CN109543734A - User portrait method and device, storage medium - Google Patents
User portrait method and device, storage medium Download PDFInfo
- Publication number
- CN109543734A CN109543734A CN201811353247.6A CN201811353247A CN109543734A CN 109543734 A CN109543734 A CN 109543734A CN 201811353247 A CN201811353247 A CN 201811353247A CN 109543734 A CN109543734 A CN 109543734A
- Authority
- CN
- China
- Prior art keywords
- user
- network
- target
- portrait
- collection
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of user's portrait method and device, storage medium.This method comprises: determining candidate user collection according to the user data of objective time interval；Wherein, the data acquisition system that the candidate user integrates as candidate user, the candidate user are that the user of first network is persistently used in the objective time interval；The candidate user collection is handled using object classifiers, obtains target user's collection；Wherein, the object classifiers are used to predict that the candidate user to concentrate the target user for reaching destination probability by the probability that the first network is converted to the second network；The target user that the target user concentrates is handled using target cluster device, obtains user's portrait of the target user；Wherein, the target cluster device is used to add user's portrait label for the target user, obtains user's portrait of the target.Method of the invention improves the success rate for turning network service to a certain extent, and flexibility is higher, reduces cost of human resources.
Description
Technical field
The present invention relates to data processing technique more particularly to a kind of user portrait method and devices, storage medium.
Background technique
The communication network provided by common carrier, may be implemented to communicate between user.With the development of communication technology, it uses
Family can convert communication network to enjoy new generation network technological service.
Currently, common carrier is recommended to turn network service to user by the way of generally pushing away using a line service personnel.Having
When body is realized, a line service personnel may turn the user of net firstly the need of determination, that is, the network used is not latest generation
Then the user of network recommends to turn network service one by one to these users, the way of recommendation is usually telephone reference or pushes away on the spot
It recommends.
It is existing to turn the network service way of recommendation, user is recommended without differentiation, the indiscriminate turn net that carries out, this unified ground
It pushes away mode and had both wasted a large amount of human resources, and the success rate for turning network service is lower.
Summary of the invention
The present invention provides a kind of user's portrait method and device, storage medium, recommends to solve the existing network service that turns
Mode has that waste of manpower resource and success rate are lower.
In a first aspect, the present invention provides a kind of user's portrait method, comprising:
According to the user data of objective time interval, candidate user collection is determined；Wherein, the candidate user integrates as candidate user
Data acquisition system, the candidate user are that the user of first network is persistently used in the objective time interval；
The candidate user collection is handled using object classifiers, obtains target user's collection；Wherein, the object classifiers are used
The target for reaching destination probability by the probability that the first network is converted to the second network is concentrated to use in the prediction candidate user
Family；
The target user that the target user concentrates is handled using target cluster device, the user for obtaining the target user draws
Picture；Wherein, the target cluster device is used to add user's portrait label for the target user, and the user for obtaining the target draws
Picture.
Second aspect, the present invention provide a kind of user's portrait device, comprising:
Determining module determines candidate user collection for the user data according to objective time interval；Wherein, the candidate user
Integrate the data acquisition system as candidate user, the candidate user is that the user of first network is persistently used in the objective time interval；
First processing module obtains target user's collection for handling the candidate user collection using object classifiers；Its
In, the object classifiers are for predicting that the candidate user concentration is reached by the probability that the first network is converted to the second network
To the target user of destination probability；
Second processing module obtains institute for handling the target user that the target user concentrates using target cluster device
State user's portrait of target user；Wherein, the target cluster device is used to add user's portrait label for the target user, obtains
User to the target draws a portrait.
The third aspect, the present invention provide a kind of user and draw a portrait device, comprising: transceiver, processor, memory and total
Line, the transceiver, the processor and the memory are connect with the bus respectively, and the memory is stored with program
Instruction, the processor operation described program instruction is to execute following method:
According to the user data of objective time interval, candidate user collection is determined；Wherein, the candidate user integrates as candidate user
Data acquisition system, the candidate user are that the user of first network is persistently used in the objective time interval；
The candidate user collection is handled using object classifiers, obtains target user's collection；Wherein, the object classifiers are used
The target for reaching destination probability by the probability that the first network is converted to the second network is concentrated to use in the prediction candidate user
Family；
The target user that the target user concentrates is handled using target cluster device, the user for obtaining the target user draws
Picture；Wherein, the target cluster device is used to add user's portrait label for the target user, and the user for obtaining the target draws
Picture.
Fourth aspect, the present invention provide a kind of computer readable storage medium, are stored thereon with computer program,
The computer program is executed by processor to realize such as the described in any item methods of first aspect.
User's portrait method provided by the invention, it is first determined the candidate user for using first network always, then, to this
A little candidate users carry out prediction classification, it is determined that have the target user that greater probability turn net in candidate user, it is in turn, right
Target user carries out cluster verifying, obtains user's portrait of target user, in this way, when specifically turn the recommendation of network service,
Business personnel can draw a portrait according to user, carry out turning network service recommendation as far as possible for target user, this can be mentioned to a certain extent
Height turns the success rate of network service, and flexibility is higher, is adaptable to a variety of turn net scene；And it is different from the existing indistinguishably side of pushing away
Formula can effectively save the workload of business personnel, reduce cost of human resources.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows the implementation for meeting the disclosure
Example, and together with specification for explaining the principles of this disclosure.
