CN110910262A - Community grouping method and device, content isolation method and device and server - Google Patents
Community grouping method and device, content isolation method and device and server Download PDFInfo
- Publication number
- CN110910262A CN110910262A CN201911054465.4A CN201911054465A CN110910262A CN 110910262 A CN110910262 A CN 110910262A CN 201911054465 A CN201911054465 A CN 201911054465A CN 110910262 A CN110910262 A CN 110910262A
- Authority
- CN
- China
- Prior art keywords
- community
- user
- grouping
- grouped
- users
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
Abstract
The disclosure provides a community grouping method and device, a content isolation method and device and a server. When community grouping is carried out, first sample data at least comprising position characteristics and culture custom characteristics of users to be grouped are obtained on the basis of user data produced by the users to be grouped on a specified platform; and finally, performing community grouping on the users to be grouped according to the first community grouping information, thereby improving the accuracy of the community grouping.
Description
Technical Field
The present disclosure relates to the field of big data processing technologies, and in particular, to a community grouping method and apparatus, a content isolation method and apparatus, and a server.
Background
In the related art, users are directly grouped into communities based on information such as IP (Internet Protocol Address), GPS (Global Positioning System), MCC (Mobile Country Code), or Country and region selected by the users. However, since many cultural customs are involved in some countries and regions, a phenomenon of wrong community grouping occurs when community grouping is performed, and particularly, the accuracy of community grouping is sharply reduced under the conditions of traveling outside, leaving a study, or being vague in the border of the country and the region.
Disclosure of Invention
The present disclosure provides a community grouping method and apparatus, a content isolation method and apparatus, and a server, so as to at least solve the problem of low community grouping accuracy in the related art. The technical scheme of the disclosure is as follows:
according to a first aspect of the embodiments of the present disclosure, there is provided a community grouping method, including:
acquiring first sample data based on user data produced by users to be grouped on a specified platform, wherein the first sample data at least comprises position characteristics and culture custom characteristics of the users to be grouped;
inputting the first sample data into a community grouping model to obtain first community grouping information of the users to be grouped, wherein the community grouping model is obtained by clustering training based on second sample data of a plurality of users on the appointed platform, and the second sample data and the first sample data have the same characteristic dimension;
and carrying out community grouping on the users to be grouped according to the first community grouping information.
According to a second aspect of the embodiments of the present disclosure, there is provided a content isolation method, which is implemented by a user community after community grouping based on the above community grouping method, and includes:
determining an isolation target based on a community tag of a user community or a community tag carried in multimedia information released by a user on a specified platform;
and when content pushing is carried out on the users in the user community, isolating the content to be pushed corresponding to the isolation target from the content to be pushed.
According to a third aspect of the embodiments of the present disclosure, there is provided a community grouping apparatus including:
the system comprises a sample data acquisition module, a data processing module and a data processing module, wherein the sample data acquisition module is configured to execute user data produced on a specified platform based on users to be grouped, and the first sample data at least comprises position characteristics and culture custom characteristics of the users to be grouped;
the grouping information analysis module is configured to input the first sample data into a community grouping model to obtain first community grouping information of the users to be grouped, wherein the community grouping model is obtained through clustering training based on second sample data of a plurality of users on the specified platform, and the second sample data and the first sample data have the same characteristic dimension;
and the community grouping module is configured to perform community grouping on the users to be grouped according to the first community grouping information.
According to a fourth aspect of the embodiments of the present disclosure, there is provided a content isolation apparatus, which is implemented based on a user community after community grouping by the community grouping apparatus, the content isolation apparatus including:
the isolation target determining module is configured to execute determining an isolation target needing content isolation based on a community tag of a user community to which the user belongs or a community tag carried in multimedia information released by the user on the specified platform;
and the content pushing module is configured to push the content to be pushed to the user based on the isolation target when the content is pushed to the user in the user community.
According to a fifth aspect of embodiments of the present disclosure, there is provided a server including:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the aforementioned community grouping method or the aforementioned content isolation method.
