WO2017211191A1 - 信息推送方法及装置 - Google Patents

信息推送方法及装置 Download PDF

Info

Publication number
WO2017211191A1
WO2017211191A1 PCT/CN2017/085653 CN2017085653W WO2017211191A1 WO 2017211191 A1 WO2017211191 A1 WO 2017211191A1 CN 2017085653 W CN2017085653 W CN 2017085653W WO 2017211191 A1 WO2017211191 A1 WO 2017211191A1
Authority
WO
WIPO (PCT)
Prior art keywords
users
relationship
preset
user
information
Prior art date
Application number
PCT/CN2017/085653
Other languages
English (en)
French (fr)
Inventor
宋念
Original Assignee
阿里巴巴集团控股有限公司
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 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Priority to MYPI2018002358A priority Critical patent/MY194098A/en
Priority to KR1020187037819A priority patent/KR20190015373A/ko
Priority to JP2018563865A priority patent/JP6991163B2/ja
Priority to EP17809630.1A priority patent/EP3467754A4/en
Priority to SG11201810809XA priority patent/SG11201810809XA/en
Publication of WO2017211191A1 publication Critical patent/WO2017211191A1/zh
Priority to US16/210,988 priority patent/US20190108552A1/en
Priority to PH12018502583A priority patent/PH12018502583A1/en
Priority to US16/721,857 priority patent/US11074623B2/en
Priority to US17/382,130 priority patent/US20210357985A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0269—Targeted advertisements based on user profile or attribute
    • 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
    • G06F16/906—Clustering; Classification
    • 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
    • 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
    • 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
    • G06Q30/0242—Determining effectiveness of advertisements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0277—Online advertisement
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • 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
    • 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
    • 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
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • H04L67/63—Routing a service request depending on the request content or context

