CN109492160A - Method and apparatus for pushed information - Google Patents
Method and apparatus for pushed information Download PDFInfo
- Publication number
- CN109492160A CN109492160A CN201811288022.7A CN201811288022A CN109492160A CN 109492160 A CN109492160 A CN 109492160A CN 201811288022 A CN201811288022 A CN 201811288022A CN 109492160 A CN109492160 A CN 109492160A
- Authority
- CN
- China
- Prior art keywords
- information
- comment
- comment information
- sample
- evaluation
- 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
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The embodiment of the present application discloses the method and apparatus for pushed information.One specific embodiment of this method includes: to obtain the comment information set for being directed to intended display information；For the comment information in comment information set, by comment information input comment identification model trained in advance, obtain the evaluation information for the comment information, wherein, comment identification model is used to characterize the corresponding relationship of comment information and evaluation information, and evaluation information is for characterizing user to the interest level of the corresponding intended display information of comment information；Based on obtained evaluation information, the comment information in comment information set is ranked up；Comment information set after sequence is pushed to the terminal of target user.The embodiment can be more acurrate and be ranked up for greater flexibility to each comment information, and the specific aim of push comment information is improved.
Description
Technical field
The invention relates to field of computer technology, and in particular to the method and apparatus for pushed information.
Background technique
With the development of internet technology, people's more and more continually various information of using terminal equipment browse.In user
While browsing information, information can be commented on, as reference when these comment informations can browse information for other users.
For example, comment information may include the scoring of user, the higher comment information that scores can be arranged in forward position, thus
User is set preferentially to read the higher comment of scoring.
Summary of the invention
The embodiment of the present application proposes the method and apparatus for pushed information.
In a first aspect, the embodiment of the present application provides a kind of method for pushed information, it is directed to this method comprises: obtaining
The comment information set of intended display information；For the comment information in comment information set, which is inputted preparatory
Trained comment identification model obtains the evaluation information for the comment information, wherein comment identification model is for characterizing comment
The corresponding relationship of information and evaluation information, evaluation information is for characterizing sense of the user to the corresponding intended display information of comment information
Level of interest；Based on obtained evaluation information, the comment information in comment information set is ranked up；By commenting after sequence
The terminal of target user is pushed to by information aggregate.
In some embodiments, the type of evaluation information includes at least two.
In some embodiments, it is based on obtained evaluation information, the comment information in comment information set is arranged
Sequence, comprising: the type based on obtained evaluation information classifies to the comment information in comment information set, obtain to
Few two comment information subclass；According to the sequence of the corresponding type of preset, evaluation information, at least two comment informations
Set is ranked up；For the subclass at least two subclass after sequence, it is based on predetermined sortord, to this
Comment information in subclass is ranked up.
In some embodiments, training obtains comment identification model as follows in advance: training sample set is obtained,
Wherein, training sample include for preset samples show information sample comment information and in advance for sample comment information into
The mark evaluation information of rower note；Using machine learning method, the sample that the training sample in training sample set includes is commented
By information as inputting, using the corresponding mark evaluation information of the sample comment information of input as desired output, training is commented
By identification model.
In some embodiments, the corresponding mark evaluation information of sample comment information obtains as follows in advance: really
The quantity of the user of sample comment information was browsed calmly as the first quantity；It determines and browsed sample comment information and for sample
Show that information carries out the quantity of the user of object run as the second quantity；Determine the ratio of the second quantity Yu the first quantity；Base
In the corresponding relationship of preset mark evaluation information and ratio, the corresponding mark evaluation information conduct of identified ratio is determined
The corresponding mark evaluation information of sample comment information.
Second aspect, the embodiment of the present application provide a kind of device for pushed information, which includes: to obtain list
Member is configured to obtain the comment information set for being directed to intended display information；Recognition unit is configured to for comment information collection
Comment information input comment identification model trained in advance is obtained commenting for the comment information by the comment information in conjunction
Valence information, wherein comment identification model is used to characterize the corresponding relationship of comment information and evaluation information, and evaluation information is for characterizing
Interest level of the user to the corresponding intended display information of comment information；Sequencing unit is configured to comment based on obtained
Valence information is ranked up the comment information in comment information set；Push unit, the comment information after being configured to sort
Set pushes to the terminal of target user.
In some embodiments, the type of evaluation information includes at least two.
In some embodiments, sequencing unit includes: categorization module, is configured to the kind based on obtained evaluation information
Class classifies to the comment information in comment information set, obtains at least two comment information subclass；First sequence mould
Block is configured to the sequence according to the corresponding type of preset, evaluation information, arranges at least two comment information subclass
Sequence；Second sorting module is configured to be based on predetermined row for the subclass at least two subclass after sequence
Sequential mode is ranked up the comment information in the subclass.
In some embodiments, training obtains comment identification model as follows in advance: training sample set is obtained,
Wherein, training sample include for preset samples show information sample comment information and in advance for sample comment information into
The mark evaluation information of rower note；Using machine learning method, the sample that the training sample in training sample set includes is commented
By information as inputting, using the corresponding mark evaluation information of the sample comment information of input as desired output, training is commented
By identification model.
In some embodiments, the corresponding mark evaluation information of sample comment information obtains as follows in advance: really
The quantity of the user of sample comment information was browsed calmly as the first quantity；It determines and browsed sample comment information and for sample
Show that information carries out the quantity of the user of object run as the second quantity；Determine the ratio of the second quantity Yu the first quantity；Base
In the corresponding relationship of preset mark evaluation information and ratio, the corresponding mark evaluation information conduct of identified ratio is determined
The corresponding mark evaluation information of sample comment information.
