C与D为好友，则A与D之间存在A-B-C-D的社交关系链)的社交网络应用，利用了社交网络应用中的用户账号之间的关联与转发操作，只需将内 容数据推送给小部分的用户账号，然后通过依次转发将内容数据传递给该小部分的用户账号上的社交关系链上的其他用户账号。将内容数据再由该小部分的用户账号扩散至较广的范围。同时，用户之间的关联程度也保证了用户点击或查看接收到的该用户关注的用户转发的内容数据的概率，从而在保证了内容数据的投放精度的和广度的前提下，减少起始需要推送内容数据的用户账号的数量，降低投放成本。
进一步的，该方法的实现可依赖于计算机程序，该计算机程序可以是社交网络应用、即时通信应用、网页游戏应用等存在社交关系链的应用程序，该计算机程序可运行于基于冯诺依曼体系的计算机系统之上，该计算机系统可以是上述社交网络应用、即时通信应用、网页游戏应用等存在社交关系链的应用程序的服务器设备。
具体的，如图1所示，该内容投放方法包括：
步骤S102：获取遍历用户账号，生成遍历到的第一用户账号的初始的内容传播评价值。
内容数据即为需要投放给用户账号的数据。从业务上进行划分，内容数据可以是广告内容数据、新闻内容数据、多媒体内容数据等。例如，在一个基于社交网络应用的广告投放系统的应用场景中，需要投放给用户账号的内容数据为广告。而在一个基于社交网络应用的新闻发布系统中，需要投放给用户账号的内容数据即为新闻资讯。
内容数据通常可包含链接以及该链接对应的完整内容的摘要描述信息。在不同的实现方式中，与实际业务所对应的，该链接可跳转至购买页面、完整新闻内容的阅读页面或者多媒体文件的播放页面。例如，在一个基于社交网络应用的广告投放系统的应用场景中，待投放的内容数据可以包含广告图片、广告的文本信息、以及对应电商购买页面的链接。若将该内容数据推送给某个用户账号，则使用该用户账号登录的用户可在终端上接收到该广告内容数据，通过点击该广告中的链接，即可跳转至相应的电商购买页面。
本发明使用了“内容传播评价值”这一度量值来评价某个用户账号传播内容数据的能力。一个用户账号的内容传播评价值越大，则表示在该用户账号接收到内容数据时，可能传播的用户数更广；一个用户账号的内容传播评价值越 小，则表示在该用户账号接收到内容数据时，可能传播的用户数较少。
在本实施例中，可预先计算社交网络应用中的每个用户账号各自的第一转发概率，并将第一转发概率作为每个用户账号初始的内容传播评价值。
用户账号转发接收到的内容数据的第一转发概率可通过以下方式计算：获取发布内容数据的次数N1，获取用户账号A转发发布的内容数据的次数N2，通过计算N2/N1得到用户账号A转发该内容数据的第一转发概率。
例如，某个新闻类的应用的某个页面定期会发布新闻，该页面上具有基于社交网络应用的分享按钮。可统计该页面发布新闻的次数N1，以及使用用户账号A转发该新闻的次数N2，然后计算N2/N1得到用户账号A转发该页面发布的欣慰的第一转发概率。
进一步的，还可预先获取输入的内容数据，提取接收到的内容数据匹配的关键字或标签(tag)，获取发布的同样匹配该关键字或标签的内容数据的发布次数N1，以及用户账号A转发该同样匹配该关键字或标签的内容数据的次数N2，然后计算N2/N1得到用户账号A转发该页面发布的欣慰的第一转发概率。对于特定的内容数据，计算特定的第一转发概率作为每个用户账号初始的内容传播评价值，从而可针对接收到的不同的内容数据确定每个用户账号与之对应的内容传播能力，使得内容数据能够传播的更加广泛。
需要说明的是，本文中使用的第一、第二...等对用户标识进行限定的词汇为用于区分本方法在执行过程中的多个遍历的过程，第一用户账号为计算初始的内容传播评价值时，遍历到的社交网络应用中的用户账号；第二用户账号为迭代更新社交网络应用中的用户账号的内容传播评价值的阶段，遍历到的社交网络应用中的用户账号；第三用户账号和第四用户账号则分别为后续对社交网络应用中的用户账号的内容传播评价值进行修正的阶段，遍历到的社交网络应用中的用户账号。但第一、第二...等用户账号指代的用户账号仍然为社交网络应用中的用户账号，以下不在赘述。
在本实施例中，在计算得到了每个社交网络应用的用户账号对接收到的内容数据的第一转发概率，即每个用户账号的初始内容传播评价值之后，则可进入迭代过程。在迭代过程中，一次迭代将遍历所有用户账号，位于前述的遍历过程加以区分，将迭代过程中遍历到的用户账号标记为第二用户账号。迭代过 程即为根据其他用户账号的内容传播评价值更新遍历到的第二用户账号的内容传播评价值的过程，直至所有用户账号的内容传播评价值收敛至不随迭代过程的演进而发生变化(或者变化量小于阈值)。
具体的，可迭代执行：
步骤S104：遍历用户账号，查找所述遍历到的第二用户账号的好友账号。
步骤S106：计算所述第二用户账号的好友账号的内容传播评价值的加权和，根据所述加权和与所述第二用户账号的第一转发概率的和更新所述第二用户账号的内容传播评价值。
如图2所示，图2展示了本方法所依赖的社交网络应用的用户账号之间的社交关系链。若社交网络应用中的用户账号为A、B、C、D、E和F，其中A与B、C、D互为好友关系，B与A、F互为好友关系，C与A、D、E互为好友关系，D与A、C、E互为好友关系，E与C、D、F互为好友关系，F与B、E互为好友关系。
在一次迭代过程中，可以A、B、C、D、E和F的页序遍历用户账号。先从用户账号A开始遍历，遍历到A时，可执行：