Definitions

  • the present invention relates to the field of information technology, and in particular, to an information push method and apparatus.
  • the user who directly pushes the preset information is directly filtered according to the preset scoring model, wherein the preset scoring model is generated according to the user result and the user attribute acquired by the application history. That is, the potential users are scored according to the preset scoring model, and then the users with the highest scores to be pushed the preset information are selected from the potential users.
  • the preset scoring model directly filtering the user to push the preset information to push the application information, only to ensure that the filtered user itself is converted into an application user for information push, and cannot obtain more applications. Users, which can not magnify the application advertising budget cost benefits, resulting in less efficient information push.
  • the embodiment of the present invention provides an information pushing method and device, and the main purpose thereof is to solve the problem that the information pushing efficiency is low.
  • the present invention provides the following technical solutions:
  • An embodiment of the present invention provides an information pushing method, including:
  • An embodiment of the present invention provides an information pushing apparatus, including:
  • An obtaining unit configured to acquire multiple users that meet preset conditions
  • a screening unit configured to filter, according to the relationship strength information between the plurality of users acquired by the acquiring unit, and the plurality of user influence strength information, the preset information to be pushed from the plurality of users user;
  • a pushing unit configured to perform information pushing on the user to be pushed by the screening unit to push the preset information.
  • An embodiment of the present invention provides a method for determining a target user, including:
  • the user whose influence strength information meets the preset influence strength condition is determined as the target user.
  • An embodiment of the present invention provides a target user determining apparatus, including:
  • An obtaining unit configured to obtain a plurality of users that meet preset conditions
  • a screening unit configured to filter, according to the influence strength information corresponding to the plurality of users acquired by the acquiring unit, a user whose influence strength information meets a preset influence strength condition from the plurality of users;
  • a determining unit configured to determine, by the user whose influence strength information filtered by the screening unit is in accordance with a preset influence strength condition, as a target user.
  • the technical solution provided by the embodiment of the present invention has at least the following advantages:
  • the information pushing method and device provided by the embodiment of the present invention firstly acquire a plurality of users that meet the preset condition; and according to the relationship strength information between the multiple users, and the influence strength information of the multiple users, The user who wants to push the preset information is filtered out from the plurality of users; and finally, the user who pushes the preset information is pushed by the information.
  • the embodiment of the present invention filters the plurality of users whose scores meet the preset conditions by using the preset scoring model, and then according to the relationship strength information between the multiple users.
  • the influence strength information of the plurality of users, and the user who selects the preset information to be pushed from the plurality of users can additionally convert the filtered user into an application user who performs information push. Get more application users, which can magnify the cost of application advertising budgets, which can improve the efficiency of information push.
  • FIG. 1 is a flowchart of an information pushing method according to an embodiment of the present invention
  • FIG. 2 is a flowchart of another information pushing method according to an embodiment of the present invention.
  • FIG. 3 is a flowchart of a method for determining a target user according to an embodiment of the present invention
  • FIG. 4 is a flowchart of another method for determining a target user according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of an information pushing apparatus according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of another information pushing apparatus according to an embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of a target user determining apparatus according to an embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of another target user determining apparatus according to an embodiment of the present invention.
  • FIG. 9 is a flowchart of a data processing method according to an embodiment of the present invention.
  • FIG. 10 is a flowchart of a method for generating a preset scoring model according to an embodiment of the present invention.
  • FIG. 11 is a flowchart of user screening of preset information to be pushed according to an embodiment of the present invention.
  • An embodiment of the present invention provides an information pushing method. As shown in FIG. 1 , the method includes:
  • the preset condition may be set according to the requirements of the user, or may be set according to the system requirements, and the examples of the present invention are not limited.
  • the plurality of users may specifically be a plurality of users whose scores meet the preset score condition. Specifically, the user whose score meets the preset score condition can be filtered by the preset scoring model.
  • the preset scoring model is generated according to user results and user attributes acquired by the application history.
  • the user results obtained by the application history may include potential users acquired by the application history and users converted to application users.
  • the preset score condition may be set according to the requirements of the user, or may be set according to the system requirements, and the examples of the present invention are not limited. For example, the preset score condition may be 750 users with the highest score among potential users, or users with scores greater than or 0.8 points among potential users.
  • the information obtained by the information push method pushes the attribute data of the sample user and the sample user, and then processes the attribute data of the sample user to obtain the feature data for generating the preset scoring model, wherein the feature data of the preset scoring model is required to be complete and lacking.
  • the data format conforms to the standard generated by the scoring model.
  • the information pushing manner of the unspecified target user is to deliver and display the application advertisement on various channels, and the advertisement delivery user is not explicitly defined.
  • the attribute data of the sample user and the sample user are collected by the program embedded in the channel, there are usually multiple advertisements of the application or product on the same channel, and the collected sample users may have other application advertisement users; or the sample Some data in the user's attribute data is not the feature data of the generated scoring model. Therefore, the attribute data of the sample user and the sample user needs to be cleaned and processed.
  • the collected user asset data is the asset data of the user in different asset accounts.
  • the asset data that generates the scoring model is the total asset data of the user, so the total asset data of the user needs to be obtained according to the asset data of the user in different asset accounts.
  • Feature transformation Since the model algorithm in the scoring model algorithm only supports sparse matrices and does not support ordinary two-dimensional tables, it is necessary to format the attribute data to obtain the feature data of the scoring model. If the format of the attribute data can be converted by logarithmic transformation, the feature data of the generated scoring model is obtained.
  • the scoring model generation algorithm may be a decision tree algorithm, a logistic regression algorithm, a random forest algorithm, an iterative decision tree algorithm (GBDT, Gradient Boosting Decision Tree), or the like.
  • Model training According to the feature data of the generated scoring model and the scoring model generating algorithm, the training generates a plurality of scoring models.
  • the scoring effects of the respective scoring models are obtained according to the parameters in the scoring model generation algorithm and the different combinations of the feature data. Specifically, the scoring effect of the scoring model can be measured according to the model scoring effect metric.
  • the metric of the model scoring effect may be the area under the curve, and the larger the area under the curve, the better the scoring effect of the scoring model.
  • the scoring model with the best scoring effect is selected from the plurality of scoring models, and the scoring model with the best scoring effect is determined as the preset scoring model of the user whose screening score meets the preset condition.
  • Potential user ratings The potential users are scored according to the preset scoring model, and then the users whose scores meet the preset conditions are screened out from the potential users.
  • the relationship strength information between the multiple users may be a similarity between multiple user attribute information.
  • the user's influence strength refers to the strength of the user's ability to influence other users. Specifically, the number of times the information posted by the user is read, shared, and forwarded by other users.
  • the preset information to be pushed may be advertisement information that the application provider pushes according to the cost of the advertisement budget.
  • advertisement information that the application provider pushes according to the cost of the advertisement budget.
  • many application providers usually budget the cost of advertising, that is, to push the user data according to the cost of the advertising budget, and only push the information to the user who pushes the user data.
  • the application's corresponding advertising cost is 5 million
  • the cost per user is 1 yuan
  • the number of users who want to push the preset information is 5 million.
  • the preset score model is used to select the highest score of 7.5 million users. Then, according to the relationship strength information between the users and the influence strength information of the user, the points filtered by the preset scoring model Among the top 7.5 million users, 500 users who want to push preset information are selected.
  • a plurality of users whose scores meet the preset conditions are first selected by using a preset scoring model, and then according to the relationship strength information between the plurality of users, and the influence strength information of the plurality of users,
  • the user who selects the preset information to be pushed from the plurality of users can ensure that the user who wants to push the preset information converts itself into an application user who pushes the information, and can utilize the influence and communication ability of the user and share , forwarding, or word of mouth, etc., to obtain more users for the application, so as to be able to amplify the cost of the application advertising budget, thereby improving the efficiency of information push.
  • the user who pushes the preset information performs information push.
  • An information pushing method provided by an embodiment of the present invention firstly acquires a plurality of users that meet a preset condition; and according to the relationship strength information between the multiple users, and the influence strength information of the multiple users, The user who wants to push the preset information is filtered out among the plurality of users; and finally, the user who pushes the preset information is pushed by the information.
  • the embodiment of the present invention filters the plurality of users whose scores meet the preset conditions by using the preset scoring model, and then according to the relationship strength information between the multiple users.
  • the influence strength information of the plurality of users, and the user who selects the preset information to be pushed from the plurality of users can additionally convert the filtered user into an application user who performs information push. Get more application users, which can magnify the cost of application advertising budgets, which can improve the efficiency of information push.
  • An embodiment of the present invention provides another information pushing method. As shown in FIG. 2, the method includes:
  • the relationship strength information between users stored in each relationship community is greater than a preset threshold.
  • the relationship community may be a relationship network community, specifically an online communication space such as a forum, a bar, a bulletin board, a group discussion, an online chat, a friend, a personal space, a wireless value-added service, and the like, and a user in the same relationship community. Have the same relationship community tag.
  • the preset threshold may be set according to the requirements of the user, or may be set according to the default mode of the system, which is not limited in the embodiment of the present invention. For example, the preset threshold may be 0.4, 0.5, or the like.
  • the method further includes: acquiring an initial relationship community label of the multiple users.
  • the step 202 may specifically include: updating an initial relationship community label of each of the multiple users to a relationship community label of a user whose relationship strength information is greater than a preset threshold; Users with the same relationship community label in the user are divided into the same relationship community.
  • the plurality of users filtered by the preset scoring model include User 1, User 2, User 3, and User 4.
  • the relationship strength between the user 1 and the user 2 is 0.1
  • the relationship strength between the user 1 and the user 3 is 0.4
  • the relationship strength between the user 1 and the user 4 is 0.2
  • the relationship strength between the user 1 and the user 3 is greater than or equal to 0.4
  • the user 1's relationship community tag is updated to the user 3's relationship community tag.
  • the relationship community tags of User 2, User 3, and User 4 can be updated in the same manner.
  • the process of dividing the relationship community is an iterative process.
  • the initial relationship community label of the multiple users is obtained, and the initial relationship community label of each of the multiple users may be The identity number (Identity, ID) of the user.
  • each user's relationship community tag is an updated relationship community tag based on the relationship strength information between the users.
  • the method further includes: acquiring the number of relationship community label updates of the user; determining the user Whether the number of relationship community tag updates is greater than the preset number threshold.
  • the preset number of times thresholds may be set according to the requirements of the user, or may be set according to the default mode of the system, which is not limited in the embodiment of the present invention.
  • the preset number threshold may be 100 times, 150 times, 200 times, and the like.
  • the step of dividing the user with the same relationship community label in the same relationship into the same relationship community includes: if the number of relationship community label updates of the user is greater than a preset number threshold, the relationship community of the multiple users Users with the same tags are grouped into the same relationship community.
  • the preset number of thresholds is 100
  • the number of times the user's relationship community is updated is 50
  • the relationship community is not divided, and the relationship community information of the user needs to be updated according to the relationship strength information between the users.
  • the number of user relationship community tags is updated 100 times, the division of the relationship community is completed.
  • the method further includes: if the number of times of updating the relationship community label of the user is less than or equal to a threshold number of preset times, And updating the relationship community label of the user to a relationship community label of the user whose relationship strength information with the user is greater than a preset threshold.
  • the step 203 may be: selecting, according to the influence strength index information, a strong influence from the different relationship communities.
  • the degree index information is greater than or equal to the user of the preset influence strength index information.
  • the impact strength index may be the number of times the information published by the user is read by other users, the number of times of sharing, or the number of times of forwarding.
  • the threshold of the preset influence index may be set according to the requirements of the user, or may be set according to the default mode of the system, and may be set according to the budget cost of the preset information to be pushed, which is not limited in the embodiment of the present invention.
  • the estimated advertising cost of the preset advertisement is 3 million, and the preset advertisement is to be distributed to 300 million users. It is necessary to select 300 million users from the relationship community, and the users in the relationship community are in accordance with the influence strength index.
  • the small order is sorted and the preset influence strength index threshold is set according to the influence strength index of the 300th user.
  • the preset influence strength index threshold may be a threshold value of the preset published information being read by other users, for example, may be 50,000 times;
  • the preset influence strength index threshold may be a threshold for the preset published information to be shared by other users, such as 100000 times;
  • the impact strength index may be used when the number of times the information published by the user is forwarded by other users.
  • the threshold of the preset influence strength index may be a threshold for the number of times the preset information is forwarded by other users, for example, it may be 80,000 times.
  • the user whose influence strength information meets the preset influence strength condition is determined as the user to push the preset information.
  • the step 204 may specifically be: determining, by the user whose influence strength index information is greater than or equal to the preset influence strength index threshold, to be pushed. The user who set the information.
  • the complete process of user screening to push the preset information is as shown in FIG. 11 : first, by using a preset scoring model, a plurality of users whose scores meet the preset conditions are filtered; and then according to the multiple users.
  • the relationship strength information between the plurality of users is divided into different relationship communities; finally, the users whose influence intensity information meets the preset influence strength condition are separately selected from the different relationship communities and the influence strength is selected.
  • the user whose information meets the preset influence strength condition is determined as the user who wants to push the preset information.
  • the user who pushes the preset information performs information push.
  • the specific application scenario may be as follows, but is not limited thereto, including: 5 million users who are to be advertised by the application advertisement, and selected from the potential users obtained by the historical information push through the preset scoring model.
  • 4 relationship communities are divided, including relationship community 1, relationship community 2, relationship community 3, and Department 4, from each of the relationship communities, select the most influential users with 2/3 of the number of users per community, and get 5 million users.
  • the 5 million users selected from the relationship community are identified as users who are to be placed with the application advertisement and push information.
  • Another information pushing method provided by the embodiment of the present invention firstly acquires a plurality of users that meet the preset conditions; and according to the relationship strength information between the plurality of users, and the influence strength information of the plurality of users, The user who wants to push the preset information is filtered out by the plurality of users; and finally, the user who pushes the preset information is pushed by the information.
  • the embodiment of the present invention filters the plurality of users whose scores meet the preset conditions by using the preset scoring model, and then according to the relationship strength information between the multiple users.
  • the influence strength information of the plurality of users, and the user who selects the preset information to be pushed from the plurality of users can additionally convert the filtered user into an application user who performs information push. Get more application users, which can magnify the cost of application advertising budgets, which can improve the efficiency of information push.
  • An embodiment of the present invention provides a method for determining a target user. As shown in FIG. 3, the method includes:
  • the preset influence strength condition may be set according to the user's needs, or may be set according to the system default mode, which is not limited in the embodiment of the present invention.
  • the preset influence strength condition may be that the influence strength level is high, or the influence strength index may be greater than or equal to 50,000 times.
  • the target user may be a user who wants to push preset information.
  • the preset influence strength condition is: the influence strength level is high.
  • 8 million users with the highest scores are obtained.
  • the 8 million users with the highest scores have a high impact strength level of 5 million users, and then 5 million users with high impact levels are determined as push-to-push preset information. user.
  • the method for determining a target user provided by the embodiment of the present invention firstly acquires a plurality of users that meet the preset condition; and according to the relationship strength information between the multiple users and the influence strength information of the multiple users, The user who wants to push the preset information is filtered out by the plurality of users; and finally, the user who pushes the preset information is pushed by the information.
  • the embodiment of the present invention presupposes that compared with the information that is currently directly filtered by the user who is to be pushed to push the preset information.
  • the user of the preset information can additionally obtain more application users while ensuring that the filtered user itself is converted into an application user who pushes the information, thereby being able to enlarge the application advertisement budget cost benefit, thereby improving the information push. effectiveness.