The third aspect, the embodiment of the present application provide a kind of server, which includes: one or more processors；
Storage device is stored thereon with one or more programs；When one or more programs are executed by one or more processors, so that
One or more processors realize the method as described in implementation any in first aspect.
Fourth aspect, the embodiment of the present application provide a kind of computer-readable medium, are stored thereon with computer program, should
The method as described in implementation any in first aspect is realized when computer program is executed by processor.
Method and apparatus provided by the embodiments of the present application for pushed information, by using comment identification model, to obtaining
The comment information that the comment information set taken includes is identified, the evaluation information of type belonging to characterization comment information is obtained,
It is based on evaluation information again, the comment information in comment information set is ranked up, finally by the comment information set after sequence
The terminal of target user is pushed to, it is more acurrate and for greater flexibility to each comment information so as to using commenting on identification model
It is ranked up, improves the specific aim of push comment information.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is that one embodiment of the application can be applied to exemplary system architecture figure therein；
Fig. 2 is the flow chart according to one embodiment of the method for pushed information of the embodiment of the present application；
Fig. 3 is the schematic diagram according to an application scenarios of the method for pushed information of the embodiment of the present application；
Fig. 4 is the flow chart according to another embodiment of the method for pushed information of the embodiment of the present application；
Fig. 5 is the structural schematic diagram according to one embodiment of the device for pushed information of the embodiment of the present application；
Fig. 6 is adapted for the structural schematic diagram for the computer system for realizing the server of the embodiment of the present application.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can the method for pushed information using the embodiment of the present application or the device for pushed information
Exemplary system architecture 100.
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out
Send message etc..Various telecommunication customer end applications can be installed, such as web browser is answered on terminal device 101,102,103
With, shopping class application, searching class application, instant messaging tools, mailbox client, social platform software etc..
Terminal device 101,102,103 can be hardware, be also possible to software.When terminal device 101,102,103 is hard
When part, it can be the various electronic equipments with display screen, including but not limited to smart phone, tablet computer, on knee portable
Computer and desktop computer etc..When terminal device 101,102,103 is software, above-mentioned cited electricity may be mounted at
In sub- equipment.Multiple softwares or software module may be implemented into (such as providing the software of Distributed Services or software mould in it
Block), single software or software module also may be implemented into.It is not specifically limited herein.
Server 105 can be to provide the server of various services, such as to showing on terminal device 101,102,103
Comment information provides the background information processing server supported.Background information processing server can be to the comment information collection of acquisition
Conjunction carries out the processing such as analyzing, and processing result (such as comment information set after sequence) is fed back to terminal device.
It should be noted that the method provided by the embodiment of the present application for pushed information is generally held by server 105
Row, correspondingly, the device for pushed information is generally positioned in server 105.
It should be noted that server can be hardware, it is also possible to software.When server is hardware, may be implemented
At the distributed server cluster that multiple servers form, individual server also may be implemented into.It, can when server is software
To be implemented as multiple softwares or software module (such as providing the software of Distributed Services or software module), also may be implemented
At single software or software module.It is not specifically limited herein.
It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
With continued reference to Fig. 2, the process of one embodiment of the method for pushed information according to the application is shown
200.This is used for the method for pushed information, comprising the following steps:
Step 201, the comment information set for being directed to intended display information is obtained.
In the present embodiment, can lead to for the executing subject of the method for pushed information (such as server shown in FIG. 1)
Wired connection mode or radio connection are crossed from comment information collection long-range or from local acquisition for intended display information
It closes.Wherein, intended display information can be preset displaying information aggregate (such as the server-side of certain application program is to user terminal
Push each displaying information) in displaying information, above-mentioned executing subject can in advance from show information aggregate in select show
Information is as intended display information.Comment information can be the information that user comments on intended display information, including but not
It is limited to following at least one: text, symbol, number etc..
Step 202, for the comment information in comment information set, comment information input comment trained in advance is known
Other model obtains the evaluation information for the comment information.
In the present embodiment, for the comment information in comment information set, above-mentioned executing subject can believe the comment
Breath input comment identification model trained in advance, obtains the evaluation information for the comment information.
Wherein, evaluation information is for characterizing user to the interest level of the corresponding intended display information of comment information.It comments
Valence information may include various forms of information, including but not limited to following at least one information: text, number, symbol etc..Make
For example, evaluation information can be the numerical value between 0 to 1, and numerical value is bigger, indicate user to the corresponding target of comment information
Show that information is interested.Evaluation information can also be the information of classification belonging to characterization comment information or grade, such as:
" good ", " in ", " poor " etc..
In the present embodiment, comment identification model is used to characterize the corresponding relationship of comment information and evaluation information.Specifically,
As an example, comment identification model may include contents extraction part and mapping table.Wherein, contents extraction part can be used
Object content is extracted in the comment information from input.For example, object content can be word, object content is contained in preset
In set of words, comment identification model can extract the word conduct for including in above-mentioned set of words from the comment information of input
The object content extracted.Mapping table can be technical staff and be based on uniting to a large amount of object content and evaluation information
Count and pre-establish, be stored with the mapping table of the corresponding relationship of multiple object contents and evaluation information.In this way, upper commentary
By identification model object content can be extracted using contents extraction part first.Later, by the object content and corresponding relationship
Multiple object contents in table are successively compared, if in some object content in mapping table and the target extracted
Hold same or similar (such as the similarity of the two is greater than preset similarity threshold), then it will be in the target in mapping table
Hold corresponding evaluation information as the obtained evaluation information of step 201.