查找A的好友账号，即B、C、D，分别获取B、C、D的内容传播评价值MPAb、MPAc、MPAd；
计算B、C、D的内容传播评价值的加权和，即：
αb×MPAb+αc×MPAc+αd×MPAd
其中，αb、αc、αd为计算加权和时B、C、D的内容传播评价值MPAb、MPAe、MPAd各自的权重系数。
然后可获取A的第一转发概率βa，计算A的第一转发概率与前述加权和的和得到更新的A的内容传播评价值：
MPAa＝βa+αb×MPAb+αc×MPAc+αd×MPAd
也就是说，若该次迭代过程为首次迭代，在遍历到A时，A的内容传播评价值由初始的第一转发概率βa更新为：
MPAa＝βa+αb×MPAb+αc×MPAc+αd×MPAd
在该次迭代过程中遍历更新了A的内容传播评价值之后，则继续遍历到B，可执行：
查找B的好友账号，即A、F，分别获取A、F的内容传播评价值MPAa、MPAf；
计算A、F的内容传播评价值的加权和，即：
αa×MPAa+αf×MPAf
然后获取B的第一转发概率βb，计算B的第一转发概率与前述加权和的和得到更新的B的内容传播评价值：
MPAb＝βb+αa×MPAa+αf×MPAf
也就是说，若该次迭代过程为首次迭代，在遍历到B时，B的内容传播评价值由初始的第一转发概率Ba更新为：
MPAb＝βb+αa×MPAa+αf×MPAf
且此时A的内容传播评价值MPAa已由初始的βa更新为前述的：
MPAa＝βa+αb×MPAb+αc×MPAc+αd×MPAd
综上所述，在一次迭代过程中，可依次按照上述方式依次遍历A、B、C、D、E和F，计算各自的内容传播评价值。然后A、B、C、D、E和F的内容传播评价值可随着迭代次数的增加而得到逐步的修正，直至收敛而不发生变化。
也就是说，对于每次迭代过程中遍历到的第i个用户账号而言，其在第t次迭代时，其内容传播评价值MPAti可根据公式：
Figure PCTCN2016082502-appb-000001
计算得到，其中Ni每次迭代过程中遍历到的第i个用户账号的好友账号的总个数，MPAtj为计算MPAti时，遍历到的第i个用户账号的Ni中的第j个好友账号的内容传播评价值，αj为该遍历到的第i个用户账号的Ni中的第j个好友账号的内容传播评价值的权重系数。
在本实施例中，可根据遍历到的第二用户账号的每个好友账号对该第二用户账号的内容转发能力设定每个好友账号的内容传播评价值的权重系数，具体的，如图3所示，可执行：
步骤S202：遍历所述第二用户账号的好友账号，计算所述遍历到的好友账号对所述第二用户账号的第二转发概率。
步骤S204：将所述第二转发概率设为所述遍历所述第二用户账号的好友账号的内容传播评价值的权重系数。
步骤S206：根据所述第二用户账号的好友账号的内容传播评价值的权重系数计算所述第二用户账号的好友账号的内容传播评价值的加权和。
第二用户账号的某个好友账号对第二用户账号的第二转发概率即为该好友账号对第二用户账号发布的内容进行转发操作的比例，具体的可执行：
获取所述第二用户账号发布内容的发布次数；获取所述遍历到的好友账号转发所述第二用户账号发布的内容的转发次数；将所述转发次数除以所述发布次数计算所述遍历到的好友账号对所述第二用户账号的第二转发概率。
如前例中，遍历到用户账号A时，查找到A的好友账号为B、C、D，若A发布内容(例如发布微博、发表朋友圈、发布博文等发布操作)的次数100次，好友B转发了20次、好友C转发了30次、好友D转发了40次，则好友B对于用户A的第二转发概率为0.2，好友C对于用户A的第二转发概率为0.3，好友D对于用户A的第二转发概率为0.4，因此前述计算A的内容传播评价值时可根据公式：
MPAa＝βa+0.2×MPAb+0.3×MPAc+0.4×MPAd
计算得到。
在本实施例中，为了判断是否结束迭代过程，可在每次迭代遍历完成所有用户账号时，如前例中，在每次迭代过程中遍历了A、B、C、D、E和F之后，可判断每个用户账号的内容传播评价值是否均收敛，若是，则执行步骤S108：根据所述内容传播评价值选择用户账号推送内容数据。若存在第二用户账号的内容传播评价值不收敛，则返回迭代，执行步骤S104，对用户账号重新进行遍历。
具体的，在本实施例中，可获取更新各个所述第二用户账号的内容传播评价值时的变化值，在各个所述变化值小于阈值时，判定各个所述第二用户账号的内容传播评价值收敛。
例如，在若在迭代执行了上述步骤S104和步骤S106达到k次时，MPAka至相对于MPAk-1a的变化值小于阈值，MPAkb至相对于MPAk-1b的变化值小于阈值......，MPAkf至相对于MPAk-1f的变化值小于阈值，则可判定各个所述第二用 户账号的内容传播评价值收敛从而结束迭代，执行步骤S108。
在另一个实施例中，可获取执行所述遍历所述已计算得到转发概率的第二用户账号的遍历次数，在所述遍历次数大于或等于阈值时，判定所述各个所述第二用户账号的内容传播评价值收敛。
例如，可预设最大迭代次数为1000次，则在迭代执行了上述步骤S104和步骤S106达到1000次时，结束迭代，并得到A、B、C、D、E和F各自的内容传播评价值为MPA1000a、MPA1000b......、MPA1000f，执行步骤S108。
如前所述，若在迭代k次后，每个用户账号的内容传播评价值均收敛，则得到每个用户账号的内容传播评价值为MPAki，则可根据MPAki的大小进行排序，然后根据该排序选择MPAki较大的用户账号作为内容数据推送的目标。