Fig. 1 is a kind of flow diagram of user's portrait method provided by the embodiment of the present invention；
Fig. 2 is a kind of set division mode schematic diagram provided by the embodiment of the present invention；
Fig. 3 is the flow diagram of the portrait method of another kind user provided by the embodiment of the present invention；
Fig. 4 is the flow diagram of the portrait method of another kind user provided by the embodiment of the present invention；
Fig. 5 is a kind of functional block diagram of user's portrait device provided by the embodiment of the present invention；
Fig. 6 is a kind of entity structure schematic diagram of user's portrait device provided by the embodiment of the present invention.
Through the above attached drawings, it has been shown that the specific embodiment of the disclosure will be hereinafter described in more detail.These attached drawings
It is not intended to limit the scope of this disclosure concept by any means with verbal description, but is by referring to specific embodiments
Those skilled in the art illustrate the concept of the disclosure.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary embodiment
Described in embodiment do not represent all implementations consistent with this disclosure.On the contrary, they be only with it is such as appended
The example of the consistent device and method of some aspects be described in detail in claims, the disclosure.
Noun according to the present invention is explained first:
Turn net: referring to and currently employed (currently providing communication service for user) network is converted into other networks.
In specific implementation, the network of newer generation is usually converted to by a generation (or mostly generation) network before, for example,
By Generation Mobile Telecommunication System technology (the 2th Generation mobile communication technology, 2G) net
Network is converted to fourth generation mobile communication technology (the 4th Generation mobile communication
Technology, 4G) network.
User's set: refer to the set being made of the user data of at least one user.
The specific application scenarios of the present invention are as follows: user is directed to the portrait scene for turning network service.Wherein, a preferred application
Scene are as follows: turn in the scene that net is recommended to user.It is, carrying out portrait point to user using offer scheme of the present invention
Analysis, user's portrait can be applied particularly to service guidance personnel and carry out turning network service recommendation to user.
In turning the application scenarios that net is recommended, existing implementation is that a line service personnel are indiscriminate to not using
The user of target network pushes away with carrying out, and there is a problem of turning net success rate lower, and waste a large amount of human resources.
And the portrait method of user provided by the embodiment of the present invention, it is intended to the technical problem as above of the prior art is solved, and
It is proposed following resolving ideas: by being trained to classifier and cluster device, thus, using trained classifier and cluster device
The candidate user user of target network (do not use) that may turn net is handled, to determine has the larger mesh that may turn net
User's portrait is marked, realizes that the differentiation to user handles and recommends.
The method in addition, it should be noted that, user provided by the invention draws a portrait, except be applied to aforementioned turn of net recommendation scene it
Outside, other scenes can also be applied to.For example, scene is analyzed for the reason of may turning network users, etc..
How to be solved with technical solution of the specifically embodiment to technical solution of the present invention and the application below above-mentioned
Technical problem is described in detail.These specific embodiments can be combined with each other below, for the same or similar concept
Or process may repeat no more in certain embodiments.Below in conjunction with attached drawing, the embodiment of the present invention is described.
Embodiment one
The embodiment of the invention provides a kind of user portrait methods.Referring to FIG. 1, this method comprises the following steps:
S102 determines candidate user collection according to the user data of objective time interval；Wherein, the candidate user collection is candidate
The data acquisition system of user, the candidate user are that the user of first network is persistently used in the objective time interval.
S104 handles candidate user collection using object classifiers, obtains target user's collection；Wherein, the object classifiers
Reach the target of destination probability by the probability that the first network is converted to the second network for predicting that the candidate user is concentrated
User.
S106 handles the target user that the target user concentrates using target cluster device, obtains the target user's
User's portrait；Wherein, the target cluster device is used to add user's portrait label for the target user, obtains the target
User's portrait.
It should be noted that only by taking first network and the second network as an example, the second network is to turn net in the embodiment of the present invention
The target network of business.
In specific implementation, it is contemplated that the type of communication network may have it is multiple, for example, 2G network, 4G network, 5G network
Deng aforementioned first network involved in the embodiment of the present invention can be the all-network in addition to the second network, alternatively, can also
To be one of other multiple networks in addition to the second network network, alternatively, being also possible to its in addition to the second network
One or more networks that he meets certain condition in network.
It illustrates, it is assumed that network type includes 2G network, 3G network, 4G network and 5G network, and the second network is 4G
Network.It is possible to using the all-network in addition to 4G network: 2G network, 3G network and 5G network are executed as first network
Aforementioned processing；Alternatively, can be by one of other networks in addition to 4G network particular network, if 2G network is as the first net
Network executes aforementioned processing；Alternatively, can by other networks in addition to 4G network in the prior-generation of 4G network or the net in mostly generation
Network is as first network, it is, first network at this time includes 2G network and 3G network.
It is on the other side, if first network it has been determined that if the second network can be the all-network in addition to first network,
Alternatively, being also possible to one of other multiple networks in addition to first network network, alternatively, being also possible to except first network
Except other networks in meet one or more networks of another condition.Wherein, another condition may are as follows: first network it
Certain emerging generation or mostly for network afterwards.