In some optional implementations of the present embodiment, comment identification model can by above-mentioned executing subject or other
Training obtains electronic equipment as follows in advance:
Firstly, obtaining training sample set.Wherein, training sample includes commenting for the sample of preset samples show information
By information and the preparatory mark evaluation information being labeled for sample comment information.Mark evaluation information can be used for characterizing and make
Interest level of the user to samples show information of sample comment information is generated with terminal.Mark evaluation information can be number
Word, text, symbol or combinations thereof.For example, mark evaluation information can be text " good " or " in " or " poor ", user is to sample for characterization
The interest level of this displaying information is from high to low.For another example mark evaluation information can also be that numerical value, such as " 0 " indicate to use
Family loses interest in samples show information, and " 1 " indicates interested in samples show information.Comment identification model after then training can
To export the numerical value between 0-1, characterization user is to the displaying interested probability of information.
Then, using machine learning method, the sample comment information for including by the training sample in training sample set is made
For input, using the corresponding mark evaluation information of the sample comment information of input as desired output, training obtains comment identification mould
Type.
Specifically, above-mentioned executing subject or other electronic equipments can select training sample from training sample set, and
Execute following training step:
Step 1, the sample comment information for including by the training sample of selection input initial model, obtain sample comment letter
Cease corresponding evaluation information.In practice, above-mentioned initial model may include various models, for example, for determine the feature of text to
The word2vec model of amount, neural network (such as convolutional neural networks) model for classifying to feature vector etc..Make
For example, when above-mentioned neural network model is depth convolutional neural networks, since depth convolutional neural networks are a multilayers
Neural network, it is therefore desirable to predefine depth convolutional neural networks include which layer (for example, convolutional layer, pond layer, Quan Lian
Connect layer, classifier etc.), order of connection relationship and each layer between layers include which network parameter (for example,
Weight, bias term, the step-length of convolution) etc..
Step 2 compares the evaluation information that the mark evaluation information that the training sample of selection includes is exported with step 1
Compared with the difference of the evaluation information of determining mark evaluation information and step 1 output.Here it is possible to utilize preset loss function meter
It calculates obtained evaluation information and marks the difference between evaluation information.For example, loss function can be cross entropy loss function,
Logarithm loss function etc..Using the available penalty values of loss function, penalty values can be used as comparison result.
Step 3 determines whether above-mentioned initial model reaches preset condition up to standard according to comparison result.On as an example,
Stating condition up to standard can be penalty values less than or equal to preset threshold.
Step 4 reaches condition up to standard in response to the above-mentioned initial model of determination, identifies above-mentioned initial model as comment
Model.
Step 5 adjusts the parameter of initial model, Yi Jicong in response to determining that initial model is not up to above-mentioned condition up to standard
Chosen again in above-mentioned training sample set and select training sample, the initial model after using adjusting parameter as initial model, after
The continuous above-mentioned steps one that execute arrive step 4.As an example, back-propagation algorithm (Back Propgation can be used
Algorithm, BP algorithm) or gradient descent method (such as stochastic gradient descent algorithm) to the network parameter of above-mentioned initial model into
Row adjustment.It should be noted that back-propagation algorithm and gradient descent method are the well-known techniques studied and applied extensively at present,
This is repeated no more.
It should be noted that above-mentioned mark evaluation information can be what technical staff was manually arranged, it is also possible to by training
The executing subject for commenting on identification model is predetermined.
In some optional implementations of the present embodiment, the corresponding mark evaluation information of sample comment information can be by
Above-mentioned executing subject or other electronic equipments obtain as follows in advance:
Firstly, determining the quantity for browsing the user of sample comment information as the first quantity.Specifically, above-mentioned execution master
Body or other electronic equipments can determine that the quantity for showing the terminal device of sample comment information was commented on as sample is browsed
The quantity of the user of information.
Then, it is determined that browsing sample comment information and carrying out the quantity of the user of object run for samples show information
As the second quantity.Specifically, above-mentioned executing subject or other electronic equipments can detecte the use for browsing sample comment information
Whether family executed object run.Wherein, object run can be various operation operations relevant to sample comment information, example
Such as, when samples show information is products propaganda category information, object run can be directed to the production of samples show information instruction with user
The various operations that product (such as the virtual class product such as entity class product or software, file) carry out, including but not limited to below at least
It is a kind of: purchase, downloading etc..
Subsequently, the ratio of the second quantity Yu the first quantity is determined.
Finally, the corresponding relationship based on preset mark evaluation information and ratio, determines the corresponding mark of identified ratio
Evaluation information is infused as the corresponding mark evaluation information of sample comment information.As an example, pair of mark evaluation information and ratio
Should be related to can be characterized by preset mapping table.Multiple ratios and each ratio are can store in the mapping table
Corresponding mark evaluation information.Above-mentioned executing subject or other electronic equipments compare determined by can searching from mapping table
It is worth corresponding mark evaluation information as the corresponding mark evaluation information of sample comment information.As another example, training comment
It can be previously provided with ratio section in the executing subject of identification model and mark the corresponding relationship of comment information, in turn, training
The executing subject of comment identification model can determine ratio section locating for determined ratio, then determine the ratio section pair
The mark evaluation information answered is as the corresponding mark evaluation information of sample comment information.Wherein, ratio section and mark evaluation letter
The corresponding relationship of breath can first pass through the foundation of the forms such as table, pointer in advance.