例如，若MPAka＞MPAkb＞MPAkc＞MPAkd＞MPAke＞MPAkf，则根据A、B、C、D、E和F的页序选择用户账号作为内容数据推送的目标。
进一步的，可获取预设的投放用户数；根据内容传播评价值对用户账号进行排序，选取处于所述排序后的用户账号的前列的投放用户数的用户账号推送内容数据。
例如，在前例中，若预设的投放用户数为2，则可选择A、B作为内容数据推送的目标。如图2所示，该内容数据将有很大几率被A和B转发从而被C、D和E接收，从而在推送内容数据给较少的用户账号时，仍然能将该内容数据投放给较多的用户账号。
进一步的，根据所述内容传播评价值选择用户账号推送内容数据的步骤之前，还可对迭代之后得到的每个用户账号的内容传播评价值进行修正，修正方式可包括多种方法，下述通过两个实施例进行阐述。
实施例一：
在该实施例中，如图4所示，通过迭代得到了每个用户账号的收敛的内容传播评价值之后，可执行：
步骤S302：遍历用户账号，获取所述遍历到的第三用户账号的内容传播评价值，查找内容传播评价值大于所述第三用户账号的内容传播评价值的用户账号的集合。
第三用户账号为该实施例的修正过程中遍历用户账号时遍历到的用户账 号。如前例中，若得到MPAka＞MPAkd＞MPAkf＞MPAkb＞MPAke＞MPAkc，则在该次遍历中，若遍历到F时，查找到的内容传播评价值大于F的内容传播评价值的用户账号即为A和D。
步骤S304：查找所述第三用户账号在社交关系链上与所述查找到的用户账号之间相隔的用户账号数的最小值。
如前例中，F与A不为好友关系，但F与B为好友关系、B与A为好友关系，因此在F到A这条社交关系链上，相隔的用户账号数为2(即B和A)。F与D不为好友关系，但F与E为好友关系、E与D为好友关系，因此在F到D这条社交关系链上，相隔的用户账号数为2(即E和D)。因此，对于F在社交关系链上与A之间相隔的用户账号数以及与D之间相隔的用户账号数的最小值为2。
步骤S306：根据预设的递增函数以所述最小值为自变量计算修正系数。
步骤S308：根据所述修正系数对所述第三用户账号的内容传播评价值进行更新。
也就是说，可根据公式：
MPA＝f(n)×MPA，且f(n)为增函数
对每个用户账号的内容传播评价值进行更新。其中n为该用户账号的对应的前述的最小值。优选的，f(n)可预设为：
Figure PCTCN2016082502-appb-000002
如前例中，大于MPAkd的内容传播评价值为MPAka，且A与D之间在社交关系链上相隔的用户账号数的最小值为1，则修正得到的D的内容传播评价值MPAkd即为：
Figure PCTCN2016082502-appb-000003
而大于MPAkf的内容传播评价值为MPAka和MPAkd，且F与A、D之间在社交关系链上相隔的用户账号数的最小值为2，则修正得到的F的内容传播评价值MPAkf即为：
Figure PCTCN2016082502-appb-000004
因此，MPAkf相对于MPAkd的修正系数较大。
也就是说，虽然MPAkd相较于MPAkf较大，但由于D与A在社交关系链上较近，因此，A和D转发之后依次传播的用户账号可能为同一批用户账号的概率较大，从而使得投放点过于集中而下降了内容数据传播的范围。而由于F与A在社交关系链上较远，因此使得投放点较分散，可通过转发扩散至更广的传播范围。加入该基于社交关系链上相隔用户账号数的最小值的修正系数之后，可平衡内容传播能力与投放点的分散程度之间的关系，从而使得内容数据传播的范围更广阔。
实施例二：
在该实施例中，如图5所示，通过迭代得到了每个用户账号的收敛的内容传播评价值之后，可执行：
步骤S402：遍历用户账号，获取所述遍历到的第四用户账号对所述内容数据的点击概率。
步骤S404：计算所述第四用户账号的内容传播评价值和所述第四用户账号对所述内容数据的点击概率的加权和。
步骤S406：根据所述加权和选择用户账号推送内容数据。
第四用户账号为该实施例的修正过程中遍历用户账号时遍历到的用户账号。在该实施例中，在选择用户账号作为内容数据推送的目标时，不仅参考了用户账号的内容传播能力，还需要参考用户账号点击或转化内容数据的概率。
例如，在前例中，虽然MPAkf相对于MPAkd较小，但若F点击该内容数据进行查看浏览或者转化的概率较大，则仍然选择F作为内容数据推送的目标。使得在选择用户账号作为内容数据推送的目标时，既参考了用户账号的内容传播能力，又参考了内容数据与用户账号的适配程度，因此，提高了内容数据投放的精度。
为了解决前述传统技术中的内容投放方式在选择推送对象时，需要对大量的用户账号推送内容，从而造成了内容投放的精度较低的问题，在一个实施例中，如图6所示，提出了一种新的内容投放装置，包括初始化模块102、好友查找模块104、内容传播评价值更新模块106、内容数据推送模块108，其中：
初始化模块102，用于遍历用户账号，生成遍历到的第一用户账号的初始的内容传播评价值。
好友查找模块104，用于遍历用户账号，查找所述遍历到的第二用户账号的好友账号。
内容传播评价值更新模块106，用于计算所述第二用户账号的好友账号的内容传播评价值的加权和，根据所述加权和与所述第二用户账号的第一转发概率的和更新所述第二用户账号的内容传播评价值。
