CN108228595A - Speculate the method and system for obtaining user property - Google Patents

Speculate the method and system for obtaining user property Download PDF

Info

Publication number
CN108228595A
CN108228595A CN201611149073.2A CN201611149073A CN108228595A CN 108228595 A CN108228595 A CN 108228595A CN 201611149073 A CN201611149073 A CN 201611149073A CN 108228595 A CN108228595 A CN 108228595A
Authority
CN
China
Prior art keywords
user
sample
game
behavioral data
grader
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.)
Pending
Application number
CN201611149073.2A
Other languages
Chinese (zh)
Inventor
丁圣勇
樊勇兵
陈楠
赖培源
陈天
黄志兰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Telecom Corp Ltd
Original Assignee
China Telecom Corp Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Telecom Corp Ltd filed Critical China Telecom Corp Ltd
Priority to CN201611149073.2A priority Critical patent/CN108228595A/en
Publication of CN108228595A publication Critical patent/CN108228595A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/955Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses a kind of method and systems for speculating acquisition user property, are related to big data technical field.This method includes:The game behavioral data of user is obtained according to games log Information Statistics;And the user property of the acquisition user is speculated according to the game behavioral data and by the grader that training obtains;Wherein, which includes the correspondence of game behavioral data and user property.This method does not need to the more sensitive data for being related to privacy of user, and cost is relatively low.