That is, the first network is at least one in 2G network, 3G network and 4G network in the embodiment of the present invention
Kind network, second network are as follows: at least one of 3G network, 4G network and 5G network network；And the first network with
Same network is not present in second network.
For example, the second network can be with if first network is 2G network are as follows: in 3G network, 4G network and 5G network at least
A kind of network；Alternatively, the second network can be with are as follows: at least one of 4G network and 5G network net if first network is 3G network
Network；Alternatively, the second network can be with are as follows: 5G network if first network is 4G network.
When implementing this programme, the network class of first network and the second network referred specifically to generation can be set as needed
Type, the present invention are not particularly limited this.Also, the present invention is not particularly limited network type, be not limited to presently, there are
2G network, 3G network, 4G network and 5G network, for the future may appear other network types involved in turn network service,
It is also applied for the portrait method of user provided by the embodiment of the present invention.
In the embodiment of the present invention, related objective time interval be may range from: the initial carrier moment of common carrier
(at the time of starting to provide communication service for first user) is to current time.It is found that when objective time interval takes aforementioned range most
When big collection, that is, objective time interval is full time range, is equivalent to is not defined to objective time interval at this time.In practical realization
When, objective time interval is set as needed.For example, objective time interval can be set as unit of year, such as objective time interval is set
It is set to: whole year in 2000 or 2000~2010 years etc..Further, it is also possible to be set as unit of month, day, hour, minute etc.
Objective time interval repeats no more.
Objective time interval is determined by two timing nodes, hereinafter, using the initial time of objective time interval as the first moment, by mesh
It was the second moment that the finish time for marking the period, which is used as, was specifically described.
In addition, the default time section of objective time interval can also be set in a feasible realization scene, at this point, if
When the executive device does not receive specific time information, this method can be executed according to the preset default time section；If connecing
Specific time information is received, then information determines objective time interval at the time of receiving.For example, setting default time section to
1 year before current time.
In addition to this, user data involved in the embodiment of the present invention can be from the data of common carrier maintenance
Library.
Specifically, user data may include following but be not limited to following information: user information, network type (terminal system
Formula), communication service used by user (which kind of set meal used), terminal price and user communication data.Wherein, user information can
At least one of to include but is not limited to: in age, gender, the networking time limit, moon spending amount and permanent residence.And user communicates number
According to can include but is not limited to: the duration of call, message registration (dealing conversation number), flow service condition (are used in net duration
At least one of the duration of family generation signaling message).
In addition, in some special application scenarios, user data can also include but is not limited in addition to aforementioned information
Following at least one information: open an account ground, mobile phone terminal brand and terminal models.
Based on this, assembly diagram shown in Fig. 2 is please referred to, the embodiment of the present invention provides a kind of side for constructing candidate user collection
Formula:
The user data at the first moment is obtained in the database, determines the user's set A for wherein using first network；With
And the user data at the second moment is obtained in the database, determine the user's set B for wherein using the second network.To obtain
The intersection of user's set A and user's set B；For candidate user collection, obtain the user's in user's set A in addition to the intersection
Set, obtains candidate user collection.
For example, if objective time interval are as follows: on January 1st, 2018 and on May 1st, 2018, first network are 2G network, the
Two networks are 4G network, then when executing S102, it is all using 2G network can to filter out on January 1st, 2018 in the database
User set A, also, filter out all user set B using 4G network on May 1st, 2018 in the database, then to user
Set A and user's set B take intersection, obtain set C, it is known that, set C is by be converted to the by first network in objective time interval
What the user of two networks was constituted, therefore, the set for removing the certain customers after set C in user's set A is obtained, can be obtained
Candidate user collection.
In the embodiment of the present invention, after determining candidate user collection, candidate user collection can be input to object classifiers, it is defeated
Result is the set that candidate user concentrates the user converted by first network with greater probability to the second network out, that is,
Target user's collection.
In the embodiment of the present invention, object classifiers can according to sample training collection training obtain, using sample training collection to point
Class device is trained can complete before this programme execution, alternatively, can also execute before S104.
At this point, in a feasible realization scene Fig. 3 can be referred to, before S104, this method further includes walking as follows
It is rapid:
S103 is trained preliminary classification device using sample of users collection using sorting algorithm, until the classification after training
The predictablity rate of device is greater than preset accuracy rate threshold value, obtains the object classifiers.
Specifically, sorting algorithm involved in the embodiment of the present invention can include but is not limited to: XGBoost algorithm or with
Machine forest algorithm.
The step in specific implementation, can obtain object classifiers, trained termination condition by taking turns training more are as follows:
The predictablity rate of classification after training is greater than preset accuracy rate threshold value.Wherein, predictablity rate refers to classifier output
Degree of agreement between prediction result and actual result.For example, if 100 user data of input to classifier, and it is known this 100
It is a with being converted by first network to the second network per family, if it is that target is used that the output result of classifier, which is 50 users therein,
Family, it is determined that the predictablity rate of the classifier at this time is 50%.In addition, accuracy rate threshold value can be set as needed, the present invention
Embodiment is not particularly limited this.