内容数据推送模块108，用于在各个所述第二用户账号的内容传播评价值收敛时，跳出迭代，根据所述内容传播评价值选择用户账号推送内容数据；在不收敛时，调用所述好友查找模块104和内容传播评价值更新模块106进行迭代。
在本实施例中，初始化模块102还用于计算所述遍历到的第一用户账号的第一转发概率，将第一转发概率设置为所述遍历到的第一用户账号的初始的内容传播评价值。
在本实施例中，内容传播评价值更新模块106还用于遍历所述第二用户账号的好友账号，计算所述遍历到的好友账号对所述第二用户账号的第二转发概率；将所述第二转发概率设为所述遍历所述第二用户账号的好友账号的内容传播评价值的权重系数；根据所述第二用户账号的好友账号的内容传播评价值的权重系数计算所述第二用户账号的好友账号的内容传播评价值的加权和。
在本实施例中，内容传播评价值更新模块106还用于获取所述第二用户账号发布内容的发布次数；获取所述遍历到的好友账号转发所述第二用户账号发布的内容的转发次数；将所述转发次数除以所述发布次数计算所述遍历到的好友账号对所述第二用户账号的第二转发概率。
在一个实施例中，如图6所示，该装置还包括第一收敛判定模块110，用于获取更新各个所述第二用户账号的内容传播评价值时的变化值，在各个所述变化值小于阈值时，判定各个所述第二用户账号的内容传播评价值收敛。
在另一个实施例中，如图6所示，该装置还包括第二收敛判定模块112，用于获取执行所述遍历所述已计算得到转发概率的第二用户账号的遍历次数，在所述遍历次数大于或等于阈值时，判定所述各个所述第二用户账号的内容传播评价值收敛。
在本实施例中，如图6所示，该装置还包括内容传播评价值修正模块114， 用于遍历用户账号，获取所述遍历到的第三用户账号的内容传播评价值，查找内容传播评价值大于所述第三用户账号的内容传播评价值的用户账号的集合；查找所述第三用户账号在社交关系链上与所述查找到的用户账号之间相隔的用户账号数的最小值；根据预设的递增函数以所述最小值为自变量计算修正系数；根据所述修正系数对所述第三用户账号的内容传播评价值进行更新。
在本实施例中，内容数据推送模块108还用于遍历用户账号，获取所述遍历到的第四用户账号对所述内容数据的点击概率；计算所述第四用户账号的内容传播评价值和所述第四用户账号对所述内容数据的点击概率的加权和；根据所述加权和选择用户账号推送内容数据。
在本实施例中，内容数据推送模块108还用于获取预设的投放用户数；根据内容传播评价值对用户账号进行排序，选取处于所述排序后的用户账号的前列的投放用户数的用户账号推送内容数据。
根据本发明的一个实施例，根据图1及图3至5所示的内容投放方法可以是由图6中所示的内容投放装置中的各个单元来执行。例如，图1所示的步骤S102、S104、S106和S108可分别由图6所示的初始化模块102、好友查找模块104、内容传播评价值更新模块106和数据内容推送模块108来执行；图3中所示的步骤S202、S204和S106可由图6所示的内容传播评价值更新模块106来执行；图4中所示的步骤S302、S304、S306和S308可由图6所示的内容传播评价值修正模块114来执行；图5中所示的步骤S402、S404和S406可由图6所示的内容数据推送模块108来执行。
根据本发明的另一个实施例，图6所示的内容投放装置中的各个单元可以分别或全部合并为一个或若干个另外的单元来构成，或者其中的某个(些)单元还可以再拆分为功能上更小的多个单元来构成，这可以实现同样的操作，而不影响本发明的实施例的技术效果的实现。上述单元是基于逻辑功能划分的，在实际应用中，一个单元的功能也可以由多个单元来实现，或者多个单元的功能由一个单元实现。在本发明的其它实施例中，终端设备也可以包括其它模块。但在实际应用中，这些功能也可以由其它单元协助实现，并且可以由多个单元协作实现。
根据本发明的另一个实施例，可以通过在包括中央处理单元(CPU)、随 机存取存储器(RAM)、只读存储器(ROM)等处理元件和存储元件的例如计算机的通用计算设备上运行能够执行如图1及图3至图5所示的内容投放方法的计算机程序(包括程序代码)，来构造如图6所示的内容投放装置，以及来实现根据本发明的实施例的内容投放方法。所述计算机程序可以记载于例如计算机可读记录介质上，并通过计算机可读记录介质装载于上述计算设备中，并在其中运行。
上述内容投放方法及装置利用了社交网络应用中的用户账号之间的关联与转发操作，只需将内容数据推送给小部分的用户账号，然后通过依次转发将内容数据传递给该小部分的用户账号上的社交关系链上的其他用户账号。将内容数据再由该小部分的用户账号扩散至较广的范围。同时，用户之间的关联程度也保证了用户点击或查看接收到的该用户关注的用户转发的内容数据的概率，从而在保证了内容数据的投放精度的和广度。
在一个实施例中，如图7所示，图7展示了一种运行上述第一种或第二种内容投放方法的基于冯诺依曼体系的计算机系统10的服务器设备。该计算机系统可以是直接执行上述内容投放方法的服务器、服务器集群或者提供虚拟机作为服务器程序运行环境执行上述内容投放方法的服务器集群设备。具体的，可包括通过系统总线连接的网络接口1001、处理器1002、存储器1003。存储器1003可包括外存储器10032(例如硬盘、光盘或软盘等)和内存储器10034。