Description

Speculate the method and system for obtaining user property
Technical field
The present invention relates to big data technical fields, more particularly to a kind of to speculate the method and system for obtaining user property.
Background technology
User draws a portrait, i.e. user information labeling.The core work of user's portrait is labelled for user, the weight to label Want the first purpose and be in order to allow people it will be appreciated that and facilitate computer disposal, such as statistic of classification can be done, such as, statistics happiness Male to female ratio in the number of users of joyous certain money game, the crowd that certain money is liked to play etc..Data mining work, example can also be done Which type of movement brand the people for such as certain money being liked to play using correlation rule calculating generally prefers that.
Big data processing, be unable to do without the operation of computer, and label provides a kind of easily mode so that computer can Procedure treatment and the relevant information of people or even people " can be understood " by algorithm, model.When computer has such ability Afterwards, either search engine, recommended engine, advertisement launch etc. various application fields, all will further promote precision, improve The efficiency of acquisition of information.
User portrait generally by relatively straightforward data (for example, user send out microblogging, user shopping record etc.) come Speculate its attribute.
Invention content
It was found by the inventors of the present invention that user's representation data generally has stronger privacy, for example user does shopping and records Or internet records etc., the procurement cost of these data are higher.
According to the first aspect of the invention, a kind of method for speculating and obtaining user property is provided, including:According to game day Will Information Statistics obtain the game behavioral data of user;And according to the game behavioral data and the classification obtained by training Device speculates the user property for obtaining the user;Wherein, it is corresponding with user property to include game behavioral data for the grader Relationship.
In one embodiment, according to games log Information Statistics obtain user game behavioral data the step of it Before, the method further includes:It is trained to obtain by the sample of users cluster of the known users attribute to desired amt described Grader.
In one embodiment, it is trained to obtain by the sample of users cluster of the known users attribute to desired amt The step of grader, includes:By gaming platform choose desired amt the user by the use of known terminal number registration as Sample of users cluster;The log-on message of the sample of users cluster is obtained by business system, the log-on message, which includes, to be used Family attribute;And the log-on message of the sample of users cluster and the game behavioral data of the sample of users cluster are instructed Practice, obtain the grader, wherein, the grader includes the correspondence of game behavioral data and user property.
In one embodiment, using selecting sample of users of the Random Forest model to the known users attribute of desired amt Cluster is trained to obtain the grader.
In one embodiment, the game behavioral data includes:For downloading the model of the mobile phone of the game and screen Curtain resolution ratio and the attribute of the game;The user property includes:Gender, the range of age and the game liked of user Type.
In the above-mentioned methods, speculate acquisition user property using user's games log information of muting sensitive sense and low cost. This method does not need to the more sensitive data for being related to privacy of user, and cost is relatively low.
According to the second aspect of the invention, a kind of system for speculating and obtaining user property is provided, including:Statistic unit, For obtaining the game behavioral data of user according to games log Information Statistics;And analytic unit, for according to the game Behavioral data and the user property that the acquisition user is speculated by the grader that training obtains;Wherein, the grader includes The correspondence of game behavioral data and user property.
In one embodiment, the system also includes:Training unit, for passing through the known users category to desired amt The sample of users cluster of property is trained to obtain the grader.
In one embodiment, the statistic unit chooses the utilization known terminal number of desired amt by gaming platform The user of registration is as sample of users cluster;And pass through the log-on message that business system obtains the sample of users cluster, institute It states log-on message and includes user property;Log-on message and the sample of users of the training unit to the sample of users cluster The game behavioral data of cluster is trained, and obtains the grader, wherein, the grader includes game behavioral data with using The correspondence of family attribute.
In one embodiment, the training unit utilizes and selects known users category of the Random Forest model to desired amt The sample of users cluster of property is trained to obtain the grader.
In one embodiment, the game behavioral data includes:For downloading the model of the mobile phone of the game and screen Curtain resolution ratio and the attribute of the game;The user property includes:Gender, the range of age and the game liked of user Type.
In above system, acquisition user property is speculated using user's games log information of muting sensitive sense and low cost. The system does not need to the more sensitive data for being related to privacy of user, and cost is relatively low.
By referring to the drawings to the detailed description of exemplary embodiment of the present invention, other feature of the invention and its Advantage will become apparent.
Description of the drawings
The attached drawing of a part for constitution instruction describes the embodiment of the present invention, and is used to solve together with the description Release the principle of the present invention.
With reference to attached drawing, according to following detailed description, the present invention can be more clearly understood, wherein:
Fig. 1 is the flow chart for showing the method according to an embodiment of the invention for speculating and obtaining user property.
Fig. 2 is the flow chart for showing the method in accordance with another embodiment of the present invention for speculating and obtaining user property.
Fig. 3 is to show that training according to an embodiment of the invention obtains the flow chart of the method for grader.