  • An embodiment of the present invention provides another method for determining a target user. As shown in FIG. 4, the method includes:
  • Filter and select, by using a preset scoring model, a plurality of users whose scores meet the preset conditions.
  • the preset influence force intensity level table stores threshold ranges of influence strength information corresponding to different influence strength levels respectively.
  • the level of influence strength can be divided into three levels: high, medium and low, and can also be divided into four levels: high, medium, medium and low.
  • the influence strength level is divided into three levels: high, medium, and low.
  • the corresponding influence strength information threshold interval is (50000, 100000); when the influence strength level is medium, the corresponding influence strength information threshold interval is (5000, 50000), the influence strength level When it is low, the corresponding influence strength information threshold interval is (0, 5000).
  • the user's influence strength is 4000, it is determined that the user influence strength level is low, and when the user's influence strength is 80000, it is determined that the user influence strength level is high.
  • the method before the step 402 further includes: establishing a mapping relationship between different influence strength levels and different influence strength information threshold intervals; and saving the mapping relationship in the preset influence strength level In the table.
  • the relationship strength information between users stored in each relationship community is greater than a preset threshold.
  • the preset threshold may be set according to the requirements of the user, or may be set according to the default mode of the system, which is not limited in the embodiment of the present invention.
  • the preset threshold may be 0.5, 0.6, or the like.
  • the method further includes: acquiring an initial relationship community tag of the multiple users.
  • the step 403 may be: updating the initial relationship community label of the multiple users to a relationship community label of the user with the plurality of household relationship strength information being greater than a preset threshold; and the relationship community label of the multiple users The same users are divided into the same relationship community.
  • the method further includes: acquiring the number of relationship community label updates of the user; determining the number of relationship community label updates of the user Whether the user having the same relationship community label among the plurality of users is divided into the same relationship community includes: if the number of relationship community label updates of the user is greater than a preset number threshold, the Users with the same relationship community label among multiple users are divided into the same relationship community.
  • the method further includes: if the number of times of updating the relationship community label of the user is less than or equal to a threshold number of preset times, And updating the relationship community label of the user to a relationship community label of the user whose relationship strength information with the user is greater than a preset threshold.
  • the preset level may be set according to the user's requirement, or may be set according to the system default mode, which is not limited in the embodiment of the present invention. For example, if the preset level is high, the user whose influence strength level is high is determined as the user whose influence strength information meets the preset influence strength condition.
  • the preset level is high, and 7.5 million users are filtered through the preset scoring model.
  • 7.5 million users are divided into five relationship communities, namely, relationship community 1, relationship community 2.
  • the high user is determined to be the user whose influence strength information meets the preset influence strength condition.
  • the target user may be a user to push preset information.
  • Another target user determining method provided by the embodiment of the present invention first obtains a plurality of users that meet the preset condition; and according to the relationship strength information between the multiple users, and the influence strength information of the multiple users, The user who wants to push the preset information is filtered out from the plurality of users; and finally, the user who pushes the preset information is pushed by the information.
  • the embodiment of the present invention filters the plurality of users whose scores meet the preset conditions by using the preset scoring model, and then according to the relationship strength information between the multiple users.
  • the influence strength information of the plurality of users, and the user who selects the preset information to be pushed from the plurality of users can additionally convert the filtered user into an application user who performs information push. Get more application users, which can magnify the cost of application advertising budgets, which can improve the efficiency of information push.
  • the embodiment of the present invention provides an information pushing device.
  • the device may include: an obtaining unit 51, a screening unit 52, and a pushing unit 53.
  • the obtaining unit 51 is configured to acquire a plurality of users that meet the preset condition.
  • the filtering unit 52 is configured to filter, according to the relationship strength information between the plurality of users acquired by the acquiring unit 51, and the plurality of user influence strength information, the preset to be pushed from the plurality of users. User of the information.
  • the pushing unit 53 is configured to perform information push on the user to be pushed by the screening unit 52 to push the preset information.
  • the device embodiment corresponds to the foregoing method embodiment.
  • the device embodiment does not describe the details in the foregoing method embodiments one by one, but it should be clear that the device in this embodiment can Corresponding to implementing all of the foregoing method embodiments.
  • An information pushing apparatus provided by an embodiment of the present invention first acquires a plurality of users that meet the preset conditions; and according to the relationship strength information between the multiple users, and the influence strength information of the multiple users, The user who wants to push the preset information is filtered out among the plurality of users; and finally, the user who pushes the preset information is pushed by the information.
  • the embodiment of the present invention filters the plurality of users whose scores meet the preset conditions by using the preset scoring model, and then according to the relationship strength information between the multiple users.
  • the influence strength information of the plurality of users, and the user who selects the preset information to be pushed from the plurality of users can additionally convert the filtered user into an application user who performs information push. Get more application users, which can magnify the cost of application advertising budgets, which can improve the efficiency of information push.
  • the embodiment of the present invention provides another information pushing device.
  • the device may include: an obtaining unit 61, a screening unit 62, and a pushing unit 63.
  • the obtaining unit 61 is configured to acquire a plurality of users that meet the preset condition.
  • the filtering unit 62 is configured to filter, according to the relationship strength information between the plurality of users acquired by the acquiring unit 61, and the plurality of user influence strength information, the preset to be pushed from the plurality of users. User of the information.
  • the pushing unit 63 is configured to perform information pushing on the user to be pushed by the screening unit 62 to push the preset information.
  • the screening unit 62 includes: a dividing subunit 621, a screening subunit 622, and a determining subunit 623.
  • the sub-unit 621 is configured to divide the multiple users into different relationship communities according to the relationship strength information between the multiple users acquired by the acquiring unit 61, and the relationship between the users saved in each relationship community Intensity The information is greater than the preset threshold.
  • the filtering sub-unit 622 is configured to filter, from the different relationship communities divided by the dividing sub-unit 621, the users whose influence strength information meets the preset influence strength condition.
  • the determining sub-unit 623 is configured to determine, according to the preset influence strength condition, the user whose influence strength information is filtered by the screening sub-unit 622 as the user to push the preset information.
  • the screening unit 62 further includes: an obtaining subunit 624.
  • the obtaining sub-unit 624 is configured to acquire an initial relationship community label of the multiple users.
  • the dividing subunit 621 includes an updating module 6211 and a dividing module 6212.
  • the update module 6211 is configured to update the initial relationship community tag of each of the multiple users acquired by the obtaining sub-unit 624 to a relationship community tag of the user whose relationship strength information is greater than a preset threshold.
  • the dividing module 6212 is configured to divide the users whose relationship community tags are the same among the plurality of users into the same relationship community.
  • the dividing subunit 621 further includes: an obtaining module 6213 and a determining module 6214.
  • the obtaining module 6213 is configured to obtain the number of relationship community label updates of the user.
  • the determining module 6214 is configured to determine whether the number of relationship community label updates of the user acquired by the acquiring module 6213 is greater than a preset number threshold.
  • the dividing module 6213 is specifically configured to: if the determining module 6214 determines that the number of relationship community label updates of the user is greater than a preset number of times threshold, the users of the plurality of users having the same relationship community label are classified into the same relationship community. in.
  • the update module 6211 is further configured to update the relationship community label of the user to be compared with the user, if the determining module 6214 determines that the number of relationship community label updates of the user is less than or equal to a preset number threshold.
  • the relationship community label of the user whose relationship strength information is greater than the preset threshold.
  • the screening sub-unit 622 is configured to: when the influence strength information is the influence strength index information, select, respectively, the influence strength index information is greater than or equal to the preset influence strength index information from the different relationship communities. user.
  • the device embodiment corresponds to the foregoing method embodiment.
  • the device embodiment does not describe the details in the foregoing method embodiments one by one, but it should be clear that the device in this embodiment can Corresponding to implementing all of the foregoing method embodiments.
  • Another information pushing apparatus provided by the embodiment of the present invention first acquires a plurality of users that meet the preset condition; and according to the relationship strength information between the plurality of users, and the influence strength information of the plurality of users, More Among the users, the user who wants to push the preset information is selected; finally, the user who pushes the preset information is pushed by the information.
  • the embodiment of the present invention filters the plurality of users whose scores meet the preset conditions by using the preset scoring model, and then according to the relationship strength information between the multiple users.
  • the influence strength information of the plurality of users, and the user who selects the preset information to be pushed from the plurality of users can additionally convert the filtered user into an application user who performs information push. Get more application users, which can magnify the cost of application advertising budgets, which can improve the efficiency of information push.
  • an embodiment of the present invention provides a target user determining apparatus.
  • the apparatus may include: an obtaining unit 71, a screening unit 72, and a determining unit 73.
  • the obtaining unit 71 is configured to acquire a plurality of users that meet the preset condition.
  • the screening unit 72 is configured to filter, according to the influence strength information corresponding to the plurality of users acquired by the acquiring unit 71, the user whose influence strength information meets the preset influence strength condition from the plurality of users.
  • the determining unit 73 is configured to determine, as the target user, the user whose influence strength information filtered by the screening unit 72 meets the preset influence strength condition.
  • the device embodiment corresponds to the foregoing method embodiment.
  • the device embodiment does not describe the details in the foregoing method embodiments one by one, but it should be clear that the device in this embodiment can Corresponding to implementing all of the foregoing method embodiments.
  • the target user determining apparatus provided by the embodiment of the present invention first acquires a plurality of users that meet the preset condition; and according to the relationship strength information between the multiple users, and the influence strength information of the multiple users, The user who wants to push the preset information is filtered out by the plurality of users; and finally, the user who pushes the preset information is pushed by the information.
  • the embodiment of the present invention filters the plurality of users whose scores meet the preset conditions by using the preset scoring model, and then according to the relationship strength information between the multiple users.
  • the influence strength information of the plurality of users, and the user who selects the preset information to be pushed from the plurality of users can additionally convert the filtered user into an application user who performs information push. Get more application users, which can magnify the cost of application advertising budgets, which can improve the efficiency of information push.
  • the embodiment of the present invention provides another target user determining apparatus.
  • the apparatus may include: an obtaining unit 81, a screening unit 82, and a determining unit 83.
  • the obtaining unit 81 is configured to acquire a plurality of users that meet the preset condition.
  • the filtering unit 82 is configured to respectively have strong influence corresponding to the plurality of users acquired by the acquiring unit 81
  • the degree information is used to filter out the users whose influence intensity information meets the preset influence strength condition from the plurality of users.
  • the determining unit 83 is configured to determine, as the target user, the user whose influence strength information selected by the screening unit 82 meets the preset influence strength condition.
  • the obtaining unit 81 is specifically configured to filter, by using a preset scoring model, a plurality of users whose scores meet the preset conditions.
  • the screening unit 82 includes a determining subunit 821.
  • a determining subunit 821 configured to determine an influence strength level of the plurality of users according to the preset influence strength level table and the influence strength information corresponding to the plurality of users respectively, and the preset influence strength level table The threshold range of the influence intensity information corresponding to different influence strength levels is saved.
  • the determining sub-unit 821 is further configured to determine, by the user that the influence strength level of the plurality of users is greater than or equal to the preset level, the user whose influence strength information meets the preset influence strength condition.
  • the screening unit further includes: a dividing subunit 822.
  • the sub-unit 822 is configured to divide the multiple users into different relationship communities according to the relationship strength information between the multiple users, where the relationship strength information between the users is greater than a preset threshold. user.
  • the determining subunit 821 is specifically configured to determine, as the target user, a user whose influence strength level in the different relationship community is greater than or equal to a preset level.
  • the apparatus further includes: an establishing unit 84 and a saving unit 85.
  • the establishing unit 84 is configured to establish a mapping relationship between different influence strength levels and different influence strength information threshold intervals
  • the saving unit 85 is configured to save the mapping relationship in the preset influence strength level table.
  • the screening unit further includes: an obtaining subunit 823.
  • the obtaining sub-unit 823 is specifically configured to acquire an initial relationship community label of the multiple users.
  • the dividing subunit 822 includes:
  • the update module 8221 is configured to update an initial relationship community label of each of the multiple users to a relationship community label of the user whose relationship strength information is greater than a preset threshold;
  • the dividing module 8222 is configured to divide users of the plurality of users whose relationship community tags are the same into the same relationship community.
  • the dividing subunit 822 further includes: an obtaining module 8223 and a determining module 8224.
  • the obtaining module 8223 is configured to acquire the number of relationship community label updates of the user
  • the determining module 8224 is configured to determine, by the acquiring module, the relationship community label update time of the user acquired by the acquiring module. Whether the number is greater than a preset number threshold;
  • the dividing module 8222 is configured to: if the determining module 8224 determines that the number of relationship community label updates of the user is greater than a preset number threshold, the user with the same relationship community label of the multiple users is divided into the same relationship community. in.
  • the update module 8221 is further configured to: if the determining module 8224 determines that the number of relationship community label updates of the multiple users is less than or equal to a preset number threshold, update the relationship community label of the user to the user The relationship community label of the user whose relationship strength information is greater than the preset threshold.
  • the device embodiment corresponds to the foregoing method embodiment.
  • the device embodiment does not describe the details in the foregoing method embodiments one by one, but it should be clear that the device in this embodiment can Corresponding to implementing all of the foregoing method embodiments.
  • Another target user determining apparatus provided by the embodiment of the present invention first acquires a plurality of users that meet the preset condition; and according to the relationship strength information between the multiple users, and the influence strength information of the multiple users, The user who wants to push the preset information is filtered out from the plurality of users; and finally, the user who pushes the preset information is pushed by the information.
  • the embodiment of the present invention filters the plurality of users whose scores meet the preset conditions by using the preset scoring model, and then according to the relationship strength information between the multiple users.
  • the influence strength information of the plurality of users, and the user who selects the preset information to be pushed from the plurality of users can additionally convert the filtered user into an application user who performs information push. Get more application users, which can magnify the cost of application advertising budgets, which can improve the efficiency of information push.
  • the information pushing device includes a processor and a memory.
  • the acquiring unit, the filtering unit, the pushing unit and the like are all stored as a program unit in a memory, and the program unit stored in the memory is executed by the processor to implement a corresponding function.
  • the processor contains a kernel, and the kernel removes the corresponding program unit from the memory.
  • the kernel can be set to one or more. By adjusting the kernel parameters, there is a problem that the display is stuck when the selected graphic is displayed in the chart.
  • the memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory (flash RAM), the memory including at least one Memory chip.
  • RAM random access memory
  • ROM read only memory
  • flash RAM flash memory
  • the present application also provides a computer program product, when executed on a data processing device, adapted to perform a program code that initializes a method of: obtaining a plurality of users that meet a preset condition; The relationship strength information, and the influence strength information of the plurality of users, the user who wants to push the preset information is selected from the plurality of users; and the information is pushed to the user who is to push the preset information.
  • embodiments of the present application can be provided as a method, system, or computer program product.
  • the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment in combination of software and hardware.
  • the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.
  • a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
  • processors CPUs
  • input/output interfaces network interfaces
  • memory volatile and non-volatile memory
  • the memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory.
  • RAM random access memory
  • ROM read only memory
  • Memory is an example of a computer readable medium.
  • Computer readable media includes both permanent and non-persistent, removable and non-removable media.
  • Information storage can be implemented by any method or technology.
  • the information can be computer readable instructions, data structures, modules of programs, or other data.
  • Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory, or other Memory technology, CD-ROM, digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape storage or other magnetic storage device or any other non-transportable medium, available for Stores information that can be accessed by the computing device.
  • PRAM phase change memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • RAM random access memory
  • ROM read only memory
  • EEPROM electrically erasable programmable read only memory
  • flash memory or other Memory technology
  • CD-ROM digital versatile