According to a sixth aspect of embodiments of the present disclosure, there is provided a storage medium whose instructions, when executed by a processor in a server, enable the server to perform a community grouping method as described above or a content isolation method as described above.
According to a seventh aspect of embodiments of the present disclosure, there is provided a computer program product comprising at least one non-transitory computer readable medium storing instructions translatable by at least one processor for implementing the aforementioned community grouping means or content isolation means.
The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects:
the method comprises the steps of obtaining the position characteristics and the culture custom characteristics of users to be grouped based on user data produced by the users to be grouped on a designated platform, learning the position characteristics and the culture custom characteristics of the users to be grouped by utilizing a community grouping model to obtain community grouping information of the users to be grouped, and finally carrying out community grouping on the users to be grouped according to the community grouping information, so that the accuracy of community grouping results is effectively improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.
FIG. 1 is a flow diagram illustrating a community grouping method in accordance with an exemplary embodiment.
FIG. 2 is a flow chart illustrating a community grouping method according to another exemplary embodiment.
FIG. 3 is a flow diagram illustrating a method of content isolation in accordance with an exemplary embodiment.
FIG. 4 illustrates a block diagram of a community grouping apparatus in accordance with an exemplary embodiment.
FIG. 5 illustrates a block diagram of a content isolation device, according to an exemplary embodiment.
FIG. 6 is a block diagram illustrating a server in accordance with an example embodiment
In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
Detailed Description
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
Example one
FIG. 1 is a flow diagram illustrating a community grouping method that may be performed by, but is not limited to, a server, according to an example embodiment. The community grouping method includes the following steps.
S11, acquiring first sample data of the users to be grouped based on the user data produced by the users to be grouped on the appointed platform.
The first sample data at least comprises position characteristics and culture custom characteristics of the users to be grouped. The designated platform may be, but is not limited to, a social platform, a live platform, or a gaming platform, among others. The user data may include, but is not limited to, user basic information produced when the user to be grouped registers on a designated platform, and multimedia information published on the platform by the user to be grouped. Alternatively, the user basic information may include, but is not limited to, language information, avatar information, identity background information, place of birth information, current location information, etc. of the users to be grouped, and the multimedia information may include, but is not limited to, information such as photos, videos, audios, comments, etc. published by the users to be grouped when using the specified platform. It should be noted that the aforementioned user data involved in acquiring the first sample data needs to be performed under the condition that the corresponding user authorization has been obtained.
As a possible implementation manner, in the first sample data obtained based on the user data, the location feature may be, but is not limited to, location information, country information, and the like of the user to be grouped, which are obtained based on IP, GPS, MCC, and the like. In addition, the user position information to be grouped corresponding to the position characteristics can be information uploaded to a specified platform by the user to be grouped through the positioning module in real time, or can be manually input to the specified platform by the user to be grouped.
Further, in the first sample data obtained based on the user data, the culture custom characteristics may include, but are not limited to, at least one of language characteristics, gender characteristics, apparel characteristics, behavior characteristics, and entertainment and hobby characteristics. As one implementation, the cultural custom features may be extracted from multimedia information contained in the user data based on image recognition techniques and the like. For example, if the multimedia information is dance video, the national information to which the user belongs can be presumed according to the dance type, and if the multimedia information is voice information, the hometown, country or national characteristics of the user can be judged according to the tone.
It should be noted that, when performing community grouping on the users to be grouped, besides the location characteristics and the culture custom characteristics of the users to be grouped, which are considered in the embodiment, the gender characteristics, the age characteristics, the preference characteristics, and the like of the users to be grouped can be considered at the same time, so as to further improve the accuracy of the community grouping result.
Further, as an optional implementation manner, for user data produced by a user to be grouped on a specified platform, the user data may be stored in a database or a local cache, and further asynchronously and synchronously stored in an HDFS (Hadoop distributed file system). When community grouping is needed, corresponding user data can be called from the HDFS, and preprocessing such as dirty data cleaning and normalization is carried out on the called user data, so that accuracy in subsequent data processing is improved. Meanwhile, the data storage pressure of the designated platform can be effectively reduced through a data caching mode, and the maintenance cost of the designated platform is reduced.