Fig. 4 is the schematic diagram for schematically showing selection Random Forest model according to an embodiment of the invention.
Fig. 5 is the structure for schematically showing the system according to an embodiment of the invention for speculating and obtaining user property Figure.
Specific embodiment
Carry out the various exemplary embodiments of detailed description of the present invention now with reference to attached drawing.It should be noted that:Unless in addition have Body illustrates that the unlimited system of component and the positioned opposite of step, numerical expression and the numerical value otherwise illustrated in these embodiments is originally The range of invention.
Simultaneously, it should be appreciated that for ease of description, the size of the various pieces shown in attached drawing is not according to reality Proportionate relationship draw.
It is illustrative to the description only actually of at least one exemplary embodiment below, is never used as to the present invention And its application or any restrictions that use.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable In the case of, the technology, method and apparatus should be considered as authorizing part of specification.
In shown here and discussion all examples, any occurrence should be construed as merely illustrative, without It is as limitation.Therefore, the other examples of exemplary embodiment can have different values.
It should be noted that:Similar label and letter represents similar terms in following attached drawing, therefore, once a certain Xiang Yi It is defined in a attached drawing, then in subsequent attached drawing does not need to that it is further discussed.
Highly refined signature identification as defined in label is usually manual, such as age bracket label:25~35 years old, region mark Label:Beijing.Label shows two important features:Semantization and short text.For semantization, people can easily understand each Meaning tag.But also user draws a portrait, model has practical significance for this, can preferably meet business demand.For example, judge to use Family preference.For short text, each label usually only represents a kind of meaning, and label is pre- without doing excessive text analyzing etc. again in itself Work is handled, this is provides convenience using machine extraction standard information.The core work of user's portrait is for user's mark Label.
It was found by the inventors of the present invention that user's portrait generally speculates its attribute by relatively straightforward data, and these User's representation data generally has stronger privacy, such as user's shopping record or internet records etc., the acquisition of these data Cost is higher.
Fig. 1 is the flow chart for showing the method according to an embodiment of the invention for speculating and obtaining user property.
In step S102, the game behavioral data of user is obtained according to games log Information Statistics.For example, the game behavior Data can include:For downloading attribute of the model of the mobile phone of the game and screen resolution and the game etc..
For example, game operation platform can generate a large amount of user's games log information, such as shown in table 1 daily.It needs Bright, table 1 is only exemplary, and illustrates only the partial game log information of certain customers, the scope of the present invention and not only It is limited to this.
1 user's games log information table of table
According to games log information, (such as the game download behavior of the recent game behavioral data of certain user can be counted Data), such as the game behavioral data can include:Mobile phone model, screen resolution and game attributes etc..
In step S104, speculate that the user for obtaining user belongs to according to game behavioral data and by the grader that training obtains Property.Wherein, which includes the correspondence of game behavioral data and user property.That is, obtaining above-mentioned game After behavioral data, can according to machine learning train come grader will need the user that draws a portrait to classify, so as to Export desired result, thus it is speculated that obtain the attribute of the user of unknown portrait (i.e. unknown properties).For example, the user property can wrap It includes:The gender of user, the range of age and type of play for liking etc..User can be obtained using obtained user property to draw Picture.
In the above-described embodiments, speculate acquisition user property using user's games log information of muting sensitive sense and low cost (such as gender, age etc.).This method does not need to more sensitive data (such as internet records, the individual for being related to privacy of user Spatial data etc.), and cost is relatively low.
In the above method of the embodiment of the present invention, the game behavior of user can be characterized (such as user The game frequency, game classification etc.), Behavior-based control feature carries out classification prediction.
In one embodiment, before step S102, the manufacturing method can also include:By to desired amt The sample of users cluster of known users attribute is trained to obtain grader.
Fig. 2 is the flow chart for showing the method in accordance with another embodiment of the present invention for speculating and obtaining user property.
In step S200, it is trained and is classified by the sample of users cluster of the known users attribute to desired amt Device.Wherein, which includes the correspondence of game behavioral data and user property.In one embodiment, it can utilize Selection Random Forest model is trained to obtain the grader to the sample of users cluster of the known users attribute of desired amt.
In step S202, the game behavioral data of user is obtained according to games log Information Statistics.
In step S204, speculate that the user for obtaining user belongs to according to game behavioral data and by the grader that training obtains Property.
In this embodiment, it is trained to obtain grader first with the sample of users cluster of known users attribute, then Statistics obtains the game behavioral data of some (unknown properties) user, and the use is obtained according to the game behavioral data and grader The attribute at family, so as to obtain the portrait of the user.
Fig. 3 is to show that training according to an embodiment of the invention obtains the flow chart of the method for grader.The Fig. 3's is each A step is illustrating to the step S200 in Fig. 2.
In step S302, the user by the use of known terminal number registration of desired amt is chosen as sample by gaming platform This user cluster.For example, the user that 20,000 telecommunication handset number registrations are chosen by gaming platform is used as sample of users cluster.