在本实施例中，本方法的运行基于计算机程序，该计算机程序的程序文件存储于前述基于冯诺依曼体系的计算机系统10的外存储器10032中，在运行时被加载到内存储器10034中，然后被编译为机器码之后传递至处理器1002中执行，从而使得基于冯诺依曼体系的计算机系统10中形成逻辑上的初始化模块102、好友查找模块104、内容传播评价值更新模块106、内容数据推送模块108。且在上述内容投放方法执行过程中，输入的参数均通过网络接口1001接收，并传递至存储器1003中缓存，然后输入到处理器1002中进行处理，处理的结果数据或缓存于存储器1003中进行后续地处理，或被传递至网络接口1001进行输出。
其中，所述存储器1003中存储一组程序代码，且所述处理器1002调用所述存储器1003中存储的程序代码，用于执行以下操作：
遍历用户账号，生成遍历到的第一用户的初始的内容传播评价值；
遍历用户账号，查找所述遍历到的第二用户账号的好友账号；
计算所述第二用户账号的好友账号的内容传播评价值的加权和，根据所述加权和与所述第二用户账号的第一转发概率的和更新所述第二用户账号的内容传播评价值；及
在各个所述第二用户账号的内容传播评价值收敛时，跳出迭代，根据所述内容传播评价值选择用户账号推送内容数据。
在另一实施例中，所述处理器1002调用所述存储器1003中存储的程序代码执行所述生成遍历到的第一用户的初始的内容传播评价值的操作，包括：
计算所述遍历到的第一用户账号的第一转发概率，将所述第一转发概率设置为所述遍历到的第一用户账号的初始的内容传播评价值。
在另一实施例中，所述处理器1002调用所述存储器1003中存储的程序代码执行所述计算所述第二用户账号的好友账号的内容传播评价值的加权和的操作，还可以包括：
遍历所述第二用户账号的好友账号，计算所述遍历到的好友账号对所述第二用户账号的第二转发概率；
将所述第二转发概率设为所述遍历所述第二用户账号的好友账号的内容传播评价值的权重系数；及
根据所述第二用户账号的好友账号的内容传播评价值的权重系数计算所述第二用户账号的好友账号的内容传播评价值的加权和。
在另一实施例中，所述处理器1002调用所述存储器1003中存储的程序代码执行所述计算所述遍历到的好友账号对所述第二用户账号的第二转发概率的操作，还可以包括：
获取所述第二用户账号发布内容的发布次数；
获取所述遍历到的好友账号转发所述第二用户账号发布的内容的转发次数；及
将所述转发次数除以所述发布次数计算所述遍历到的好友账号对所述第二用户账号的第二转发概率。
在一可选实施例中，处理器1002调用存储器1003中存储的程序代码，还 用于执行以下操作：
获取更新各个所述第二用户账号的内容传播评价值时的变化值，在各个所述变化值小于阈值时，判定各个所述第二用户账号的内容传播评价值收敛。
在一可选实施例中，处理器1002调用存储器1003中存储的程序代码，还用于执行以下操作：
获取执行所述遍历所述已计算得到转发概率的第二用户账号的遍历次数，在所述遍历次数大于或等于阈值时，判定所述各个所述第二用户账号的内容传播评价值收敛。
在另一实施例中，所述处理器1002调用所述存储器1003中存储的程序代码执行所述根据所述内容传播评价值选择用户账号推送内容数据的操作之前，还可以包括：
遍历用户账号，获取所述遍历到的第三用户账号的内容传播评价值，查找内容传播评价值大于所述第三用户账号的内容传播评价值的用户账号的集合；
查找所述第三用户账号在社交关系链上与所述查找到的用户账号之间相隔的用户账号数的最小值；
根据预设的递增函数以所述最小值为自变量计算修正系数；
根据所述修正系数对所述第三用户账号的内容传播评价值进行更新。
在另一实施例中，所述处理器1002调用所述存储器1003中存储的程序代码执行所述根据所述内容传播评价值选择用户账号推送内容数据的操作，还可以包括：
遍历用户账号，获取所述遍历到的第四用户账号对所述内容数据的点击概率；
计算所述第四用户账号的内容传播评价值和所述第四用户账号对所述内容数据的点击概率的加权和；
根据所述加权和选择用户账号推送内容数据。
在另一实施例中，所述处理器1002调用所述存储器1003中存储的程序代码执行所述根据所述内容传播评价值选择用户账号推送内容数据的操作，还可以包括：
获取预设的投放用户数；
根据内容传播评价值对用户账号进行排序，选取处于所述排序后的用户账号的前列的投放用户数的用户账号推送内容数据。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。就本说明书而言，″计算机可读介质″可以是任何可以包含、存储、通信、传播或传输程序以供指令执行系统、装置或设备或结合这些指令执行系统、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本发明的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。例如，如果用硬件来实现，和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
上述提到的存储介质可以是只读存储器，磁盘或光盘等。尽管上面已经示出和描述了本发明的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本发明的限制，本领域的普通技术人员在本发明的范围内可以对上述实施例进行变化、修改、替换和变型。