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Databases & Information Systems (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

本发明公开了一种信息推送方法及装置，涉及信息技术领域。本发明主要用于解决信息推送效率较低的问题。所述方法包括：获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；对所述待推送预置信息的用户进行信息推送。

Description

信息推送方法及装置
本申请要求2016年06月06日递交的申请号为201610395230.1、发明名称为“信息推送方法及装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及信息技术领域，尤其涉及一种信息推送方法及装置。
背景技术
随着信息技术和互联网技术的不断发展，各种各样的应用程序也不断出现，而应用程序拥有的用户数量和质量是其得以生存和发展的前提。为了吸引和获取更多的用户，许多应用程序提供商会在新应用程序前期投放广告，即向用户推荐应用程序信息。
目前，在进行信息推送时，通常根据预置评分模型，直接筛选待推送预置信息的用户进行信息推送，其中，预置评分模型是根据应用程序历史获取的用户结果和用户属性生成的。即根据预置评分模型为潜在用户进行评分，然后从潜在用户中筛选出分数最高的待推送预置信息的用户。然而，若根据预置评分模型，直接筛选待推送预置信息的用户进行应用程序信息推送，只能保证筛选出的用户本身转化为进行信息推送的应用程序用户，无法额外获取更多的应用程序用户，导致无法放大应用程序广告预算成本收益，从而导致信息推送效率较低。
发明内容
有鉴于此，本发明实施例提供一种信息推送方法及装置，主要目的是解决信息推送效率较低的问题。
为达到上述目的，本发明提供如下技术方案：
本发明实施例提供一种信息推送方法，包括：
获取符合预置条件的多个用户；
根据所述多个用户之间的关系强度信息，以及所述多个用户影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；
对所述待推送预置信息的用户进行信息推送。
本发明实施例提供一种信息推送装置，包括：
获取单元，用于获取符合预置条件的多个用户；
筛选单元，用于根据所述获取单元获取的所述多个用户之间的关系强度信息，以及所述多个用户影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；
推送单元，用于对所述筛选单元筛选的所述待推送预置信息的用户进行信息推送。
本发明实施例提供一种目标用户确定方法，包括：
获取符合预置条件的多个用户；
根据所述多个用户分别对应的影响力强度信息，从所述多个用户中筛选出影响力强度信息符合预置影响力强度条件的用户；
将所述影响力强度信息符合预置影响力强度条件的用户确定为目标用户。
本发明实施例提供一种目标用户确定装置，包括：
获取单元，用于通过获取符合预置条件的多个用户；
筛选单元，用于根据所述获取单元获取的所述多个用户分别对应的影响力强度信息，从所述多个用户中筛选出影响力强度信息符合预置影响力强度条件的用户；
确定单元，用于将所述筛选单元筛选的所述影响力强度信息符合预置影响力强度条件的用户确定为目标用户。
借由上述技术方案，本发明实施例提供的技术方案至少具有下列优点：
本发明实施例提供的一种信息推送方法及装置，首先获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；最后对所述待推送预置信息的用户进行信息推送。与目前直接筛选待推送预置信息的用户进行信息推送相比，本发明实施例通过预置评分模型，筛选得到分数符合预置条件的多个用户，然后再根据多个用户之间关系强度信息以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够在保证筛选出的用户本身转化为进行信息推送的应用程序用户的同时能够额外获取更多的应用程序用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
上述说明仅是本发明技术方案的概述，为了能够更清楚了解本发明的技术手段，而可依照说明书的内容予以实施，并且为了让本发明的上述和其它目的、特征和优点能够更明显易懂，以下特举本发明的具体实施方式。
附图说明
通过阅读下文优选实施方式的详细描述，各种其他的优点和益处对于本领域普通技 术人员将变得清楚明了。附图仅用于示出优选实施方式的目的，而并不认为是对本发明的限制。而且在整个附图中，用相同的参考符号表示相同的部件。在附图中：
图1示出了本发明实施例提供的一种信息推送方法的流程图；
图2示出了本发明实施例提供的另一种信息推送方法的流程图；
图3示出了本发明实施例提供的一种目标用户确定方法的流程图；
图4示出了本发明实施例提供的另一种目标用户确定方法的流程图；
图5示出了本发明实施例提供的一种信息推送装置的结构示意图；
图6示出了本发明实施例提供的另一种信息推送装置的结构示意图；
图7示出了本发明实施例提供的一种目标用户确定装置的结构示意图；
图8示出了本发明实施例提供的另一种目标用户确定装置的结构示意图；
图9示出了本发明实施例提供的数据处理方法流程图；
图10示出了本发明实施例提供的预置评分模型生成方法流程图；
图11示出了本发明实施例提供的待推送预置信息的用户筛选的流程图。
具体实施方式
下面将参照附图更详细地描述本公开的示例性实施例。虽然附图中显示了本公开的示例性实施例，然而应当理解，可以以各种形式实现本公开而不应被这里阐述的实施例所限制。相反，提供这些实施例是为了能够更透彻地理解本公开，并且能够将本公开的范围完整的传达给本领域的技术人员。
本发明实施例提供一种信息推送方法，如图1所示，所述方法包括：
101、获取符合预置条件的多个用户。
其中，所述预置条件可以根据用户的需求进行设置，也可以根据系统需求进行设置，本发明实例不做限定。所述多个用户具体可以为分数符合预置分数条件的多个用户。具体地，可以通过预置评分模型，筛选分数符合预置分数条件的用户。
其中，所述预置评分模型是根据应用程序历史获取的用户结果和用户属性生成的。所述应用程序历史获取的用户结果可以包括应用程序历史获取的潜在用户和转化为应用程序用户的用户。所述预置分数条件可以根据用户的需求进行设置，也可以根据系统需求进行设置，本发明实例不做限定。例如，预置分数条件可以为潜在用户中分数最高的750个用户，也可以为潜在用户中分数大于或者0.8分的用户等。
对于本发明实施例，在生成预置评分模型之前，需要通过不明确指定目标用户的信 息推送方式获取的信息推送样本用户以及样本用户的属性数据，然后对样本用户的属性数据进行处理，获取生成预置评分模型的特征数据，其中，生成预置评分模型的特征数据要求完备无缺失、准确无异常、数据格式符合评分模型生成的标准。
其中，所述不明确指定目标用户的信息推送方式为在各种渠道上投放和展示应用程序广告，不明确限定广告投放用户。通过不明确指定目标用户的信息推送方式可以获取较多的信息推送用户，在推送信息后可以通过广告渠道上嵌入的程序采集信息推送用户的属性数据。
对于本发明实施例，针对用户的属性数据，在此对数据处理方法进行说明，如图9：
1、数据清洗和加工。由于样本用户以及样本用户的属性数据是通过渠道上嵌入的程序采集的，在同一渠道上通常会存在多个应用程序或者产品的广告，采集的样本用户可能会存在其他应用程序广告用户；或者样本用户的属性数据中的某些数据不是生成评分模型的特征数据，因此需要对样本用户以及样本用户的属性数据进行清洗和加工，如采集到的用户资产数据为用户在不同资产账号中的资产数据，而生成评分模型的资产数据是用户的总资产数据，因此需要根据用户在不同资产账号中的资产数据，获取用户的总资产数据。
2、异常检测。由于样本用户以及样本用户的属性数据是通过渠道上嵌入的程序采集的日志信息解析出来的数据，从日志信息中解析出来的数据可能会存在一些脏数据，例如，用户使用手机的品牌数据中可能会存在一些乱码、用户访问应用程序广告次数过多、其中，访问应用程序广告次数过多可能是由于机器恶意访问攻击导致的等，脏数据会影响评分模型的生成，因此需要检测样本用户的属性数据的是否存在脏数据。
3、数据拆分→数据采样→数据探索。由于样本用户中转化为应用程序用户的用户和应用程序潜在用户的比例、或者应用程序潜在用户的比例和转化为应用程序用户的用户的比例可能会很大，因此需要对样本用户的属性数据进行过采样、或者欠采样，以便于评分模型生成算法能够有效的学习出转化为应用程序用户的属性特征。
4、特征剔除→缺失处理→异常处理。由于大部分评分模型生成算法不支持有缺失的样本数据也不支持属性特征数据过多的样本数据，对于属性特征数据缺失过多、属性特征类型过多的样本数据进行剔除；而针对属性特征数据缺失较少的样本数据进行属性特征数据处理，即将缺失的属性特征数据填充上，其中，数值型的数据可以用中位数填充、离散型的数据可以用众数填充，如用户的属性数据缺失某用户的年龄数据，可以根据样本用户中年龄数据的中位数作为该用户的年龄数据。其次，为了避免评分模型受一些异 常数据的影响，通常需要对异常数据进行处理，例如，用99分位数替代异常数据等。
5、特征变换。由于评分模型算法中的模型算法只支持稀疏矩阵，不支持普通二维表，因此需要对属性数据进行格式转换，得到生成评分模型的特征数据。如可以通过对数变换转换属性数据的格式，得到生成评分模型的特征数据。
对于本发明实施例，图10所示，在此对预置评分模型的生成方法进行说明：
1、特征选择。获取生成评分模型的特征数据和评分模型生成算法。所述评分模型生成算法可以为决策树算法、逻辑回归算法、随机森林算法、迭代的决策树算法(GBDT，Gradient Boosting Decision Tree)等。
2、模型训练。根据生成评分模型的特征数据和评分模型生成算法，训练生成多个评分模型。
3、参数调优。根据评分模型生成算法中的参数以及不同的所述特征数据组合，获取各个评分模型的评分效果。具体可以根据模型评分效果度量标准度量评分模型的评分效果。所述模型评分效果度量标准可以为曲线下面积，曲线下面积越大，评分模型的评分效果越佳。
4、预置评分模型。从多个评分模型中选择评分效果最好的评分模型并将评分效果最好的评分模型确定为筛选分数符合预置条件的用户的预置评分模型。
5、潜在用户评分。根据预置评分模型对潜在用户进行评分，然后从潜在用户中筛选出分数符合预置条件的用户。
102、根据多个用户之间的关系强度信息，以及多个用户的影响力强度信息，从多个用户中筛选出待推送预置信息的用户。
其中，所述多个用户之间的关系强度信息可以为多个用户属性信息之间的相似度。用户的影响力强度是指用户影响其他用户的能力强度，具体可以为用户发布的信息被其他用户阅读、分享以及转发的次数。
其中，待推送预置信息可以为应用程序提供商根据广告预算成本推送的广告信息。为了避免在应用程序广告上投入较高成本，许多的应用程序提供商通常会对广告成本进行预算，即根据广告预算成本确定信息推送用户数据，只将信息推送给信息推送用户数据的用户。
例如，应用程序对应的广告投放成本为500万，每个用户的广告成本为1元，待推送预置信息的用户有500万个，通过预置评分模型筛选出分数最高的750万个用户，则根据用户之间的关系强度信息和用户的影响力强度信息，从通过预置评分模型筛选的分 数最高的750万个用户中筛选出待推送预置信息的500个用户。
对于本发明实施例，首先通过预置评分模型筛选得到分数符合预置条件的多个用户，然后根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够保证待推送预置信息的用户自身转化为进行信息推送的应用程序用户，同时能够利用用户的影响力和传播能力以及通过分享、转发、或者口碑的等方式为应用程序获取更多的用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
103、对待推送预置信息的用户进行信息推送。
本发明实施例提供的一种信息推送方法，首先获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；最后对所述待推送预置信息的用户进行信息推送。与目前直接筛选待推送预置信息的用户进行信息推送相比，本发明实施例通过预置评分模型，筛选得到分数符合预置条件的多个用户，然后再根据多个用户之间关系强度信息以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够在保证筛选出的用户本身转化为进行信息推送的应用程序用户的同时能够额外获取更多的应用程序用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
本发明实施例提供另一种信息推送方法，如图2所示，所述方法包括：
201、获取符合预置条件的多个用户。