In step S304, the log-on message of sample of users cluster is obtained by business system, which, which includes, uses Family attribute.For example, obtaining the log-on message of above-mentioned 20,000 sample of users by telecommunication system backstage, which can wrap User property containing sample of users, such as gender, age etc..
In step S306, the game behavioral data of log-on message and sample of users cluster to sample of users cluster (for example, The game behavior number of these sample of users clusters can be obtained by counting the games log information of these sample of users clusters According to) be trained, grader is obtained, wherein, which includes the correspondence of game behavioral data and user property.Example Such as, by the log-on message of above-mentioned 20,000 users and in the recent period, game behavioral data (such as behavioral data is downloaded in game) is input to machine Classified automatically in device learning software, train grader.
In this embodiment, the grader that machine learning obtains is to utilize a certain number of known users attributes (or portrait) The training of user's cluster, such as training sample may come from the identified use using phone number of gaming platform Family can obtain the accurate portrait information of these sample of users by the client management system of common carrier, so as to this A little sample datas can obtain required grader after training.
Fig. 4 is the schematic diagram for schematically showing selection Random Forest model according to an embodiment of the invention.
In the selection Random Forest model, the game behavioral data that will can need to speculate the user of user property is from top End node enters, and is classified according to the class condition that training obtains to the game behavioral data of the user, until reaching most bottom At some node of layer (wherein, each node of the bottom can correspond to a kind of user property), on speculating to obtain State the user property of user.
Fig. 5 is the structure for schematically showing the system according to an embodiment of the invention for speculating and obtaining user property Figure.As shown in figure 5, the system can include:Statistic unit 502 and analytic unit 504.
The statistic unit 502 is used to obtain the game behavioral data of user according to games log Information Statistics.For example, game Behavioral data can include:For downloading attribute of the model of the mobile phone of the game and screen resolution and the game etc..
The analytic unit 504 is used to speculate the acquisition use according to game behavioral data and by the grader that training obtains The user property at family.Wherein, which can include the correspondence of game behavioral data and user property.For example, user Attribute can include:The gender of user, the range of age and type of play for liking etc..
In the above-described embodiments, speculate acquisition user property using user's games log information of muting sensitive sense and low cost (such as gender, age etc.).The system does not need to more sensitive data (such as internet records, the individual for being related to privacy of user Spatial data etc.), and cost is relatively low.
In an embodiment of the present invention, the user for draw a portrait and (needing to obtain user property) obtained will be counted Game behavioral data be input in the obtained grader of training, which exports the user data namely the use of the user The corresponding portrait in family.
In one embodiment, as shown in figure 5, the system can also include:Training unit 506.The training unit 506 is used It is trained to obtain grader in the sample of users cluster by the known users attribute to desired amt.For example, the training list Member 506, which can utilize, selects Random Forest model to be trained the sample of users cluster of the known users attribute of desired amt To the grader.
In one embodiment, statistic unit 502 can choose the utilization known terminal of desired amt by gaming platform The user of number registration is as sample of users cluster.And the statistic unit 502 can obtain the sample by business system The log-on message of user's cluster.The log-on message can include user property.Training unit 506 can be to the sample of users cluster Log-on message and the game behavioral data of the sample of users cluster be trained, obtain grader.Wherein, which can be with Correspondence comprising game behavioral data and user property.
In the above-described embodiments, the grader that machine learning obtains is the sample using a certain number of known users attributes The training of user's cluster.For example, training sample may come from the identified user using phone number of gaming platform, The accurate portrait information of these sample of users can be obtained by the client management system of common carrier, so as to these samples Notebook data can obtain required grader after training.
So far, the present invention is described in detail.In order to avoid the design of the masking present invention, it is public that this field institute is not described Some details known.Those skilled in the art as described above, can be appreciated how to implement technology disclosed herein completely Scheme.
The method and system of the present invention may be achieved in many ways.For example, can by software, hardware, firmware or Software, hardware, firmware any combinations come realize the present invention method and system.The said sequence of the step of for the method Merely to illustrate, the step of method of the invention, is not limited to sequence described in detail above, special unless otherwise It does not mentionlet alone bright.In addition, in some embodiments, the present invention can be also embodied as recording program in the recording medium, these programs Including being used to implement machine readable instructions according to the method for the present invention.Thus, the present invention also covering stores to perform basis The recording medium of the program of the method for the present invention.
Although some specific embodiments of the present invention are described in detail by example, the skill of this field Art personnel it should be understood that above example merely to illustrating, the range being not intended to be limiting of the invention.The skill of this field Art personnel are it should be understood that can without departing from the scope and spirit of the present invention modify to above example.This hair Bright range is defined by the following claims.