By above-mentioned optional implementation, mark evaluation information can be objectively obtained depending on the user's operation, thus
It is more accurate to be conducive to the comment identification model for obtaining training.
Step 203, it is based on obtained evaluation information, the comment information in comment information set is ranked up.
In the present embodiment, above-mentioned executing subject can be based on obtained evaluation information, in comment information set
Comment information is ranked up.As an example, evaluation information can be characterization user to the corresponding intended display information of comment information
Interest level numerical value.For example, evaluation information can be the numerical value between 0 to 1, which can characterize user to target
Show the interested probability of information.Above-mentioned executing subject can be according to the size of probability, to the comment in comment information set
Information is ranked up.
Step 204, the comment information set after sequence is pushed to the terminal of target user.
In the present embodiment, the comment information set after sequence can be pushed to the end of target user by above-mentioned executing subject
End, to be shown in the terminal of target user.Wherein, after target user can be the terminal that it is used browsing sequence to be utilized
The user of comment information.For example, target user, which can be, opens target application in the terminal that it is used to browse above-mentioned target
Show the user of information.
With continued reference to the signal that Fig. 3, Fig. 3 are according to the application scenarios of the method for pushed information of the present embodiment
Figure.In the application scenarios of Fig. 3, server 301 obtains the comment information set 302 that user is directed to intended display information first.
Wherein, comment information is text information.Then, server 301 inputs each comment information in comment information set 302 pre-
First trained comment identification model 303, obtains the corresponding evaluation information of each comment information, wherein evaluation information is between 0-1
Numerical value (such as 0.4,0.6,0.9 ...), the interest level of the intended display information for characterizing user couple.Subsequently,
Server 301 is based on obtained evaluation information, is ranked up, is sorted to the comment information in comment information set 302
Comment information set 304 afterwards.For example, being ranked up according to the descending sequence of the numerical value of evaluation information to comment information.
Finally, the comment information set 304 after sequence to be pushed to the terminal 305 of target user.According to above-mentioned sequence side in terminal 305
Formula shows the comment information in comment information set 304.
The method provided by the above embodiment of the application, by using comment identification model, to the comment information collection of acquisition
The comment information that conjunction includes is identified, obtains the evaluation information of type belonging to characterization comment information, then be based on evaluation information,
Comment information in comment information set is ranked up, the comment information set after sequence is finally pushed to target user's
Terminal, it is more acurrate and each comment information is ranked up for greater flexibility so as to using commenting on identification model, it improves and pushes away
Send the specific aim of comment information.
With further reference to Fig. 4, it illustrates the processes 400 of another embodiment of the method for pushed information.The use
In the process 400 of the method for pushed information, comprising the following steps:
Step 401, the comment information set for being directed to intended display information is obtained.
In the present embodiment, step 401 and the step 201 in Fig. 2 corresponding embodiment are almost the same, and which is not described herein again.
Step 402, for the comment information in comment information set, comment information input comment trained in advance is known
Other model obtains the evaluation information for the comment information.
In the present embodiment, comment identification model is used to characterize the corresponding relationship of comment information and evaluation information.It is acquired
The type of evaluation information can be at least two.For example, evaluation information may include " good ", " in ", " poor ", i.e. evaluation information
Type may include favorable comment class, in comment class, difference to comment class.Intended display is believed for another example evaluation information can be characterization user
The interest level of breath, numerical value between 0-1, then each type of evaluation information can correspond to a numerical value area
Between, such as [0,0.3] corresponding difference comments class, comments class during (0.3,0.7) is corresponding, [0.7,1] corresponding favorable comment class.
Step 403, the type based on obtained evaluation information, divides the comment information in comment information set
Class obtains at least two comment information subclass.
It in the present embodiment, can be with base for the executing subject of the method for pushed information (such as server shown in FIG. 1)
In the type of obtained evaluation information, classify to the comment information in comment information set, obtains at least two comments
Information subset is closed.Specifically, above-mentioned executing subject can be true by the corresponding comment information of evaluation information for belonging to the same type
It is set to a classification, that is, the comment information for belonging to the same classification is a comment information subclass.
Step 404, according to the sequence of the corresponding type of preset, evaluation information, at least two comment information subclass
It is ranked up.
In the present embodiment, above-mentioned executing subject can be according to the sequence of the corresponding type of preset, evaluation information, to extremely
Few two comment information subclass are ranked up.Specifically, the type of evaluation information can be used for characterizing user to intended display
The interest level of information.For example, the type of evaluation information include favorable comment class, in comment class, difference to comment class.The type of evaluation information
Sequence can for favorable comment class, in comment class, difference to comment class, above-mentioned executing subject can be according to this sequentially to each comment information subclass
It is ranked up.
Step 405, for the subclass at least two subclass after sequence, it is based on predetermined sortord,
Comment information in the subclass is ranked up.
In the present embodiment, for the subclass at least two subclass after sequence, it is based on predetermined sequence
Mode, above-mentioned executing subject can be ranked up the comment information in the subclass.Wherein, predetermined sortord can
To be in a manner of technical staff is pre-set, or the sortord selected by target user.Wherein, target user can be
The user of comment information after the terminal that it is used browsing sequence to be utilized.Sortord can include but is not limited to it is following at least
A kind of: the generations time-sequencing based on comment information, the numerical value based on the popular degree for characterizing comment information (such as thumb up number
Amount, reply quantity etc.) sequence etc..It, can be for greater flexibility to each comment information by using predetermined sortord
The comment information that set includes is ranked up, and enriches the mode of sequence.