其中，符合预置条件的用户的解释以及获取符合预置条件的多个用户的过程，在步骤101中已进行了详细的描述，在此本发明实施例不进行赘述。
202、根据多个用户之间的关系强度信息，将多个用户划分为不同关系社区。
其中，每一个关系社区中保存的用户之间关系强度信息大于预设阈值。所述关系社区可以为关系网络社区，具体可以为论坛、贴吧、公告栏、群组讨论、在线聊天、交友、个人空间、无线增值服务等形式在内的网上交流空间，同一关系社区中的用户具有相同的关系社区标签。所述预设阈值可以根据用户的需求进行设置，也可以根据系统默认模式进行设置，本发明实施例不做限定。例如，预设阈值可以为0.4、0.5等。
对于本发明实施例，步骤202之前，所述方法还包括：获取所述多个用户的初始关系社区标签。所述步骤202具体可以包括：将所述多个用户中每个用户的初始关系社区标签更新为与所述用户关系强度信息大于预设阈值的用户的关系社区标签；将所述多个 用户中关系社区标签相同的用户划分到同一关系社区中。
例如，若预置阈值为0.4，通过预置评分模型筛选的多个用户有用户1、用户2、用户3、用户4。用户1与用户2之间的关系强度为0.1、用户1与用户3之间的关系强度为0.4、用户1与用户4的关系强度为0.2，其中，用户1与用户3的关系强度大于或者等于0.4，则将用户1的关系社区标签更新为用户3的关系社区标签。按照同样的方式可以对用户2、用户3、用户4的关系社区标签进行更新。
其中，划分关系社区的过程是不断迭代的过程，进行第一次迭代时，获取所述多个用户的初始关系社区标签，所述多个用户中每个用户的初始关系社区标签，可以为所述用户的身份标识号(Identity，ID)。进行第一次迭代以后的迭代时，每个用户的关系社区标签是根据用户之间的关系强度信息，更新后的关系社区标签。
对于本发明实施例，所述将所述多个用户中关系社区标签相同的用户划分到同一关系社区中之前，所述方法还包括：获取所述用户的关系社区标签更新次数；判断所述用户的关系社区标签更新次数是否大于预置次数阈值。
其中，所述预置次数阈值可以根据用户的需求进行设置，也可以根据系统的默认模式进行设置，本发明实施例不做限定。例如，预置次数阈值可以为100次、150次、200次等。
所述将所述多个用户中关系社区标签相同的用户划分到同一关系社区中步骤包括：若所述用户的关系社区标签更新次数大于预置次数阈值，则将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
例如，若预置次数阈值为100次，当用户的关系社区标签更新次数为50次时，说明关系社区未划分完，需要继续根据用户之间的关系强度信息，更新用户的关系社区标签，当用户的关系社区标签更新次数为100次时，说明关系社区的划分完成。
对于本发明实施例，所述判断所述用户的关系社区标签更新次数是否大于预置次数阈值之后，所述方法还包括：若所述用户的关系社区标签更新次数小于或者等于预置次数阈值，则将所述用户的关系社区标签更新为与所述用户之间关系强度信息大于预设阈值的用户的关系社区标签。
203、从不同关系社区中分别筛选出影响力强度信息符合预置影响力强度条件的用户。
对于本发明实施例，当所述影响力强度信息为影响力强度指数信息时，步骤203具体可以为：根据所述影响力强度指数信息，从所述不同关系社区中分别筛选出影响力强 度指数信息大于或者等于预置影响力强度指数信息的用户。
其中，所述影响力强度指数可以为用户发布的信息被其他用户阅读次数、分享次数或者转发次数。所述预置影响力指数阈值可以根据用户的需求进行设置，也可以根据系统的默认模式进行设置，具体地，可以根据待推送预置信息的预算成本设置，本发明实施例不做限定。
例如，预置广告的广告预算成本为300万，预置广告待投放用户为300个万用户，需从关系社区中筛选出300个万用户，将关系社区中的用户按照影响力强度指数从大到小的排序并根据第300个用户的影响力强度指数设置预置影响力强度指数阈值。
又例如，当所述影响力强度指数可以为用户发布的信息被其他用户阅读次数时，预置影响力强度指数阈值可以为预置发布的信息被其他用户阅读次数阈值，如可以为50000次；当所述影响力强度指数可以为用户发布的信息被其他用户分享次数时，预置影响力强度指数阈值可以为预置发布的信息被其他用户分享次数阈值，如可以为100000次；当所述影响力强度指数可以为用户发布的信息被其他用户转发次数时，预置影响力强度指数阈值可以为预置发布的信息被其他用户转发次数阈值，如可以为80000次。
204、将影响力强度信息符合预置影响力强度条件的用户确定为待推送预置信息的用户。
对于本发明实施例，当所述影响力强度信息为影响力强度指数信息时，步骤204具体可以为：将影响力强度指数信息大于或者等于预置影响力强度指数阈值的用户确定为待推送预置信息的用户。
对于本发明实施例，待推送预置信息的用户筛选的完整过程，如图11所示：首先通过预置评分模型，筛选得到分数符合预置条件的多个用户；然后根据所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区；最后从所述不同关系社区中分别筛选出影响力强度信息符合预置影响力强度条件的用户并将所述影响力强度信息符合预置影响力强度条件的用户确定为待推送预置信息的用户。
205、对待推送预置信息的用户进行信息推送。
对于本发明实施例，具体的应用场景可以如下所示，但不限于此，包括：待投放应用程序广告的用户有500万个，通过预置评分模型从历史信息推送获取的潜在用户中筛选出分数最高的750万个用户，其中，750万是500万的1.5倍，并且获取750万用户中每个用户的影响力以及每个用户与其他用户之间的关系强度。根据每个用户与其他用户之间的关系强度，划分得到4关系社区，包括关系社区1、关系社区2、关系社区3、关 系社区4，从每个关系社区中分别筛选出影响力最强的且数量为每个社区用户数量2/3的用户，得到500万个用户。将从关系社区中筛选出的500万个用户确定为待投放应用程序广告的用户并进行信息推送。
本发明实施例提供的另一种信息推送方法，首先获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；最后对所述待推送预置信息的用户进行信息推送。与目前直接筛选待推送预置信息的用户进行信息推送相比，本发明实施例通过预置评分模型，筛选得到分数符合预置条件的多个用户，然后再根据多个用户之间关系强度信息以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够在保证筛选出的用户本身转化为进行信息推送的应用程序用户的同时能够额外获取更多的应用程序用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
本发明实施例提供一种目标用户确定方法，如图3所示，所述方法包括：
301、获取符合预置条件的多个用户。
其中，符合预置条件的用户的解释以及获取符合预置条件的多个用户的过程，在步骤101中已进行了详细的描述，在此本发明实施例不进行赘述。
302、根据多个用户分别对应的影响力强度信息，从多个用户中筛选出影响力强度信息符合预置影响力强度条件的用户。
其中，预置影响力强度条件可以根据用户需要进行设置，也可以根据系统默认模式进行设置，本发明实施例不做限定。如预置影响力强度条件可以为影响力强度等级为高，也可以为影响力强度指数大于或者等于50000次。
303、将影响力强度信息符合预置影响力强度条件的用户确定为目标用户。
其中，目标用户可以为待推送预置信息的用户。
例如，预置影响力强度条件为：影响力强度等级为高。首先获取800万个分数最高的用户，800万个分数最高的用户影响力强度等级为高的用户有500万个，然后将影响力等级为高的500万个用户确定为待推送预置信息的用户。
本发明实施例提供的一种目标用户确定方法，首先获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；最后对所述待推送预置信息的用户进行信息推送。与目前直接筛选待推送预置信息的用户进行信息推送相比，本发明实施例通过预 置评分模型，筛选得到分数符合预置条件的多个用户，然后再根据多个用户之间关系强度信息以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够在保证筛选出的用户本身转化为进行信息推送的应用程序用户的同时能够额外获取更多的应用程序用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
本发明实施例提供另一种目标用户确定方法，如图4所示，所述方法包括：
401、通过预置评分模型，筛选得到分数符合预置条件的多个用户。
其中，预置评分模型的解释以及通过预置评分模型筛选出分数符合预置条件的多个用户的过程，在步骤101中已进行了详细的描述，在此本发明实施例不进行赘述。
402、根据预置影响力强度等级表和多个用户分别对应的影响力强度信息，确定多个用户的影响力强度等级。
其中，所述预置影响力强度等级表中保存有不同影响力强度等级分别对应的影响力强度信息阈值区间。影响力强度等级可以分为高、中、低三个等级、也可以分为高、中、次中、低四个等级等。
例如，影响力强度等级分为高、中、低三个等级。影响力强度等级为高时，对应的影响力强度信息阈值区间为(50000，100000)；影响力强度等级为中时，对应的影响力强度信息阈值区间为(5000，50000)，影响力强度等级为低时，对应的影响力强度信息阈值区间为(0，5000)。当用户的影响力强度为4000时，确定用户影响力强度等级为低，当用户的影响力强度为80000时，确定用户影响力强度等级高。
对于本发明实施例，步骤402之前所述方法还包括：建立不同影响力强度等级与不同影响力强度信息阈值区间之间的映射关系；将所述映射关系保存在所述预置影响力强度等级表中。
403、根据多个用户之间的关系强度信息，将多个用户划分为不同关系社区。
其中，每一个关系社区中保存的用户之间关系强度信息大于预设阈值。其中，所述预设阈值可以根据用户的需求进行设置，也可以根据系统默认模式进行设置，本发明实施例不做限定。例如，预设阈值可以为0.5、0.6等。
对于本发明实施例，步骤403之前，所述方法还包括：获取所述多个用户的初始关系社区标签。步骤403具体可以为：将所述多个用户的初始关系社区标签更新为与所述用多个户关系强度信息大于预设阈值的用户的关系社区标签；将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
所述将所述多个用户中关系社区标签相同的用户划分到同一关系社区中之前，所述方法还包括：获取所述用户的关系社区标签更新次数；判断所述用户的关系社区标签更新次数是否大于预置次数阈值；所述将所述多个用户中关系社区标签相同的用户划分到同一关系社区中包括：若所述用户的关系社区标签更新次数大于预置次数阈值，则将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
对于本发明实施例，所述判断所述用户的关系社区标签更新次数是否大于预置次数阈值之后，所述方法还包括：若所述用户的关系社区标签更新次数小于或者等于预置次数阈值，则将所述用户的关系社区标签更新为与所述用户之间关系强度信息大于预设阈值的用户的关系社区标签。
404、将不同关系社区中影响力强度等级大于或等于预置等级的用户确定为影响力强度信息符合预置影响力强度条件的用户。
所述预置等级可以根据用户需求进行设置，也可以根据系统默认模式进行设置，本发明实施例不做限定。例如，预置等级为高，则将影响力强度等级为高的用户确定为影响力强度信息符合预置影响力强度条件的用户。
例如，预置等级为高，通过预置评分模型筛选得到750万个用户，根据用户之间的关系强度信息将750万个用户划分为5个关系社区、分别为关系社区1、关系社区2、关系社区3、关系社区4、关系社区5，则获取关系社区1、关系社区2、关系社区3、关系社区4、关系社区5中影响力强度等级分别为高的用户，将影响力强度等级为高的用户确定为影响力强度信息符合预置影响力强度条件的用户。
405、将影响力强度信息符合预置影响力强度条件的用户确定为目标用户。
其中，所述目标用户可以为待推送预置信息的用户。
本发明实施例提供的另一种目标用户确定方法，首先获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；最后对所述待推送预置信息的用户进行信息推送。与目前直接筛选待推送预置信息的用户进行信息推送相比，本发明实施例通过预置评分模型，筛选得到分数符合预置条件的多个用户，然后再根据多个用户之间关系强度信息以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够在保证筛选出的用户本身转化为进行信息推送的应用程序用户的同时能够额外获取更多的应用程序用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
进一步地，作为图1所示方法的具体实现，本发明实施例提供一种信息推送装置，如图5所示，所述装置可以包括：获取单元51、筛选单元52、推送单元53。