Claims (10)

1. a kind of speculate the method for obtaining user property, which is characterized in that including:
The game behavioral data of user is obtained according to games log Information Statistics;And
The user property of the acquisition user is speculated according to the game behavioral data and by the grader that training obtains;Its In, the grader includes the correspondence of game behavioral data and user property.
2. according to the method described in claim 1, it is characterized in that, in the game that user is obtained according to games log Information Statistics Before the step of behavioral data, the method further includes:
It is trained to obtain the grader by the sample of users cluster of the known users attribute to desired amt.
3. according to the method described in claim 2, it is characterized in that, the sample by the known users attribute to desired amt is used The step of family cluster is trained to obtain the grader includes:
The user by the use of known terminal number registration of desired amt is chosen as sample of users cluster by gaming platform;
The log-on message of the sample of users cluster is obtained by business system, the log-on message includes user property;With And
The game behavioral data of log-on message and the sample of users cluster to the sample of users cluster is trained, and is obtained The grader, wherein, the grader includes the correspondence of game behavioral data and user property.
4. according to the method described in claim 2, it is characterized in that,
The sample of users cluster of the known users attribute of desired amt is trained to obtain institute using Random Forest model is selected State grader.
5. according to the method described in claim 1, it is characterized in that,
The game behavioral data includes:For downloading the model of the mobile phone of the game and screen resolution and the trip The attribute of play;
The user property includes:Gender, the range of age and the type of play liked of user.
6. a kind of speculate the system for obtaining user property, which is characterized in that including:
Statistic unit, for obtaining the game behavioral data of user according to games log Information Statistics;And
Analytic unit, for speculating the acquisition user's according to the game behavioral data and by the grader that training obtains User property;Wherein, the grader includes the correspondence of game behavioral data and user property.
7. system according to claim 6, which is characterized in that further include:
Training unit, for being trained to obtain described point by the sample of users cluster of the known users attribute to desired amt Class device.
8. system according to claim 7, which is characterized in that
The statistic unit chooses the user by the use of known terminal number registration of desired amt as sample by gaming platform User's cluster;And passing through the log-on message that business system obtains the sample of users cluster, the log-on message includes user Attribute;
The training unit is to the log-on message of the sample of users cluster and the game behavioral data of the sample of users cluster It is trained, obtains the grader, wherein, the grader includes the correspondence of game behavioral data and user property.
9. system according to claim 7, which is characterized in that
The training unit using select Random Forest model to the sample of users cluster of the known users attribute of desired amt into Row training obtains the grader.
10. system according to claim 6, which is characterized in that
The game behavioral data includes:For downloading the model of the mobile phone of the game and screen resolution and the trip The attribute of play;
The user property includes:Gender, the range of age and the type of play liked of user.
CN201611149073.2A 2016-12-14 2016-12-14 Speculate the method and system for obtaining user property Pending CN108228595A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611149073.2A CN108228595A (en) 2016-12-14 2016-12-14 Speculate the method and system for obtaining user property

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611149073.2A CN108228595A (en) 2016-12-14 2016-12-14 Speculate the method and system for obtaining user property

Publications (1)

Publication Number Publication Date
CN108228595A true CN108228595A (en) 2018-06-29

Family

ID=62638191

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611149073.2A Pending CN108228595A (en) 2016-12-14 2016-12-14 Speculate the method and system for obtaining user property

Country Status (1)

Country Link
CN (1) CN108228595A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110400106A (en) * 2019-06-17 2019-11-01 天津五八到家科技有限公司 Information acquisition method, device and electronic equipment
CN110489453A (en) * 2019-07-02 2019-11-22 广东工业大学 User's game real-time recommendation method and system based on big data log analysis