By being ranked up to above-mentioned at least two comment informations subclass, then to the comment that comment information subclass includes
Information is ranked up, and can refine the sortord of comment information, helps more targetedly to show comment information to user.
Step 406, the comment information set after sequence is pushed to the terminal of target user.
In the present embodiment, step 406 and the step 204 in Fig. 2 corresponding embodiment are almost the same, and which is not described herein again.
Figure 4, it is seen that the method for pushed information compared with the corresponding embodiment of Fig. 2, in the present embodiment
Process 400 highlight the step of comment information set is classified and sorted.The scheme of the present embodiment description can be with as a result,
It further refines and neatly each comment information is ranked up, to further improve the specific aim of information push.
With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, this application provides one kind for pushing letter
One embodiment of the device of breath, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, which can specifically answer
For in various electronic equipments.
As shown in figure 5, the device 500 for pushed information of the present embodiment includes: acquiring unit 501, it is configured to obtain
Take the comment information set for intended display information；Recognition unit 502 is configured to for the comment in comment information set
Comment information input comment identification model trained in advance is obtained the evaluation information for the comment information by information,
In, comment identification model is used to characterize the corresponding relationship of comment information and evaluation information, and evaluation information is for characterizing user to commenting
By the interest level of the corresponding intended display information of information；Sequencing unit 503 is configured to believe based on obtained evaluation
Breath, is ranked up the comment information in comment information set；Push unit 504, the comment information after being configured to sort
Set pushes to the terminal of target user.
In the present embodiment, acquiring unit 501 can by wired connection mode or radio connection from long-range or
From the local comment information set obtained for intended display information.Wherein, intended display information can be preset displaying letter
Displaying information in breath set (such as the server-side of certain application program to each displaying information that user terminal pushes), it is above-mentioned to obtain
Take unit 501 that can select to show information as intended display information from displaying information aggregate in advance.Comment information can be
The information that user comments on intended display information, it is including but not limited to following at least one: text, symbol, number etc..
In the present embodiment, for the comment information in comment information set, above-mentioned recognition unit 502 can be by the comment
The comment identification model that information input is trained in advance obtains the evaluation information for the comment information.
Wherein, evaluation information is for characterizing user to the interest level of the corresponding intended display information of comment information.It comments
Valence information may include various forms of information, including but not limited to following at least one information: text, number, symbol etc..Make
For example, evaluation information can be the numerical value between 0 to 1, and numerical value is bigger, indicate user to the corresponding target of comment information
Show that information is interested.Evaluation information can also be the information of classification belonging to characterization comment information or grade, such as:
" good ", " in ", " poor " etc..
In the present embodiment, comment identification model is used to characterize the corresponding relationship of comment information and evaluation information.Specifically,
As an example, comment identification model may include contents extraction part and mapping table.Wherein, contents extraction part can be used
Object content is extracted in the comment information from input.For example, object content can be word, object content is contained in preset
In set of words, comment identification model can extract the word conduct for including in above-mentioned set of words from the comment information of input
The object content extracted.Mapping table can be technical staff and be based on uniting to a large amount of object content and evaluation information
Count and pre-establish, be stored with the mapping table of the corresponding relationship of multiple object contents and evaluation information.In this way, upper commentary
By identification model object content can be extracted using contents extraction part first.Later, by the object content and corresponding relationship
Multiple object contents in table are successively compared, if in some object content in mapping table and the target extracted
Hold same or similar (such as the similarity of the two is greater than preset similarity threshold), then it will be in the target in mapping table
Hold corresponding evaluation information as the obtained evaluation information of recognition unit 502.
In the present embodiment, sequencing unit 503 can be based on obtained evaluation information, to commenting in comment information set
It is ranked up by information.As an example, evaluation information can be characterization user to the corresponding intended display information of comment information
The numerical value of interest level.For example, evaluation information can be the numerical value between 0 to 1, which can characterize user to target exhibition
Show the interested probability of information.Above-mentioned sequencing unit 503 can be according to the size of probability, to the comment in comment information set
Information is ranked up.
In the present embodiment, the comment information set after sequence can be pushed to the end of target user by push unit 504
End, to be shown in the terminal of target user.Wherein, after target user can be the terminal that it is used browsing sequence to be utilized
The user of comment information.For example, target user, which can be, opens target application in the terminal that it is used to browse above-mentioned target
Show the user of information.
In some optional implementations of the present embodiment, the type of evaluation information may include at least two.
In some optional implementations of the present embodiment, sequencing unit 503 may include: categorization module (in figure not
Show), it is configured to the type based on obtained evaluation information, is classified to the comment information in comment information set,
Obtain at least two comment information subclass；First sorting module (not shown) is configured to according to preset, evaluation letter
The sequence for ceasing corresponding type is ranked up at least two comment information subclass；Second sorting module (not shown),
It is configured to predetermined sortord is based on, to the subset for the subclass at least two subclass after sequence
Comment information in conjunction is ranked up.
In some optional implementations of the present embodiment, comment identification model is trained as follows in advance
It arrives: obtaining training sample set, wherein training sample includes for the sample comment information of preset samples show information and pre-
The mark evaluation information being first labeled for sample comment information；It, will be in training sample set using machine learning method
Sample comment information that training sample includes as input, using the corresponding mark evaluation information of the sample comment information of input as
Desired output, training obtain comment identification model.