获取单元51，用于获取符合预置条件的多个用户。
筛选单元52，用于根据所述获取单元51获取的所述多个用户之间的关系强度信息，以及所述多个用户影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户。
推送单元53，用于对所述筛选单元52筛选的所述待推送预置信息的用户进行信息推送。
需要说明的是，该装置实施例与前述方法实施例对应，为便于阅读，本装置实施例不再对前述方法实施例中的细节内容进行逐一赘述，但应当明确，本实施例中的装置能够对应实现前述方法实施例中的全部内容。
本发明实施例提供的一种信息推送装置，首先获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；最后对所述待推送预置信息的用户进行信息推送。与目前直接筛选待推送预置信息的用户进行信息推送相比，本发明实施例通过预置评分模型，筛选得到分数符合预置条件的多个用户，然后再根据多个用户之间关系强度信息以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够在保证筛选出的用户本身转化为进行信息推送的应用程序用户的同时能够额外获取更多的应用程序用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
进一步地，作为图2所示方法的具体实现，本发明实施例提供另一种信息推送装置，如图6所示，所述装置可以包括：获取单元61、筛选单元62、推送单元63。
获取单元61，用于获取符合预置条件的多个用户。
筛选单元62，用于根据所述获取单元61获取的所述多个用户之间的关系强度信息，以及所述多个用户影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户。
推送单元63，用于对所述筛选单元62筛选的所述待推送预置信息的用户进行信息推送。
进一步地，所述筛选单元62包括：划分子单元621、筛选子单元622、确定子单元623。
划分子单元621，用于根据所述获取单元61获取的所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区，每一个关系社区中保存的用户之间关系强度 信息大于预设阈值。
筛选子单元622，用于从所述划分子单元621划分的所述不同关系社区中分别筛选出影响力强度信息符合预置影响力强度条件的用户。
确定子单元623，用于将所述筛选子单元622筛选的所述影响力强度信息符合预置影响力强度条件的用户确定为待推送预置信息的用户。
进一步地，所述筛选单元62还包括：获取子单元624。
所述获取子单元624，用于获取所述多个用户的初始关系社区标签。
所述划分子单元621包括：更新模块6211和划分模块6212。
更新模块6211，用于将所述获取子单元624获取的所述多个用户中每个用户的初始关系社区标签更新为与所述用户关系强度信息大于预设阈值的用户的关系社区标签。
划分模块6212，用于根将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
所述划分子单元621还包括：获取模块6213和判断模块6214。
所述获取模块6213，用于获取所述用户的关系社区标签更新次数。
所述判断模块6214，用于判断所述获取模块6213获取的所述用户的关系社区标签更新次数是否大于预置次数阈值。
所述划分模块6213，具体用于若所述判断模块6214判断所述用户的关系社区标签更新次数大于预置次数阈值，则将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
所述更新模块6211，还用于若所述判断模块6214判断所述用户的关系社区标签更新次数小于或者等于预置次数阈值，则将所述用户的关系社区标签更新为与所述用户之间关系强度信息大于预设阈值的用户的关系社区标签。
所述筛选子单元622，用于当所述影响力强度信息为影响力强度指数信息时，从所述不同关系社区中分别筛选出影响力强度指数信息大于或者等于预置影响力强度指数信息的用户。
需要说明的是，该装置实施例与前述方法实施例对应，为便于阅读，本装置实施例不再对前述方法实施例中的细节内容进行逐一赘述，但应当明确，本实施例中的装置能够对应实现前述方法实施例中的全部内容。
本发明实施例提供的另一种信息推送装置，首先获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多 个用户中筛选出待推送预置信息的用户；最后对所述待推送预置信息的用户进行信息推送。与目前直接筛选待推送预置信息的用户进行信息推送相比，本发明实施例通过预置评分模型，筛选得到分数符合预置条件的多个用户，然后再根据多个用户之间关系强度信息以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够在保证筛选出的用户本身转化为进行信息推送的应用程序用户的同时能够额外获取更多的应用程序用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
进一步地，作为图3所示方法的具体实现，本发明实施例提供一种目标用户确定装置，如图7所示，所述装置可以包括：获取单元71、筛选单元72、确定单元73。
获取单元71，用于获取符合预置条件的多个用户。
筛选单元72，用于根据所述获取单元71获取的所述多个用户分别对应的影响力强度信息，从所述多个用户中筛选出影响力强度信息符合预置影响力强度条件的用户。
确定单元73，用于将所述筛选单元72筛选的所述影响力强度信息符合预置影响力强度条件的用户确定为目标用户。
需要说明的是，该装置实施例与前述方法实施例对应，为便于阅读，本装置实施例不再对前述方法实施例中的细节内容进行逐一赘述，但应当明确，本实施例中的装置能够对应实现前述方法实施例中的全部内容。
本发明实施例提供的一种目标用户确定装置，首先获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；最后对所述待推送预置信息的用户进行信息推送。与目前直接筛选待推送预置信息的用户进行信息推送相比，本发明实施例通过预置评分模型，筛选得到分数符合预置条件的多个用户，然后再根据多个用户之间关系强度信息以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够在保证筛选出的用户本身转化为进行信息推送的应用程序用户的同时能够额外获取更多的应用程序用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
进一步地，作为图4所示方法的具体实现，本发明实施例提供另一种目标用户确定装置，如图8所示，所述装置可以包括：获取单元81、筛选单元82、确定单元83。
获取单元81，用于获取符合预置条件的多个用户。
筛选单元82，用于根据所述获取单元81获取的所述多个用户分别对应的影响力强 度信息，从所述多个用户中筛选出影响力强度信息符合预置影响力强度条件的用户。
确定单元83，用于将所述筛选单元82筛选的所述影响力强度信息符合预置影响力强度条件的用户确定为目标用户。
所述获取单元81，具体用于通过预置评分模型，筛选得到分数符合预置条件的多个用户。
所述筛选单元82包括：确定子单元821。
确定子单元821，用于根据预置影响力强度等级表和所述多个用户分别对应的影响力强度信息，确定所述多个用户的影响力强度等级，所述预置影响力强度等级表中保存有不同影响力强度等级分别对应的影响力强度信息阈值区间。
所述确定子单元821，还用于将所述多个用户中影响力强度等级大于或等于预置等级的用户确定为影响力强度信息符合预置影响力强度条件的用户。
进一步地，所述筛选单元还包括：划分子单元822。
划分子单元822，用于根据所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区，所述关系社区中保存有用户之间关系强度信息大于预设阈值的用户。
所述确定子单元821，具体用于将所述不同关系社区中影响力强度等级大于或等于预置等级的用户确定为目标用户。
进一步地，所述装置还包括：建立单元84和保存单元85。
建立单元84，用于建立不同影响力强度等级与不同影响力强度信息阈值区间之间的映射关系；
保存单元85，用于将所述映射关系保存在所述预置影响力强度等级表中。
进一步地，所述筛选单元还包括：获取子单元823。
所述获取子单元823，具体用于获取所述多个用户的初始关系社区标签。
所述划分子单元822包括：
更新模块8221，用于将所述多个用户中每个用户的初始关系社区标签更新为与所述用户关系强度信息大于预设阈值的用户的关系社区标签；
划分模块8222，用于将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
所述划分子单元822还包括：获取模块8223和判断模块8224。
所述获取模块8223，用于获取所述用户的关系社区标签更新次数；
所述判断模块8224，用于判断所述获取模块获取的所述用户的关系社区标签更新次 数是否大于预置次数阈值；
所述划分模块8222，具体用于若所述判断模块8224判断所述用户的关系社区标签更新次数大于预置次数阈值，则将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
所述更新模块8221，还用于若所述判断模块8224判断所述多个用户的关系社区标签更新次数小于或者等于预置次数阈值，则将所述用户的关系社区标签更新为与所述用户之间关系强度信息大于预设阈值的用户的关系社区标签。
需要说明的是，该装置实施例与前述方法实施例对应，为便于阅读，本装置实施例不再对前述方法实施例中的细节内容进行逐一赘述，但应当明确，本实施例中的装置能够对应实现前述方法实施例中的全部内容。
本发明实施例提供的另一种目标用户确定装置，首先获取符合预置条件的多个用户；根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；最后对所述待推送预置信息的用户进行信息推送。与目前直接筛选待推送预置信息的用户进行信息推送相比，本发明实施例通过预置评分模型，筛选得到分数符合预置条件的多个用户，然后再根据多个用户之间关系强度信息以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户，能够在保证筛选出的用户本身转化为进行信息推送的应用程序用户的同时能够额外获取更多的应用程序用户，从而能够放大应用程序广告预算成本收益，进而能够提高信息推送效率。
所述信息推送装置包括处理器和存储器，上述获取单元、筛选单元和推送单元等均作为程序单元存储在存储器中，由处理器执行存储在存储器中的上述程序单元来实现相应的功能。
处理器中包含内核，由内核去存储器中调取相应的程序单元。内核可以设置一个或以上，通过调整内核参数来解决在对图表中被选中的图形进行显示时存在显示卡顿的问题。
存储器可能包括计算机可读介质中的非永久性存储器，随机存取存储器(RAM)和/或非易失性内存等形式，如只读存储器(ROM)或闪存(flash RAM)，存储器包括至少一个存储芯片。
本申请还提供了一种计算机程序产品，当在数据处理设备上执行时，适于执行初始化有如下方法步骤的程序代码：获取符合预置条件的多个用户；根据所述多个用户之间 的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；对所述待推送预置信息的用户进行信息推送。
本领域内的技术人员应明白，本申请的实施例可提供为方法、系统、或计算机程序产品。因此，本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请是参照根据本实施例的图表中图形的显示方法、装置、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
在一个典型的配置中，计算设备包括一个或多个处理器(CPU)、输入/输出接口、网络接口和内存。
存储器可能包括计算机可读介质中的非永久性存储器，随机存取存储器(RAM)和/或非易失性内存等形式，如只读存储器(ROM)或闪存(flash RAM)。存储器是计算机可读介质的示例。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括，但不限于相变内存(PRAM)、静态随机存取存储器 (SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。按照本文中的界定，计算机可读介质不包括暂存电脑可读媒体(transitory media)，如调制的数据信号和载波。
以上仅为本申请的实施例而已，并不用于限制本申请。对于本领域技术人员来说，本申请可以有各种更改和变化。凡在本申请的精神和原理之内所作的任何修改、等同替换、改进等，均应包含在本申请的权利要求范围之内。