So, after each training, predictablity rate to the prediction result of classifier output and preset accurate
Rate threshold value is judged, if predictablity rate is more than or equal to preset accuracy rate threshold value, by the classification after current training
Device is as object classifiers.Conversely, continuing to instruct classifier if predictablity rate is less than preset accuracy rate threshold value
Practice.
Sample of users collection involved in the embodiment of the present invention, which can use to be completed in database, turns net (first network conversion
To the second network) user data constitute.
In a preferred realization scene, it can be in the preceding aim period and converted by first network to the second network
Subscriber data set.At this point, referring to FIG. 4, this method can also include the following steps: before executing aforementioned S103
S101 determines the sample of users collection according to the user data of the objective time interval, wherein sample of users collection is
The data acquisition system of sample of users, the sample of users are to be converted to second net by the first network in the objective time interval
The user of network.
In order to make it easy to understand, can be with reference to assembly diagram shown in Fig. 2, in this implementation, user's set A and use
The set C for taking intersection to constitute between the set B of family, as sample of users collection.
The embodiment of the present invention is unlimited for the execution sequence of S101 and S102, and Fig. 4 show a kind of possible realization side
Formula, for example, in specific implementation, S101 and S102 can also first carry out S102 and execute S101 again, or be performed simultaneously.
Based on foregoing schemes, target user can be determined in each candidate user, at this point, determining target user is concentrated
Each user data input to trained target cluster device, can be obtained target user user portrait, it is, obtaining
The user data of user's portrait label is added.
In the embodiment of the present invention, user's portrait shows in the form of user draws a portrait label.
For turning to net this specific application scenarios, in the embodiment of the present invention, user's label of drawing a portrait may include but unlimited
In following at least one: old user, vice card user, the second not yet covered with networks user and Internet of Things network users.
In order to make it easy to understand, drawing a portrait and marking to aforementioned user so that first network is 2G network, the second network is 4G network as an example
Label are illustrated.
Firstly, first kind user holds 2G terminal mostly, and it is nearly free from every month or generates and is total lower than 20M
Flow in conjunction with its specific communication data, determines that the user of such user draws a portrait label there are also the features such as the universal age is higher
Are as follows: " old user " label；Second class user holds dual-card dual-standby terminal mostly, alternatively, possess mostly in user two and with
On mobile phone card, terminal generates a large amount of 4G flow, but this connection card used is nearly free from flow, determines such user
User draw a portrait label are as follows: " vice card user " label；The obvious characteristic of third class user be leave permanent residence certain day or certain
4G flow can be generated in the section time, but uses 2G flow always in its permanent residence, determines user's portrait label of such user
Are as follows: " 4G does not cover user " label；4th class user characteristics are usually expressed as fixed number section, are merely creating call and message note
Record etc. determines user's portrait label of such user are as follows: " Internet of Things " user tag.
Based on the mode classification for being directed to aforementioned label, sample can be utilized before executing S106 or before executing this programme
This user collection is trained initial clustering device, obtains target cluster device.
It can also include such as before executing S106 step as shown in Figure 3 or Figure 4 in a preferred realization scene
Lower step:
S105 is trained initial clustering device using sample of users collection, is obtained the target using fuzzy clustering algorithm
Cluster device.
Wherein, fuzzy clustering algorithm involved in the embodiment of the present invention can include but is not limited to: fuzzy clustering C algorithm
(fuzzy C-means, FCM).
In addition, the sample of users collection for training cluster device, which can use in database, to be completed with described in S103 step
The user data for turning net (first network is converted to the second network) is constituted.It, can also be such as Fig. 4 institute in a preferred realization scene
Show, and for training sample of users collection used by classifier identical, to be converted by first network to the in the preceding aim period
The subscriber data set of two networks.
In the embodiment of the present invention, mode shown in Fig. 4 be a kind of possible implementation, the embodiment of the present invention for S103 with
The successive execution sequence of S105 is not particularly limited, and in addition to the mode shown in Fig. 4, can also be first carried out S105 and be executed S103 again, alternatively, together
Shi Zhihang.And S103 and S105 can also be executed before S102 step.
Technical solution provided by the embodiment of the present invention at least has following technical effect:
User's portrait method provided by the invention, it is first determined the candidate user for using first network always, then, to this
A little candidate users carry out prediction classification, it is determined that have the target user that greater probability turn net in candidate user, it is in turn, right
Target user carries out cluster verifying, obtains user's portrait of target user, in this way, when specifically turn the recommendation of network service,
Business personnel can draw a portrait according to user, carry out turning network service recommendation as far as possible for target user, this can be mentioned to a certain extent
Height turns the success rate of network service, and flexibility is higher, is adaptable to a variety of turn net scene；And it is different from the existing indistinguishably side of pushing away
Formula can effectively save the workload of business personnel, reduce cost of human resources.
Embodiment two
The portrait of user provided by one method based on the above embodiment, the embodiment of the present invention, which further provides, realizes above-mentioned side
The Installation practice of each step and method in method embodiment.