In some optional implementations of the present embodiment, the corresponding mark evaluation information of sample comment information leads in advance
It crosses following steps to obtain: determining the quantity for browsing the user of sample comment information as the first quantity；Determination browsed sample
Comment information and for samples show information carry out object run user quantity as the second quantity；Determine the second quantity with
The ratio of first quantity；Based on the corresponding relationship of preset mark evaluation information and ratio, determine that identified ratio is corresponding
Evaluation information is marked as the corresponding mark evaluation information of sample comment information.
The device provided by the above embodiment of the application, by using comment identification model, to the comment information collection of acquisition
The comment information that conjunction includes is identified, obtains the evaluation information of type belonging to characterization comment information, then be based on evaluation information,
Comment information in comment information set is ranked up, the comment information set after sequence is finally pushed to target user's
Terminal, it is more acurrate and each comment information is ranked up for greater flexibility so as to using commenting on identification model, it improves and pushes away
Send the specific aim of comment information.
Below with reference to Fig. 6, it illustrates the computer systems 600 for the server for being suitable for being used to realize the embodiment of the present application
Structural schematic diagram.Server shown in Fig. 6 is only an example, should not function and use scope band to the embodiment of the present application
Carry out any restrictions.
As shown in fig. 6, computer system 600 includes central processing unit (CPU) 601, it can be read-only according to being stored in
Program in memory (ROM) 602 or be loaded into the program in random access storage device (RAM) 603 from storage section 608 and
Execute various movements appropriate and processing.In RAM 603, also it is stored with system 600 and operates required various programs and data.
CPU 601, ROM 602 and RAM 603 are connected with each other by bus 604.Input/output (I/O) interface 605 is also connected to always
Line 604.
I/O interface 605 is connected to lower component: the importation 606 including keyboard, mouse etc.；Including such as liquid crystal
Show the output par, c 607 of device (LCD) etc. and loudspeaker etc.；Storage section 608 including hard disk etc.；And including such as LAN
The communications portion 609 of the network interface card of card, modem etc..Communications portion 609 is executed via the network of such as internet
Communication process.Driver 610 is also connected to I/O interface 605 as needed.Detachable media 611, such as disk, CD, magneto-optic
Disk, semiconductor memory etc. are mounted on as needed on driver 610, in order to from the computer program root read thereon
According to needing to be mounted into storage section 608.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communications portion 609, and/or from detachable media
611 are mounted.When the computer program is executed by central processing unit (CPU) 601, limited in execution the present processes
Above-mentioned function.
It should be noted that computer-readable medium described herein can be computer-readable signal media or meter
Calculation machine readable medium either the two any combination.Computer-readable medium for example may be-but not limited to-
Electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.It is computer-readable
The more specific example of medium can include but is not limited to: have electrical connection, the portable computer magnetic of one or more conducting wires
Disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM or sudden strain of a muscle
Deposit), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or above-mentioned appoint
The suitable combination of meaning.In this application, computer-readable medium can be any tangible medium for including or store program, the journey
Sequence can be commanded execution system, device or device use or in connection.And in this application, it is computer-readable
Signal media may include in a base band or as carrier wave a part propagate data-signal, wherein carrying computer can
The program code of reading.The data-signal of this propagation can take various forms, including but not limited to electromagnetic signal, optical signal or
Above-mentioned any appropriate combination.Computer-readable signal media can also be any calculating other than computer-readable medium
Machine readable medium, the computer-readable medium can be sent, propagated or transmitted for by instruction execution system, device or device
Part uses or program in connection.The program code for including on computer-readable medium can use any Jie appropriate
Matter transmission, including but not limited to: wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
The calculating of the operation for executing the application can be write with one or more programming languages or combinations thereof
Machine program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
It further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be complete
It executes, partly executed on the user computer on the user computer entirely, being executed as an independent software package, part
Part executes on the remote computer or executes on a remote computer or server completely on the user computer.It is relating to
And in the situation of remote computer, remote computer can pass through the network of any kind --- including local area network (LAN) or extensively
Domain net (WAN)-be connected to subscriber computer, or, it may be connected to outer computer (such as utilize ISP
To be connected by internet).
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard
The mode of part is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor packet
Include acquiring unit, recognition unit, sequencing unit and push unit.Wherein, the title of these units not structure under certain conditions
The restriction of the pairs of unit itself, for example, push unit is also described as " pushing to the comment information set after sequence
The unit of the terminal of target user ".
As on the other hand, present invention also provides a kind of computer-readable medium, which be can be
Included in server described in above-described embodiment；It is also possible to individualism, and without in the supplying server.It is above-mentioned
Computer-readable medium carries one or more program, when said one or multiple programs are executed by the server,
So that the server: obtaining the comment information set for being directed to intended display information；For the comment information in comment information set,
By comment information input comment identification model trained in advance, the evaluation information for the comment information is obtained, wherein comment
Identification model is used to characterize the corresponding relationship of comment information and evaluation information, and evaluation information is for characterizing user to comment information pair
The interest level for the intended display information answered；Based on obtained evaluation information, to the comment information in comment information set
It is ranked up；Comment information set after sequence is pushed to the terminal of target user.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from foregoing invention design, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed herein
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (12)
1. a kind of method for pushed information, comprising:
Obtain the comment information set for being directed to intended display information；
For the comment information in the comment information set, the comment identification model that comment information input is trained in advance,
Obtain the evaluation information for the comment information, wherein the comment identification model is for characterizing comment information and evaluation information
Corresponding relationship, evaluation information is for characterizing user to the interest level of the corresponding intended display information of comment information；
Based on obtained evaluation information, the comment information in the comment information set is ranked up；
Comment information set after sequence is pushed to the terminal of target user.