S13, inputting the first sample data into the community grouping model to obtain the first community grouping information of the users to be grouped.
The community grouping model is obtained by training the machine learning model based on second sample data of a plurality of users on the appointed platform, and the first community grouping information of the users to be grouped can be calculated, and the second sample data and the first sample data have the same characteristic dimension, so that the accuracy of the first community grouping information obtained based on the first sample data is effectively ensured. For example, the first sample data and the second sample data may each contain two feature dimensions, a location feature and a cultural custom feature. Alternatively, the aforementioned machine learning model may generally employ a neural network model, a support vector machine, a logistic regression model, a random forest model, or the like.
In addition, the first community grouping information may include, but is not limited to, community information of a user community to which the user to be grouped belongs, and a probability value of the user community to which the user to be grouped belongs. For example, the second community grouping information may include that the user X to be grouped belongs to the a1 community, and the corresponding probability value is B1%; the user Y to be grouped belongs to the A2 community, and the corresponding probability value is B2% … ….
It should be noted that the obtained first community grouping information is different according to the grouping standard of the user community, for example, when grouping is performed by country, the user community may be china, usa, korea, etc.; for example, when the users are grouped according to their popular culture, the user community may be a ramen culture community, a seal-land culture community, a Chinese culture community, a Japanese-Korean culture community, or the like, which is not limited in this embodiment.
And S15, carrying out community grouping on the users to be grouped according to the first community grouping information.
Wherein, one user to be grouped may correspond to at least one user community. For example, the user X to be grouped may belong to both the ramee culture community and the landed culture community, and the probability that the user X to be grouped belongs to the ramee culture community may be C1, the probability that the user X belongs to the landed culture community may be C2, and the like.
It should be noted that, when performing community grouping on the users to be grouped based on the above S11 to S15, the first sample data may be acquired based on the multimedia information generated by the users to be grouped in the last year time on the specified platform, or the first sample data may be acquired based on the multimedia information generated by the users to be grouped in the last three year time on the specified platform, and the like, which is not limited in this embodiment.
In addition, as a possible implementation manner, after the user community of the user to be grouped is obtained based on the community grouping method, content recommendation can be performed on the user to be grouped according to the community label of the user community to which the user to be grouped belongs, so as to improve the content click rate, for example, for 80% of the users to be grouped belonging to the Chinese culture community, content corresponding to the Chinese culture community can be recommended to the user to be grouped.
Further, as a possible implementation manner, the community grouping method may further include a step of training a community grouping model, for example, as shown in fig. 2, the aforementioned training process of the community grouping model may include the following steps.
S100, acquiring second sample data based on historical user data produced by a plurality of users on a specified platform. The second sample data and the first sample data have the same feature dimension, so the second sample data may refer to the aforementioned description of the first sample data, and the historical user data may refer to the aforementioned description of the user data, which is not described herein again. It should be noted that, similarly to the first sample data, the aforementioned acquisition of the historical user data involved in the second sample data is also performed on the condition that the corresponding user authorization has been obtained.
And S101, inputting second sample data into a preset machine learning model to obtain second community grouping information corresponding to the historical user data. In step S101, preprocessing, such as dirty data cleaning, data normalization, etc., may be performed on the second sample data to improve the accuracy of subsequent data processing.
S102, carrying out community grouping on the plurality of users based on the second community grouping information to obtain a community grouping prediction result. It should be noted that in the present embodiment, one user may correspond to one or more user communities, for example, the user X may belong to the a1 community or the a2 community, but the probability values of the user X belonging to the a1 community and the a2 community may be different.
S103, determining a training target based on the difference between the community grouping prediction result and the community grouping expected result.