Claims (19)

  1. 一种内容投放方法，其特征在于，包括：
    遍历用户账号，生成遍历到的第一用户的初始的内容传播评价值；
    迭代执行：
    遍历用户账号，查找所述遍历到的第二用户账号的好友账号；
    计算所述第二用户账号的好友账号的内容传播评价值的加权和，根据所述加权和与所述第二用户账号的第一转发概率的和更新所述第二用户账号的内容传播评价值；
    在各个所述第二用户账号的内容传播评价值收敛时，跳出迭代，根据所述内容传播评价值选择用户账号推送内容数据。
  2. 根据权利要求1所述的内容投放方法，其特征在于，所述生成遍历到的第一用户的初始的内容传播评价值的步骤为
    计算所述遍历到的第一用户账号的第一转发概率，将所述第一转发概率设置为所述遍历到的第一用户账号的初始的内容传播评价值。
  3. 根据权利要求1所述的内容投放方法，其特征在于，所述计算所述第二用户账号的好友账号的内容传播评价值的加权和的步骤为：
    遍历所述第二用户账号的好友账号，计算所述遍历到的好友账号对所述第二用户账号的第二转发概率；
    将所述第二转发概率设为所述遍历所述第二用户账号的好友账号的内容传播评价值的权重系数；
    根据所述第二用户账号的好友账号的内容传播评价值的权重系数计算所述第二用户账号的好友账号的内容传播评价值的加权和。
  4. 根据权利要求3所述的内容投放方法，其特征在于，所述计算所述遍历到的好友账号对所述第二用户账号的第二转发概率的步骤为：
    获取所述第二用户账号发布内容的发布次数；
    获取所述遍历到的好友账号转发所述第二用户账号发布的内容的转发次数；
    将所述转发次数除以所述发布次数计算所述遍历到的好友账号对所述第二用户账号的第二转发概率。
  5. 根据权利要求1所述的内容投放方法，其特征在于，所述方法还包括：
    获取更新各个所述第二用户账号的内容传播评价值时的变化值，在各个所述变化值小于阈值时，判定各个所述第二用户账号的内容传播评价值收敛。
  6. 根据权利要求1所述的内容投放方法，其特征在于，所述方法还包括：
    获取执行所述遍历所述已计算得到转发概率的第二用户账号的遍历次数，在所述遍历次数大于或等于阈值时，判定所述各个所述第二用户账号的内容传播评价值收敛。
  7. 根据权利要求1至6任一项所述的内容投放方法，其特征在于，根据所述内容传播评价值选择用户账号推送内容数据的步骤之前还包括：
    遍历用户账号，获取所述遍历到的第三用户账号的内容传播评价值，查找内容传播评价值大于所述第三用户账号的内容传播评价值的用户账号的集合；
    查找所述第三用户账号在社交关系链上与所述查找到的用户账号之间相隔的用户账号数的最小值；
    根据预设的递增函数以所述最小值为自变量计算修正系数；
    根据所述修正系数对所述第三用户账号的内容传播评价值进行更新。
  8. 根据权利要求1至6任一项所述的内容投放方法，其特征在于，所述根据所述内容传播评价值选择用户账号推送内容数据的步骤为：
    遍历用户账号，获取所述遍历到的第四用户账号对所述内容数据的点击概率；
    计算所述第四用户账号的内容传播评价值和所述第四用户账号对所述内容数据的点击概率的加权和；
    根据所述加权和选择用户账号推送内容数据。
  9. 根据权利要求1至6任一项所述的内容投放方法，其特征在于，所述根据所述内容传播评价值选择用户账号推送内容数据的步骤为：
    获取预设的投放用户数；
    根据内容传播评价值对用户账号进行排序，选取处于所述排序后的用户账号的前列的投放用户数的用户账号推送内容数据。
  10. 一种内容投放装置，其特征在于，包括：
    初始化模块，用于遍历用户账号，生成遍历到的第一用户账号的初始的内容传播评价值；
    好友查找模块，用于遍历用户账号，查找所述遍历到的第二用户账号的好友账号；
    内容传播评价值更新模块，用于计算所述第二用户账号的好友账号的内容传播评价值的加权和，根据所述加权和与所述第二用户账号的第一转发概率的和更新所述第二用户账号的内容传播评价值；
    内容数据推送模块，用于在各个所述第二用户账号的内容传播评价值收敛时，跳出迭代，根据所述内容传播评价值选择用户账号推送内容数据；在不收敛时，调用所述好友查找模块和内容传播评价值更新模块进行迭代。
  11. 根据权利要求10所述的内容投放装置，其特征在于，所述初始化模块用于计算所述遍历到的第一用户账号的第一转发概率，将所述第一转发概率设置为所述遍历到的第一用户账号的初始的内容传播评价值。
  12. 根据权利要求10所述的内容投放装置，其特征在于，所述内容传播评价值更新模块还用于遍历所述第二用户账号的好友账号，计算所述遍历到的好友账号对所述第二用户账号的第二转发概率；将所述第二转发概率设为所述遍历所述第二用户账号的好友账号的内容传播评价值的权重系数；根据所述第二用户账号的好友账号的内容传播评价值的权重系数计算所述第二用户账号 的好友账号的内容传播评价值的加权和。
  13. 根据权利要求12所述的内容投放装置，其特征在于，所述内容传播评价值更新模块还用于获取所述第二用户账号发布内容的发布次数；获取所述遍历到的好友账号转发所述第二用户账号发布的内容的转发次数；将所述转发次数除以所述发布次数计算所述遍历到的好友账号对所述第二用户账号的第二转发概率。
  14. 根据权利要求10所述的内容投放装置，其特征在于，所述装置还包括第一收敛判定模块，用于获取更新各个所述第二用户账号的内容传播评价值时的变化值，在各个所述变化值小于阈值时，判定各个所述第二用户账号的内容传播评价值收敛。
  15. 根据权利要求10所述的内容投放装置，其特征在于，所述装置还包括第二收敛判定模块，用于获取执行所述遍历所述已计算得到转发概率的第二用户账号的遍历次数，在所述遍历次数大于或等于阈值时，判定所述各个所述第二用户账号的内容传播评价值收敛。
  16. 根据权利要求10至15任一项所述的内容投放装置，其特征在于，所述装置还包括内容传播评价值修正模块，用于遍历用户账号，获取所述遍历到的第三用户账号的内容传播评价值，查找内容传播评价值大于所述第三用户账号的内容传播评价值的用户账号的集合；查找所述第三用户账号在社交关系链上与所述查找到的用户账号之间相隔的用户账号数的最小值；根据预设的递增函数以所述最小值为自变量计算修正系数；根据所述修正系数对所述第三用户账号的内容传播评价值进行更新。
  17. 根据权利要求10至15任一项所述的内容投放装置，其特征在于，所述内容数据推送模块还用于遍历用户账号，获取所述遍历到的第四用户账号对所述内容数据的点击概率；计算所述第四用户账号的内容传播评价值和所述第 四用户账号对所述内容数据的点击概率的加权和；根据所述加权和选择用户账号推送内容数据。
  18. 根据权利要求10至15任一项所述的内容投放装置，其特征在于，所述内容数据推送模块还用于获取预设的投放用户数；根据内容传播评价值对用户账号进行排序，选取处于所述排序后的用户账号的前列的投放用户数的用户账号推送内容数据。
  19. 一种内容投放装置，其特征在于，包括：至少一个处理器及连接于所述至少一个处理器的存储器，所述处理器调用所述存储器中存储的程序代码用于执行根据权利要求1至9任一项所述的内容投放方法的操作指令。
PCT/CN2016/082502 2015-09-17 2016-05-18 内容投放方法及装置 WO2017045415A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2018514335A JP6661754B2 (ja) 2015-09-17 2016-05-18 コンテンツ配信方法および装置
EP16845529.3A EP3352121A1 (en) 2015-09-17 2016-05-18 Content delivery method and device
US15/842,275 US10621516B2 (en) 2015-09-17 2017-12-14 Content delivery method, apparatus, and storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510593480.1A CN105262794B (zh) 2015-09-17 2015-09-17 内容投放方法及装置
CN201510593480.1 2015-09-17