2. according to the method described in claim 1, wherein, the type of evaluation information includes at least two.
3. it is described to be based on obtained evaluation information according to the method described in claim 2, wherein, to the comment information collection
Comment information in conjunction is ranked up, comprising:
Based on the type of obtained evaluation information, classify to the comment information in the comment information set, obtain to
Few two comment information subclass；
According to the sequence of the corresponding type of preset, evaluation information, at least two comment informations subclass is ranked up；
For the subclass at least two subclass after sequence, it is based on predetermined sortord, to the subset
Comment information in conjunction is ranked up.
4. method described in one of -3 according to claim 1, wherein the comment identification model is trained as follows in advance
It obtains:
Obtain training sample set, wherein training sample include for preset samples show information sample comment information and
The mark evaluation information being labeled in advance for sample comment information；
Using machine learning method, the sample comment information for including using the training sample in the training sample set is as defeated
Enter, using the corresponding mark evaluation information of the sample comment information of input as desired output, training obtains comment identification model.
5. according to the method described in claim 4, wherein, the corresponding mark evaluation information of sample comment information first passes through as follows in advance
Step obtains:
Determine the quantity for browsing the user of sample comment information as the first quantity；
It determines and browsed sample comment information and carried out the quantity of the user of object run as second for samples show information
Quantity；
Determine the ratio of the second quantity Yu the first quantity；
Based on the corresponding relationship of preset mark evaluation information and ratio, the corresponding mark evaluation letter of identified ratio is determined
Breath is used as the corresponding mark evaluation information of sample comment information.
6. a kind of device for pushed information, comprising:
Acquiring unit is configured to obtain the comment information set for being directed to intended display information；
Recognition unit, is configured to for the comment information in the comment information set, which is inputted instruction in advance
Experienced comment identification model obtains the evaluation information for the comment information, wherein the comment identification model is commented for characterizing
By the corresponding relationship of information and evaluation information, evaluation information is for characterizing user to the corresponding intended display information of comment information
Interest level；
Sequencing unit is configured to carry out the comment information in the comment information set based on obtained evaluation information
Sequence；
Push unit is configured to push to the comment information set after sequence the terminal of target user.
7. device according to claim 6, wherein the type of evaluation information includes at least two.
8. device according to claim 7, wherein the sequencing unit includes:
Categorization module is configured to the type based on obtained evaluation information, believes the comment in the comment information set
Breath is classified, and at least two comment information subclass are obtained；
First sorting module is configured to the sequence according to the corresponding type of preset, evaluation information, comments described at least two
It is ranked up by information subset conjunction；
Second sorting module is configured to for the subclass at least two subclass after sequence, based on true in advance
Fixed sortord is ranked up the comment information in the subclass.
9. the device according to one of claim 6-8, wherein the comment identification model is trained as follows in advance
It obtains:
Obtain training sample set, wherein training sample include for preset samples show information sample comment information and
The mark evaluation information being labeled in advance for sample comment information；
Using machine learning method, the sample comment information for including using the training sample in the training sample set is as defeated
Enter, using the corresponding mark evaluation information of the sample comment information of input as desired output, training obtains comment identification model.
10. device according to claim 9, wherein the corresponding mark evaluation information of sample comment information first pass through in advance as
Lower step obtains:
Determine the quantity for browsing the user of sample comment information as the first quantity；
It determines and browsed sample comment information and carried out the quantity of the user of object run as second for samples show information
Quantity；
Determine the ratio of the second quantity Yu the first quantity；
Based on the corresponding relationship of preset mark evaluation information and ratio, the corresponding mark evaluation letter of identified ratio is determined
Breath is used as the corresponding mark evaluation information of sample comment information.
11. a kind of server, comprising:
One or more processors；
Storage device is stored thereon with one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as method as claimed in any one of claims 1 to 5.