The training objective is data for measuring the quality of the model, and may be generally a function of the difference between the actual output of the model and the expected output of the model as an argument. For example, a loss function may be used to measure the difference between the actual output of the model and the expected output of the model, the loss function may be a squared loss function, an exponential loss function, a negative likelihood function, or the like. In this embodiment, a difference between the community grouping prediction result and the community grouping expected result may be used as an argument to generate a corresponding loss function, which is not described herein again.
And S104, adjusting model parameters or/and second sample data of the machine learning model according to the direction of the optimized training target, and continuing training until the training stopping condition is met.
The direction of optimizing the training target may be a direction of minimizing the training target, etc., such as a direction of optimizing the training target with a difference between an actual output of the model and a desired output of the model reduced. The training stopping condition is a condition for finishing the model training, such as whether a preset iteration number is reached, whether the output of the model after model parameter adjustment reaches a preset index, and the like. In addition, the manner of adjusting the model parameters of the machine learning model and the second sample data may include replacing the type of the machine learning model, adjusting the number of convolution kernels of the machine learning model, the number of samples of the second sample data, and the like.
Further, as yet another possible implementation manner, based on the community grouping result of the users to be grouped given in the above S11 to S15, the user community grouping method may further include the following steps.
And S17, when the user to be grouped needs to release new multimedia information on the appointed platform, adding the community label of the user community to which the user to be grouped belongs in the multimedia information.
And S19, releasing the multimedia information after the community label is added.
In S17 to S19, by adding the community tag, the content recommendation system can recommend the corresponding content to the users to be grouped according to the community tag, so as to improve the user experience and ensure the click rate of the recommended content. In addition, other users browsing multimedia information can know culture preferences of users to be grouped according to the community tags, and the like, so that the smoothness of communication among the users is improved, and the occurrence of conflict of culture and preference is avoided.
Optionally, when the community tag is added, the community tag may be implemented in a watermark, a text, a picture, or the like, which is not limited in this embodiment. It should be noted that the act of adding the community tag to the multimedia information to be published by the user is performed on the condition that the user authorization has been obtained.
Further, the following describes an implementation process of the community grouping method in the first embodiment with reference to several possible application scenarios. It should be noted that the user data involved in the following scenarios are all performed under the condition that user authorization has been obtained.
Scene 1
After a user M to be grouped logs in a designated platform, user data produced on the designated platform comprises a video with a language of the user M to be grouped being portuguese, a video with a region of the user M to be grouped being St Paul and Baba dance, a position feature in first sample data obtained based on the user data can be St Paul, a cultural custom feature can be Portuguese and Samba dance, first sample data containing the portuguese Paul, the portuguese and the Samba dance are input into a community grouping model to be analyzed, first community grouping information of the user M to be grouped can be obtained, the first community grouping information comprises a Lame community, and the corresponding probability value is 80%, and the user M to be grouped can be divided into the Lame culture community by the first community grouping information.
Based on the user community division result, assuming that the user M to be grouped needs to release multimedia content such as self-portrait, a community label of a ramen culture community can be added to the self-portrait multimedia content.
Scene 2
Assuming a user N to be grouped, the user N to be grouped is grouped into the Chinese culture community according to user data produced by the user N on a specified platform, but the user N to be grouped can travel to Saint Paul as time goes on, the first sample data obtained based on the user data of the user N to be grouped in the last two years produced on the specified platform comprises the position characteristic of Saint Paul, the Chinese culture community can be included in the obtained first community grouping information of the user to be grouped through learning the first sample data through a community grouping model, the corresponding probability value is 80%, the Lameian culture community is 5%, and the like.
Then, as the stay time of the user N to be grouped in saint paul is extended (for example, a long-time residence), data related to the ramen culture community included in the user data of the user N to be grouped in the last two years produced on the specified platform will gradually increase, and then, in the user grouping information to be grouped obtained by learning the first sample data included in the user data through the community grouping model, the corresponding probability value will gradually increase, for example, from 5% to 20%, 50%, and the like.