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/842,275 Continuation US10621516B2 (en) 2015-09-17 2017-12-14 Content delivery method, apparatus, and storage medium

Publications (1)

Publication Number Publication Date
WO2017045415A1 true WO2017045415A1 (zh) 2017-03-23

Family

ID=55102298

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/082502 WO2017045415A1 (zh) 2015-09-17 2016-05-18 内容投放方法及装置

Country Status (5)

Country Link
US (1) US10621516B2 (zh)
EP (1) EP3352121A1 (zh)
JP (1) JP6661754B2 (zh)
CN (1) CN105262794B (zh)
WO (1) WO2017045415A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105262794B (zh) * 2015-09-17 2018-08-17 腾讯科技（深圳）有限公司 内容投放方法及装置
CN106202168A (zh) * 2016-06-24 2016-12-07 武汉斗鱼网络科技有限公司 一种网站内容展示方法与装置
CN108322317B (zh) * 2017-01-16 2022-07-29 腾讯科技（深圳）有限公司 一种账号识别关联方法及服务器
CN107104883B (zh) * 2017-04-21 2019-05-03 腾讯科技（深圳）有限公司 一种社交关系链中的信息分享方法、客户端及服务器
CN110287688B (zh) * 2019-06-28 2020-11-24 京东数字科技控股有限公司 关联账号分析方法、装置和计算机可读存储介质
CN110675197B (zh) * 2019-09-30 2023-09-26 北京达佳互联信息技术有限公司 评估数据的方法、装置、设备及存储介质
JP7373249B1 (ja) * 2023-05-24 2023-11-02 株式会社PocketRD コンテンツ評価システム、コンテンツ評価方法及びコンテンツ評価プログラム

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101520878A (zh) * 2009-04-03 2009-09-02 华为技术有限公司 向用户进行广告推送的方法、装置和系统
WO2009154430A2 (en) * 2008-06-19 2009-12-23 Samsung Electronics Co., Ltd. System and method for managing shared/forwarded advertisement
US20100100415A1 (en) * 2008-10-17 2010-04-22 Yahoo!,Inc. Common tag format for ad serving and information tracking in internet advertising
CN102110265A (zh) * 2009-12-23 2011-06-29 深圳市腾讯计算机系统有限公司 一种预估网络广告效果的方法及系统
CN103294800A (zh) * 2013-05-27 2013-09-11 华为技术有限公司 一种信息推送方法及装置
CN104902294A (zh) * 2015-06-18 2015-09-09 北京酷云互动科技有限公司 一种用户特征信息的形成方法及其形成设备
CN105262794A (zh) * 2015-09-17 2016-01-20 腾讯科技（深圳）有限公司 内容投放方法及装置

Family Cites Families (48)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5694546A (en) * 1994-05-31 1997-12-02 Reisman; Richard R. System for automatic unattended electronic information transport between a server and a client by a vendor provided transport software with a manifest list
US5696965A (en) * 1994-11-03 1997-12-09 Intel Corporation Electronic information appraisal agent
US6006218A (en) * 1997-02-28 1999-12-21 Microsoft Methods and apparatus for retrieving and/or processing retrieved information as a function of a user's estimated knowledge
US6981040B1 (en) * 1999-12-28 2005-12-27 Utopy, Inc. Automatic, personalized online information and product services
US20040122735A1 (en) * 2002-10-09 2004-06-24 Bang Technologies, Llc System, method and apparatus for an integrated marketing vehicle platform
SE0300555D0 (sv) * 2003-02-24 2003-02-24 Ericsson Telefon Ab L M Improvements in or relating to push-to-talk services
US20060053047A1 (en) * 2004-08-30 2006-03-09 Garcia Rita M System and method for selecting targets for sales and marketing campaigns
US20080134043A1 (en) * 2006-05-26 2008-06-05 Sony Corporation System and method of selective media content access through a recommednation engine
US8438062B2 (en) * 2006-12-29 2013-05-07 Google Inc. Network node ad targeting
US20080281711A1 (en) * 2007-05-11 2008-11-13 Bridges Thomas L System and method for displaying advertisements according to business and consumer relevance
CN101136837A (zh) * 2007-09-21 2008-03-05 华为技术有限公司 推送消息的控制方法、装置和系统
US20100088130A1 (en) * 2008-10-07 2010-04-08 Yahoo! Inc. Discovering Leaders in a Social Network
JP2010128882A (ja) * 2008-11-28 2010-06-10 Kddi Corp コンテンツ配信方法およびシステム
JP5148525B2 (ja) * 2009-02-18 2013-02-20 Ｋｄｄｉ株式会社 コンテンツ配信方法およびシステム
BRPI0924542A2 (pt) * 2009-03-03 2015-06-30 Google Inc Método e sistema para fornecimento de publicidade para usuários de rede social
CN101505311B (zh) * 2009-03-18 2012-06-13 腾讯科技（深圳）有限公司 一种基于社会化网络的信息传播方法及系统
US8301132B1 (en) * 2009-03-31 2012-10-30 Sprint Communications Company L.P. Push registry application programming interface for capturing events
US20120036080A1 (en) * 2009-04-06 2012-02-09 Tracx Systems Ltd. Method and system for tracking online social interactions
KR101048572B1 (ko) * 2009-05-07 2011-07-11 한양대학교 산학협력단 수동형 가입자 네트워크에서 푸시형 멀티캐스트의 인터넷 그룹 관리 시스템 및 그 방법
US20100306249A1 (en) * 2009-05-27 2010-12-02 James Hill Social network systems and methods
US20110015987A1 (en) * 2009-07-20 2011-01-20 International Business Machines Corporation Systems and methods for marketing to mobile devices
US20110314017A1 (en) * 2010-06-18 2011-12-22 Microsoft Corporation Techniques to automatically manage social connections
US8706750B2 (en) * 2010-08-19 2014-04-22 Google Inc. Predictive query completion and predictive search results
KR101187742B1 (ko) * 2010-12-09 2012-10-05 주식회사 이너스커뮤니티 소셜 네트워크 서비스 장치 및 그 서비스 방법
JP2012150652A (ja) * 2011-01-19 2012-08-09 Kddi Corp インフルエンサー抽出装置、インフルエンサー抽出方法およびプログラム
US20120245980A1 (en) * 2011-03-22 2012-09-27 Accenture Global Services Limited Productivity prediction technique and system
US9037700B2 (en) * 2011-04-29 2015-05-19 International Business Machines Corporation Predictive placement of content through network analysis
US20130138662A1 (en) * 2011-11-29 2013-05-30 Dr. ZHIJIANG HE Method for assigning user-centric ranks to database entries within the context of social networking
WO2013082501A1 (en) * 2011-12-02 2013-06-06 Notice Software, Llc Systems and methods for facilitating push notification
US9378287B2 (en) * 2011-12-14 2016-06-28 Patrick Frey Enhanced search system and method based on entity ranking
CN103200073B (zh) * 2012-01-04 2016-08-10 微梦创科网络科技（中国）有限公司 一种信息处理方法和装置
CN103248639B (zh) * 2012-02-06 2016-11-23 阿里巴巴集团控股有限公司 一种用于确定信息传播能力的方法及系统
US20140066044A1 (en) * 2012-02-21 2014-03-06 Manoj Ramnani Crowd-sourced contact information and updating system using artificial intelligence
US9922344B1 (en) * 2012-02-24 2018-03-20 Google Llc Serving advertisements based on partial queries
CN103294730B (zh) * 2012-03-05 2017-03-01 富士通株式会社 一种数据推送系统及方法
US20130253940A1 (en) * 2012-03-20 2013-09-26 Zilla Collections Llc System and method for diagnosis involving crowdsourcing
US20130254213A1 (en) * 2012-03-26 2013-09-26 Heyning Cheng Techniques for identifying and presenting connection paths
US20130275320A1 (en) * 2012-04-16 2013-10-17 Google Inc. Identifying Prospective Employee Candidates via Employee Connections
US9460237B2 (en) * 2012-05-08 2016-10-04 24/7 Customer, Inc. Predictive 411
US10380629B2 (en) * 2012-05-25 2019-08-13 Microsoft Technology Licensing, Llc Leveraging a social graph to deliver relevant recommendations
CA2893960C (en) * 2012-12-05 2020-09-15 Grapevine6 Inc. System and method for finding and prioritizing content based on user specific interest profiles
US9497234B2 (en) * 2012-12-27 2016-11-15 Facebook, Inc. Implicit social graph connections
CN103914468B (zh) * 2012-12-31 2018-01-09 阿里巴巴集团控股有限公司 一种投放信息搜索的方法和装置
CN103179025B (zh) * 2013-03-20 2016-05-11 微梦创科网络科技（中国）有限公司 一种基于用户传播力的微博推送方法及装置
US20140372203A1 (en) * 2013-06-14 2014-12-18 Microsoft Corporation Quality-weighted second-price auctions for advertisements
US20150019588A1 (en) * 2013-07-11 2015-01-15 Drexel University Identifying Implicit Relationships Between Social Media Users To Support Social Commerce
US9898743B2 (en) * 2013-10-09 2018-02-20 Salesforce.Com, Inc. Systems and methods for automatic generation of a relationship management system
US10068008B2 (en) * 2014-08-28 2018-09-04 Microsoft Technologies Licensing, LLC Spelling correction of email queries