12. a kind of computer-readable medium, is stored thereon with computer program, wherein the realization when program is executed by processor
Such as method as claimed in any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811288022.7A CN109492160A (en) | 2018-10-31 | 2018-10-31 | Method and apparatus for pushed information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811288022.7A CN109492160A (en) | 2018-10-31 | 2018-10-31 | Method and apparatus for pushed information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109492160A true CN109492160A (en) | 2019-03-19 |
Family
ID=65693434
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811288022.7A Pending CN109492160A (en) | 2018-10-31 | 2018-10-31 | Method and apparatus for pushed information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109492160A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110730387A (en) * | 2019-11-13 | 2020-01-24 | 腾讯科技（深圳）有限公司 | Video playing control method and device, storage medium and electronic device |
| CN111046941A (en) * | 2019-12-09 | 2020-04-21 | 腾讯科技（深圳）有限公司 | Target comment detection method and device, electronic equipment and storage medium |
| CN111460281A (en) * | 2020-02-27 | 2020-07-28 | 浙江口碑网络技术有限公司 | Information push optimization method and device, storage medium and terminal |
| CN111522940A (en) * | 2020-04-08 | 2020-08-11 | 百度在线网络技术（北京）有限公司 | Method and device for processing comment information |
| CN111784377A (en) * | 2019-04-25 | 2020-10-16 | 北京京东尚科信息技术有限公司 | Method and apparatus for generating information |
| CN111833080A (en) * | 2019-04-15 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Information pushing method and device, electronic equipment and computer-readable storage medium |
| CN112883270A (en) * | 2021-02-26 | 2021-06-01 | 北京金堤科技有限公司 | Public opinion information processing method and device and computer readable storage medium |
| CN112989276A (en) * | 2019-12-12 | 2021-06-18 | 北京沃东天骏信息技术有限公司 | Evaluation method and device of information push system |
| CN113282853A (en) * | 2021-05-26 | 2021-08-20 | 北京字跳网络技术有限公司 | Comment preloading method and device, storage medium and electronic equipment |
| CN113361264A (en) * | 2021-06-25 | 2021-09-07 | 上海哔哩哔哩科技有限公司 | Data processing method and device |
| CN113741759A (en) * | 2021-11-06 | 2021-12-03 | 腾讯科技（深圳）有限公司 | Comment information display method and device, computer equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704941A (en) * | 2016-08-08 | 2018-02-16 | 华为软件技术有限公司 | A kind of method and device for showing goods review |
| CN108667875A (en) * | 2017-03-31 | 2018-10-16 | 北京京东尚科信息技术有限公司 | information updating method and device |
-
2018
- 2018-10-31 CN CN201811288022.7A patent/CN109492160A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704941A (en) * | 2016-08-08 | 2018-02-16 | 华为软件技术有限公司 | A kind of method and device for showing goods review |
| CN108667875A (en) * | 2017-03-31 | 2018-10-16 | 北京京东尚科信息技术有限公司 | information updating method and device |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111833080A (en) * | 2019-04-15 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Information pushing method and device, electronic equipment and computer-readable storage medium |
| CN111784377A (en) * | 2019-04-25 | 2020-10-16 | 北京京东尚科信息技术有限公司 | Method and apparatus for generating information |
| CN110730387A (en) * | 2019-11-13 | 2020-01-24 | 腾讯科技（深圳）有限公司 | Video playing control method and device, storage medium and electronic device |
| CN111046941B (en) * | 2019-12-09 | 2023-08-15 | 腾讯科技（深圳）有限公司 | Target comment detection method and device, electronic equipment and storage medium |
| CN111046941A (en) * | 2019-12-09 | 2020-04-21 | 腾讯科技（深圳）有限公司 | Target comment detection method and device, electronic equipment and storage medium |
| CN112989276A (en) * | 2019-12-12 | 2021-06-18 | 北京沃东天骏信息技术有限公司 | Evaluation method and device of information push system |
| CN111460281A (en) * | 2020-02-27 | 2020-07-28 | 浙江口碑网络技术有限公司 | Information push optimization method and device, storage medium and terminal |
| CN111522940A (en) * | 2020-04-08 | 2020-08-11 | 百度在线网络技术（北京）有限公司 | Method and device for processing comment information |
| CN112883270A (en) * | 2021-02-26 | 2021-06-01 | 北京金堤科技有限公司 | Public opinion information processing method and device and computer readable storage medium |
| CN112883270B (en) * | 2021-02-26 | 2024-04-19 | 北京金堤科技有限公司 | Public opinion information processing method, apparatus and computer readable storage medium |
| CN113282853A (en) * | 2021-05-26 | 2021-08-20 | 北京字跳网络技术有限公司 | Comment preloading method and device, storage medium and electronic equipment |
| CN113282853B (en) * | 2021-05-26 | 2024-04-16 | 北京字跳网络技术有限公司 | Comment preloading method and device, storage medium and electronic equipment |
| CN113361264A (en) * | 2021-06-25 | 2021-09-07 | 上海哔哩哔哩科技有限公司 | Data processing method and device |
| CN113741759A (en) * | 2021-11-06 | 2021-12-03 | 腾讯科技（深圳）有限公司 | Comment information display method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109492160A (en) | Method and apparatus for pushed information | |
| CN108898185A (en) | Method and apparatus for generating image recognition model | |
| CN107463704A (en) | Searching method and device based on artificial intelligence | |
| CN107908789A (en) | Method and apparatus for generating information | |
| CN108932220A (en) | article generation method and device | |
| CN106484766B (en) | Searching method and device based on artificial intelligence | |
| CN109086719A (en) | Method and apparatus for output data | |
| CN109976997A (en) | Test method and device | |
| CN107992478A (en) | The method and apparatus for determining focus incident | |
| CN109325121A (en) | Method and apparatus for determining the keyword of text | |
| CN109271556A (en) | Method and apparatus for output information | |
| CN109190114A (en) | Method and apparatus for generating return information | |
| CN109165344A (en) | Method and apparatus for pushed information | |
| CN109495552A (en) | Method and apparatus for updating clicking rate prediction model | |
| CN110413872A (en) | Method and apparatus for showing information | |
| CN107145485A (en) | Method and apparatus for compressing topic model | |
| CN109785072A (en) | Method and apparatus for generating information | |
| CN109214501A (en) | The method and apparatus of information for identification | |
| CN109815365A (en) | Method and apparatus for handling video | |
| CN107943895A (en) | Information-pushing method and device | |
| CN108446659A (en) | Method and apparatus for detecting facial image | |
| CN109389182A (en) | Method and apparatus for generating information | |
| CN109920431A (en) | Method and apparatus for output information | |
| CN108595448A (en) | Information-pushing method and device | |
| CN109582825A (en) | Method and apparatus for generating information |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190319 |