The description of the community grouping method can be used for obtaining the community grouping information, and therefore the accuracy of the community grouping result is improved.
Example two
With the continuous development of globalization, when content recommendation is performed on users to be grouped on a specified platform, if content recommendation is performed only based on interests or positions of the users to be grouped, cultural conflicts easily occur, and the experience of the users to be grouped is poor. In this embodiment, a content isolation method is provided, which is implemented by a user community after community grouping is performed based on the community grouping method provided in embodiment one. As shown in fig. 3, the content isolation method includes the following steps.
Step 21, determining an isolation target based on a community tag of a user community or a community tag carried in multimedia information released by users to be grouped on a designated platform.
And step 22, isolating the content to be pushed corresponding to the isolation target from the content to be pushed when pushing the content to the user to be grouped in the user community.
For example, if the community tag is a Hui nationality Chinese community, it may be determined that the isolation target at least includes content related to the phrase "eat pork" according to the community tag of the Hui nationality Chinese community because the eating culture of Hui nationality is Islamic. For another example, assuming that the community tag is an african community, since ethnic discrimination is a contraindication of african human, it can be determined that the isolation target at least includes content related to 'ethnic discrimination' according to the community tag of the african community, otherwise, when content recommendation is performed on the user, a cultural conflict is inevitably generated, and the user experience is reduced.
It should be noted that the isolation target is mainly determined according to the taboo culture of the user community, and the like, and the description thereof is omitted in this embodiment.
Further, as an optional implementation manner, the step of determining the isolation target includes determining the isolation target according to the probability of the user community to which the user belongs when a plurality of user communities corresponding to the user exist. For example, assuming that the user X belongs to both the user community a and the user community B, but the probability value of the user community a is a1, and the probability value corresponding to the user community B is B1, then in the determined isolation target, the user community a proportion is a1, and the user community B proportion is B1.
In addition, as a possible implementation manner, the community grouping model can be optimized according to the click rate of the user on the isolated recommended content and the like, so that the accuracy of model output and content isolation is improved, and the user experience is ensured.
According to the content isolation method provided by the embodiment, when content recommendation is performed, content isolation can be performed based on the user community to which the user belongs, so that the content recommended to the user is ensured to contain the content which conflicts with the culture customs and the like of the user as little as possible, the user experience is improved, the culture conflict is avoided, and the click rate of the recommended content is ensured.
EXAMPLE III
Fig. 4 is a block diagram illustrating a community grouping apparatus 40 according to an example embodiment. Referring to fig. 4, the apparatus includes a sample data acquisition module 41, a grouping information analysis module 43, and a community grouping module 45.
The sample data acquisition module 41 is configured to execute acquisition of first sample data of the users to be grouped based on user data produced by the users to be grouped on a specified platform, wherein the first sample data at least comprises position characteristics and culture custom characteristics of the users to be grouped. Optionally, the user data includes user basic information registered by the user to be grouped on a specified platform and multimedia information released by the user to be grouped on the platform, and the culture custom characteristics include at least one of language characteristics, gender characteristics, clothing characteristics, behavior characteristics and entertainment and hobby characteristics.
And the grouping information analysis module 43 is configured to perform input of the first sample data into a community grouping model to obtain first community grouping information of the users to be grouped, wherein the community grouping model is obtained by clustering training based on second sample data of a plurality of users of a specified platform, and the second sample data has the same characteristic dimension as the first sample data. Optionally, the first community grouping information includes community information of a user community to which the user to be grouped belongs, and a probability value of the user community to which the user to be grouped belongs, where one user to be grouped corresponds to at least one user community.
A community grouping module 45 configured to perform community grouping on the users to be grouped according to the first community grouping information.
Further, as an optional implementation manner, the community grouping apparatus may further include:
and the sample acquisition module is configured to acquire second sample data based on historical user data produced by a plurality of users on the specified platform.
And the grouping information prediction module is configured to input the second sample data into a preset machine learning model to obtain second community grouping information corresponding to the historical user data.
