WO2019007187A1 - 信息推送方法、装置及服务器、计算设备和存储介质 - Google Patents

信息推送方法、装置及服务器、计算设备和存储介质 Download PDF

Info

Publication number
WO2019007187A1
WO2019007187A1 PCT/CN2018/090500 CN2018090500W WO2019007187A1 WO 2019007187 A1 WO2019007187 A1 WO 2019007187A1 CN 2018090500 W CN2018090500 W CN 2018090500W WO 2019007187 A1 WO2019007187 A1 WO 2019007187A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
user
interest
users
obtaining
Prior art date
Application number
PCT/CN2018/090500
Other languages
English (en)
French (fr)
Inventor
潘岸腾
Original Assignee
广州优视网络科技有限公司
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 广州优视网络科技有限公司 filed Critical 广州优视网络科技有限公司
Publication of WO2019007187A1 publication Critical patent/WO2019007187A1/zh

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
    • 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

Definitions

  • the present invention relates to the field of computer applications, and in particular, to an information push method and apparatus, and a server, a computing device, and a storage medium.
  • an object of the present invention is to provide an information push method, apparatus, and server, which can effectively improve push accuracy.
  • an embodiment of the present invention provides an information pushing method, where the method includes: pushing, to a pre-selected user of a user group, first information that the pre-selected user has not browsed, where the first information is related to the user group. Corresponding information in the information pool associated with the interest tag; obtaining response data of the pre-selected user for the first information, and obtaining, according to the response data, the corresponding user in the user group according to the response data And transmitting the second information to the corresponding remaining users, wherein the number of the pre-selected users is less than the number of the remaining users.
  • an embodiment of the present invention further provides an information pushing device, where the device includes: a first pushing module, an obtaining module, and a second pushing module.
  • the first pushing module is configured to push the first information that the pre-selected user has not browsed to the pre-selected user of the user group, where the first information is information in an information pool associated with the interest tag corresponding to the user group.
  • an obtaining module configured to obtain response data of the pre-selected user to the first information, where the second information corresponding to the remaining users in the user group is obtained according to the response data.
  • the second pushing module is configured to push the second information to the corresponding remaining users, wherein the number of the pre-selected users is less than the number of the remaining users.
  • an embodiment of the present invention further provides a server, where the server includes a processor and a memory.
  • the memory is coupled to the processor, the memory storing instructions.
  • the server is configured to: perform, to a pre-selected user of the user group, first information that the pre-selected user has not browsed, the first information being corresponding to the user group Obtaining information in the information pool associated with the interest tag; obtaining response data of the preselected user for the first information, and obtaining, in the first information, second information corresponding to remaining users in the user group according to the response data And pushing the second information to the corresponding remaining users, wherein the number of the pre-selected users is less than the number of the remaining users.
  • an embodiment of the present invention further provides a computing device, including a processor and a memory, where executable code is stored, and when the executable code is executed by the processor, the processor is caused to execute The method of any of the preceding claims.
  • the embodiment of the present invention further provides a non-transitory machine readable storage medium, where executable code is stored, and when the executable code is executed by a processor of an electronic device, the processor is The method of any of the preceding claims is performed.
  • the information pushing method and device, the server, the computing device, and the storage medium provided by the embodiment of the present invention first push the first information that the pre-selected user has not browsed to the pre-selected user of the user group, and then according to the pre-selection The user responds to the first information, obtains the second information corresponding to the remaining users in the user group, and then sends the second information to the corresponding remaining users, where the number of pre-selected users is less than Remaining users.
  • the solution pushes the first information to the pre-selected user and explores the high-quality information according to the response data of the pre-selected user, and pushes the obtained high-quality information as the second information to the remaining users, which is beneficial to push higher quality information for the remaining users and effectively improve The accuracy of pushing information to the remaining users, thereby increasing the activity of the remaining users on the application.
  • FIG. 1 is a schematic diagram of a server interacting with a user terminal according to an embodiment of the present invention
  • FIG. 2 is a structural block diagram of a server according to an embodiment of the present invention.
  • FIG. 3 is a flowchart of a method for pushing an information according to a first embodiment of the present invention
  • FIG. 4 is a flowchart of step S120 in an information pushing method according to a first embodiment of the present invention
  • FIG. 5 is a flowchart of a method for pushing information according to a second embodiment of the present invention.
  • FIG. 6 is a flowchart of step S210 in an information pushing method according to a second embodiment of the present invention.
  • FIG. 7 is a flowchart of step S211 in an implementation manner of an information pushing method according to a second embodiment of the present invention.
  • FIG. 8 is a flowchart of step S211 in another implementation manner of an information pushing method according to a second embodiment of the present invention.
  • FIG. 9 is a block diagram showing a module of an information pushing apparatus according to a third embodiment of the present invention.
  • FIG. 10 is a block diagram of a message pushing apparatus according to a fourth embodiment of the present invention.
  • FIG. 11 is a block diagram of a computing device provided by a sixth embodiment of the present invention.
  • FIG. 1 is a schematic diagram of the server 100 interacting with the user terminal 200 according to an embodiment of the present invention.
  • the server 100 is in communication with one or more user terminals 200 over a network for data communication or interaction.
  • the server 100 can be a web server, a database server, or the like.
  • the user terminal 200 may be a personal computer (PC), a tablet computer, a smart phone, a wearable device, or the like.
  • FIG. 2 shows a block diagram of a structure applicable to a server in an embodiment of the present invention.
  • the server 100 includes a memory 110, a processor 120, and a network module 130.
  • the memory 110 can be used to store software programs and modules, such as the information pushing method and the program instructions/modules corresponding to the device in the embodiment of the present invention.
  • the processor 120 executes various functions by running software programs and modules stored in the memory 110. Application and data processing, that is, the information pushing method in the embodiment of the present invention is implemented.
  • Memory 110 may include high speed random access memory and may also include non-volatile memory such as one or more magnetic storage devices, flash memory, or other non-volatile solid state memory.
  • the software program and module in the above memory 110 may further include an operating system 111 and a service module 112.
  • the operating system 111 may be LINUX, UNIX, WINDOWS, which may include various software components and/or drivers for managing system tasks (eg, memory management, storage device control, power management, etc.), and may be various Hardware or software components communicate with one another to provide an operating environment for other software components.
  • the service module 112 runs on the basis of the operating system 111, and listens for requests from the network through the network service of the operating system 111, completes corresponding data processing according to the request, and returns the processing result to the client. That is, the service module 112 is configured to provide network services to clients.
  • the network module 130 is configured to receive and transmit network signals.
  • the above network signal may include a wireless signal or a wired signal.
  • the structure shown in Fig. 2 is merely illustrative, and the server 100 may further include more or less components than those shown in Fig. 2 or have a different configuration than that shown in Fig. 2.
  • the components shown in Figure 2 can be implemented in hardware, software, or a combination thereof.
  • the information pushing method and apparatus can be applied to other terminal devices in addition to the server 100.
  • the pushed information may be news information or video information, such as short video information.
  • the pushed information may be information other than news and video, for example, advertisement information and the like.
  • the technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
  • the components of the embodiments of the invention, which are generally described and illustrated in the figures herein, may be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the invention in the claims All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.
  • FIG. 3 is a flowchart of an information pushing method according to a first embodiment of the present invention, where the information pushing method runs in the server. Referring to FIG. 3, the method includes:
  • Step S110 Push the first information that the pre-selected user has not browsed to the pre-selected user of the user group, where the first information is information in an information pool associated with the interest tag corresponding to the user group;
  • the user group includes multiple users, and a preset number of users are selected from the user group as pre-selected users, and users other than the pre-selected users are used as remaining users.
  • the preset number can be set as needed. It should be noted that the number of pre-selected users is less than the number of remaining users.
  • the pre-selected user may be selected by randomly selecting a preset number of users from the user group as the pre-selected users. For example, 10% of users can be randomly selected in the user group as pre-selected users.
  • the user in each user group can set a corresponding number, and the user who selects the preset number range is selected as the pre-selected user.
  • the interest tag can indicate the category to which the information belongs and can be preset. For example, a description of a category of information such as military, basketball, or soccer can be used as an interest tag.
  • Different user groups correspond to different interest tags.
  • the user included in each user group and the interest tag corresponding to the user group may be obtained by querying the user list.
  • the user list may be stored in a database in the form of a list. For the specific storage manner, refer to Table 1 below.
  • Each interest tag is associated with a pool of information.
  • the information pool associated with each interest tag includes information that needs to be pushed to the user group corresponding to the interest tag.
  • the information pool may be preset, or may be used to construct information used to construct an information pool associated with the interest tag according to the number of times that all information corresponding to each interest tag is browsed within a preset time period in the preset information resource library. .
  • the information pool that has the most viewed times of the day under each interest tag in the information resource library can be selected to form an information pool.
  • the information in the information pool associated with each interest tag may be obtained by querying the information association list.
  • the information association list may be stored in a database in the form of a list. For the specific storage manner, refer to Table 2 below.
  • the foregoing user list and information association list may also be stored in a database in the form of a list shown in Table 3.
  • the information included in the information pool is the first information.
  • the foregoing manner of pushing the first information that the pre-selected user has not browsed to the pre-selected user of the user group is: randomly pushing the first information that the pre-selected user has not browsed to the pre-selected user of the user group.
  • the push of the first information can be pushed by the message.
  • the step of randomly pushing the first information that the pre-selected user has not browsed to the pre-selected user of the user group may include: assigning target information to the pre-selected user in the user group, where the target information is associated with the interest tag corresponding to the user group.
  • step S120 the response data of the pre-selected user to the first information is obtained, and the second information corresponding to the remaining users in the user group is obtained according to the response data in the first information;
  • the response data includes click behavior data of the pre-selected user for the received first information.
  • the pre-selected user selects the first information received according to the preference to view the corresponding content or ignores the received first information.
  • the click behavior data is fed back to the server.
  • the server selects the second information corresponding to the remaining users in the user group according to the click behavior data of the received first information by the pre-selected user.
  • the second information corresponding to the remaining users is the information to be pushed to each of the remaining users.
  • the foregoing step of obtaining the second information corresponding to the remaining users in the user group according to the response data includes:
  • Step S121 obtaining, according to the response data and the number of push users, a click rate of each of the first information pushed by the preselected user;
  • the number of push users of the information is the number of users to whom the information is pushed. For example, for each information in the information pool associated with the interest tag corresponding to the user group, the number of push users of the information is that the information is pushed to the user. The number of users.
  • the click behavior data of the first information received by the pre-selected user can be obtained, and then the number of pre-selected users of each first information pushed by the click can be obtained.
  • the method for obtaining the click rate of the first information may be: dividing the number of pre-selected users of the first information by the number of push users of the first information.
  • the method for obtaining the click rate of each of the first information that is pushed by the pre-selected user according to the response data and the number of the push users may be: obtaining the number of push users of each first information corresponding to the user group. Obtaining, according to the response data, the number of users of each of the first information pushed by the click; according to the number of users of each of the first information pushed by the click, and the number of push users of each of the first information, Obtaining a click rate of each of the first information pushed by the pre-selected user.
  • the information pool associated with an interest tag includes 50 pieces of information, and the number of users of the user group corresponding to the interest tag is 10,000, of which 1000 are pre-selected users, and the remaining 9000 are remaining users.
  • the first information A selected in the information pool is randomly pushed to 20 users in the pre-selected users, that is, the number of push users of the first information A is 20. Assuming that 10 of the 20 users clicked on the first information A, the first information A has a click rate of 0.5.
  • Step S122 Obtain a second corresponding to the user in the first information according to the historical behavior data of each user in the remaining users and the click rate of each of the first information pushed by the preselected user. information.
  • the click rate of the pre-selected user for each of the first information pushed reflects the pros and cons of the first information.
  • the higher the click rate the more users who are interested in the first information among the pre-selected users who receive the first information, which means that the first information meets the needs of the user. Pushing the first information that best meets the user's needs as the second information to the remaining users is beneficial to improving the accuracy of the information push.
  • the second information pushed to each of the remaining users is preferably information that the user has not viewed.
  • whether the information to be pushed to the user is information that the user has browsed may be determined according to historical behavior data of each user in the remaining users. For example, if there is browsing data of a certain information in a user's historical behavior data, the user has browsed the information.
  • the historical behavior data of each user in the remaining users and the click rate of each of the first information pushed by the pre-selected user may be the highest click rate that the user has not browsed.
  • the information serves as the second information corresponding to the user. That is to say, the second information corresponding to each user of the remaining users is the first information with the highest click rate among the first information that the user has not browsed.
  • the method for using the information with the highest click rate that the user has not browsed as the second information corresponding to the user in the embodiment may include: selecting, by the pre-selected user, the click rate of each first information that is pushed by the pre-selected user Sorting from high to low constitutes a click rate sequence; obtaining second information corresponding to each user of the remaining users according to the preset pushing rule and the click rate sequence.
  • the pushing rule may be: according to the highest to lowest order of the click rate sequence, obtaining the first information with the click rate ranked first, and determining whether the target user has browsed according to the historical behavior data of the target user.
  • the first information if the target user has not browsed the first information, the first information is used as the second information corresponding to the target user, and if the target user browses the information, the click rate is ranked next.
  • the first information until the first information that the target user has not browsed is found and the first information is used as the second information corresponding to the target user.
  • the target user is one of the remaining users.
  • the second information corresponding to each user of the remaining users may also include multiple pieces of information, for example, two or three pieces of information may be included.
  • the first information of the first two digits of the click rate that the user has not browsed may be used as the second information corresponding to the user.
  • the top two click rate refers to the first information in which the pre-selected user ranks the first and second clicks after the click rate of each first information pushed is ranked from high to low.
  • the amount of information included in the second information can be set as needed. Since pushing information to a user is an act of disturbing the user, the frequency of pushing information to the user every day needs to be controlled in consideration of the user experience.
  • the user may not have the historical behavior data of each user in the remaining users of the user group and the profit rate of each first information of the pre-selected user pushed to the user group.
  • the information with the highest rate of profit viewed is used as the second information corresponding to the user.
  • the yield of the first information pushed to the pre-selected user of the user group may be a pre-selected user's click rate of the first information multiplied by a single click of the first information.
  • the average revenue of a single click on the first piece of information may be the average traffic revenue of the first piece of information for a single click.
  • Step S130 Push the second information to the corresponding remaining users.
  • the second information corresponding to each user of the remaining users of the user group acquired in the above step S120 is pushed to the corresponding user.
  • the information pushing method provided in this embodiment pushes the first information to the pre-selected user and searches for the high-quality information according to the response data of the pre-selected user, and pushes the searched high-quality information as the second information to the remaining users, which is beneficial to pushing the remaining users for better quality.
  • the information effectively improves the accuracy of pushing information to the remaining users, thereby increasing the activity of the remaining users on the application product.
  • FIG. 5 is a flowchart of an information pushing method according to a second embodiment of the present invention, where the information pushing method runs in the server. Referring to FIG. 5, the method includes at least the following steps S210 to S250.
  • the user group since the historical behavior data of the user changes in real time, in order to improve the accuracy of the user group division and improve the accuracy of the information push, the user group may be divided into real-time according to the current historical behavior data of all users. .
  • the information pushing method provided in this embodiment further includes a user grouping step.
  • Each user is grouped by a user grouping step to obtain a plurality of user groups, and different user groups correspond to different interest tags.
  • the user grouping step includes step S210 and step S220.
  • Step S210 Obtain an interest tag corresponding to each user according to historical behavior data of each user;
  • An information resource library is stored in the server, and the information resource library includes multiple pieces of information.
  • the interest tag corresponding to the information included in the information resource library needs to be acquired first.
  • a keyword may be extracted from a title of the information by setting a plurality of standard tags in advance, and the extracted keyword is mapped to a corresponding standard tag, and the standard tag is used as an interest tag corresponding to the information.
  • the interest tag corresponding to the information may be obtained by manually marking the information by the operator.
  • the step of obtaining the interest tag corresponding to each user according to the historical behavior data of each user may include:
  • Step S211 obtaining, according to historical behavior data of each user, the degree of interest of the user for each of the interest tags;
  • the pushed information may be news information or video information.
  • it can also be information other than news and video, for example, advertising information.
  • the degree of interest is used to characterize the user's level of interest in each tag of interest.
  • the step of obtaining the user's interest level for each of the interest tags according to each user's historical behavior data may include: :
  • Step S301 obtaining, according to the historical behavior data of each user, the degree of attention of the user to each information in the information pool associated with each of the interest tags;
  • each user's historical behavior data is related browsing data of each news browsed by each user recorded by the server.
  • the manner in which the user's attention to each information in the information pool associated with each interest tag is obtained according to the historical behavior data of each user may be: obtaining the historical behavior data according to each user.
  • the browsing duration of the information, the comment index, the sharing index, and the preset parameters get the user's attention to the information.
  • the browsing duration is the length of time the user stays when browsing the news.
  • the comment index is used to indicate whether the user has commented on the news. For example, when the comment index is 0, it means that no comment is made, and when the comment index is 1, it means that there is a comment.
  • the sharing index means that users share news to other social circles (friends, WeChat friends, etc.). For example, when the sharing index is 0, it means there is no sharing, and when the sharing index is 1, it means sharing.
  • the average browsing duration of a user is the average of the user's browsing duration for all information.
  • the preset parameters may include a first parameter, a second parameter, and a third parameter.
  • the first parameter, the second parameter and the third parameter respectively correspond to the influence of the three factors of the browsing time ratio, the comment index and the sharing index on the degree of attention.
  • the first parameter, the second parameter, and the third parameter are all constants greater than 0, and the sum of the first parameter, the second parameter, and the third parameter is equal to or approximately equal to 1.
  • the specific values of the three parameters may be determined according to actual service requirements. For example, when browsing length, comment index, and sharing index are just as important, you can set all three parameters to 0.333.
  • u denotes a unique identifier of the user
  • a unique identifier of the user terminal may be adopted
  • I represents a set of all information in the preset information resource library
  • focus u i represents the degree of attention of the user u to the information i
  • Avg(viewtime u,i ) represents the average browsing duration of all the information viewed by the user u
  • iscomment u,i represents the comment index of the user u for the information i
  • isshare u,i represents the sharing index of the user u for the information i
  • the first parameter, ⁇ represents the second parameter, and ⁇ represents the third parameter.
  • the comment index and the sharing index may be disregarded, and the average browsing time of the user and the browsing duration of the user for each information are obtained.
  • the user's attention to the information Specifically, according to the formula Get the user's attention to each message.
  • Step S302 Obtain the degree of interest of the user for each of the interest tags according to the degree of attention of the user to each information in the information pool associated with each of the interest tags.
  • the implementation manner of obtaining the interest degree of the user for each interest tag according to the degree of attention of the user to each information in the information pool associated with each of the interest tags may be: obtaining the user according to the historical behavior data of the user.
  • the browsing time index for each information; the degree of interest of the user for each interest tag is obtained according to the degree of attention of the user for each information in the information pool associated with each interest tag, the browsing time index, and a preset attenuation factor.
  • the browsing time index indicates the time when the user browses the news distance. According to the user's historical behavior data, the browsing time of the news viewed by the user can be obtained.
  • the unit of the browsing time index may be days. For example, for the news that the user browsed the day, the browsing time index is taken as 1, and for the user to browse the news the previous day, the browsing time index is taken as 2, and so on. For news that the user has not viewed, the browsing time index can be represented by 0.
  • the above attenuation factor is a time attenuation factor, and the attenuation factor has a value range of [0, 1], and the specific value of the attenuation factor can be configured according to actual needs.
  • the weight of attention is For the news that the user browses the day ago, the weight of attention is For news that users browse two days ago, the weight of attention is For news that users browse three days ago, the weight of attention is ....
  • interest u, t indicates the degree of interest of the user u for the interest tag t
  • tag i, t indicates whether the information i is associated with the interest tag t
  • Viewdays u,i represents the browsing time index of the user u for the information i
  • lapse represents the attenuation factor.
  • the foregoing method for obtaining the user's interest level for each of the interest tags according to each user's historical behavior data may be used.
  • Step S401 Acquire a resolution of each interest tag corresponding to the preset information resource library
  • Resolution is a measure of the ability of a set of interest tags to pinpoint the type of information content.
  • the resolution of each of the interest tags may be obtained according to the quantity of information included in the preset information resource library and the amount of information associated with each interest tag.
  • Step S402 obtaining, according to historical behavior data of each user, a play completion rate of information associated with each of the interest tags that the user clicks to play;
  • each information associated with each interest tag in each user play information resource library is obtained.
  • Play completion rate Specifically, according to the following formula:
  • u denotes a unique identifier of the user, for example, a unique identifier of the user terminal may be adopted; i ⁇ I, I represents a collection of all information in the information resource library; precent u, i indicates that the user u clicks on the played information i is completed. Rate; playtime u, i represents the duration of the user u playing information i; videotime i represents the total duration of the information i.
  • Step S403 Obtain an interest degree of the user on the interest tag according to the resolution of each of the interest tags and the play completion rate of the information associated with the interest tag.
  • interest u, t represents the degree of interest of the user u for the interest tag t.
  • Step S212 Comparing the degree of interest of the user to each of the interest tags to obtain an interest tag corresponding to the user.
  • the interest tag with the highest interest degree may be selected according to the user's interest degree of each interest tag as the interest tag corresponding to the user. At this point, one user corresponds to an interest tag.
  • an interest tag whose interest degree exceeds a preset threshold may be selected according to the degree of interest of the user for each interest tag, and is used as the interest tag corresponding to the user.
  • the preset threshold may be set according to a specific situation to ensure that each user corresponds to at least one interest tag.
  • Step S220 The each user is grouped according to the interest tag corresponding to each user to obtain a plurality of user groups, one user corresponds to at least one user group, and different user groups correspond to different interest tags.
  • Users corresponding to the same interest tag are divided into one user group, and different user groups correspond to different interest tags. Since each user corresponds to at least one interest tag, one user is at least divided into one user group. Of course, if a user only corresponds to one interest tag, the user belongs to only one user group.
  • the grouping situation of the user and the correspondence between the user group and the interest tag may be stored in the database in the form of a list, for example, may be stored in the form of a list as shown in Table 1 or Table 3 above.
  • Step S230 the first information that is not browsed by the pre-selected user is pushed to the pre-selected user of the user group, where the first information is information in an information pool associated with the interest tag corresponding to the user group;
  • Step S240 obtaining response data of the pre-selected user to the first information, and obtaining second information corresponding to the remaining users in the user group according to the response data;
  • Step S250 Push the second information to the corresponding remaining users, where the number of the pre-selected users is less than the number of the remaining users.
  • step S230 to the step S250 is similar to the embodiment of the step S110 to the step S130 in the first embodiment, and details are not described herein again.
  • the division of the user group may be divided according to the preset time interval in addition to the real-time division manner described above.
  • the preset time interval can be set as needed.
  • the user group can be divided once according to the historical behavior data of all users every interval of one day, and the specific division method is similar to the above-mentioned user grouping step.
  • the user group may also be divided according to the interest tag selected by the user.
  • a plurality of interest tags may be preset, and each of the interest tags is associated with an information pool.
  • the user may select at least one interest tag.
  • the user information of the user includes the user.
  • the information of the selected interest tag can also modify the selected interest tag.
  • users who select the same interest tag can be divided into one user group.
  • the information pushing method provided by the embodiment obtains the interest tag corresponding to the user according to the historical behavior data of the user, and groups the user according to the interest tag corresponding to the user, and then pushes the first information to the preselected user of the user group according to the preselected user.
  • the response data explores high-quality information, and the high-quality information obtained by the exploration is pushed to the remaining users as the second information, which is beneficial to pushing higher quality information for the remaining users, effectively improving the accuracy of pushing information to the remaining users, thereby improving the remaining users. Activeness on the application.
  • a third embodiment of the present invention provides an information pushing apparatus, which is used in the server 100 to implement the information pushing method proposed in the first embodiment.
  • the information pushing device 10 includes a first pushing module 11 , an obtaining module 12 , and a second pushing module 13 .
  • the first pushing module 11 is configured to push, to the pre-selected user of the user group, the first information that the pre-selected user has not browsed, where the first information is information in an information pool associated with the interest tag corresponding to the user group. .
  • the first pushing module 11 is specifically configured to randomly push, to the pre-selected user of the user group, the first information that the pre-selected user has not browsed.
  • the first pushing module 11 may be specifically configured to allocate target information to the pre-selected users in the user group, where the target information is randomly selected from the information pool associated with the interest tags corresponding to the user group. Determining whether the target information is information that the pre-selected user has browsed, and if so, randomly selecting information from the information pool associated with the interest tag corresponding to the user group as the target information and assigning the information to the pre-selected user until the allocation
  • the target information for the pre-selected user is information that the pre-selected user has not browsed; the target information that is assigned to the pre-selected user and that has not been browsed by the pre-selected user is pushed to the pre-selected user as the first information.
  • the obtaining module 12 is configured to obtain response data of the pre-selected user to the first information, and obtain, in the first information, second information corresponding to the remaining users in the user group according to the response data.
  • the obtaining module 12 may include a click rate acquisition sub-module 121 and a second information acquiring sub-module 122 .
  • the click rate acquisition sub-module 121 is configured to obtain, according to the response data and the number of push users, a click rate of each of the first information pushed by the pre-selected user, where the number of push users of the information is The number of users who pushed this information. For example, for each information in the information pool associated with the interest tag corresponding to the user group, the number of push users of the information is the number of users to whom the information is pushed.
  • a second information acquiring sub-module configured to obtain second information corresponding to the user according to the historical behavior data of each of the remaining users and the click rate of each of the first information pushed by the pre-selected user .
  • the click rate acquisition sub-module 121 is configured to acquire the number of push users of each of the first information corresponding to the user group, and obtain, according to the response data, each of the first information that is pushed by the click. The number of users; according to the number of users of each of the first information pushed by the click and the number of push users of each of the first information, obtaining the first information that the pre-selected user pushes each of the first information Clickthrough rate.
  • the second information obtaining sub-module 122 is configured to: according to the historical behavior data of each user of the remaining users and the click rate of each of the first information pushed by the pre-selected user, the user has not browsed The first information with the highest click rate is used as the second information corresponding to the user.
  • the second pushing module 13 is configured to push the second information to the corresponding remaining users, wherein the number of the pre-selected users is less than the number of the remaining users.
  • each module may be implemented by software code.
  • each module described above may be stored in the memory 110 of the server 100.
  • the above modules can also be implemented by hardware such as an integrated circuit chip.
  • a fourth embodiment of the present invention provides an information pushing apparatus, which is used in the server 100 to implement the information pushing method proposed in the second embodiment.
  • the information pushing device 20 includes a first acquiring module 21, a grouping module 22, a first pushing module 23, a second acquiring module 24, and a second pushing module 25.
  • the first obtaining module 21 is configured to obtain, according to historical behavior data of each user, an interest tag corresponding to each user.
  • the grouping module 22 is configured to group each user according to the interest tag corresponding to each user to obtain a plurality of user groups, one user corresponding to at least one user group, and different user groups corresponding to different interest tags ;
  • the first pushing module 23 is configured to: push the first information that the pre-selected user has not browsed to the pre-selected user of the user group, where the first information is information in an information pool associated with the interest tag corresponding to the user group;
  • the second obtaining module 24 is configured to obtain response data of the pre-selected user to the first information, and obtain second information corresponding to the remaining users in the user group according to the response data;
  • the second pushing module 25 is configured to push the second information to the corresponding remaining users, wherein the number of the pre-selected users is less than the number of the remaining users.
  • the first obtaining module 21 is configured to obtain, according to the historical behavior data of each user, the degree of interest of the user for each of the interest tags; and compare the interest of the user to each of the interest tags. The user's corresponding interest tag.
  • each module may be implemented by software code.
  • each module described above may be stored in the memory 110 of the server 100.
  • the above modules can also be implemented by hardware such as an integrated circuit chip.
  • a fifth embodiment of the present invention provides a server, the server including a memory and a processor, the memory coupled to the processor, the memory storing instructions. When the instructions are executed by the processor to cause the processor to perform the following operations:
  • FIG. 11 is a block diagram showing the structure of a computing device that can be used to implement data processing of the above information pushing method according to an embodiment of the present invention.
  • computing device 1000 includes a memory 1010 and a processor 1020.
  • the processor 1020 can be a multi-core processor or can include multiple processors.
  • processor 1020 can include a general purpose main processor and one or more special coprocessors, such as a graphics processing unit (GPU), a digital signal processor (DSP), and the like.
  • the processor 1020 can be implemented using a customized circuit, such as an Application Specific Integrated Circuit (ASIC) or Field Programmable Gate Array (FPGA).
  • ASIC Application Specific Integrated Circuit
  • FPGA Field Programmable Gate Array
  • Memory 1010 can include various types of storage units, such as system memory, read only memory (ROM), and persistent storage.
  • the ROM can store static data or instructions required by the processor 1020 or other modules of the computer.
  • the persistent storage device can be a readable and writable storage device.
  • the persistent storage device may be a non-volatile storage device that does not lose stored instructions and data even after the computer is powered off.
  • the persistent storage device employs a mass storage device (eg, magnetic or optical disk, flash memory) as the permanent storage device.
  • the persistent storage device can be a removable storage device (eg, a floppy disk, an optical drive).
  • the system memory can be a readable and writable storage device or a volatile read/write storage device, such as dynamic random access memory.
  • System memory can store instructions and data that some or all of the processors need at runtime.
  • the memory 1010 may comprise any combination of computer readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read only memory), and magnetic disks and/or optical disks may also be employed.
  • the memory 1010 can include a removable storage device readable and/or writable, such as a compact disc (CD), a read-only digital versatile disc (eg, a DVD-ROM, a dual layer DVD-ROM), Read-only Blu-ray discs, ultra-density discs, flash cards (such as SD cards, min SD cards, Micro-SD cards, etc.), magnetic floppy disks, and so on.
  • a removable storage device readable and/or writable such as a compact disc (CD), a read-only digital versatile disc (eg, a DVD-ROM, a dual layer DVD-ROM), Read-only Blu-ray discs, ultra-density discs, flash cards (such as SD cards, min SD cards, Micro-SD cards, etc.), magnetic floppy disks, and so on.
  • the computer readable storage medium does not include a carrier wave and an instantaneous electronic signal transmitted by wireless or wire.
  • a processable code is stored on the memory 1010, and when the processable code is processed by the processor 1020, the processor 1020 can be caused to perform the information push method described above.
  • each block of the flowchart or block diagram can represent a module, a program segment, or a portion of code that includes one or more of the Executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may also occur in a different order than those illustrated in the drawings.
  • each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts can be implemented in a dedicated hardware-based system that performs the specified function or action. Or it can be implemented by a combination of dedicated hardware and computer instructions.
  • each functional module in each embodiment of the present invention may be integrated to form a separate part, or each module may exist separately, or two or more modules may be integrated to form a separate part.
  • the functions, if implemented in the form of software functional modules and sold or used as separate products, may be stored in a computer readable storage medium.
  • the technical solution of the present invention which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .
  • the present invention may also be embodied as a non-transitory machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) having executable code (or computer program, or computer instruction code) stored thereon.
  • executable code or computer program, or computer instruction code
  • the processor is caused to perform the various steps of the above described method in accordance with the present invention.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

本发明提供了一种信息推送方法、装置及服务器、计算设备和存储介质，属于计算机应用领域。该信息推送方法包括：向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；获取所述预选用户对所述第一信息的响应数据，根据所述响应数据在所述第一信息中获得所述用户组中剩余用户对应的第二信息；将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。本发明提供的信息推送方法有利于为剩余用户推送更优质的信息，有效地提高了向剩余用户推送信息的准确性，从而提升剩余用户在应用产品上的活跃度。

Description

信息推送方法、装置及服务器、计算设备和存储介质 技术领域
本发明涉及计算机应用领域，具体而言，涉及一种信息推送方法、装置及服务器、计算设备和存储介质。
背景技术
在视频或新闻等信息的应用运营过程中，主动向用户推荐视频或新闻等信息是提升用户活跃度的重要手段。现有推荐方案一般是由运营人员圈定不同人群，然后对用户推送不同的视频或新闻等信息。然而，这种推送方案推送的信息往往不符合用户的需求，推送准确性较低。
发明内容
有鉴于此，本发明的目的在于提供一种信息推送方法、装置及服务器，能够有效地提高推送准确性。
为了实现上述目的，本发明采用的技术方案如下：
第一方面，本发明实施例提供了一种信息推送方法，所述方法包括：向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；获取所述预选用户对所述第一信息的响应数据，根据所述响应数据在所述第一信息中获得所述用户组中剩余用户对应的第二信息；将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
第二方面，本发明实施例还提供了一种信息推送装置，所述装置包括：第一推送模块、获取模块以及第二推送模块。其中，第一推送模块，用于 向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息。获取模块，用于获取所述预选用户对所述第一信息的响应数据，在所述第一信息中根据所述响应数据获得所述用户组中剩余用户对应的第二信息。第二推送模块，用于将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
第三方面，本发明实施例还提供了一种服务器，所述服务器包括处理器以及存储器。所述存储器耦接到所述处理器，所述存储器存储指令。当所述指令由所述处理器执行时使所述服务器执行以下操作：向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；获取所述预选用户对所述第一信息的响应数据，在所述第一信息中根据所述响应数据获得所述用户组中剩余用户对应的第二信息；将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
第四方面，本发明实施例还提供了一种计算设备，包括处理器以及存储器，其上存储有可执行代码，当所述可执行代码被所述处理器执行时，使所述处理器执行如上任何一项所述的方法。
第五方面，本发明实施例还提供了一种非暂时性机器可读存储介质，其上存储有可执行代码，当所述可执行代码被电子设备的处理器执行时，使所述处理器执行如上任一项所述的方法。
相比于现有技术，本发明实施例提供的信息推送方法、装置及服务器、计算设备和存储介质，通过先向用户组的预选用户推送该预选用户没有浏览过的第一信息，然后根据预选用户对第一信息的响应数据，在所述第一信息中得到该用户组中剩余用户对应的第二信息，再将该第二信息推送给对应的剩余用户，其中，预选用户的数量少于剩余用户。本方案通过向预选用户推送第一信息并根据预选用户的响应数据探索优质信息，将探索得 到的优质信息作为第二信息推送给剩余用户，有利于为剩余用户推送更优质的信息，有效地提高了向剩余用户推送信息的准确性，从而提升剩余用户在应用产品上的活跃度。
为使本发明的上述目的、特征和优点能更明显易懂，下文特举较佳实施例，并配合所附附图，作详细说明如下。
附图说明
为了更清楚地说明本发明实施例的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，应当理解，以下附图仅示出了本发明的某些实施例，因此不应被看作是对范围的限定，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相关的附图。
图1示出了本发明实施例提供的一种服务器与用户终端进行交互的示意图；
图2示出了本发明实施例提供的一种服务器的结构框图；
图3示出了本发明第一实施例提供的一种信息推送方法的方法流程图；
图4示出了本发明第一实施例提供的一种信息推送方法中步骤S120的流程图；
图5示出了本发明第二实施例提供的一种信息推送方法的方法流程图；
图6示出了本发明第二实施例提供的一种信息推送方法中步骤S210的流程图；
图7示出了本发明第二实施例提供的一种信息推送方法的一种实施方式中步骤S211的流程图；
图8示出了本发明第二实施例提供的一种信息推送方法的另一种实施方式中步骤S211的流程图；
图9示出了本发明第三实施例提供的一种信息推送装置的模块框图；
图10示出了本发明第四实施例提供的一种信息推送装置的模块框图。
图11示出了本发明第六实施例提供的一种计算设备的模块框图。
具体实施方式
图1示出了本发明实施例提供的服务器100与用户终端200进行交互的示意图。所述服务器100通过网络与一个或多个用户终端200进行通信连接，以进行数据通信或交互。所述服务器100可以是网络服务器、数据库服务器等。所述用户终端200可以是个人电脑(personal computer，PC)、平板电脑、智能手机、穿戴设备等。
图2示出了一种可应用于本发明实施例中的服务器的结构框图。如图2所示，服务器100包括：存储器110、处理器120以及网络模块130。
存储器110可用于存储软件程序以及模块，如本发明实施例中的信息推送方法及装置对应的程序指令/模块，处理器120通过运行存储在存储器110内的软件程序以及模块，从而执行各种功能应用以及数据处理，即实现本发明实施例中的信息推送方法。存储器110可包括高速随机存储器，还可包括非易失性存储器，如一个或者多个磁性存储装置、闪存、或者其他非易失性固态存储器。进一步地，上述存储器110内的软件程序以及模块还可包括：操作系统111以及服务模块112。其中操作系统111，例如可为LINUX、UNIX、WINDOWS，其可包括各种用于管理系统任务(例如内存管理、存储设备控制、电源管理等)的软件组件和/或驱动，并可与各种硬件或软件组件相互通讯，从而提供其他软件组件的运行环境。服务模块112运行在操作系统111的基础上，并通过操作系统111的网络服务监听来自网络的请求，根据请求完成相应的数据处理，并返回处理结果给客户端。也就是说，服务模块112用于向客户端提供网络服务。
网络模块130用于接收以及发送网络信号。上述网络信号可包括无线信号或者有线信号。
可以理解，图2所示的结构仅为示意，服务器100还可包括比图2中 所示更多或者更少的组件，或者具有与图2所示不同的配置。图2中所示的各组件可以采用硬件、软件或其组合实现。
需要说明的是，本发明实施例提供的信息推送方法及装置除了应用于服务器100外，还可以应用于其它终端设备。本实施例中，推送的信息可以是新闻信息，也可以为视频信息，如短视频信息。当然，也可以是除新闻和视频以外的信息，例如，广告信息等。下面将结合本发明实施例中附图，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。通常在此处附图中描述和示出的本发明实施例的组件可以以各种不同的配置来布置和设计。因此，以下对在附图中提供的本发明的实施例的详细描述并非旨在限制要求保护的本发明的范围，而是仅仅表示本发明的选定实施例。基于本发明的实施例，本领域技术人员在没有做出创造性劳动的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。
第一实施例
图3示出了本发明第一实施例提供的一种信息推送方法的流程图，该信息推送方法运行于上述服务器中。请参阅图3，所述方法包括：
步骤S110，向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；
本实施例中，用户组包括多个用户，从用户组中选取预设数量的用户作为预选用户，除预选用户以外的用户作为剩余用户。其中，预设数量可以根据需要设置。需要说明的是，预选用户的数量少于剩余用户的数量。
具体的，预选用户的选取方式可以为：从用户组中随机抽取预设数量的用户作为预选用户。例如，可以在用户组中随机抽取10％的用户作为预选用户。或者，每个用户组中的用户均可以设置相应的编号，选取预设编号范围的用户作为预选用户。
兴趣标签可以表示信息所属的类别，可以预先设定。例如，可以将军 事、篮球、足球等表示信息类别的描述作为兴趣标签。不同用户组对应于不同的兴趣标签。本实施例中，可以通过查询用户列表来获取每个用户组包括的用户以及该用户组对应的兴趣标签。作为一种实施方式，该用户列表可以以列表的形式存储于数据库中，具体存储方式请参见下表1。
表1
用户组 兴趣标签
用户n1、用户n2、用户n3、… 军事
用户m1、用户m2、用户m3、… 篮球
… …
每个兴趣标签均关联有信息池。每个兴趣标签关联的信息池包括需要推送给与该兴趣标签对应的用户组的信息。信息池可以预先设置，也可以根据预设的信息资源库中，每个兴趣标签对应的所有信息在预设时间段内被浏览的次数，获取用于构建与该兴趣标签关联的信息池的信息。例如，可以选取信息资源库中每个兴趣标签下的当天被浏览次数最多的100个信息构成信息池。
本实施例中，可以通过查询信息关联列表来获取每个兴趣标签关联的信息池中的信息。作为一种实施方式，该用信息关联列表可以以列表的形式存储于数据库中，具体存储方式请参见下表2。
表2
兴趣标签 信息池
军事 军事类信息1、军事类信息2、军事类信息3、…
篮球 篮球类信息1、篮球类信息2、篮球类信息3、…
… …
或者，作为另一种实施方式，上述用户列表和信息关联列表还可以以表3所示的列表形式存储于数据库中。
表3
用户组 兴趣标签 信息池
用户n1、用户n2、… 军事 军事类信息1、军事类信息2、军事类信息3、…
用户m1、用户m2、… 篮球 篮球类信息1、篮球类信息2、篮球类信息3、…
… … …
信息池中包括的信息为第一信息。本实施例中，上述的向用户组的预选用户推送该预选用户没有浏览过的第一信息的方式为：向用户组的预选用户随机推送该预选用户没有浏览过的第一信息。本实施例中，第一信息的推送可以通过消息推送。
具体的，向用户组的预选用户随机推送该预选用户没有浏览过的第一信息的步骤，可以包括：为用户组中的预选用户分配目标信息，目标信息为从该用户组对应的兴趣标签关联的信息池中随机选取的信息；判断目标信息是否为该预选用户浏览过的信息，若该目标信息为该预选用户浏览过的信息，则重新从该用户组对应的兴趣标签关联的信息池中随机选取信息作为目标信息并分配给该预选用户，直至分配给该预选用户的目标信息为该预选用户没有浏览过的信息；将分配给所述预选用户且该预选用户没有浏览过的目标信息作为所述第一信息推送给该预选用户。
步骤S120，获取所述预选用户对所述第一信息的响应数据，在所述第一信息中根据所述响应数据获得所述用户组中剩余用户对应的第二信息；
响应数据包括预选用户对接收到的第一信息的点击行为数据。预选用户通过用户终端接收到第一信息后，会根据自己的喜好选择点击接收到的第一信息查看相应内容或者是忽视接收到的第一信息。用户通过用户终端点击接收到的第一信息后会形成点击行为数据反馈给服务器。服务器接收到预选用户的点击行为数据后，根据预选用户对接收到的第一信息的点击行为数据，在第一信息中选取该用户组中剩余用户对应的第二信息。剩余用户对应的第二信息也就是要推送给剩余用户中每个用户的信息。
本实施例中，如图4所示，上述的根据响应数据获得用户组中剩余用户对应的第二信息的步骤包括：
步骤S121，根据所述响应数据以及推送用户数量获取所述预选用户对所推送的每个所述第一信息的点击率；
其中，信息的推送用户数量为向其推送了该信息的用户的数量，例如对于用户组对应的兴趣标签关联的信息池中的每个信息，该信息的推送用户数量为向其推送了该信息的用户的数量。根据预选用户反馈的响应数据可以得到预选用户对接收到的第一信息的点击行为数据，进而可以得到点击所推送的每个第一信息的预选用户的数量。第一信息的点击率的获取方式可以为：点击该第一信息的预选用户的数量除以该第一信息的推送用户数量。
具体的，根据所述响应数据以及推送用户数量获取所述预选用户对所推送的每个所述第一信息的点击率的方法可以为：获取用户组对应的每个第一信息的推送用户数量；根据所述响应数据获取点击所推送的每个所述第一信息的用户数量；根据点击所推送的每个所述第一信息的用户数量以及每个所述第一信息的推送用户数量，得到所述预选用户对所推送的每个所述第一信息的点击率。
例如，某兴趣标签关联的信息池中包括50个信息，与该兴趣标签对应的用户组的用户人数为10000人，其中1000人为预选用户，其余的9000人则为剩余用户。该信息池中选取的某第一信息A随机推送给了预选用户中的20个用户，即该第一信息A的推送用户数量为20。假设这20个用户中有10个用户点击了该第一信息A，此时该第一信息A的点击率为0.5。
步骤S122，根据所述剩余用户中每个用户的历史行为数据以及所述预选用户对所推送的每个所述第一信息的点击率，在所述第一信息中得到该用户对应的第二信息。
预选用户对所推送的每个第一信息的点击率反映了该第一信息的优劣性。点击率越高表示接收到第一信息的预选用户中对该第一信息感兴趣的用户越多，也就说明该第一信息越符合用户的需求。将最符合用户需求的第一信息作为第二信息推送给剩余用户有利于提高信息推送的准确性。
此外，可以理解的是，推送给剩余用户中每个用户的第二信息优选为 该用户没有浏览过的信息。本实施例中，可以根据剩余用户中每个用户的历史行为数据判定待推送给该用户的信息是否为该用户浏览过的信息。例如，若某用户的历史行为数据中存在某信息的浏览数据，则说明该用户已浏览过该信息。
因此，本实施例中，可以根据所述剩余用户中每个用户的历史行为数据以及所述预选用户对所推送的每个第一信息的点击率，将该用户没有浏览过的点击率最高的信息作为该用户对应的第二信息。也就是说，剩余用户中每个用户对应的第二信息为该用户没有浏览过的第一信息中点击率最高的第一信息。
具体的，本实施例中将该用户没有浏览过的所述点击率最高的信息作为该用户对应的第二信息的方法可以包括：将预选用户对所推送的每个第一信息的点击率由高到低进行排序，构成点击率序列；根据预设推送规则以及该点击率序列获取剩余用户中的每个用户对应的第二信息。
本实施例中，上述推送规则可以为：按照点击率序列的由高到低的排序，获取点击率排在第一位的第一信息，根据目标用户的历史行为数据判断该目标用户是否浏览过该第一信息，若该目标用户没有浏览过该第一信息，则将该第一信息作为该目标用户对应的第二信息，若该目标用户浏览过该信息，获取点击率排在下一位的第一信息，直至找到该目标用户未浏览过的第一信息并将该第一信息作为该目标用户对应的第二信息。其中，目标用户为剩余用户中的其中一个用户。
当然，在本发明的其他实施例中，剩余用户中每个用户对应的第二信息也可以包括多个信息，例如，可以包括两个或三个信息。当每个用户对应的第二信息为两个信息时，也可以将该用户没有浏览过的点击率排在前两位的第一信息作为该用户对应的第二信息。其中，点击率排在前两位是指将预选用户对所推送的每个第一信息的点击率由高到低进行排序后，点击率排在第一位和第二位的第一信息。需要说明的是，第二信息所包括的 信息数量可以根据需要设置。由于给用户推送信息是打扰用户的一种行为，因此考虑到用户体验，每天给用户推送信息的频率需要有所控制。
另外，在本发明的其他实施方式中，也可以根据用户组的剩余用户中每个用户的历史行为数据以及推送给该用户组的预选用户的每个第一信息的收益率，将该用户没有浏览过的收益率最高的信息作为该用户对应的第二信息。其中，推送给该用户组的预选用户的第一信息的收益率可以为预选用户对该第一信息的点击率乘以单次点击该第一信息的平均收益。例如，单次点击该第一信息的平均收益可以为单次点击该第一信息的平均流量收益。
步骤S130，将所述第二信息推送给对应的剩余用户。
将通过上述步骤S120获取到的用户组的剩余用户中每个用户对应的第二信息推送给相应的用户。
本实施例提供的信息推送方法通过向预选用户推送第一信息并根据预选用户的响应数据探索优质信息，将探索得到的优质信息作为第二信息推送给剩余用户，有利于为剩余用户推送更优质的信息，有效地提高了向剩余用户推送信息的准确性，从而提升剩余用户在应用产品上的活跃度。
第二实施例
图5示出了本发明第二实施例提供的一种信息推送方法的流程图，该信息推送方法运行于上述服务器中。请参阅图5，所述方法至少包括以下步骤S210至步骤S250。
本实施例中，由于用户的历史行为数据是实时变化的，因此为了提高用户组划分的准确性，以提高信息推送的准确性，用户组可以为服务器根据所有用户当前的历史行为数据实时划分的。
此时，在向用户组的预选用户推送该预选用户没有浏览过的第一信息之前，本实施例提供的信息推送方法还包括用户分组步骤。通过用户分组步骤对每个用户进行分组以获得多个用户组，且不同用户组对应于不同的 兴趣标签。本实施例中，上述用户分组步骤包括步骤S210和步骤S220。
步骤S210，根据每个用户的历史行为数据得到所述每个用户对应的兴趣标签；
服务器中存储有信息资源库，信息资源库中包括多个信息。本实施例中，获取每个用户对应的兴趣标签之前，需要先获取信息资源库中包括的信息对应的兴趣标签。
信息对应的兴趣标签的获取途径很多。例如：可以通过预先设置多个标准标签，从信息的标题中提取关键字，将所提取的关键词映射到相应的标准标签，将该标准标签作为该信息对应的兴趣标签。又例如，可以通过运营人员人工对信息打上标签等方式获取信息对应的兴趣标签。
作为一种实施方式，请参阅图6，根据每个用户的历史行为数据得到所述每个用户对应的兴趣标签的步骤可以包括：
步骤S211，根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签的兴趣度；
本实施例中，推送的信息可以是新闻信息，也可以为视频信息。当然，也可以是除新闻和视频以外的信息，例如，广告信息。
兴趣度用于表征用户对每个兴趣标签的感兴趣程度。作为一种具体实施方式，请参阅图7，当推送的信息为新闻信息时，上述的根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签的兴趣度的步骤可以包括：
步骤S301，根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签关联的信息池中的每个信息的关注度；
当推送的信息为新闻信息时，每个用户的历史行为数据为服务器记录的每个用户浏览的每一条新闻的相关浏览数据。作为一种实施方式，上述根据每个用户的历史行为数据得到该用户对每个兴趣标签关联的信息池中的每个信息的关注度的方式可以为：根据每个用户的历史行为数据得到该 用户对每个兴趣标签关联的信息池中每个信息的浏览时长、评论指数以及分享指数；根据用户对所有信息的浏览时长得到该用户的平均浏览时长；根据用户的平均浏览时长、用户对每个信息的浏览时长、评论指数、分享指数以及预设参数得到该用户对该信息的关注度。
其中，浏览时长为用户浏览新闻时的停留时长。评论指数用于表示用户是否对新闻发表评论。例如，当评论指数为0时表示没有发表评论，当评论指数为1时表示有发表评论。分享指数表示用户是把新闻分享到其他社交圈(朋友圈、微信好友等)。例如，当分享指数为0时表示没有分享，当分享指数为1时表示有分享。用户的平均浏览时长为用户对所有信息的浏览时长的平均值。
本实施例中，预设参数可以包括第一参数、第二参数和第三参数。第一参数、第二参数和第三参数分别对应于浏览时长占比、评论指数以及分享指数这三种因素对关注度的影响占比。其中，第一参数、第二参数和第三参数均为大于0的常数，且第一参数、第二参数和第三参数之和等于或近似等于1。本实施例中，这三个参数的具体值可以根据实际业务需要确定。例如，当浏览时长占比、评论指数以及分享指数一样重要时，可以将三个参数均设置为0.333。
具体的，可以根据以下公式：
Figure PCTCN2018090500-appb-000001
计算得到用户对每个信息的关注度。其中，u表示用户唯一标识，例如，可以采用用户终端的唯一身份标识；i∈I，I表示预设信息资源库中所有信息的集合；focus u,i表示用户u对信息i的关注度；avg(viewtime u,i)表示用户u浏览过的所有信息的平均浏览时长；iscomment u,i表示用户u对信息i的评论指数；isshare u,i表示用户u对信息i的分享指数；α表示第一参数，β表 示第二参数，γ表示第三参数。
当然，当推送的信息为新闻信息时，为了简化计算量，作为另一种实施方式，也可以不考虑评论指数和分享指数，根据用户的平均浏览时长以及用户对每个信息的浏览时长得到该用户对该信息的关注度。具体的，可以根据公式
Figure PCTCN2018090500-appb-000002
得到用户对每个信息的关注度。
步骤S302，根据所述用户对每个所述兴趣标签关联的信息池中的每个信息的关注度得到该用户对每个所述兴趣标签的兴趣度。
具体的，上述的根据用户对每个兴趣标签关联的信息池中的每个信息的关注度得到该用户对每个兴趣标签的兴趣度的实施方式可以为：根据用户的历史行为数据得到该用户对每个信息的浏览时间指数；根据用户对每个兴趣标签关联的信息池中的每个信息的关注度、浏览时间指数以及预设的衰减因子得到该用户对每个兴趣标签的兴趣度。
其中，浏览时间指数表示用户浏览新闻距离当前的时间。根据用户的历史行为数据可以得到用户浏览过的新闻的浏览时间。本实施例中，浏览时间指数的单位可以为天，例如：对于用户当天浏览过的新闻，浏览时间指数则取1，对于用户前一天浏览过新闻，浏览时间指数则取2，以此类推。而对于用户没有浏览过的新闻，浏览时间指数可以用0表示。
此外，上述的衰减因子为时间衰减因子，衰减因子的取值范围为[0，1]，衰减因子的具体值可以根据实际需要配置。衰减因子配置的越大表示越重视用户最近浏览新闻的行为，取0时表示对用户所有时间的浏览行为都同等对待。实际业务中通常可以取1，此时，对于用户当天浏览的新闻，关注度的权重为
Figure PCTCN2018090500-appb-000003
对于用户一天前浏览的新闻，关注度的权重为
Figure PCTCN2018090500-appb-000004
对于用户两天前浏览的新闻，关注度的权重为
Figure PCTCN2018090500-appb-000005
对于用户三天前浏览的新闻，关注 度的权重为
Figure PCTCN2018090500-appb-000006
……。
具体的，可以根据以下公式：
Figure PCTCN2018090500-appb-000007
计算得到用户对每个兴趣标签的兴趣度。其中，interest u,t表示用户u对兴趣标签t的兴趣度；tag i,t表示信息i是否与兴趣标签t关联；tag i,t＝0表示不关联，tag i,t＝1表示关联；viewdays u,i表示用户u对信息i的浏览时间指数；lapse表示所述衰减因子。
作为另一种具体实施方式，请参阅图8，当推送的信息为视频信息时，上述的根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签的兴趣度的方法可以包括：
步骤S401，获取预设信息资源库对应的各个兴趣标签的分辨率；
分辨率用于衡量所设置的兴趣标签精准定位信息内容的类型的能力。本实施例中，可以根据预设的信息资源库包括的信息数量以及与每个兴趣标签关联的信息数量得到每个所述兴趣标签的分辨率。
具体的，本实施例中可以根据以下公式：
Figure PCTCN2018090500-appb-000008
计算得到每个兴趣标签的分辨率。其中，t表示兴趣标签的标识，rp t表示兴趣标签t的分辨率，I表示所述预设信息资源库中所有信息的集合，i∈I，|I|表示信息资源库中包括的信息的数量；tag i,t表示信息i是否与兴趣标签t关联，tag i,t＝0表示不关联，tag i,t＝1表示关联。
步骤S402，根据每个所述用户的历史行为数据得到该用户点击播放过的每个所述兴趣标签所关联的信息的播放完成率；
本实施例中，根据每个用户播放信息资源库中每个兴趣标签所关联的 每个信息的时长以及该信息的总时长，得到该用户点击播放过的每个兴趣标签所关联的每个信息的播放完成率。具体的，根据以下公式：
Figure PCTCN2018090500-appb-000009
计算得到用户点击播放过的每个兴趣标签所关联的信息的播放完成率。其中，u表示用户唯一标识，例如，可以采用用户终端的唯一身份标识；i∈I，I表示信息资源库中所有信息的集合；precent u,i表示用户u点击播放过的信息i的播放完成率；playtime u,i表示用户u播放信息i的时长；videotime i表示信息i的总时长。
步骤S403，根据每个所述兴趣标签的分辨率以及该兴趣标签所关联的信息的播放完成率得到该用户对该兴趣标签的兴趣度。
具体的，本实施例中，可以根据以下公式：
Figure PCTCN2018090500-appb-000010
得到用户对每个兴趣标签的兴趣度。其中，interest u,t表示用户u对兴趣标签t的兴趣度。
步骤S212，比较所述用户对各个所述兴趣标签的兴趣度得到该用户对应的兴趣标签。
本实施例中，作为一种实施方式，可以根据用户对各个兴趣标签的兴趣度大小选择兴趣度最大的兴趣标签，作为该用户对应的兴趣标签。此时，一个用户对应于一个兴趣标签。
作为另一种实施方式，可以根据用户对各个兴趣标签的兴趣度大小选择兴趣度超过预设阈值的兴趣标签，均作为该用户对应的兴趣标签。其中，预设阈值可以根据具体情况设置，以保证每个用户至少对应于一个兴趣标签。
步骤S220，根据所述每个用户对应的兴趣标签对所述每个用户进行分 组以获得多个用户组，一个用户至少对应于一个用户组，且不同用户组对应于不同的兴趣标签。
将对应于同一个兴趣标签的用户划分为一个用户组，不同用户组对应于不同的兴趣标签。由于每个用户至少对应于一个兴趣标签，因此一个用户至少被划分到一个用户组。当然，若一个用户仅对应于一个兴趣标签，则该用户仅属于一个用户组。
本实施例中，可以将用户的分组情况以及用户组与兴趣标签的对应关系可以以列表的形式存储于数据库中，例如，可以采用上述表1或表3所示的列表形式存储。
步骤S230，向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；
步骤S240，获取所述预选用户对所述第一信息的响应数据，根据所述响应数据获得所述用户组中剩余用户对应的第二信息；
步骤S250，将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
本实施例中，步骤S230至步骤S250的实施方式与第一实施例中步骤S110至步骤S130的实施方式相似，此处不再赘述。
当然，在本发明的其他实施例中，用户组的划分除了上述的实时划分方式外，还可以按照预设的时间间隔划分。其中，预设的时间间隔可以根据需要设置。例如，可以每间隔1天根据所有用户的历史行为数据划分一次用户组，具体的划分方法与上述的用户分组步骤类似。
或者，在本发明的其他实施例中，用户组也可以根据用户所选择兴趣标签划分。例如，可以预先设置多个兴趣标签，每个兴趣标签关联有信息池，用户通过用户终端首次使用应用软件时，可以选择至少一个兴趣标签，此时，用户的用户信息中则包括了该用户所选择的兴趣标签的信息。当然，用户后续也可以修改所选择的兴趣标签。进而，可以将选择同一兴趣标签 的用户划分为一个用户组。
本实施例提供的信息推送方法根据用户的历史行为数据得到用户对应的兴趣标签，并根据用户对应的兴趣标签对用户进行分组，然后，通过向用户组的预选用户推送第一信息并根据预选用户的响应数据探索优质信息，将探索得到的优质信息作为第二信息推送给剩余用户，有利于为剩余用户推送更优质的信息，有效地提高了向剩余用户推送信息的准确性，从而提升剩余用户在应用产品上的活跃度。
第三实施例
请参阅图9，本发明第三实施例提供了一种信息推送装置，运行于服务器100，用于实现上述第一实施例提出的信息推送方法。所述信息推送装置10包括：第一推送模块11、获取模块12以及第二推送模块13。
其中，第一推送模块11，用于向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息。
作为一种实施方式，第一推送模块11具体用于向用户组的预选用户随机推送该预选用户没有浏览过的第一信息。
作为另一种实施方式，第一推送模块11还可以具体用于为所述用户组中的预选用户分配目标信息，所述目标信息为从该用户组对应的兴趣标签关联的信息池中随机选取的信息；判断所述目标信息是否为该预选用户浏览过的信息，若是，则重新从该用户组对应的兴趣标签关联的信息池中随机选取信息作为目标信息并分配给该预选用户，直至分配给该预选用户的目标信息为该预选用户没有浏览过的信息；将分配给所述预选用户且该预选用户没有浏览过的目标信息作为所述第一信息推送给该预选用户。
获取模块12，用于获取所述预选用户对所述第一信息的响应数据，在所述第一信息中根据所述响应数据获得所述用户组中剩余用户对应的第二信息。
具体的，如图9所示，获取模块12可以包括点击率获取子模块121和第二信息获取子模块122。
其中，点击率获取子模块121，用于根据所述响应数据以及推送用户数量获取所述预选用户对所推送的每个所述第一信息的点击率，其中，信息的推送用户数量为向其推送了该信息的用户的数量。例如，对于用户组对应的兴趣标签关联的信息池中的每个信息，该信息的推送用户数量为向其推送了该信息的用户的数量。第二信息获取子模块，用于根据所述剩余用户中每个用户的历史行为数据以及所述预选用户对所推送的每个所述第一信息的点击率，得到该用户对应的第二信息。
具体的，点击率获取子模块121，具体用于获取所述用户组对应的每个所述第一信息的推送用户数量；根据所述响应数据获取点击所推送的每个所述第一信息的用户数量；根据所述点击所推送的每个所述第一信息的用户数量以及每个所述第一信息的推送用户数量，得到所述预选用户对所推送的每个所述第一信息的点击率。
第二信息获取子模块122，具体用于根据所述剩余用户中每个用户的历史行为数据以及所述预选用户对所推送的每个所述第一信息的点击率，将该用户没有浏览过的所述点击率最高的第一信息作为该用户对应的第二信息。
第二推送模块13，用于将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
本实施例中，各模块可以是由软件代码实现，此时，上述的各模块可存储于服务器100的存储器110内。以上各模块同样可以由硬件例如集成电路芯片实现。
本实施例所提供的信息推送装置10，其实现原理及产生的技术效果和前述方法实施例相同，为简要描述，装置实施例部分未提及之处，可参考前述方法实施例中相应内容。
第四实施例
请参阅图10，本发明第四实施例提供了一种信息推送装置，运行于服务器100，用于实现上述第二实施例提出的信息推送方法。所述信息推送装置20包括：第一获取模块21、分组模块22、第一推送模块23、第二获取模块24以及第二推送模块25。
其中，第一获取模块21，用于根据每个用户的历史行为数据得到所述每个用户对应的兴趣标签；
分组模块22，用于根据所述每个用户对应的兴趣标签对所述每个用户进行分组以获得多个用户组，一个用户至少对应于一个用户组，且不同用户组对应于不同的兴趣标签；
第一推送模块23，用于向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；
第二获取模块24，用于获取所述预选用户对所述第一信息的响应数据，根据所述响应数据获得所述用户组中剩余用户对应的第二信息；
第二推送模块25，用于将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
具体的，上述第一获取模块21，用于根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签的兴趣度；比较所述用户对各个所述兴趣标签的兴趣度得到该用户对应的兴趣标签。
本实施例中，各模块可以是由软件代码实现，此时，上述的各模块可存储于服务器100的存储器110内。以上各模块同样可以由硬件例如集成电路芯片实现。
本实施例所提供的信息推送装置20，其实现原理及产生的技术效果和前述方法实施例相同，为简要描述，装置实施例部分未提及之处，可参考前述方法实施例中相应内容。
第五实施例
本发明第五实施例提供了一种服务器，所述服务器包括存储器及处理器，所述存储器耦接到所述处理器，所述存储器存储指令。当所述指令由所述处理器执行时以使所述处理器执行以下操作：
向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；
获取所述预选用户对所述第一信息的响应数据，根据所述响应数据获得所述用户组中剩余用户对应的第二信息；
将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
需要说明的是，本说明书中的各个实施例均采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似的部分互相参见即可。
第六实施例
图11示出了根据本发明一实施例可用于实现上述信息推送方法的数据处理的计算设备的结构示意图。
参见图11，计算设备1000包括存储器1010和处理器1020。
处理器1020可以是一个多核的处理器，也可以包含多个处理器。在一些实施例中，处理器1020可以包含一个通用的主处理器以及一个或多个特殊的协处理器，例如图形处理器(GPU)、数字信号处理器(DSP)等等。在一些实施例中，处理器1020可以使用定制的电路实现，例如特定用途集成电路(ASIC，Application Specific Integrated Circuit)或者现场可编程逻辑门阵列(FPGA，Field Programmable Gate Arrays)。
存储器1010可以包括各种类型的存储单元，例如系统内存、只读存储器(ROM)，和永久存储装置。其中，ROM可以存储处理器1020或者计算机的其他模块需要的静态数据或者指令。永久存储装置可以是可读写的存储装置。永久存储装置可以是即使计算机断电后也不会失去存 储的指令和数据的非易失性存储设备。在一些实施方式中，永久性存储装置采用大容量存储装置(例如磁或光盘、闪存)作为永久存储装置。另外一些实施方式中，永久性存储装置可以是可移除的存储设备(例如软盘、光驱)。系统内存可以是可读写存储设备或者易失性可读写存储设备，例如动态随机访问内存。系统内存可以存储一些或者所有处理器在运行时需要的指令和数据。此外，存储器1010可以包括任意计算机可读存储媒介的组合，包括各种类型的半导体存储芯片(DRAM，SRAM，SDRAM，闪存，可编程只读存储器)，磁盘和/或光盘也可以采用。在一些实施方式中，存储器1010可以包括可读和/或写的可移除的存储设备，例如激光唱片(CD)、只读数字多功能光盘(例如DVD-ROM，双层DVD-ROM)、只读蓝光光盘、超密度光盘、闪存卡(例如SD卡、min SD卡、Micro-SD卡等等)、磁性软盘等等。计算机可读存储媒介不包含载波和通过无线或有线传输的瞬间电子信号。
存储器1010上存储有可处理代码，当可处理代码被处理器1020处理时，可以使处理器1020执行上文述及的信息推送方法。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置和方法，也可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，附图中的流程图和框图显示了根据本发明的多个实施例的装置、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或代码的一部分，所述模块、程序段或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现方式中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
另外，在本发明各个实施例中的各功能模块可以集成在一起形成一个独立的部分，也可以是各个模块单独存在，也可以两个或两个以上模块集成形成一个独立的部分。
所述功能如果以软件功能模块的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
本发明还可以实施为一种非暂时性机器可读存储介质(或计算机可读存储介质、或机器可读存储介质)，其上存储有可执行代码(或计算机程序、或计算机指令代码)，当所述可执行代码(或计算机程序、或计算机指令代码)被电子设备(或计算设备、服务器等)的处理器执行时，使所述处理器执行根据本发明的上述方法的各个步骤。
需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
以上所述仅为本发明的优选实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明 的保护范围之内。应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步定义和解释。

Claims (18)

  1. 一种信息推送方法，其特征在于，所述方法包括：
    向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；
    获取所述预选用户对所述第一信息的响应数据，根据所述响应数据在所述第一信息中获得所述用户组中剩余用户对应的第二信息；
    将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
  2. 根据权利要求1所述的方法，其特征在于，所述的向用户组的预选用户推送该预选用户没有浏览过的第一信息，包括：
    向用户组的预选用户随机推送该预选用户没有浏览过的第一信息。
  3. 根据权利要求1所述的方法，其特征在于，所述的向用户组的预选用户推送该预选用户没有浏览过的第一信息，包括：
    为所述用户组中的预选用户分配目标信息，所述目标信息为从该用户组对应的兴趣标签关联的信息池中随机选取的信息；
    判断所述目标信息是否为该预选用户浏览过的信息，若是，则重新从该用户组对应的兴趣标签关联的信息池中随机选取信息作为目标信息并分配给该预选用户，直至分配给该预选用户的目标信息为该预选用户没有浏览过的信息；
    将分配给所述预选用户且该预选用户没有浏览过的目标信息作为所述第一信息推送给该预选用户。
  4. 根据权利要求1所述的方法，其特征在于，所述根据所述响应数据在所述第一信息中获得所述用户组中剩余用户对应的第二信息，包括：
    根据所述响应数据以及推送用户数量获取所述预选用户对所推送的每个所述第一信息的点击率，其中，信息的推送用户数量为向其推送了该信 息的用户的数量；
    根据所述剩余用户中每个用户的历史行为数据以及所述预选用户对所推送的每个所述第一信息的点击率，在所述第一信息中得到该用户对应的第二信息。
  5. 根据权利要求4所述的方法，其特征在于，所述的根据所述响应数据以及推送用户数量获取所述预选用户对所推送的每个所述第一信息的点击率，包括：
    获取所述用户组对应的每个所述第一信息的推送用户数量；
    根据所述响应数据获取点击所推送的每个所述第一信息的用户数量；
    根据所述点击所推送的每个所述第一信息的用户数量以及每个所述第一信息的推送用户数量，得到所述预选用户对所推送的每个所述第一信息的点击率。
  6. 根据权利要求4所述的方法，其特征在于，所述的根据所述剩余用户中每个用户的历史行为数据以及所述预选用户对所推送的每个所述第一信息的点击率，在所述第一信息中得到该用户对应的第二信息，包括：
    根据所述剩余用户中每个用户的历史行为数据以及所述预选用户对所推送的每个所述第一信息的点击率，将该用户没有浏览过的所述点击率最高的第一信息作为该用户对应的第二信息。
  7. 根据权利要求1所述的方法，其特征在于，所述的向用户组的预选用户推送该预选用户没有浏览过的第一信息的步骤之前，所述方法还包括：
    根据每个用户的历史行为数据得到所述每个用户对应的兴趣标签；
    根据所述每个用户对应的兴趣标签对所述每个用户进行分组以获得多个用户组，一个用户至少对应于一个用户组，且不同用户组对应于不同的兴趣标签。
  8. 根据权利要求7所述的方法，其特征在于，所述的根据每个用户的历史行为数据得到所述每个用户对应的兴趣标签，包括：
    根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签的兴趣度；
    比较所述用户对各个所述兴趣标签的兴趣度得到该用户对应的兴趣标签。
  9. 根据权利要求8所述的方法，其特征在于，所述的根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签的兴趣度，包括：
    根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签关联的信息池中的每个信息的关注度；
    根据所述用户对每个所述兴趣标签关联的信息池中的每个信息的关注度得到该用户对每个所述兴趣标签的兴趣度。
  10. 根据权利要求9所述的方法，其特征在于，所述的根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签关联的信息池中的每个信息的关注度，包括：
    根据每个所述用户的历史行为数据得到该用户对每个所述兴趣标签关联的信息池中每个信息的浏览时长、评论指数以及分享指数；
    根据所述用户对所有信息的所述浏览时长得到该用户的平均浏览时长；
    根据所述用户的所述平均浏览时长、所述用户对每个所述信息的浏览时长、所述评论指数、所述分享指数以及预设参数得到该用户对该信息的关注度。
  11. 根据权利要求10所述的方法，其特征在于，所述的根据所述用户的所述平均浏览时长、所述用户对每个所述信息的浏览时长、所述评论指数、所述分享指数以及预设参数得到该用户对该信息的关注度，包括：
    根据公式：
    Figure PCTCN2018090500-appb-100001
    计算得到所述用户对每个所述信息的关注度，其中，u表示用户唯一标识，i ∈I，I表示预设信息资源库中所有信息的集合，focus u,i表示用户u对信息i的关注度，avg(viewtime u,i)表示用户u浏览过的所有信息的平均浏览时长，iscomment u,i表示用户u对信息i的评论指数，isshare u,i表示用户u对信息i的分享指数，α、β、γ表示预设参数。
  12. 根据权利要求9所述的方法，其特征在于，所述的根据所述用户对每个所述兴趣标签关联的信息池中的每个信息的关注度得到该用户对每个所述兴趣标签的兴趣度，包括：
    根据所述用户的历史行为数据得到该用户对每个所述信息的浏览时间指数；
    根据所述用户对每个所述兴趣标签关联的信息池中的每个所述信息的所述关注度、所述浏览时间指数以及预设的衰减因子得到该用户对每个所述兴趣标签的兴趣度。
  13. 根据权利要求12所述的方法，其特征在于，所述的根据所述用户对每个所述兴趣标签关联的信息池中的每个所述信息的所述关注度、所述浏览时间指数以及预设的衰减因子得到该用户对每个所述兴趣标签的兴趣度，包括：
    根据公式：
    Figure PCTCN2018090500-appb-100002
    得到所述用户对每个所述兴趣标签的兴趣度；其中，u表示用户唯一标识，i∈I，I表示预设信息资源库中所有信息的集合；t∈T，T表示所述信息资源库对应的所有所述兴趣标签的集合，interest u,t表示用户u对兴趣标签t的兴趣度，focus u,i表示用户u对信息i的关注度；tag i,t表示信息i是否与兴趣标签t关联，tag i,t＝0表示不关联，tag i,t＝1表示关联；viewdays u,i表示用户u对信息i的浏览时间指数；lapse表示所述衰减因子。
  14. 根据权利要求8所述的方法，其特征在于，所述的比较所述用户对 各个所述兴趣标签的兴趣度得到该用户对应的兴趣标签，包括：
    根据所述用户对各个所述兴趣标签的兴趣度大小选择兴趣度最大的兴趣标签作为该用户对应的兴趣标签。
  15. 一种信息推送装置，其特征在于，所述装置包括：
    第一推送模块，用于向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；
    获取模块，用于获取所述预选用户对所述第一信息的响应数据，在所述第一信息中根据所述响应数据获得所述用户组中剩余用户对应的第二信息；
    第二推送模块，用于将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
  16. 一种服务器，其特征在于，所述服务器包括处理器以及存储器，所述存储器耦接到所述处理器，所述存储器存储指令，当所述指令由所述处理器执行时使所述服务器执行以下操作：
    向用户组的预选用户推送该预选用户没有浏览过的第一信息，所述第一信息为与所述用户组对应的兴趣标签关联的信息池中的信息；
    获取所述预选用户对所述第一信息的响应数据，在所述第一信息中根据所述响应数据获得所述用户组中剩余用户对应的第二信息；
    将所述第二信息推送给对应的剩余用户，其中，所述预选用户的数量少于所述剩余用户的数量。
  17. 一种计算设备，包括：
    处理器；以及
    存储器，其上存储有可执行代码，当所述可执行代码被所述处理器执行时，使所述处理器执行如权利要求1-14中任何一项所述的方法。
  18. 一种非暂时性机器可读存储介质，其上存储有可执行代码，当所述可执行代码被电子设备的处理器执行时，使所述处理器执行如权利要求1至14中任一项所述的方法。
PCT/CN2018/090500 2017-07-06 2018-06-08 信息推送方法、装置及服务器、计算设备和存储介质 WO2019007187A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710548188.7 2017-07-06
CN201710548188.7A CN107222566B (zh) 2017-07-06 2017-07-06 信息推送方法、装置及服务器

Publications (1)

Publication Number Publication Date
WO2019007187A1 true WO2019007187A1 (zh) 2019-01-10

Family

ID=59952651

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/090500 WO2019007187A1 (zh) 2017-07-06 2018-06-08 信息推送方法、装置及服务器、计算设备和存储介质

Country Status (2)

Country Link
CN (1) CN107222566B (zh)
WO (1) WO2019007187A1 (zh)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109977321A (zh) * 2019-02-19 2019-07-05 平安科技（深圳）有限公司 消息中心消息内容推荐方法、装置及计算机可读存储介质
CN110489646A (zh) * 2019-08-15 2019-11-22 中国平安人寿保险股份有限公司 用户画像构建方法及终端设备
CN111026968A (zh) * 2019-12-09 2020-04-17 湖南快乐阳光互动娱乐传媒有限公司 对象推送方法、装置及系统
CN111178950A (zh) * 2019-12-19 2020-05-19 车智互联（北京）科技有限公司 一种用户画像构建方法、装置及计算设备
CN111768260A (zh) * 2019-09-11 2020-10-13 北京京东尚科信息技术有限公司 推荐同兴趣用户的方法、装置及设备
CN111915148A (zh) * 2020-07-10 2020-11-10 北京科技大学 一种基于信息技术的课堂教学评价方法和系统
CN114547417A (zh) * 2022-02-25 2022-05-27 北京百度网讯科技有限公司 一种媒体资源排序方法及电子设备
CN115795172A (zh) * 2023-02-06 2023-03-14 湖南出头科技有限公司 一种基于大数据的信息推送引流系统
CN116522005A (zh) * 2023-07-04 2023-08-01 之江实验室 一种实时整合多波段暂现源报导的信息推送方法及装置

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108259546A (zh) * 2017-01-16 2018-07-06 广州市动景计算机科技有限公司 消息推送方法、设备及可编程设备
CN107222566B (zh) * 2017-07-06 2020-09-04 阿里巴巴（中国）有限公司 信息推送方法、装置及服务器
CN110334270B (zh) * 2018-03-21 2023-07-11 深圳市雅阅科技有限公司 一种新闻排序方法、服务器及计算机可读存储介质
CN109377394A (zh) * 2018-10-22 2019-02-22 中国平安人寿保险股份有限公司 保险业务数据处理方法、装置及存储介质、计算机设备
CN109697629B (zh) * 2018-11-15 2023-02-24 平安科技（深圳）有限公司 产品数据推送方法及装置、存储介质、计算机设备
CN109241452B (zh) * 2018-11-19 2022-03-22 天津网之易创新科技有限公司 信息推荐方法及装置、存储介质和电子设备
CN109522141B (zh) * 2018-11-27 2023-07-25 苏州我的打工人力资源有限公司 信息推送方法及装置、服务器、设备及存储介质
CN110753247B (zh) * 2019-10-18 2021-11-09 重庆空间视创科技有限公司 一种信息推送策略生成系统及方法
CN110753248B (zh) * 2019-10-18 2021-11-05 重庆空间视创科技有限公司 一种用于iptv用户端数据收集系统及方法
CN111683154B (zh) * 2020-06-17 2023-11-14 腾讯科技（深圳）有限公司 一种内容推送的方法、装置、介质及电子设备
CN111598630A (zh) * 2020-06-30 2020-08-28 成都新潮传媒集团有限公司 小区画像构建方法、装置及存储介质
CN112163180A (zh) * 2020-11-02 2021-01-01 北京明略软件系统有限公司 关联活跃度计算方法、装置、计算机设备及可读存储介质
CN113905037A (zh) * 2021-06-18 2022-01-07 武汉理工数字传播工程有限公司 一种文件传输管理方法、装置、设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103377250A (zh) * 2012-04-27 2013-10-30 杭州载言网络技术有限公司 基于邻域的top-k推荐方法
CN104537115A (zh) * 2015-01-21 2015-04-22 北京字节跳动科技有限公司 用户兴趣的探索方法和装置
CN105159910A (zh) * 2015-07-03 2015-12-16 安一恒通（北京）科技有限公司 信息推荐方法和装置
CN106055617A (zh) * 2016-05-26 2016-10-26 乐视控股（北京）有限公司 一种数据推送方法及装置
US9680959B2 (en) * 2012-08-30 2017-06-13 Google Inc. Recommending content based on intersecting user interest profiles
CN107222566A (zh) * 2017-07-06 2017-09-29 广州优视网络科技有限公司 信息推送方法、装置及服务器

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090259621A1 (en) * 2008-04-11 2009-10-15 Concert Technology Corporation Providing expected desirability information prior to sending a recommendation
CN102411596A (zh) * 2010-09-21 2012-04-11 阿里巴巴集团控股有限公司 一种信息推荐方法及系统
WO2013138969A1 (en) * 2012-03-17 2013-09-26 Beijing Haipu Wangju Technology Limited Method and system for recommending content to a user
CN104038517A (zh) * 2013-03-05 2014-09-10 腾讯科技（深圳）有限公司 基于群组关系的信息推送方法以及服务器
CN104239338A (zh) * 2013-06-19 2014-12-24 阿里巴巴集团控股有限公司 信息推荐方法及装置
CN106503269A (zh) * 2016-12-08 2017-03-15 广州优视网络科技有限公司 应用推荐的方法、装置及服务器
CN106651533A (zh) * 2016-12-29 2017-05-10 合肥华凌股份有限公司 一种基于用户行为的个性化商品推荐方法及装置
CN106919692B (zh) * 2017-03-07 2021-02-19 阿里巴巴（中国）有限公司 一种推送消息的方法和装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103377250A (zh) * 2012-04-27 2013-10-30 杭州载言网络技术有限公司 基于邻域的top-k推荐方法
US9680959B2 (en) * 2012-08-30 2017-06-13 Google Inc. Recommending content based on intersecting user interest profiles
CN104537115A (zh) * 2015-01-21 2015-04-22 北京字节跳动科技有限公司 用户兴趣的探索方法和装置
CN105159910A (zh) * 2015-07-03 2015-12-16 安一恒通（北京）科技有限公司 信息推荐方法和装置
CN106055617A (zh) * 2016-05-26 2016-10-26 乐视控股（北京）有限公司 一种数据推送方法及装置
CN107222566A (zh) * 2017-07-06 2017-09-29 广州优视网络科技有限公司 信息推送方法、装置及服务器

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109977321A (zh) * 2019-02-19 2019-07-05 平安科技（深圳）有限公司 消息中心消息内容推荐方法、装置及计算机可读存储介质
CN110489646A (zh) * 2019-08-15 2019-11-22 中国平安人寿保险股份有限公司 用户画像构建方法及终端设备
CN110489646B (zh) * 2019-08-15 2024-02-13 中国平安人寿保险股份有限公司 用户画像构建方法及终端设备
CN111768260A (zh) * 2019-09-11 2020-10-13 北京京东尚科信息技术有限公司 推荐同兴趣用户的方法、装置及设备
CN111026968A (zh) * 2019-12-09 2020-04-17 湖南快乐阳光互动娱乐传媒有限公司 对象推送方法、装置及系统
CN111178950B (zh) * 2019-12-19 2023-08-29 车智互联（北京）科技有限公司 一种用户画像构建方法、装置及计算设备
CN111178950A (zh) * 2019-12-19 2020-05-19 车智互联（北京）科技有限公司 一种用户画像构建方法、装置及计算设备
CN111915148A (zh) * 2020-07-10 2020-11-10 北京科技大学 一种基于信息技术的课堂教学评价方法和系统
CN111915148B (zh) * 2020-07-10 2023-11-03 北京科技大学 一种基于信息技术的课堂教学评价方法和系统
CN114547417A (zh) * 2022-02-25 2022-05-27 北京百度网讯科技有限公司 一种媒体资源排序方法及电子设备
CN115795172A (zh) * 2023-02-06 2023-03-14 湖南出头科技有限公司 一种基于大数据的信息推送引流系统
CN116522005B (zh) * 2023-07-04 2023-08-25 之江实验室 一种实时整合多波段暂现源报导的信息推送方法及装置
CN116522005A (zh) * 2023-07-04 2023-08-01 之江实验室 一种实时整合多波段暂现源报导的信息推送方法及装置

Also Published As

Publication number Publication date
CN107222566B (zh) 2020-09-04
CN107222566A (zh) 2017-09-29

Similar Documents

Publication Publication Date Title
WO2019007187A1 (zh) 信息推送方法、装置及服务器、计算设备和存储介质
AU2018200833B2 (en) Providing content to a user across multiple devices
JP6827515B2 (ja) ビデオ検索に対する視聴時間クラスタリング
US20230004594A1 (en) Media selection
US10305851B1 (en) Network-based content discovery using messages of a messaging platform
TWI529642B (zh) Promotion method and equipment of product information
US10672036B2 (en) Bidding based on the relative value of identifiers
JP6334696B2 (ja) ハッシュタグおよびコンテンツ提示
US8688984B2 (en) Providing content to a user across multiple devices
CA2871785A1 (en) Privacy management across multiple devices
US10164936B2 (en) Providing content to devices in a cluster
US9449052B1 (en) Trend based distribution parameter suggestion
US9092463B2 (en) Keyword generation
US20230252011A1 (en) Method and system for data indexing and reporting
CN110708361A (zh) 数字内容发布用户的等级确定系统、方法、装置及服务器
WO2018205642A1 (zh) 视频收益计算建模装置、方法和视频推荐装置、方法和服务器及存储介质
US8666810B1 (en) System and method for online unique users and frequency estimation for advertisements on a video sharing website based on auction history results
US10346519B1 (en) Selecting content based on entities

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: 18827736

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18827736

Country of ref document: EP

Kind code of ref document: A1