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103020412A (en) * 2011-09-28 2013-04-03 深圳市快播科技有限公司 Method and system for counting user attributes of game platform
CN104281882A (en) * 2014-09-16 2015-01-14 中国科学院信息工程研究所 Method and system for predicting social network information popularity on basis of user characteristics
CN105045916A (en) * 2015-08-20 2015-11-11 广东顺德中山大学卡内基梅隆大学国际联合研究院 Mobile game recommendation system and recommendation method thereof
CN105477860A (en) * 2015-12-22 2016-04-13 北京奇虎科技有限公司 Game activity recommending method and device
CN105933425A (en) * 2016-05-18 2016-09-07 北京奇虎科技有限公司 Application recommendation method and device
CN106126597A (en) * 2016-06-20 2016-11-16 乐视控股(北京)有限公司 User property Forecasting Methodology and device

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103020412A (en) * 2011-09-28 2013-04-03 深圳市快播科技有限公司 Method and system for counting user attributes of game platform
CN104281882A (en) * 2014-09-16 2015-01-14 中国科学院信息工程研究所 Method and system for predicting social network information popularity on basis of user characteristics
CN105045916A (en) * 2015-08-20 2015-11-11 广东顺德中山大学卡内基梅隆大学国际联合研究院 Mobile game recommendation system and recommendation method thereof
CN105477860A (en) * 2015-12-22 2016-04-13 北京奇虎科技有限公司 Game activity recommending method and device
CN105933425A (en) * 2016-05-18 2016-09-07 北京奇虎科技有限公司 Application recommendation method and device
CN106126597A (en) * 2016-06-20 2016-11-16 乐视控股(北京)有限公司 User property Forecasting Methodology and device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110400106A (en) * 2019-06-17 2019-11-01 天津五八到家科技有限公司 Information acquisition method, device and electronic equipment
CN110489453A (en) * 2019-07-02 2019-11-22 广东工业大学 User's game real-time recommendation method and system based on big data log analysis

Similar Documents

Publication Publication Date Title
US11875273B2 (en) Machine learning classification of digital content for mobile devices
US10762299B1 (en) Conversational understanding
CN105512326B (en) A kind of method and system that picture is recommended
CN105224623B (en) The training method and device of data model
CN109145204B (en) Portrait label generation and use method and system
EP3288275B1 (en) Methods and systems of providing visual content editing functions
CN109155136A (en) The computerized system and method for highlight are detected and rendered automatically from video
CN108885639A (en) Properties collection navigation and automatic forwarding
CN108062375A (en) A kind of processing method, device, terminal and the storage medium of user's portrait
US20140095308A1 (en) Advertisement distribution apparatus and advertisement distribution method
JP2015181025A (en) Leveraging context to present content on communication device
CN110059255A (en) A kind of browser air navigation aid, device and medium
CN108932451A (en) Audio-video frequency content analysis method and device
CN109286850A (en) A kind of video labeling method and terminal based on barrage
CN106688215A (en) Automated click type selection for content performance optimization
US9104507B2 (en) Developer platform
CN107909466A (en) Customer relationship network display method, apparatus, equipment and readable storage medium storing program for executing
CN101668176A (en) Multimedia content-on-demand and sharing method based on social interaction graph
CN104484435B (en) The method of alternate analysis user behavior
CN109688479A (en) A kind of barrage display methods, barrage display device and barrage display server
US20150287069A1 (en) Personal digital engine for user empowerment and method to operate the same
CN111722766A (en) Multimedia resource display method and device
CN106649830A (en) Information showing method and device
CN112199296A (en) Page testing method and device, computer equipment and storage medium
US20070094335A1 (en) Systems and methods for providing a visual indicator of magnitude

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20180629

RJ01 Rejection of invention patent application after publication