And the grouping result prediction module is configured to perform community grouping on the plurality of users based on the second community grouping information to obtain a community grouping prediction result.
A training goal determination module configured to perform determining a training goal based on a difference between the community grouping prediction result and the community grouping expected result.
And the model training module is configured to adjust the model parameters of the machine learning model or/and the second sample data according to the direction of the optimized training target and continue training until the training is finished when the training stopping condition is met.
Further, as yet another optional implementation manner, the community grouping device 40 may further include an information publishing module configured to add a community tag of a user community to which the user to be grouped belongs to the multimedia information when the user to be grouped needs to publish new multimedia information on a specified platform; and releasing the multimedia information after the community label is added.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here. It should be noted that the aforementioned user data involved in acquiring the first sample data, and the historical user data involved in acquiring the second sample data are both performed under the condition that the corresponding user authorization is obtained.
Example four
Fig. 5 is a block diagram illustrating a content isolation device 50 according to an exemplary embodiment. Referring to fig. 5, the content isolation apparatus 50 device includes an isolation target determination module 51 and a content push module 53.
An isolation target determination module 51, configured to execute determining an isolation target that needs content isolation based on a community tag of a user community to which a user to be grouped belongs or a community tag carried in multimedia information issued by the user to be grouped on a specified platform;
and the content pushing module 53 is configured to perform pushing of the content to be pushed to the user to be grouped based on the isolation target when the content is pushed to the user to be grouped in the user community. Optionally, the content pushing module may be further specifically configured to determine, for different users to be grouped, an isolation target according to a probability of a user community to which the users to be grouped belong.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
EXAMPLE five
Fig. 6 is a block diagram illustrating one type of server 10 according to an example embodiment. The server 10 may perform, but is not limited to, the community grouping method or/and the content isolation method provided by the present embodiment. It should be noted that, since the community grouping method or/and the content isolation method executed by the server 10 in this embodiment has the same or corresponding technical features as the community grouping method or/and the content isolation method in the foregoing embodiment one or second, the detailed description of the community grouping method or/and the content isolation method in this embodiment may refer to the description of the community grouping method or/and the content isolation method in the foregoing embodiment one or second, and the description of this embodiment is not repeated here.
Further, in one possible implementation, the server 10 may include, but is not limited to, the processor 20 and the memory 30 shown in fig. 6. The processor 20 and the memory 30 are electrically connected directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines.
The memory 30 is used for storing programs or data, such as instructions executable by the processor 20. The Memory 30 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
The processor 20 is used to read/write data or programs stored in the memory 30 and perform corresponding functions.
As one possible implementation, the server 10 may also include a power component configured to perform power management of the server, a wired or wireless network interface configured to connect the server to a network, and an input output (I/O) interface. The server may operate based on an operating system stored in memory, such as Windows Server, MacOS XTM, UnixTM, LinuxTM, FreeBSDTM, or the like.
It should be understood that the configuration shown in fig. 6 is merely a schematic diagram of the configuration of the server 10, and that the server 10 may include more or fewer components than shown in fig. 6, or have a different configuration than shown in fig. 6. The components shown in fig. 6 may be implemented in hardware, software, or a combination thereof. In addition, in the present embodiment, the server 10 may be, but is not limited to, a computer, a mobile phone, an IPad, a mobile internet device, and the like.
EXAMPLE six
In an exemplary embodiment, there is further provided a storage medium, and instructions in the storage medium, when executed by a processor in a server, enable the server to execute to implement the community grouping method or/and the content isolation method in the first embodiment or the second embodiment. Alternatively, the storage medium may be a non-transitory storage medium, for example, the non-transitory storage medium may be a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
It should be noted that, since the community grouping method or/and the content isolation method executed by the server 10 in this embodiment has the same or corresponding technical features as the community grouping method or/and the content isolation method in the foregoing embodiment one or second, the detailed description of the community grouping method or/and the content isolation method in this embodiment may refer to the description of the community grouping method or/and the content isolation method in the foregoing embodiment one or second, and the description of this embodiment is not repeated here.