Claims (28)

  1. 一种信息推送方法，其特征在于，包括：
    获取符合预置条件的多个用户；
    根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；
    对所述待推送预置信息的用户进行信息推送。
  2. 根据权利要求1所述的方法，其特征在于，所述根据所述多个用户之间的关系强度信息，以及所述多个用户的影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户包括：
    根据所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区，每一个关系社区中保存的用户之间关系强度信息大于预设阈值；
    从所述不同关系社区中分别筛选出影响力强度信息符合预置影响力强度条件的用户并将所述影响力强度信息符合预置影响力强度条件的用户确定为待推送预置信息的用户。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区之前，所述方法还包括：
    获取所述多个用户的初始关系社区标签；
    所述根据所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区包括：
    将所述多个用户中每个用户的初始关系社区标签更新为与所述用户关系强度信息大于预设阈值的用户的关系社区标签；
    将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
  4. 根据权利要求3所述的方法，其特征在于，所述将所述多个用户中关系社区标签相同的用户划分到同一关系社区中之前，所述方法还包括：
    获取所述用户的关系社区标签更新次数；
    判断所述用户的关系社区标签更新次数是否大于预置次数阈值；
    所述将所述多个用户中关系社区标签相同的用户划分到同一关系社区中包括：
    若所述用户的关系社区标签更新次数大于预置次数阈值，则将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
  5. 根据权利要求4所述的方法，其特征在于，所述判断所述多个用户的关系社区标 签更新次数是否大于预置次数阈值之后，所述方法还包括：
    若所述用户的关系社区标签更新次数小于或者等于预置次数阈值，则将所述用户的关系社区标签更新为与所述用户之间关系强度信息大于预设阈值的用户的关系社区标签。
  6. 根据权利要求2所述的方法，其特征在于，所述影响力强度信息为影响力强度指数信息，所述从所述不同关系社区中分别筛选出影响力强度信息符合预置影响力强度条件的用户包括：
    从所述不同关系社区中分别筛选出影响力强度指数信息大于或者等于预置影响力强度指数信息的用户。
  7. 一种信息推送装置，其特征在于，包括：
    获取单元，用于获取符合预置条件的多个用户；
    筛选单元，用于根据所述获取单元获取的所述多个用户之间的关系强度信息，以及所述多个用户影响力强度信息，从所述多个用户中筛选出待推送预置信息的用户；
    推送单元，用于对所述筛选单元筛选的所述待推送预置信息的用户进行信息推送。
  8. 根据权利要求7所述的装置，其特征在于，所述筛选单元包括：
    划分子单元，用于根据所述获取单元获取的所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区，每一个关系社区中保存的用户之间关系强度信息大于预设阈值；
    筛选子单元，用于从所述划分子单元划分的所述不同关系社区中分别筛选出影响力强度信息符合预置影响力强度条件的用户；
    确定子单元，用于将所述筛选子单元筛选的所述影响力强度信息符合预置影响力强度条件的用户确定为待推送预置信息的用户。
  9. 根据权利要求8所述的装置，其特征在于，所述筛选单元还包括：获取子单元，
    所述获取子单元，用于获取所述多个用户的初始关系社区标签；
    所述划分子单元包括：
    更新模块，用于将所述多个用户中每个用户的初始关系社区标签更新为与所述用户关系强度信息大于预设阈值的用户的关系社区标签；
    划分模块，用于将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
  10. 根据权利要求9所述的装置，其特征在于，所述划分子单元还包括：获取模块和 判断模块，
    所述获取模块，用于获取所述用户的关系社区标签更新次数；
    所述判断模块，用于判断所述获取模块获取的所述用户的关系社区标签更新次数是否大于预置次数阈值；
    所述划分模块，具体用于若所述判断模块判断所述用户的关系社区标签更新次数大于预置次数阈值，则将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
  11. 根据权利要求10所述的装置，其特征在于，
    所述更新模块，还用于若所述判断模块判断所述用户的关系社区标签更新次数小于或者等于预置次数阈值，则将所述用户的关系社区标签更新为与所述用户之间关系强度信息大于预设阈值的用户的关系社区标签。
  12. 根据权利要求8所述的装置，其特征在于，
    所述筛选子单元，用于当所述影响力强度信息为影响力强度指数信息时，从所述不同关系社区中分别筛选出影响力强度指数信息大于或者等于预置影响力强度指数信息的用户。
  13. 一种目标用户确定方法，其特征在于，包括：
    获取符合预置条件的多个用户；
    根据所述多个用户分别对应的影响力强度信息，从所述多个用户中筛选出影响力强度信息符合预置影响力强度条件的用户；
    将所述影响力强度信息符合预置影响力强度条件的用户确定为目标用户。
  14. 根据权利要求13所述的方法，其特征在于，所述获取符合预置条件的多个用户包括：
    通过预置评分模型，筛选得到分数符合预置条件的多个用户。
  15. 根据权利要求13所述的方法，其特征在于，所述从所述多个用户中筛选出影响力强度信息符合预置影响力强度条件的用户包括：
    根据预置影响力强度等级表和所述多个用户分别对应的影响力强度信息，确定所述多个用户的影响力强度等级，所述预置影响力强度等级表中保存有不同影响力强度等级分别对应的影响力强度信息阈值区间；
    将所述多个用户中影响力强度等级大于或等于预置等级的用户确定为影响力强度信息符合预置影响力强度条件的用户。
  16. 根据权利要求15所述的方法，所述将所述多个用户中影响力强度等级大于或等于预置等级的用户确定为影响力强度信息符合预置影响力强度条件的用户之前，所述方法还包括：
    根据所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区，每一个关系社区中保存的用户之间关系强度信息大于预设阈值；
    所述将所述多个用户中影响力强度等级大于或等于预置等级的用户确定为影响力强度信息符合预置影响力强度条件的用户包括：
    将所述不同关系社区中影响力强度等级大于或等于预置等级的用户确定为所述影响力强度信息符合预置影响力强度条件的用户。
  17. 根据权利要求15所述的方法，其特征在于，所述获取符合预置条件的多个用户之前，所述方法还包括：
    建立不同影响力强度等级与不同影响力强度信息阈值区间之间的映射关系；
    将所述映射关系保存在所述预置影响力强度等级表中。
  18. 根据权利要求16所述的方法，其特征在于，所述根据所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区之前，所述方法还包括：
    获取所述多个用户的初始关系社区标签；
    所述根据多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区包括：
    将所述多个用户中每个用户的初始关系社区标签更新为与所述用户关系强度信息大于预设阈值的用户的关系社区标签；
    将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
  19. 根据权利要求18所述的方法，其特征在于，所述将所述多个用户中关系社区标签相同的用户划分到同一关系社区中之前，所述方法还包括：
    获取所述用户的关系社区标签更新次数；
    判断所述用户的关系社区标签更新次数是否大于预置次数阈值；
    所述将所述多个用户中关系社区标签相同的用户划分到同一关系社区中包括：
    若所述用户的关系社区标签更新次数大于预置次数阈值，则将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
  20. 根据权利要求19所述的方法，其特征在于，所述判断所述多个用户的关系社区标签更新次数是否大于预置次数阈值之后，所述方法还包括：
    若所述用户的关系社区标签更新次数小于或者等于预置次数阈值，则将所述用户的关系社区标签更新为与所述用户之间关系强度信息大于预设阈值的用户的关系社区标签。
  21. 一种目标用户确定装置，其特征在于，包括：
    获取单元，用于获取符合预置条件的多个用户；
    筛选单元，用于根据所述获取单元获取的所述多个用户分别对应的影响力强度信息，从所述多个用户中筛选出影响力强度信息符合预置影响力强度条件的用户；
    确定单元，用于将所述筛选单元筛选的所述影响力强度信息符合预置影响力强度条件的用户确定为目标用户。
  22. 根据权利要求21所述的装置，其特征在于，
    所述获取单元，具体用于通过预置评分模型，筛选得到分数符合预置条件的多个用户。
  23. 根据权利要求21所述的装置，其特征在于，所述筛选单元包括：
    确定子单元，用于根据预置影响力强度等级表和所述多个用户分别对应的影响力强度信息，确定所述多个用户的影响力强度等级，所述预置影响力强度等级表中保存有不同影响力强度等级分别对应的影响力强度信息阈值区间；
    所述确定子单元，还用于将所述多个用户中影响力强度等级大于或等于预置等级的用户确定为影响力强度信息符合预置影响力强度条件的用户。
  24. 根据权利要求23所述的装置，其特征在于，所述筛选单元还包括：划分子单元，划分子单元，用于根据所述多个用户之间的关系强度信息，将所述多个用户划分为不同关系社区，所述关系社区中保存的用户之间关系强度信息大于预设阈值；
    所述确定子单元，具体用于将所述不同关系社区中影响力强度等级大于或等于预置等级的用户确定为所述目标用户。
  25. 根据权利要求23所述的装置，其特征在于，所述装置还包括：
    建立单元，用于建立不同影响力强度等级与不同影响力强度信息阈值区间之间的映射关系；
    保存单元，用于将所述映射关系保存在所述预置影响力强度等级表中。
  26. 根据权利要求24所述的装置，其特征在于，所述筛选单元还包括：获取子单元，所述获取子单元，具体用于获取所述多个用户的初始关系社区标签；
    所述划分子单元包括：
    更新模块，用于将所述多个用户中每个用户的初始关系社区标签更新为与所述用户关系强度信息大于预设阈值的用户的关系社区标签；
    划分模块，用于将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
  27. 根据权利要求26所述的装置，其特征在于，所述划分子单元还包括：获取模块和判断模块，
    所述获取模块，用于获取所述用户的关系社区标签更新次数；
    所述判断模块，用于判断所述获取模块获取的所述用户的关系社区标签更新次数是否大于预置次数阈值；
    所述划分模块，具体用于若所述判断模块判断所述用户的关系社区标签更新次数大于预置次数阈值，则将所述多个用户中关系社区标签相同的用户划分到同一关系社区中。
  28. 根据权利要求27所述的装置，其特征在于，
    所述更新模块，还用于若所述判断模块判断所述用户的关系社区标签更新次数小于或者等于预置次数阈值，则将所述用户的关系社区标签更新为与所述用户之间关系强度信息大于预设阈值的用户的关系社区标签。
PCT/CN2017/085653 2016-06-06 2017-05-24 信息推送方法及装置 WO2017211191A1 (zh)

Priority Applications (9)

Application Number Priority Date Filing Date Title
MYPI2018002358A MY194098A (en) 2016-06-06 2017-05-24 Method and device for pushing information
KR1020187037819A KR20190015373A (ko) 2016-06-06 2017-05-24 정보 푸싱 방법 및 장치
JP2018563865A JP6991163B2 (ja) 2016-06-06 2017-05-24 情報をプッシュする方法及びデバイス
EP17809630.1A EP3467754A4 (en) 2016-06-06 2017-05-24 METHOD AND DEVICE FOR SUBMITTING INFORMATION
SG11201810809XA SG11201810809XA (en) 2016-06-06 2017-05-24 Method and device for pushing information
US16/210,988 US20190108552A1 (en) 2016-06-06 2018-12-05 Method and device for pushing information
PH12018502583A PH12018502583A1 (en) 2016-06-06 2018-12-06 Method and device for pushing information
US16/721,857 US11074623B2 (en) 2016-06-06 2019-12-19 Method and device for pushing information
US17/382,130 US20210357985A1 (en) 2016-06-06 2021-07-21 Method and device for pushing information

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610395230.1 2016-06-06
CN201610395230.1A CN106909594B (zh) 2016-06-06 2016-06-06 信息推送方法及装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/210,988 Continuation US20190108552A1 (en) 2016-06-06 2018-12-05 Method and device for pushing information

Publications (1)

Publication Number Publication Date
WO2017211191A1 true WO2017211191A1 (zh) 2017-12-14

Family

ID=59206137

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/085653 WO2017211191A1 (zh) 2016-06-06 2017-05-24 信息推送方法及装置

Country Status (10)

Country Link
US (3) US20190108552A1 (zh)
EP (1) EP3467754A4 (zh)
JP (1) JP6991163B2 (zh)
KR (1) KR20190015373A (zh)
CN (1) CN106909594B (zh)
MY (1) MY194098A (zh)
PH (1) PH12018502583A1 (zh)
SG (1) SG11201810809XA (zh)
TW (1) TWI665607B (zh)
WO (1) WO2017211191A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110069700A (zh) * 2019-03-13 2019-07-30 深圳壹账通智能科技有限公司 金融社区信息推送方法、装置、介质、电子设备

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105677881B (zh) * 2016-01-12 2020-07-03 腾讯科技（深圳）有限公司 一种信息推荐方法、装置及服务器
CN106909594B (zh) 2016-06-06 2020-05-05 阿里巴巴集团控股有限公司 信息推送方法及装置
CN107679239B (zh) * 2017-10-27 2020-12-29 天津理工大学 一种基于用户行为的个性化社区推荐方法
CN109189935B (zh) * 2018-07-02 2021-01-12 北京市天元网络技术股份有限公司 一种基于知识图谱的app传播分析方法及系统
CN109309715B (zh) * 2018-09-20 2021-11-30 江苏满运软件科技有限公司 基于用户群的信息分组推送方法、系统、设备及存储介质
CN109993573A (zh) * 2019-02-19 2019-07-09 拉扎斯网络科技（上海）有限公司 信息投放方法、装置、电子设备和非易失性存储介质
CN111818114B (zh) * 2019-04-12 2023-05-30 财付通支付科技有限公司 信息分享方法和装置
CN110817636B (zh) * 2019-11-20 2021-09-21 上海电气集团股份有限公司 一种电梯门系统故障诊断方法、装置、介质和设备
JP6934040B2 (ja) * 2019-12-27 2021-09-08 株式会社ビデオリサーチ 顧客推定装置及び顧客推定方法
CN111291264B (zh) * 2020-01-23 2023-06-23 腾讯科技（深圳）有限公司 基于机器学习的访问对象预测方法、装置和计算机设备
CN111339468B (zh) * 2020-02-28 2023-08-04 北京嘀嘀无限科技发展有限公司 信息推送方法、装置、电子设备和存储介质
CN115023934A (zh) * 2020-03-05 2022-09-06 深圳市欢太科技有限公司 信息推送方法、装置、服务器及存储介质
CN111476657B (zh) * 2020-04-07 2023-07-18 度小满科技(北京)有限公司 信息推送方法、装置及系统
CN111815368B (zh) * 2020-07-24 2024-05-03 深圳市欢太科技有限公司 广告推送方法、装置、终端及存储介质
TWI808454B (zh) * 2021-07-22 2023-07-11 沛倫設計股份有限公司 廣告投放方法
KR102360969B1 (ko) * 2021-08-17 2022-02-10 (주)브이플랫폼 광고 제공 방법 및 시스템
CN117496669B (zh) * 2023-11-17 2024-09-17 成都市美幻科技有限公司 灾害预警和紧急信息的靶向发布与闭环管理的方法及装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103745105A (zh) * 2014-01-06 2014-04-23 中国科学院计算技术研究所 一种社交网络中用户属性的预测方法及系统
CN104885114A (zh) * 2012-12-04 2015-09-02 阿尔卡特朗讯 用于对通过交互相似性相联系的社区之间的信息扩散进行优化的方法和设备
CN105095256A (zh) * 2014-05-07 2015-11-25 阿里巴巴集团控股有限公司 基于用户之间相似度进行信息推送的方法及装置

Family Cites Families (39)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5872952A (en) * 1995-04-17 1999-02-16 Synopsys, Inc. Integrated circuit power net analysis through simulation
US20050228778A1 (en) * 2004-04-05 2005-10-13 International Business Machines Corporation System and method for retrieving documents based on mixture models
US20060010029A1 (en) * 2004-04-29 2006-01-12 Gross John N System & method for online advertising
US20050246391A1 (en) * 2004-04-29 2005-11-03 Gross John N System & method for monitoring web pages
GB0504568D0 (en) * 2005-03-04 2005-04-13 Vida Software S L User interfaces for electronic devices
US20060282282A1 (en) * 2005-06-03 2006-12-14 Spechler Brent A Bonus tracking system and method
US20080140506A1 (en) * 2006-12-08 2008-06-12 The Procter & Gamble Corporation Systems and methods for the identification, recruitment, and enrollment of influential members of social groups
US8438062B2 (en) * 2006-12-29 2013-05-07 Google Inc. Network node ad targeting
US9946975B2 (en) * 2007-08-24 2018-04-17 At&T Intellectual Property I, L.P. Method and apparatus to identify influencers
US20100198660A1 (en) * 2009-01-30 2010-08-05 Bank Of America Corporation Subcontractor compliance measurement
BRPI0924431A2 (pt) * 2009-03-03 2016-01-26 Google Inc modelo de anúncio adheat para rede social
BRPI0924542A2 (pt) 2009-03-03 2015-06-30 Google Inc Método e sistema para fornecimento de publicidade para usuários de rede social
US9183287B2 (en) * 2010-01-29 2015-11-10 E-Therapeutics Plc Social media analysis system
US8639692B2 (en) * 2010-06-28 2014-01-28 Alcatel Lucent High-dimensional stratified sampling
KR101297271B1 (ko) 2011-04-11 2013-08-14 에스케이씨앤씨 주식회사 소셜 네트워크 분석 서비스 방법 및 장치
WO2012166660A2 (en) * 2011-05-27 2012-12-06 Groupon Zappedy, Inc. Computing early adopters and potential influencers using transactional data and network analysis
WO2013123462A1 (en) * 2012-02-15 2013-08-22 Bottlenose, Inc. Systems and methods for recommending advertisement placement based on cross network online activity analysis
US9262791B2 (en) * 2012-03-23 2016-02-16 Facebook, Inc. Targeting stories based on influencer scores
US8566165B1 (en) * 2012-03-30 2013-10-22 Palo Alto Research Center Incorporated Performance-efficient system and method for sampling users for population-centric activity modeling
US9558277B2 (en) * 2012-04-04 2017-01-31 Salesforce.Com, Inc. Computer implemented methods and apparatus for identifying topical influence in an online social network
CN103426042B (zh) 2012-05-15 2016-04-13 腾讯科技（深圳）有限公司 社交网络的分组方法及系统
US9558273B2 (en) * 2012-09-21 2017-01-31 Appinions Inc. System and method for generating influencer scores
US20140189000A1 (en) * 2013-01-02 2014-07-03 Microsoft Corporation Social media impact assessment
US9294576B2 (en) * 2013-01-02 2016-03-22 Microsoft Technology Licensing, Llc Social media impact assessment
US9489864B2 (en) * 2013-01-07 2016-11-08 Educational Testing Service Systems and methods for an automated pronunciation assessment system for similar vowel pairs
US10559013B2 (en) 2013-03-07 2020-02-11 Facebook, Inc. Identifying users for advertising opportunities based on paired identifiers
US20140288999A1 (en) * 2013-03-12 2014-09-25 Correlor Technologies Ltd Social character recognition (scr) system
US9443193B2 (en) * 2013-04-19 2016-09-13 Educational Testing Service Systems and methods for generating automated evaluation models
KR20160079830A (ko) * 2013-10-25 2016-07-06 시소모스 엘.피. 소셜 데이터 네트워크의 인플루언서들을 가중된 분석을 이용하여 동적 결정하는 시스템 및 방법
CN104811428B (zh) * 2014-01-28 2019-04-12 阿里巴巴集团控股有限公司 利用社交关系数据验证客户端身份的方法、装置及系统
US10459767B2 (en) * 2014-03-05 2019-10-29 International Business Machines Corporation Performing data analytics utilizing a user configurable group of reusable modules
US9754503B2 (en) * 2014-03-24 2017-09-05 Educational Testing Service Systems and methods for automated scoring of a user's performance
US20160019579A1 (en) * 2014-05-21 2016-01-21 Jamie Reardon Sims Systems and methods to coordinate communication programs
US10438206B2 (en) * 2014-05-27 2019-10-08 The Toronto-Dominion Bank Systems and methods for providing merchant fraud alerts
US20160034974A1 (en) * 2014-07-31 2016-02-04 Ravi Neb Method, a system and a mobile app for providing an influencer with an offer to endorse a brand's product or service
WO2016095133A1 (en) * 2014-12-17 2016-06-23 Yahoo! Inc. Method and system for determining user interests based on a correspondence graph
CN106815732A (zh) * 2015-11-27 2017-06-09 英业达科技有限公司 计算方法及计算系统
CN105608179B (zh) * 2015-12-22 2019-03-12 百度在线网络技术（北京）有限公司 确定用户标识的关联性的方法和装置
CN106909594B (zh) 2016-06-06 2020-05-05 阿里巴巴集团控股有限公司 信息推送方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104885114A (zh) * 2012-12-04 2015-09-02 阿尔卡特朗讯 用于对通过交互相似性相联系的社区之间的信息扩散进行优化的方法和设备
CN103745105A (zh) * 2014-01-06 2014-04-23 中国科学院计算技术研究所 一种社交网络中用户属性的预测方法及系统
CN105095256A (zh) * 2014-05-07 2015-11-25 阿里巴巴集团控股有限公司 基于用户之间相似度进行信息推送的方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3467754A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110069700A (zh) * 2019-03-13 2019-07-30 深圳壹账通智能科技有限公司 金融社区信息推送方法、装置、介质、电子设备

Also Published As

Publication number Publication date
EP3467754A1 (en) 2019-04-10
CN106909594A (zh) 2017-06-30
US20190108552A1 (en) 2019-04-11
MY194098A (en) 2022-11-11
PH12018502583A1 (en) 2019-09-30
KR20190015373A (ko) 2019-02-13
TWI665607B (zh) 2019-07-11
TW201743204A (zh) 2017-12-16
CN106909594B (zh) 2020-05-05
US20200134674A1 (en) 2020-04-30
JP2019519042A (ja) 2019-07-04
SG11201810809XA (en) 2018-12-28
US11074623B2 (en) 2021-07-27
EP3467754A4 (en) 2019-04-10
US20210357985A1 (en) 2021-11-18
JP6991163B2 (ja) 2022-01-12

Similar Documents

Publication Publication Date Title
WO2017211191A1 (zh) 信息推送方法及装置
Sarstedt et al. Treating unobserved heterogeneity in PLS-SEM: A multi-method approach
CN105427129B (zh) 一种信息的投放方法及系统
US20160357870A1 (en) Entity classification and/or relationship identification
US8620718B2 (en) Industry specific brand benchmarking system based on social media strength of a brand
WO2019169978A1 (zh) 资源推荐方法及装置
CN104281622A (zh) 一种社交媒体中的信息推荐方法和装置
US9286379B2 (en) Document quality measurement
CN103761296A (zh) 移动终端用户网络行为分析方法及系统
US20190266619A1 (en) Behavior pattern search system and behavior pattern search method
JP5253519B2 (ja) スマートテキストを生成するための方法、装置、及び記憶媒体
US20150262238A1 (en) Techniques for Topic Extraction Using Targeted Message Characteristics
US20120042267A1 (en) Inferring that a message has been read
CN107918618B (zh) 数据处理方法及装置
CN112925973B (zh) 数据处理方法和装置
CN106708871B (zh) 一种社交业务特征用户的识别方法和装置
CN110969473B (zh) 一种用户标签生成方法及装置
CN110224859A (zh) 用于识别团伙的方法和系统
CN110674404A (zh) 链接信息生成方法、装置、系统、存储介质及电子设备
US20230131759A1 (en) Processing messages for value extraction
WO2018161677A1 (zh) 一种推送消息的方法、装置及存储介质
CN111787042B (zh) 用于推送信息的方法和装置
US8700756B2 (en) Systems, methods and devices for extracting and visualizing user-centric communities from emails
US20170300937A1 (en) System and method for inferring social influence networks from transactional data
US10354313B2 (en) Emphasizing communication based on past interaction related to promoted items

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: 17809630

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2018563865

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20187037819

Country of ref document: KR

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2017809630

Country of ref document: EP

Effective date: 20190107