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009154430A2 (en) * 2008-06-19 2009-12-23 Samsung Electronics Co., Ltd. System and method for managing shared/forwarded advertisement
US20100100415A1 (en) * 2008-10-17 2010-04-22 Yahoo!,Inc. Common tag format for ad serving and information tracking in internet advertising
CN101520878A (zh) * 2009-04-03 2009-09-02 华为技术有限公司 向用户进行广告推送的方法、装置和系统
CN102110265A (zh) * 2009-12-23 2011-06-29 深圳市腾讯计算机系统有限公司 一种预估网络广告效果的方法及系统
CN103294800A (zh) * 2013-05-27 2013-09-11 华为技术有限公司 一种信息推送方法及装置
CN104902294A (zh) * 2015-06-18 2015-09-09 北京酷云互动科技有限公司 一种用户特征信息的形成方法及其形成设备
CN105262794A (zh) * 2015-09-17 2016-01-20 腾讯科技（深圳）有限公司 内容投放方法及装置

Also Published As

Publication number Publication date
CN105262794A (zh) 2016-01-20
JP6661754B2 (ja) 2020-03-11
EP3352121A4 (en) 2018-07-25
JP2018530827A (ja) 2018-10-18
CN105262794B (zh) 2018-08-17
EP3352121A1 (en) 2018-07-25
US10621516B2 (en) 2020-04-14
US20180107953A1 (en) 2018-04-19

Similar Documents

Publication Publication Date Title
WO2017045415A1 (zh) 内容投放方法及装置
US10789311B2 (en) Method and device for selecting data content to be pushed to terminal, and non-transitory computer storage medium
US10552488B2 (en) Social media user recommendation system and method
WO2020073673A1 (zh) 一种文本分析方法及终端
US9594826B2 (en) Co-selected image classification
US20200412675A1 (en) Network based data traffic latency reduction
US9098819B1 (en) Identifying social network accounts belonging to the same user
US20160321281A1 (en) System and method for measuring the virality of internet content using indices
JP2019532445A (ja) 多義コードを使用した類似度検索
US10249070B2 (en) Dynamic interaction graphs with probabilistic edge decay
US10269048B2 (en) Xpath related and other techniques for use in native advertisement placement
WO2016101811A1 (zh) 一种信息排序方法及装置
TW201928724A (zh) 使用深度神經語言模型於自動出價之系統與方法
US20220237220A1 (en) Template generation using directed acyclic word graphs
TW201316191A (zh) 資訊查詢方法和裝置
US20120246219A1 (en) Shared cache for potentially repetitive message data in a publish-subscription environment
US20140365571A1 (en) Automatically Determining Veracity of Documents Posted in a Public Forum
Liang et al. Analyzing user influence of microblog
US20160140454A1 (en) User Interest Learning through Hierarchical Interest Graphs
CN111723201A (zh) 一种用于文本数据聚类的方法和装置
CN110019671B (zh) 一种处理实时消息的方法和系统
CN112015924A (zh) 流媒体缓存方法及其装置、计算机可存储介质
CN111858917A (zh) 文本分类方法和装置
CN114449031B (zh) 信息获取方法、装置、设备和存储介质
CN113051483A (zh) 一种帖子热度的计算方法和装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16845529

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2018514335

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2016845529

Country of ref document: EP