EXAMPLE seven
In an exemplary embodiment, there is also provided a computer program product comprising at least one non-transitory computer readable medium storing instructions translatable by at least one processor for implementing the community grouping means or/and the content isolation means of the third preceding embodiment.
It should be noted that, since the community grouping method or/and the content isolation method executed by the server in this embodiment has the same or corresponding technical features as the community grouping method or/and the content isolation method in the foregoing embodiment one or embodiment two, the detailed description of the community grouping method or/and the content isolation method in this embodiment may refer to the description of the community grouping method or/and the content isolation method in the foregoing embodiment one or embodiment two, and the description of this embodiment is not repeated here.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
Claims (10)
1. A community grouping method, comprising:
acquiring first sample data based on user data produced by users to be grouped on a specified platform, wherein the first sample data at least comprises position characteristics and culture custom characteristics of the users to be grouped;
inputting the first sample data into a community grouping model to obtain first community grouping information of the users to be grouped, wherein the community grouping model is obtained by clustering training based on second sample data of a plurality of users on the appointed platform, and the second sample data and the first sample data have the same characteristic dimension;
and carrying out community grouping on the users to be grouped according to the first community grouping information.
2. The community grouping method according to claim 1, wherein the user data includes user basic information registered by the user to be grouped on a specified platform and multimedia information released by the user to be grouped on the specified platform, and the cultural custom characteristics include at least one of language characteristics, gender characteristics, clothing characteristics, behavior characteristics and entertainment and hobby characteristics.
3. The community grouping method according to claim 1, wherein the training process for obtaining the community grouping model based on the second sample data cluster training of the plurality of users on the designated platform comprises:
acquiring second sample data based on historical user data produced by a plurality of users on the specified platform;
inputting the second sample data into a preset machine learning model to obtain second community grouping information corresponding to the historical user data;
performing community grouping on the plurality of users based on the second community grouping information to obtain a community grouping prediction result;
determining a training objective based on a difference between the community grouping prediction result and a community grouping expected result;
and according to the direction of optimizing the training target, adjusting the model parameters of the machine learning model or/and the second sample data and continuing training until the training stopping condition is met, and finishing the training.
4. The community grouping method according to claim 1, wherein the first community grouping information includes community information of user communities to which the users to be grouped belong and probability values of the user communities to which the users to be grouped belong, wherein one user to be grouped corresponds to at least one user community.
5. The community grouping method according to claim 1, wherein the method further comprises:
when the user to be grouped needs to release new multimedia information on the appointed platform, adding a community label of a user community to which the user to be grouped belongs to the multimedia information;
and releasing the multimedia information after the community label is added.
6. A content isolation method, implemented by a user community after community grouping based on the community grouping method of any one of claims 1 to 5, the content isolation method comprising:
determining an isolation target based on a community tag of a user community or a community tag carried in multimedia information released by a user on a specified platform;
and when content pushing is carried out on the users in the user community, isolating the content to be pushed corresponding to the isolation target from the content to be pushed.
7. A community grouping apparatus, comprising:
the system comprises a sample data acquisition module, a data processing module and a data processing module, wherein the sample data acquisition module is configured to execute user data produced on a specified platform based on users to be grouped, and the first sample data at least comprises position characteristics and culture custom characteristics of the users to be grouped;
the grouping information analysis module is configured to input the first sample data into a community grouping model to obtain first community grouping information of the users to be grouped, wherein the community grouping model is obtained through clustering training based on second sample data of a plurality of users on the specified platform, and the second sample data and the first sample data have the same characteristic dimension;
and the community grouping module is configured to perform community grouping on the users to be grouped according to the first community grouping information.
8. A content isolation apparatus, based on the user community implementation after community grouping performed by the community grouping apparatus of claim 7, the content isolation apparatus comprising:
the isolation target determining module is configured to execute a community tag based on a user community to which the user belongs or a community tag carried in multimedia information released by the user on a specified platform, and determine an isolation target needing content isolation;
and the content pushing module is configured to push the content to be pushed to the user based on the isolation target when the content is pushed to the user in the user community.