The embodiment of the invention provides a kind of user portrait device, the device 500 referring to FIG. 5, user draws a portrait, comprising:
Determining module 51 determines candidate user collection for the user data according to objective time interval；Wherein, the candidate use
The data acquisition system that family integrates as candidate user, the candidate user are that the use of first network is persistently used in the objective time interval
Family；
First processing module 52 obtains target user's collection for handling the candidate user collection using object classifiers；Its
In, the object classifiers are for predicting that the candidate user concentration is reached by the probability that the first network is converted to the second network
To the target user of destination probability；
Second processing module 53 is obtained for handling the target user that the target user concentrates using target cluster device
The user of the target user draws a portrait；Wherein, the target cluster device is used to add user's portrait label for the target user,
Obtain user's portrait of the target.
User's portrait label involved in the embodiment of the present invention can include but is not limited to following at least one: old
Year user, vice card user, the second not yet covered with networks user and Internet of Things network users.
Specifically, first processing module 52, is specifically used for: fuzzy clustering algorithm is used, using sample of users collection to initial
Cluster device is trained, and obtains the target cluster device.
Specifically, Second processing module 53, is specifically used for: using XGBoost algorithm or random forests algorithm, utilize sample
This user collection is trained preliminary classification device, until the predictablity rate of the classifier after training is greater than preset accuracy rate threshold
Value, obtains the object classifiers.
In addition, determining module 51 is also used in a preferred realization scene:
According to the user data of the objective time interval, the sample of users collection is determined；
Wherein, the data acquisition system that the sample of users integrates as sample of users, the sample of users are in the objective time interval
The user of second network is converted to by the first network.
In the embodiment of the present invention, the first network is at least one of 2G network, 3G network and 4G network network, institute
State the second network are as follows: at least one of 3G network, 4G network and 5G network network；And the first network and second net
Same network is not present in network.
It should be understood that user shown in figure 5 above draws a portrait, the division of the modules of device is only a kind of drawing for logic function
Point, it can completely or partially be integrated on a physical entity in actual implementation, it can also be physically separate.And these modules can
All to be realized by way of processing element calls with software；It can also all realize in the form of hardware；It can also part
Module realizes that part of module passes through formal implementation of hardware with software by way of processing element calls.For example, determining module
It can be the processing element individually set up, also can integrate in user's portrait device, such as in some chip of server
It realizes, in addition it is also possible to be stored in the form of program in the memory of user's portrait device, by a certain of user's portrait device
A processing element calls and executes the function of the above modules.The realization of other modules is similar therewith.Furthermore these modules are complete
Portion or part can integrate together, can also independently realize.Processing element described here can be a kind of integrated circuit, tool
There is the processing capacity of signal.During realization, each step of the above method or the above modules can pass through processor member
The integrated logic circuit of hardware in part or the instruction of software form are completed.
For example, the above module can be arranged to implement one or more integrated circuits of above method, such as:
One or more specific integrated circuits (Application Specific Integrated Circuit, ASIC), or, one
Or multi-microprocessor (digital singnal processor, DSP), or, one or more field programmable gate array
(Field Programmable Gate Array, FPGA) etc..For another example, when some above module dispatches journey by processing element
When the form of sequence is realized, which can be general processor, such as central processing unit (Central Processing
Unit, CPU) or it is other can be with the processor of caller.For another example, these modules can integrate together, with system on chip
The form of (system-on-a-chip, SOC) is realized.
In addition, the embodiment of the invention provides a kind of user portrait device, the device 600 referring to FIG. 6, user draws a portrait,
Including: includes: transceiver 610, processor 620, memory 630 and bus, transceiver 610, processor 620 and memory
630 connect with bus respectively, and memory 630 is stored with program instruction, and processor 620 runs program instruction to execute such as lower section
Method:
According to the user data of objective time interval, candidate user collection is determined；Wherein, the data that candidate user integrates as candidate user
Set, candidate user are that the user of first network is persistently used in objective time interval；
Candidate user collection is handled using object classifiers, obtains target user's collection；Wherein, object classifiers are used for pre- astronomical observation
Family is selected to concentrate the target user for reaching destination probability by the probability that first network is converted to the second network；
The target user concentrated using target cluster device processing target user obtains user's portrait of target user；Wherein,
Target clusters device and is used to add user's portrait label for target user, obtains user's portrait of target.
In the embodiment of the present invention, user's portrait label can include but is not limited to following at least one: old user, vice card
User, the second not yet covered with networks user and Internet of Things network users.
During a concrete implementation, processor 620 runs program instruction to execute following method:
Using fuzzy clustering algorithm, initial clustering device is trained using sample of users collection, obtains target cluster device.
During another concrete implementation, processor 620 runs program instruction to execute following method:
Using XGBoost algorithm or random forests algorithm, preliminary classification device is trained using sample of users collection, directly
The predictablity rate of classifier after to training is greater than preset accuracy rate threshold value, obtains object classifiers.
Preferably, processor 620 runs program instruction to execute following method:
According to the user data of objective time interval, sample of users collection is determined；
Wherein, the data acquisition system that sample of users integrates as sample of users, sample of users are to be turned in objective time interval by first network
It is changed to the user of the second network.
In the embodiment of the present invention, first network is at least one of 2G network, 3G network and 4G network network, the second net
Network are as follows: at least one of 3G network, 4G network and 5G network network；And same net is not present in first network and the second network
Network.
It may include one or more processors 620, the place in user portrait device 600 in the embodiment of the present invention
Reason device 620 is referred to as processing unit or processing element, and certain control function may be implemented.The processor 620 can be with
It is general processor or application specific processor etc..
Here processing element can be general processor, such as central processing unit (Central Processing
Unit, CPU), it can also be the one or more integrated circuits for being configured to implement above method, such as: it is one or more special
Integrated circuit (Application Specific Integrated Circuit, ASIC) is determined, or, one or more micro processs
Device (digital singnal processor, DSP), or, one or more field programmable gate array (Field
Programmable Gate Array, FPGA) etc..Memory element can be a memory, be also possible to multiple memory elements
General designation.
Optionally, it may include one or more memories 630 in user's portrait device 600, have calculating thereon
Machine instruction or intermediate data, the computer instruction can be run on the processor, the device so that user draws a portrait
600 execute method described in above method embodiment.Optionally, other dependency numbers can also be stored in the memory
According to.It optionally also can store instruction and/or data in processor.The processor and memory can be separately provided, can also
To integrate.
Optionally, the communication device 600 can also include transceiver 610.It is single that the transceiver 610 is properly termed as transmitting-receiving
Member, transceiver, transmission circuit or transceiver etc., for realizing the transmission-receiving function of communication device.It is mentioned in the embodiment of the present invention
In the preceding method of confession, transceiver 610 can be used for exporting user's portrait result.
Processor and transceiver described in this application may be implemented in integrated circuit (integrated circuit, IC),
Analog IC, RF IC RFIC, mixed-signal IC, specific integrated circuit (application specific
Integrated circuit, ASIC), printed circuit board (printed circuit board, PCB), on electronic equipment etc..
The processor and transceiver can also be manufactured with various 1C technologies, such as complementary metal oxide semiconductor
(complementary metal oxide semiconductor, CMOS), N-type metal-oxide semiconductor (MOS) (n Metal-
Oxide-semiconductor, NMOS), P type metal oxide semiconductor (positive channel metal oxide
Semiconductor, PMOS), bipolar junction transistor (Bipolar Junction Transistor, BJT), bipolar CMOS
(BiCMOS), SiGe (SiGe), GaAs (GaAs) etc..
Optionally, user's portrait device 600 can be independent equipment or can be a part of larger equipment.Such as
The equipment may is that
(1) independent Integrated circuit IC or chip, or, chip system or subsystem；
(2) with one or more IC set, optionally, the IC gather also may include for storing data and/or
The storage unit of instruction；
(3) ASIC, such as modem (MSM)；
(4) the embeddable module in other equipment；
(5) receiver, terminal, cellular phone, wireless device, hand-held set, mobile unit, network equipment etc.；
(6) other etc..
In addition, it is stored thereon with computer program the embodiment of the invention provides a kind of readable storage medium storing program for executing,
The computer program is executed by processor to realize the method as described in embodiment one.
In addition, the embodiment of the present application also provides a kind of computer program product, which includes computer
Program, when run on a computer, so that computer executes method described in above-described embodiment.
Method shown in embodiment one is able to carry out as each module in this present embodiment, what the present embodiment was not described in detail
Part can refer to the related description to embodiment one.
Technical solution provided by the embodiment of the present invention at least has following technical effect:
User's portrait device and storage medium provided by the invention, it is first determined used using the candidate of first network always
Then family carries out prediction classification to these candidate users, it is determined that have the target that greater probability turn net in candidate user
User carries out cluster verifying to target user in turn, user's portrait of target user is obtained, in this way, carrying out turning net industry specifically
When the recommendation of business, business personnel can draw a portrait according to user, carry out turning network service recommendation as far as possible for target user, this can be one
Determine to improve in degree and turn the success rate of network service, flexibility is higher, is adaptable to a variety of turn net scene；And it is different from existing nothing
Mode is differentially pushed away, the workload of business personnel can be effectively saved, reduces cost of human resources.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.The computer program
Product includes one or more computer instructions.When loading on computers and executing the computer program instructions, all or
It partly generates according to process described herein or function.The computer can be general purpose computer, special purpose computer, meter
Calculation machine network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or
It is transmitted from a computer readable storage medium to another computer readable storage medium, for example, the computer instruction can
To pass through wired (such as coaxial cable, optical fiber, digital subscriber from a web-site, computer, server or data center
Line) or wireless (such as infrared, wireless, microwave etc.) mode to another web-site, computer, server or data center into
Row transmission.The computer readable storage medium can be any usable medium or include one that computer can access
Or the data storage devices such as integrated server, data center of multiple usable mediums.The usable medium can be magnetic medium,
(for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk Solid State
Disk) etc..
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to its of the disclosure
Its embodiment.The present invention is directed to cover any variations, uses, or adaptations of the disclosure, these modifications, purposes or
Person's adaptive change follows the general principles of this disclosure and including the undocumented common knowledge in the art of the disclosure
Or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the disclosure are by following
Claims are pointed out.
It should be understood that the present disclosure is not limited to the precise structures that have been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.The scope of the present disclosure is only limited by appended claims
System.
Claims (13)
- A kind of method 1. user draws a portrait characterized by comprisingAccording to the user data of objective time interval, candidate user collection is determined；Wherein, the data that the candidate user integrates as candidate user Set, the candidate user are that the user of first network is persistently used in the objective time interval；The candidate user collection is handled using object classifiers, obtains target user's collection；Wherein, the object classifiers are for pre- It surveys the candidate user and concentrates the target user for reaching destination probability by the probability that the first network is converted to the second network；The target user that the target user concentrates is handled using target cluster device, obtains user's portrait of the target user； Wherein, the target cluster device is used to add user's portrait label for the target user, obtains user's portrait of the target.
- 2. label comprises at least one of the following the method according to claim 1, wherein the user draws a portrait: old Year user, vice card user, the second not yet covered with networks user and Internet of Things network users.
- 3. the method according to claim 1, wherein the method also includes:Using fuzzy clustering algorithm, initial clustering device is trained using sample of users collection, obtains the target cluster device.
- 4. the method according to claim 1, wherein the method also includes:Using XGBoost algorithm or random forests algorithm, preliminary classification device is trained using sample of users collection, until instruction The predictablity rate of classifier after white silk is greater than preset accuracy rate threshold value, obtains the object classifiers.
- 5. the method according to claim 3 or 4, which is characterized in that the method also includes:According to the user data of the objective time interval, the sample of users collection is determined；Wherein, the data acquisition system that the sample of users integrates as sample of users, the sample of users in the objective time interval by institute State the user that first network is converted to second network.
- 6. the method according to claim 1, whereinThe first network is at least one of 2G network, 3G network and 4G network network, second network are as follows: 3G net At least one of network, 4G network and 5G network network；And same net is not present in the first network and second network Network.
- The device 7. a kind of user draws a portrait characterized by comprising transceiver, processor, memory and bus, the transmitting-receiving Device, the processor and the memory are connect with the bus respectively, and the memory is stored with program instruction, the place Device operation described program instruction is managed to execute following method:According to the user data of objective time interval, candidate user collection is determined；Wherein, the data that the candidate user integrates as candidate user Set, the candidate user are that the user of first network is persistently used in the objective time interval；The candidate user collection is handled using object classifiers, obtains target user's collection；Wherein, the object classifiers are for pre- It surveys the candidate user and concentrates the target user for reaching destination probability by the probability that the first network is converted to the second network；The target user that the target user concentrates is handled using target cluster device, obtains user's portrait of the target user； Wherein, the target cluster device is used to add user's portrait label for the target user, obtains user's portrait of the target.
- 8. device according to claim 7, which is characterized in that user's portrait label comprises at least one of the following: old Year user, vice card user, the second not yet covered with networks user and Internet of Things network users.
- 9. device according to claim 7, which is characterized in that the processor operation described program instruction is as follows to execute Method:Using fuzzy clustering algorithm, initial clustering device is trained using sample of users collection, obtains the target cluster device.
- 10. device according to claim 7, which is characterized in that the processor operation described program instruction is to execute such as Lower method:Using XGBoost algorithm or random forests algorithm, preliminary classification device is trained using sample of users collection, until instruction The predictablity rate of classifier after white silk is greater than preset accuracy rate threshold value, obtains the object classifiers.
- 11. device according to claim 9 or 10, which is characterized in that the processor operation described program instruction is to hold The following method of row:According to the user data of the objective time interval, the sample of users collection is determined；Wherein, the data acquisition system that the sample of users integrates as sample of users, the sample of users in the objective time interval by institute State the user that first network is converted to second network.
- 12. device according to claim 7, which is characterized in that the first network is 2G network, 3G network and 4G network At least one of network, second network are as follows: at least one of 3G network, 4G network and 5G network network；And it is described Same network is not present in first network and second network.
- 13. a kind of computer readable storage medium, which is characterized in that it is stored thereon with computer program,The computer program is executed by processor to realize such as method as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811353247.6A CN109543734A (en) | 2018-11-14 | 2018-11-14 | User portrait method and device, storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811353247.6A CN109543734A (en) | 2018-11-14 | 2018-11-14 | User portrait method and device, storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109543734A true CN109543734A (en) | 2019-03-29 |
Family
ID=65847255
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811353247.6A Pending CN109543734A (en) | 2018-11-14 | 2018-11-14 | User portrait method and device, storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109543734A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111191122A (en) * | 2019-12-20 | 2020-05-22 | 重庆邮电大学 | Learning resource recommendation system based on user portrait |
| CN111275245A (en) * | 2020-01-13 | 2020-06-12 | 宜通世纪物联网研究院(广州)有限公司 | Potential network switching user identification method, system, message pushing method, device and medium |
| CN111538751A (en) * | 2020-03-23 | 2020-08-14 | 重庆特斯联智慧科技股份有限公司 | Tagged user portrait generation system and method for Internet of things data |
| CN113496288A (en) * | 2020-04-08 | 2021-10-12 | 中移动信息技术有限公司 | User stability determination method, device, equipment and storage medium |
| CN113543117A (en) * | 2020-04-22 | 2021-10-22 | 中国移动通信集团重庆有限公司 | Prediction method and device for number portability user and computing equipment |
| CN114900553A (en) * | 2022-04-28 | 2022-08-12 | 中国联合网络通信集团有限公司 | Information pushing method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389713A (en) * | 2015-10-15 | 2016-03-09 | 南京大学 | Mobile data traffic package recommendation algorithm based on user historical data |
| CN107194727A (en) * | 2017-05-22 | 2017-09-22 | 贵州大学 | The user's request factor based on Kano model is characterized and extracting method |
| CN107609708A (en) * | 2017-09-25 | 2018-01-19 | 广州赫炎大数据科技有限公司 | A kind of customer loss Forecasting Methodology and system based on mobile phone games shop |
| CN107944913A (en) * | 2017-11-21 | 2018-04-20 | 重庆邮电大学 | High potential user's purchase intention Forecasting Methodology based on big data user behavior analysis |
-
2018
- 2018-11-14 CN CN201811353247.6A patent/CN109543734A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389713A (en) * | 2015-10-15 | 2016-03-09 | 南京大学 | Mobile data traffic package recommendation algorithm based on user historical data |
| CN107194727A (en) * | 2017-05-22 | 2017-09-22 | 贵州大学 | The user's request factor based on Kano model is characterized and extracting method |
| CN107609708A (en) * | 2017-09-25 | 2018-01-19 | 广州赫炎大数据科技有限公司 | A kind of customer loss Forecasting Methodology and system based on mobile phone games shop |
| CN107944913A (en) * | 2017-11-21 | 2018-04-20 | 重庆邮电大学 | High potential user's purchase intention Forecasting Methodology based on big data user behavior analysis |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111191122A (en) * | 2019-12-20 | 2020-05-22 | 重庆邮电大学 | Learning resource recommendation system based on user portrait |
| CN111275245A (en) * | 2020-01-13 | 2020-06-12 | 宜通世纪物联网研究院(广州)有限公司 | Potential network switching user identification method, system, message pushing method, device and medium |
| CN111538751A (en) * | 2020-03-23 | 2020-08-14 | 重庆特斯联智慧科技股份有限公司 | Tagged user portrait generation system and method for Internet of things data |
| CN113496288A (en) * | 2020-04-08 | 2021-10-12 | 中移动信息技术有限公司 | User stability determination method, device, equipment and storage medium |
| CN113496288B (en) * | 2020-04-08 | 2024-04-12 | 中移动信息技术有限公司 | User stability determining method, device, equipment and storage medium |
| CN113543117A (en) * | 2020-04-22 | 2021-10-22 | 中国移动通信集团重庆有限公司 | Prediction method and device for number portability user and computing equipment |
| CN113543117B (en) * | 2020-04-22 | 2022-10-04 | 中国移动通信集团重庆有限公司 | Prediction method and device for number portability user and computing equipment |
| CN114900553A (en) * | 2022-04-28 | 2022-08-12 | 中国联合网络通信集团有限公司 | Information pushing method, device, equipment and storage medium |
| CN114900553B (en) * | 2022-04-28 | 2023-04-25 | 中国联合网络通信集团有限公司 | Information pushing method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109543734A (en) | User portrait method and device, storage medium | |
| Do et al. | Where and what: Using smartphones to predict next locations and applications in daily life | |
| CN111190939A (en) | User portrait construction method and device | |
| CN104348855B (en) | Processing method, mobile terminal and the server of user information | |
| CN105007171A (en) | User data analysis system and method based on big data in communication field | |
| CN110049079A (en) | Information push and model training method, device, equipment and storage medium | |
| CN106791235B (en) | A kind of method, apparatus and system selecting service agent | |
| CN110782318A (en) | Marketing method and device based on audio interaction and storage medium | |
| CN101163268A (en) | Service information providing apparatus and service information providing method | |
| CN110472154A (en) | A kind of resource supplying method, apparatus, electronic equipment and readable storage medium storing program for executing | |
| US20160210647A1 (en) | Method of determining segmentations of subscribers, network entity using the same, and server using the same | |
| CN112214677B (en) | Point of interest recommendation method and device, electronic equipment and storage medium | |
| CN110298405A (en) | Classification recognition methods and device, storage medium and terminal | |
| CN109982293A (en) | Flow product method for pushing, system, electronic equipment and storage medium | |
| CN109741108A (en) | Streaming application recommended method, device and electronic equipment based on context aware | |
| CN104598543B (en) | A kind of Social Match data digging system | |
| CN109784947A (en) | Worksheet processing method, equipment, storage medium and device are rescued after sale | |
| CN117611272A (en) | Commodity recommendation method and device and electronic equipment | |
| CN108924915A (en) | Data-updating method and device based on offline mode | |
| CN106161575A (en) | User matching method and device | |
| CN109886738B (en) | Intelligent exhibition user prediction method and equipment | |
| CN110852338A (en) | User portrait construction method and device | |
| CN113806634B (en) | Service package recommending method, device and server | |
| CN110781878B (en) | Target area determination method and device, storage medium and electronic device | |
| CN113592557A (en) | Attribution method and device of advertisement putting result, storage medium and electronic equipment |
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: 20190329 |
|
| RJ01 | Rejection of invention patent application after publication |