9. A server, comprising:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the community grouping method of any one of claims 1 to 5 or the content isolation method of claim 6.
10. A storage medium, wherein instructions in the storage medium, when executed by a processor in a server, enable the server to perform the community grouping method of any one of claims 1 to 5 or the content isolation method of claim 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911054465.4A CN110910262B (en) | 2019-10-31 | 2019-10-31 | Community grouping method and device, content isolation method and device and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911054465.4A CN110910262B (en) | 2019-10-31 | 2019-10-31 | Community grouping method and device, content isolation method and device and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110910262A true CN110910262A (en) | 2020-03-24 |
| CN110910262B CN110910262B (en) | 2023-02-17 |
Family
ID=69816206
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911054465.4A Active CN110910262B (en) | 2019-10-31 | 2019-10-31 | Community grouping method and device, content isolation method and device and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110910262B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902772A (en) * | 2012-09-27 | 2013-01-30 | 福建师范大学 | Web community discovery method based on multi-objective optimization |
| CN102929942A (en) * | 2012-09-27 | 2013-02-13 | 福建师范大学 | Social network overlapping community finding method based on ensemble learning |
| CN106251230A (en) * | 2016-07-22 | 2016-12-21 | 福建师范大学 | A kind of community discovery method propagated based on election label |
-
2019
- 2019-10-31 CN CN201911054465.4A patent/CN110910262B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102902772A (en) * | 2012-09-27 | 2013-01-30 | 福建师范大学 | Web community discovery method based on multi-objective optimization |
| CN102929942A (en) * | 2012-09-27 | 2013-02-13 | 福建师范大学 | Social network overlapping community finding method based on ensemble learning |
| CN106251230A (en) * | 2016-07-22 | 2016-12-21 | 福建师范大学 | A kind of community discovery method propagated based on election label |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110910262B (en) | 2023-02-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108108821B (en) | Model training method and device | |
| US20240086971A1 (en) | Systems, methods, and storage media for training a machine learning model | |
| CN107451199B (en) | Question recommendation method, device and equipment | |
| WO2018223719A1 (en) | Method for predicting insurance purchasing behavior of a user, device, computing apparatus, and medium | |
| KR101999471B1 (en) | Information recommendation methods and devices | |
| CN109816483B (en) | Information recommendation method and device and readable storage medium | |
| CN107463580B (en) | Click rate estimation model training method and device and click rate estimation method and device | |
| US20200356781A1 (en) | Method and apparatus for performing categorised matching of videos, and selection engine | |
| CN115917535A (en) | Recommendation model training method, recommendation device and computer readable medium | |
| CN107967280B (en) | Method and system for recommending songs by tag | |
| CN111708948B (en) | Content item recommendation method, device, server and computer readable storage medium | |
| US11727082B2 (en) | Machine-learning based personalization | |
| US20210012363A1 (en) | Device, method and computer-readable medium for analyzing customer attribute information | |
| CN111782941A (en) | Information recommendation method and device and server | |
| CN117409419A (en) | Image detection method, device and storage medium | |
| CN111680213B (en) | Information recommendation method, data processing method and device | |
| CN111294620A (en) | Video recommendation method and device | |
| CN113034168A (en) | Content item delivery method and device, computer equipment and storage medium | |
| CN110910262B (en) | Community grouping method and device, content isolation method and device and server | |
| KR20210052746A (en) | Method, apparatus and computer program for estimating sales volume of content to be productized | |
| KR20210052745A (en) | Method, apparatus and computer program for selecting promising content | |
| JP6896594B2 (en) | Music selection device, music selection method and program | |
| CN116739665A (en) | Information delivery method and device, electronic equipment and storage medium | |
| CN113010788A (en) | Information pushing method and device, electronic equipment and computer readable storage medium | |
| CN113194360B (en) | Video recommendation method and device, server and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |