CN107766580A - The method for pushing and device of message - Google Patents
The method for pushing and device of message Download PDFInfo
- Publication number
- CN107766580A CN107766580A CN201711160588.7A CN201711160588A CN107766580A CN 107766580 A CN107766580 A CN 107766580A CN 201711160588 A CN201711160588 A CN 201711160588A CN 107766580 A CN107766580 A CN 107766580A
- Authority
- CN
- China
- Prior art keywords
- user
- message
- push
- portrait
- forecast model
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention discloses a kind of method for pushing of message and device, it is related to a kind of technical field of data processing, the user that main purpose is to solve the existing message push screening being directed under different scenes is randomness, or tap into capable push by sending chain of messages on a large scale so that the user individual of push is weaker, without directionality, when user does not click on the message of push, can also wasting flow, and clicking rate is relatively low, the problem of causing useless push.Main technical schemes:Obtain the user's portrait for treating PUSH message;The attributive character in user's portrait is extracted, and combines user and screens the user that PUSH message is treated in forecast model determination, the user screens forecast model and is used for the expected user that click behavior be present after being pushed according to the attributive character Filter messages;PUSH message is carried out to the user.It is mainly used in the push of message.
Description
Technical field
The present invention relates to a kind of technical field of data processing, more particularly to the method for pushing and device of a kind of message.
Background technology
Message push refers to that service end sends message link to client, is linked when user clicks on message by client
When, push the process of particular message.Especially, with push function in terminal device different application usage scenario it is universal
Change, such as certain application platform can push red packet, advertisement message to user, be easy to user to receive the message of push in time, it is determined that pushing away
Send the user of message to carry out push be vital.
At present, the existing user for being directed to message push screening under different scenes is randomness, or by sending out on a large scale
Chain of messages is sent to tap into capable push, still, randomness or the large-scale PUSH message that carries out make it that the user individual of push is weaker,
There is no directionality, when user does not click on the message of push, can also wasting flow, and clicking rate is relatively low, causes nothing
With push.
The content of the invention
In view of this, the present invention provides a kind of method for pushing and device of message, and main purpose is solve existing be directed to
The user of message push screening under different scenes is randomness, or taps into capable push by sending chain of messages on a large scale,
So that the user individual of push is weaker, without directionality, when user does not click on the message of push, wasting can also flow
Amount, and clicking rate is relatively low, the problem of causing useless push.
According to one aspect of the invention, there is provided a kind of method for pushing of message, including:
Obtain the user's portrait for treating PUSH message;
The attributive character in user's portrait is extracted, and combines user and screens the use that PUSH message is treated in forecast model determination
Family, the user screen forecast model and are used for the expected use that click behavior be present after being pushed according to the attributive character Filter messages
Family;
PUSH message is carried out to the user.
Further, the attributive character in extraction user's portrait, and combine user and screen forecast model determination
Treating the user of PUSH message includes:
Identify the attributive character of user's portrait corresponding with treating PUSH message;
The input of forecast model will be screened as the attributive character of forecast sample as user, and perform model calculation, obtained
To the user for treating PUSH message.
Further, before user's portrait of PUSH message is treated in the acquisition, methods described also includes:
History propelling data is extracted, user is established according to the history propelling data and screens forecast model.
Further, the extraction history propelling data, user is established according to the history propelling data and screens prediction mould
Type includes:
Parse user in history propelling data and click on behavior and the attributive character of user's portrait;
The user is clicked on into behavior and the attributive character carries out classification based training, and is entered by preset logic regression model
Row returns, and obtains user and screens forecast model.
Further, the extraction history propelling data, user is established according to the history propelling data and screens prediction mould
After type, methods described also includes:
Judge to screen whether the assessment result after forecast model is assessed accords with to the user according to preset feature curve
Default assessed value is closed, if not meeting default assessed value, user is redefined and screens forecast model.
Further, described obtain treats that user's portrait of PUSH message includes:
Operation is marked in the user that PUSH message is treated according to active condition, is defined as any active ues;
The user's portrait for extracting any active ues in the first prefixed time interval.
Further, after the progress PUSH message to the user, methods described also includes:
The clicking rate that behavior carries out monitoring in real time and real-time record produces click behavior user is clicked on to the user,
And push daily record is generated, according to history propelling data of second prefixed time interval in the push daily record to the user
Screening forecast model is updated.
According to one aspect of the invention, there is provided a kind of pusher of message, including:
Acquiring unit, user's portrait of PUSH message is treated for obtaining;
Screening unit, for extracting the attributive character in user's portrait, and combine user and screen forecast model determination
The user of PUSH message is treated, the user screens forecast model and is used for expected after being pushed according to the attributive character Filter messages deposit
In the user of the behavior of click;
Push unit, for carrying out PUSH message to the user.
Further, the screening unit includes:
Identification module, for identifying the attributive character of user's portrait corresponding with treating PUSH message;
Module is run, for the input of forecast model will to be screened as the attributive character of forecast sample as user, and is held
Row model calculation, obtain treating the user of PUSH message.
Further, described device also includes:
Unit is established, for extracting history propelling data, user is established according to the history propelling data and screens prediction mould
Type.
Further, the unit of establishing includes:
Parsing module, behavior and the attributive character of user's portrait are clicked on for parsing user in history propelling data;
Training module, for the user to be clicked on into behavior and attributive character progress classification based training, and by preset
Logic Regression Models are returned, and are obtained user and are screened forecast model.
Further, described device also includes:
Judging unit, for judging to screen the user commenting after forecast model is assessed according to preset feature curve
Estimate whether result meets default assessed value, if not meeting default assessed value, redefine user and screen forecast model.
Further, the acquiring unit includes:
Determining module, the user for treating PUSH message according to active condition are marked operation, are defined as enlivening use
Family;
Extraction module, for extracting user's portrait of any active ues in the first prefixed time interval.
Further, described device also includes:
Updating block, carry out monitoring in real time for the click behavior to the user and record produces click behavior in real time
The clicking rate of user, and push daily record is generated, pushed according to history of second prefixed time interval in the push daily record
Data are screened forecast model to the user and are updated.
According to another aspect of the invention, there is provided a kind of storage medium, being stored with least one in the storage medium can
Execute instruction, the executable instruction make computing device be operated as corresponding to the method for pushing of above-mentioned message.
In accordance with a further aspect of the present invention, there is provided a kind of terminal, including:Processor, memory, communication interface and communication
Bus, the processor, the memory and the communication interface complete mutual communication by the communication bus;
The memory is used to deposit an at least executable instruction, and the executable instruction makes the computing device above-mentioned
Operated corresponding to the method for pushing of message.
By above-mentioned technical proposal, technical scheme provided in an embodiment of the present invention at least has following advantages:
The invention provides a kind of method for pushing of message and device, obtains the user's portrait for treating PUSH message first, so
The attributive character in user's portrait is extracted afterwards, and is combined user and screened the user that PUSH message is treated in forecast model determination, institute
State user's screening forecast model and be used for the expected user that click behavior be present after being pushed according to the attributive character Filter messages, most
PUSH message is carried out to the user afterwards.User with the existing message push screening being directed under different scenes is randomness,
Or tap into capable push by sending chain of messages on a large scale so that the user individual of push is weaker, without directionality, works as user
When not clicking on the message of push, can also wasting flow, and clicking rate is relatively low, causes useless push to compare, the present invention
Embodiment is used as sample data by the attributive character in being drawn a portrait according to user, then screens forecast model classification instruction by user
The user of click behavior be present after getting PUSH message, and pushed, the blindness for avoiding message from pushing, increase user couple
The probability that message is clicked on, the efficiency of message push is improved, save PUSH message flow, so as to strengthen the individual character of message push
Change.
Described above is only the general introduction of technical solution of the present invention, in order to better understand the technological means of the present invention,
And can be practiced according to the content of specification, and in order to allow above and other objects of the present invention, feature and advantage can
Become apparent, below especially exemplified by the embodiment of the present invention.
Brief description of the drawings
By reading the detailed description of hereafter preferred embodiment, it is various other the advantages of and benefit it is common for this area
Technical staff will be clear understanding.Accompanying drawing is only used for showing the purpose of preferred embodiment, and is not considered as to the present invention
Limitation.And in whole accompanying drawing, identical part is denoted by the same reference numerals.In the accompanying drawings:
Fig. 1 shows a kind of method for pushing flow chart of message provided in an embodiment of the present invention;
Fig. 2 shows the method for pushing flow chart of another message provided in an embodiment of the present invention;
Fig. 3 shows the schematic flow sheet that a kind of model provided in an embodiment of the present invention is established and pushed;
Fig. 4 shows a kind of pusher block diagram of message provided in an embodiment of the present invention;
Fig. 5 shows the pusher block diagram of another message provided in an embodiment of the present invention;
Fig. 6 shows a kind of terminal schematic diagram provided in an embodiment of the present invention.
Embodiment
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although the disclosure is shown in accompanying drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
Limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
Completely it is communicated to those skilled in the art.
The embodiments of the invention provide a kind of method for pushing of message, as shown in figure 1, methods described includes:
101st, the user's portrait for treating PUSH message is obtained.
Wherein, user portrait for user's registration or when logging in different application for embodying the user of user identity
Portrait, in general, in order to embody the identity of user, the user's portrait for uploading or downloading is the true portrait of user, same use
Family can be drawn a portrait in different application programs using no user, and the embodiment of the present invention is not specifically limited.
It should be noted that different application is directed to owned user, can be analyzed by big data will be all
User's portrait of user is arranged and analyzed, and specific user's portrait can be stored in the database with massive storage space
In, when carrying out message push, to obtain user's portrait of user directly from database, the embodiment of the present invention is not done specifically
Limit.
102nd, the attributive character in user's portrait is extracted, and combines user's screening forecast model and determines to treat PUSH message
User.
Wherein, the user screens forecast model and is used to be existed a little according to expection after attributive character Filter messages push
Hit the user of behavior, the attributive character in user's portrait is the information for embodying user characteristics, can include population category
Property information, label information, installation list information, mobile phone hardware information, income letter reference breath, for example, sex, age, geographical position
Put, finance and money management data etc., the embodiment of the present invention is not specifically limited.In general, the specific determination mode in step 102 is will
Attributive character screens the input of forecast model as user, then screens use of the forecast model to the attributive character by user
Family is trained classification, obtains carrying out the user of message push.It can be disaggregated model that the user, which screens forecast model,
Obtained model is trained, such as GBDT models and XGBOOST models, the embodiment of the present invention are not specifically limited.
It should be noted that the process of extraction attributive character can be to identify user's head portrait picture number in user's portrait respectively
Determine the attributive character of user's portrait according to information such as the terminal devices of, background image data and upload user portrait, in general,
One user's portrait can correspond to registration or the log-on message for determining user, therefore, can be drawn a portrait according to user and determine to use
The sex at family, age, geographical position, finance and money management data etc..Furthermore it is also possible to the specific character features in being drawn a portrait for user
Matched with the character features image being stored in advance in background system to determine the attributive character of user's portrait, for example,
User's portrait has personage a, a background a in being, character features image is femaleness, masculinity, child's feature in background system,
Geographical position is Pekinese's background, the background in Shanghai, user's portrait is matched one by one, it is then determined that it is female to go out attributive character
Property, geographical position are Beijing.
103rd, PUSH message is carried out to the user.
It should be noted that the user for carrying out PUSH message is to carry out clicking on the big use of behavior probability after carrying out PUSH message
Family, in general, clicking on probability can be configured by user's screening forecast model, so that carry out the user of PUSH message
Clicking rate increase, and make PUSH message that there is directionality.
The invention provides a kind of method for pushing of message, the use with the existing message push screening being directed under different scenes
Family is randomness, or taps into capable push by sending chain of messages on a large scale so that the user individual of push is weaker, does not have
Directionality, when user does not click on the message of push, can also wasting flow, and clicking rate is relatively low, causes useless push away
Send and compare, the embodiment of the present invention is used as sample data by the attributive character in being drawn a portrait according to user, is then screened by user
There is the user of click behavior in forecast model classification based training, and pushed after obtaining PUSH message, avoid message from pushing blind
Mesh, the probability that increase user is clicked on to message, the efficiency of message push is improved, PUSH message flow is saved, so as to increase
The personalization of strong message push.
The embodiments of the invention provide the method for pushing of another message, as shown in Fig. 2 methods described includes:
201st, history propelling data is extracted, user is established according to the history propelling data and screens forecast model.
Wherein, the history propelling data is to carry out clicking on message after user carries out message push relative to this before
Whether data, including user carry out information, the embodiment of the present invention such as click message, the number clicked on and are not specifically limited.In addition,
It is a disaggregated model because user screens forecast model, i.e., the user of propelling data will be clicked in history propelling data with not having
The user for clicking on propelling data is classified, and forecast model is screened so as to obtain user, so as to when being predicted, according to training
Obtained user screens forecast model and determines that the higher user of clicking rate carries out PUSH message.
For the embodiment of the present invention, step 201 is specifically as follows:Parse user in history propelling data and click on behavior with using
The attributive character of family portrait;The user is clicked on into behavior and the attributive character carries out classification based training, and passes through preset logic
Regression model is returned, and is obtained user and is screened forecast model.
Wherein, when history propelling data includes PUSH message there is the user of click behavior, and user's portrait in user,
Then attributive character, ascribed characteristics of population information, label information, installation list letter are identified for the view data in user's portrait
Breath, mobile phone hardware information, income letter reference breath, the ascribed characteristics of population include sex, age etc., and tag attributes include click behavior being present
1st, in the absence of behavior 0 is clicked on, installation list information includes the time of installation current application, the system for installing current application, installation
Current application downloads source etc., CPU of the mobile phone hardware information including mobile phone, memory, screen, camera, battery etc., income letter
Reference breath includes finance, the financing information being used for.Due to the quick variability of user behavior, the history of 2 weeks or 3 weeks can be chosen
Propelling data is not specifically limited as the training sample data for establishing model, the embodiment of the present invention.
It should be noted that in the embodiment of the present invention, the click behavior of user is respectively labeled as 0 or 1, that is, clicks on behavior
For 1, behavior is not clicked on as 0, the attributive character parsed in being drawn a portrait with reference to user is used as training sample, progress classification based training, is
The accuracy of training pattern is improved, training pattern can select gradient lifting decision tree GBDT (GradientBoosting
DecisionTree) model or to the improved XGBOOST models of GBDT, then obtains the grader of a binary tree structure,
In order that classification results ensure between 0 to 1, the model after training passes through LR (Logistic Regression) logistic regression
Model carries out recurrence processing, finally obtains user and screens forecast model.
It can include for the embodiment of the present invention, the step of after step 201:Judge according to preset feature curve to described
User screens whether the assessment result after forecast model is assessed meets default assessed value, if not meeting default assessed value,
Redefine user and screen forecast model.
For the embodiment of the present invention, in order to ensure the user of training screen forecast model accuracy and high efficiency, it is necessary to
Forecast model is screened to obtained user to assess.Wherein, the preset feature curve is used for classification of assessment device, can be to connect
Under receipts person's operating characteristics ROC (Receiver Operating Characteristic) curves and ROC curve between 0.1 and 1 it
Between area AUC (Area under Curve), in general can be according to different application PUSH message for default assessed value
Scene is set, scene can include send red packet, push application download etc., in general, send red packet when message when move
State, consuming flow is bigger, higher to the accuracy requirement of user's screening forecast model, therefore can set default assessed value
It is set to higher numerical value, the embodiment of the present invention is not specifically limited.
202nd, operation is marked in the user that PUSH message is treated according to active condition, is defined as any active ues.
For the embodiment of the present invention, for screening out as the user's conduct that may not click on PUSH message further
User screens the forecast sample of forecast model, reduces user and screens the computational burden of forecast model, it is necessary to the multi-user that comforms in advance
In find the user that there may be click behavior.Wherein, the active condition is for judging whether user is used as sample data
It is predicted, so as to the condition of PUSH message, for example, active condition can be to log in or use within a certain period of time application program
Number exceedes certain amount, can also be whether user triggers corresponding operation when using application program, and the present invention is implemented
Example is not specifically limited.In general, it can be judged using the user of application program according to active condition all, work will be met
The user's mark of jump condition is any active ues, so as to using the information of any active ues as effective information.
203rd, user's portrait of any active ues in the first prefixed time interval is extracted.
For the embodiment of the present invention, in order to be predicted to the user of effective information, improve screening PUSH message user's
Accuracy is, it is necessary to extract user's portrait of any active ues.Further, since any active ues are ineffectiveness in the presence of having, cross for a long time
The information of any active ues may lose validity, therefore set the first prefixed time interval as the use for extracting any active ues
The time interval of family portrait, it can be 3 days, or 5 days, the embodiment of the present invention was not specifically limited.
204th, the attributive character of user's portrait corresponding with treating PUSH message is identified.
For the embodiment of the present invention, in order to reduce the quantity of the attributive character as sample, increase user screens prediction mould
The accuracy that type is predicted, can be when obtaining attributive character, and Direct Recognition is with treating that PUSH message has the category of corresponding relation
Property feature, for example, user portrait in can parse attributive character ascribed characteristics of population information, label information, installation list information,
Mobile phone hardware information, income letter reference breath, in general, attributive character corresponding with giving bonus are population attribute information, label letter
Breath, income reference information, then ascribed characteristics of population information, label information, income reference information in being drawn a portrait with Direct Recognition user
Particular content, to reduce the wasting of resources of data processing.
205th, the input of forecast model will be screened as the attributive character of forecast sample as user, and performs model fortune
Calculate, obtain treating the user of PUSH message.
It should be noted that it is to treat to be predicted sample as the grader trained because user screens forecast model
The algorithm of classification is predicted, therefore, is sieved according to the attributive character obtained in step 204 as forecast sample input value user
Select and model calculation is carried out in forecast model, obtain the regressand value between 0 to 1, so that it is determined that it is that possible treat to push away which, which goes out,
Message trigger is sent to click on the user of behavior.
206th, PUSH message is carried out to the user.
This step is identical with step 103 method shown in Fig. 1, will not be repeated here.
207th, the point that behavior carries out monitoring in real time and real-time record produces click behavior user is clicked on to the user
Rate is hit, and generates push daily record, according to history propelling data of second prefixed time interval in the push daily record to institute
User's screening forecast model is stated to be updated.
For the embodiment of the present invention, the accuracy of forecast model is screened in order to further improve user, and strengthen user
The directionality of PUSH message, it is necessary to monitor whether user has carried out point to the message of push in real time after PUSH message is sent
Behavior is hit, carries out recording click situation after click behavior is generated, obtains pushing daily record, to screen prediction mould to user
Type is updated.Wherein, second prefixed time interval is the time interval that forecast model is screened for updating user, specifically
Time can be set according to different application scenarios, different application programs, different any active ues amounts etc., for example, can
Think 1 month, 3 months etc., the embodiment of the present invention is not specifically limited.
For the above embodiment of the present invention, specific application scenarios can be with as follows, but not limited to this, including:Such as
Shown in Fig. 3, user's portrait is obtained in the user determined from the history click behavior of push red packet user, in being drawn a portrait according to user
The attributive character selection ascribed characteristics of population, label information, installation list, mobile phone hardware information, income reference information, raw raw training sample
This, is trained according to GBDT models and XGBOOST models, after the user trained, which screens forecast model, to be assessed by AUC
It is stand-by as forecast model, any active ues information of nearly three days is then obtained, the user's portrait for extracting any active ues, is selected above-mentioned
Feature generates forecast sample, then screens forecast model using these forecast samples as input, feeding user and is trained recurrence,
Obtain carrying out the user of message push, carry out push red packet, then record the clicking rate of user, and it is high to recall clicking rate
User, forecast model is screened to update user again according to the user of the high user of clicking rate portrait attributive character.
The invention provides the method for pushing of another message, the embodiment of the present invention according to history propelling data by extracting
Drawn a portrait in the presence of the user for clicking on behavior user, the attributive character training classification in being drawn a portrait according to user, establish user and screen prediction
Model, user's portrait of any active ues in time interval is then obtained, parse the attributive character in user's portrait as prediction
Sample data, screened with reference to user after forecast model predicts progress message push and the user of click behavior be present, and pushed away
Send, realize the personalized push of message, the combination that forecast model and user's portrait are screened using user can be predicted preferably
Which user, which exists, is clicked on the possibility of PUSH message, so as to increase the clicking rate that user is clicked on to message, is improved message and is pushed away
The efficiency sent, reduce the waste of PUSH message flow.
Further, as the realization to method shown in above-mentioned Fig. 1, the embodiments of the invention provide a kind of push of message
Device, as shown in figure 4, the device includes:Acquiring unit 31, screening unit 32, push unit 33.
Acquiring unit 31, user's portrait of PUSH message is treated for obtaining;The acquiring unit 31 fills for the push of message
Put the program module for performing and obtaining the user's portrait for treating PUSH message.
Screening unit 32, for extracting the attributive character in user's portrait, and it is true to combine user's screening forecast model
Surely the user of PUSH message is treated, the user screens forecast model and is used to be expected after pushing according to the attributive character Filter messages
In the presence of the user of the behavior of click;The screening unit 32 performs the attribute extracted in user's portrait for the pusher of message
Feature, and combine user and screen the program module that the user of PUSH message is treated in forecast model determination.
Push unit 33, for carrying out PUSH message to the user.The push unit 33 is the pusher of message
Perform the program module that PUSH message is carried out to the user.
The invention provides a kind of pusher of message, the use with the existing message push screening being directed under different scenes
Family is randomness, or taps into capable push by sending chain of messages on a large scale so that the user individual of push is weaker, does not have
Directionality, when user does not click on the message of push, can also wasting flow, and clicking rate is relatively low, causes useless push away
Send and compare, the embodiment of the present invention is used as sample data by the attributive character in being drawn a portrait according to user, is then screened by user
There is the user of click behavior in forecast model classification based training, and pushed after obtaining PUSH message, avoid message from pushing blind
Mesh, the probability that increase user is clicked on to message, the efficiency of message push is improved, PUSH message flow is saved, so as to increase
The personalization of strong message push.
Further, as the realization to method shown in above-mentioned Fig. 2, the embodiments of the invention provide pushing away for another message
Device is sent, as shown in figure 5, the device includes:Acquiring unit 41, screening unit 42, push unit 43, establish unit 44, judge
Unit 45, updating block 46.
Acquiring unit 41, user's portrait of PUSH message is treated for obtaining;
Screening unit 42, for extracting the attributive character in user's portrait, and it is true to combine user's screening forecast model
Surely the user of PUSH message is treated, the user screens forecast model and is used to be expected after pushing according to the attributive character Filter messages
In the presence of the user of the behavior of click;
Push unit 43, for carrying out PUSH message to the user.
Specifically, the screening unit 42 includes:
Identification module 4201, for identifying the attributive character of user's portrait corresponding with treating PUSH message;
Module 4202 is run, for the input of forecast model will to be screened as the attributive character of forecast sample as user,
And model calculation is performed, obtain treating the user of PUSH message.
Further, described device also includes:
Unit 44 is established, for extracting history propelling data, user is established according to the history propelling data and screens prediction
Model.
Specifically, the unit 44 of establishing includes:
Parsing module 4401, behavior and the attributive character of user's portrait are clicked on for parsing user in history propelling data;
Training module 4402, for the user to be clicked on into behavior and attributive character progress classification based training, and pass through
Preset logic regression model is returned, and is obtained user and is screened forecast model.
Further, described device also includes:
Judging unit 45, after judging that screening forecast model to the user according to preset feature curve assesses
Whether assessment result meets default assessed value, if not meeting default assessed value, redefines user and screens forecast model.
Specifically, the acquiring unit 41 includes:
Determining module 4101, the user for treating PUSH message according to active condition are marked operation, are defined as living
Jump user;
Extraction module 4102, for extracting user's portrait of any active ues in the first prefixed time interval.
Further, described device also includes:
Updating block 46, carry out monitoring in real time for the click behavior to the user and row is clicked in record generation in real time
For the clicking rate of user, and push daily record is generated, pushed away according to history of second prefixed time interval in the push daily record
Send data to screen forecast model to the user to be updated.
The invention provides the pusher of another message, the embodiment of the present invention according to history propelling data by extracting
Drawn a portrait in the presence of the user for clicking on behavior user, the attributive character training classification in being drawn a portrait according to user, establish user and screen prediction
Model, user's portrait of any active ues in time interval is then obtained, parse the attributive character in user's portrait as prediction
Sample data, screened with reference to user after forecast model predicts progress message push and the user of click behavior be present, and pushed away
Send, realize the personalized push of message, the combination that forecast model and user's portrait are screened using user can be predicted preferably
Which user, which exists, is clicked on the possibility of PUSH message, so as to increase the clicking rate that user is clicked on to message, is improved message and is pushed away
The efficiency sent, reduce the waste of PUSH message flow.
Based on above-mentioned method as shown in Figure 1, accordingly, the embodiment of the present invention additionally provides a kind of computer-readable storage medium
Matter, computer program is stored thereon with, the program realizes following steps when being executed by processor:Obtain the user for treating PUSH message
Portrait;The attributive character in user's portrait is extracted, and combines user and screens the user that PUSH message is treated in forecast model determination,
The user screens forecast model and is used for the expected user that click behavior be present after being pushed according to the attributive character Filter messages;
PUSH message is carried out to the user.
Embodiment based on above-mentioned method as shown in Figure 1 and device as shown in Figure 4, the one of one embodiment of the invention offer
The structural representation of kind terminal, as shown in fig. 6, the specific embodiment of the invention is not limited the specific implementation of terminal device.
As shown in fig. 6, the terminal can include:Processor (processor) 91, communication interface (Communications
Interface) 52, memory (memory) 53 and communication bus 54.Wherein:Processor 51, communication interface 52, Yi Jicun
Reservoir 53 completes mutual communication by communication bus 54.Communication interface 54, for miscellaneous equipment such as client or its
The network element communication of its server etc..Processor 51, for configuration processor, the method for pushing that can specifically perform above-mentioned message is implemented
Correlation step in example.Specifically, program can include program code, and the program code includes computer-managed instruction.Processing
Device 51 is probably central processor CPU, or specific integrated circuit ASIC (Application Specific
Integrated Circuit), or it is arranged to implement one or more integrated circuits of the embodiment of the present invention.
The one or more processors that terminal includes, can be same type of processor, such as one or more CPU;
Can be different types of processor, such as one or more CPU and one or more ASIC.Memory 53, for depositing journey
Sequence.Memory 53 may include high-speed RAM memory, it is also possible to also including nonvolatile memory (non-volatile
Memory), a for example, at least magnetic disk storage.Program specifically can be used for so that processor 51 performs following operation:Obtain
Treat user's portrait of PUSH message;The attributive character in user's portrait is extracted, and combines user and screens forecast model determination
The user of PUSH message is treated, the user screens forecast model and is used for expected after being pushed according to the attributive character Filter messages deposit
In the user of the behavior of click;PUSH message is carried out to the user.
Algorithm and display be not inherently related to any certain computer, virtual system or miscellaneous equipment provided herein.
Various general-purpose systems can also be used together with teaching based on this.As described above, required by constructing this kind of system
Structure be obvious.In addition, the present invention is not also directed to any certain programmed language.It should be understood that it can utilize various
Programming language realizes the content of invention described herein, and the description done above to language-specific is to disclose this hair
Bright preferred forms.
In the specification that this place provides, numerous specific details are set forth.It is to be appreciated, however, that the implementation of the present invention
Example can be put into practice in the case of these no details.In some instances, known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this description.
Similarly, it will be appreciated that in order to simplify the disclosure and help to understand one or more of each inventive aspect,
Above in the description to the exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the method for the disclosure should be construed to reflect following intention:I.e. required guarantor
The application claims of shield features more more than the feature being expressly recited in each claim.It is more precisely, such as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following embodiment are expressly incorporated in the embodiment, wherein each claim is in itself
Separate embodiments all as the present invention.
Those skilled in the art, which are appreciated that, to be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more equipment different from the embodiment.Can be the module or list in embodiment
Member or component be combined into a module or unit or component, and can be divided into addition multiple submodule or subelement or
Sub-component.In addition at least some in such feature and/or process or unit exclude each other, it can use any
Combination is disclosed to all features disclosed in this specification (including adjoint claim, summary and accompanying drawing) and so to appoint
Where all processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification (including adjoint power
Profit requires, summary and accompanying drawing) disclosed in each feature can be by providing the alternative features of identical, equivalent or similar purpose come generation
Replace.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included some features rather than further feature, but the combination of the feature of different embodiments means in of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
One of meaning mode can use in any combination.
The all parts embodiment of the present invention can be realized with hardware, or to be run on one or more processor
Software module realize, or realized with combinations thereof.It will be understood by those of skill in the art that it can use in practice
Microprocessor or digital signal processor (DSP) are realized in the method for pushing of message according to embodiments of the present invention and device
Some or all parts some or all functions.The present invention is also implemented as being used to perform side as described herein
The some or all equipment or program of device (for example, computer program and computer program product) of method.It is such
Realizing the program of the present invention can store on a computer-readable medium, or can have the shape of one or more signal
Formula.Such signal can be downloaded from internet website and obtained, and either be provided or with any other shape on carrier signal
Formula provides.
It should be noted that the present invention will be described rather than limits the invention for above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between bracket should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of some different elements and being come by means of properly programmed computer real
It is existing.In if the unit claim of equipment for drying is listed, several in these devices can be by same hardware branch
To embody.The use of word first, second, and third does not indicate that any order.These words can be explained and run after fame
Claim.
Embodiments of the invention include:
A1, a kind of method for pushing of message, including:
Obtain the user's portrait for treating PUSH message;
The attributive character in user's portrait is extracted, and combines user and screens the use that PUSH message is treated in forecast model determination
Family, the user screen forecast model and are used for the expected use that click behavior be present after being pushed according to the attributive character Filter messages
Family;
PUSH message is carried out to the user.
A2, the method according to A1, the attributive character extracted in user's portrait, and it is pre- to combine user's screening
Model is surveyed to determine to treat that the user of PUSH message includes:
Identify the attributive character of user's portrait corresponding with treating PUSH message;
The input of forecast model will be screened as the attributive character of forecast sample as user, and perform model calculation, obtained
To the user for treating PUSH message.
A3, the method according to A1 or A2, before user's portrait of PUSH message is treated in the acquisition, methods described is also wrapped
Include:
History propelling data is extracted, user is established according to the history propelling data and screens forecast model.
A4, the method according to A3, the extraction history propelling data, user is established according to the history propelling data
Screening forecast model includes:
Parse user in history propelling data and click on behavior and the attributive character of user's portrait;
The user is clicked on into behavior and the attributive character carries out classification based training, and is entered by preset logic regression model
Row returns, and obtains user and screens forecast model.
A5, the method according to A4, the extraction history propelling data, user is established according to the history propelling data
After screening forecast model, methods described also includes:
Judge to screen whether the assessment result after forecast model is assessed accords with to the user according to preset feature curve
Default assessed value is closed, if not meeting default assessed value, user is redefined and screens forecast model.
A6, the method according to any one of A1-A5, described obtain treat that user's portrait of PUSH message includes:
Operation is marked in the user that PUSH message is treated according to active condition, is defined as any active ues;
The user's portrait for extracting any active ues in the first prefixed time interval.
A7, the method according to A6, it is described to the user carry out PUSH message after, methods described also includes:
The clicking rate that behavior carries out monitoring in real time and real-time record produces click behavior user is clicked on to the user,
And push daily record is generated, according to history propelling data of second prefixed time interval in the push daily record to the user
Screening forecast model is updated.
B8, a kind of pusher of message, including:
Acquiring unit, user's portrait of PUSH message is treated for obtaining;
Screening unit, for extracting the attributive character in user's portrait, and combine user and screen forecast model determination
The user of PUSH message is treated, the user screens forecast model and is used for expected after being pushed according to the attributive character Filter messages deposit
In the user of the behavior of click;
Push unit, for carrying out PUSH message to the user.
B9, the device according to B8, the screening unit include:
Identification module, for identifying the attributive character of user's portrait corresponding with treating PUSH message;
Module is run, for the input of forecast model will to be screened as the attributive character of forecast sample as user, and is held
Row model calculation, obtain treating the user of PUSH message.
B10, the device according to B8 or B9, described device also include:
Unit is established, for extracting history propelling data, user is established according to the history propelling data and screens prediction mould
Type.
B11, the device according to B10, the unit of establishing include:
Parsing module, behavior and the attributive character of user's portrait are clicked on for parsing user in history propelling data;
Training module, for the user to be clicked on into behavior and attributive character progress classification based training, and by preset
Logic Regression Models are returned, and are obtained user and are screened forecast model.
B12, the device according to B11, described device also include:
Judging unit, for judging to screen the user commenting after forecast model is assessed according to preset feature curve
Estimate whether result meets default assessed value, if not meeting default assessed value, redefine user and screen forecast model.
B13, the device according to any one of B8-B12, the acquiring unit include:
Determining module, the user for treating PUSH message according to active condition are marked operation, are defined as enlivening use
Family;
Extraction module, for extracting user's portrait of any active ues in the first prefixed time interval.
B14, the device according to B13, described device also include:
Updating block, carry out monitoring in real time for the click behavior to the user and record produces click behavior in real time
The clicking rate of user, and push daily record is generated, pushed according to history of second prefixed time interval in the push daily record
Data are screened forecast model to the user and are updated.
C15, a kind of storage medium, an at least executable instruction, the executable instruction are stored with the storage medium
Make operation corresponding to the method for pushing of message of the computing device as any one of A1-A7.
D16, a kind of terminal, including:Processor, memory, communication interface and communication bus, the processor, described deposit
Reservoir and the communication interface complete mutual communication by the communication bus;
The memory is used to deposit an at least executable instruction, and the executable instruction makes the computing device such as
Operated corresponding to the method for pushing of message any one of A1-A7.
Claims (10)
- A kind of 1. method for pushing of message, it is characterised in that including:Obtain the user's portrait for treating PUSH message;The attributive character in user's portrait is extracted, and combines user and screens the user that PUSH message is treated in forecast model determination, The user screens forecast model and is used for the expected user that click behavior be present after being pushed according to the attributive character Filter messages;PUSH message is carried out to the user.
- 2. according to the method for claim 1, it is characterised in that the attributive character in extraction user's portrait, and Forecast model is screened with reference to user to determine to treat that the user of PUSH message includes:Identify the attributive character of user's portrait corresponding with treating PUSH message;The input of forecast model will be screened as the attributive character of forecast sample as user, and perform model calculation, treated The user of PUSH message.
- 3. method according to claim 1 or 2, it is characterised in that before user's portrait of PUSH message is treated in the acquisition, Methods described also includes:History propelling data is extracted, user is established according to the history propelling data and screens forecast model.
- 4. according to the method for claim 3, it is characterised in that the extraction history propelling data, pushed away according to the history Sending data to establish user's screening forecast model includes:Parse user in history propelling data and click on behavior and the attributive character of user's portrait;The user is clicked on into behavior and the attributive character carries out classification based training, and is returned by preset logic regression model Return, obtain user and screen forecast model.
- A kind of 5. pusher of message, it is characterised in that including:Acquiring unit, user's portrait of PUSH message is treated for obtaining;Screening unit, for extracting the attributive character in user's portrait, and combine user's screening forecast model and determine to wait to push away The user of message is sent, the user screens forecast model and is used to exist a little according to expection after attributive character Filter messages push Hit the user of behavior;Push unit, for carrying out PUSH message to the user.
- 6. device according to claim 5, it is characterised in that the screening unit includes:Identification module, for identifying the attributive character of user's portrait corresponding with treating PUSH message;Module is run, for the input of forecast model will to be screened as the attributive character of forecast sample as user, and performs mould Type computing, obtain treating the user of PUSH message.
- 7. the device according to claim 5 or 6, it is characterised in that described device also includes:Unit is established, for extracting history propelling data, user is established according to the history propelling data and screens forecast model.
- 8. device according to claim 7, it is characterised in that the unit of establishing includes:Parsing module, behavior and the attributive character of user's portrait are clicked on for parsing user in history propelling data;Training module, for the user to be clicked on into behavior and attributive character progress classification based training, and pass through preset logic Regression model is returned, and is obtained user and is screened forecast model.
- 9. a kind of storage medium, an at least executable instruction is stored with the storage medium, the executable instruction makes processing Device performs operation corresponding to the method for pushing of the message as any one of claim 1-4.
- 10. a kind of terminal, including:Processor, memory, communication interface and communication bus, the processor, the memory and The communication interface completes mutual communication by the communication bus;The memory is used to deposit an at least executable instruction, and the executable instruction makes the computing device such as right will Ask and operated corresponding to the method for pushing of the message any one of 1-4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711160588.7A CN107766580A (en) | 2017-11-20 | 2017-11-20 | The method for pushing and device of message |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711160588.7A CN107766580A (en) | 2017-11-20 | 2017-11-20 | The method for pushing and device of message |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107766580A true CN107766580A (en) | 2018-03-06 |
Family
ID=61279075
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711160588.7A Pending CN107766580A (en) | 2017-11-20 | 2017-11-20 | The method for pushing and device of message |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107766580A (en) |
Cited By (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108920717A (en) * | 2018-07-27 | 2018-11-30 | 百度在线网络技术(北京)有限公司 | For showing the method and device of information |
CN109166012A (en) * | 2018-09-21 | 2019-01-08 | 苏州创旅天下信息技术有限公司 | The method and apparatus of classification and information push for stroke predetermined class user |
CN109167816A (en) * | 2018-08-03 | 2019-01-08 | 广州虎牙信息科技有限公司 | Information-pushing method, device, equipment and storage medium |
CN109218390A (en) * | 2018-07-12 | 2019-01-15 | 北京比特智学科技有限公司 | User's screening technique and device |
CN109218411A (en) * | 2018-08-22 | 2019-01-15 | 中国平安人寿保险股份有限公司 | Data processing method and device, computer readable storage medium, electronic equipment |
CN109241408A (en) * | 2018-08-14 | 2019-01-18 | 深圳开黑科技有限公司 | User's portrait processing method, equipment and the system for playing platform are accompanied based on game |
CN109684554A (en) * | 2018-12-26 | 2019-04-26 | 腾讯科技(深圳)有限公司 | The determination method and news push method of the potential user of news |
CN109726322A (en) * | 2019-01-28 | 2019-05-07 | 北京百度网讯科技有限公司 | Method and apparatus for pushed information |
CN109741113A (en) * | 2019-01-10 | 2019-05-10 | 博拉网络股份有限公司 | A kind of user's purchase intention prediction technique based on big data |
CN110135942A (en) * | 2019-04-12 | 2019-08-16 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus and computer readable storage medium |
CN110399550A (en) * | 2019-03-14 | 2019-11-01 | 腾讯科技(深圳)有限公司 | A kind of information recommendation method and device |
CN110532467A (en) * | 2019-08-22 | 2019-12-03 | 上海易点时空网络有限公司 | Activity recommendation method and device, equipment, storage medium based on push model |
CN110782291A (en) * | 2019-10-29 | 2020-02-11 | 加和(北京)信息科技有限公司 | Advertisement delivery user determination method and device, storage medium and electronic device |
CN110875949A (en) * | 2018-09-04 | 2020-03-10 | 京东数字科技控股有限公司 | Method and device for pushing information |
CN111061979A (en) * | 2018-10-17 | 2020-04-24 | 北京字节跳动网络技术有限公司 | User label pushing method and device, electronic equipment and medium |
CN111090815A (en) * | 2019-12-31 | 2020-05-01 | 恩亿科(北京)数据科技有限公司 | Label generation method and device |
CN111159561A (en) * | 2019-12-31 | 2020-05-15 | 青梧桐有限责任公司 | Method for constructing recommendation engine according to user behaviors and user portrait |
CN111292107A (en) * | 2018-12-06 | 2020-06-16 | 北京嘀嘀无限科技发展有限公司 | User data processing method and device, electronic equipment and storage medium |
CN111612366A (en) * | 2020-05-27 | 2020-09-01 | 中国联合网络通信集团有限公司 | Channel quality evaluation method and device, electronic equipment and storage medium |
CN111681050A (en) * | 2020-06-05 | 2020-09-18 | 深圳市卡牛科技有限公司 | Advertisement pushing method, device, equipment and storage medium |
CN111754287A (en) * | 2020-02-14 | 2020-10-09 | 北京沃东天骏信息技术有限公司 | Article screening method, apparatus, device and storage medium |
CN111860878A (en) * | 2019-12-31 | 2020-10-30 | 北京骑胜科技有限公司 | Repair reporting method for shared bicycle, server, electronic device and storage medium |
CN112015975A (en) * | 2020-07-15 | 2020-12-01 | 北京淇瑀信息科技有限公司 | Financial user-oriented information pushing method and device based on Newton's cooling law |
CN112148994A (en) * | 2020-10-23 | 2020-12-29 | 腾讯科技(深圳)有限公司 | Information push effect evaluation method and device, electronic equipment and storage medium |
CN112532748A (en) * | 2020-12-24 | 2021-03-19 | 北京百度网讯科技有限公司 | Message pushing method, device, equipment, medium and computer program product |
CN113110942A (en) * | 2021-03-26 | 2021-07-13 | 支付宝(杭州)信息技术有限公司 | Message distribution method and device, computer equipment and storage medium |
CN113360768A (en) * | 2021-06-29 | 2021-09-07 | 平安信托有限责任公司 | Product recommendation method, device and equipment based on user portrait and storage medium |
CN113742571A (en) * | 2021-08-03 | 2021-12-03 | 大箴(杭州)科技有限公司 | Message pushing method and device based on big data and storage medium |
CN113761003A (en) * | 2021-04-15 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | User portrait data processing method and device, electronic equipment and readable storage medium |
CN114707097A (en) * | 2022-05-31 | 2022-07-05 | 每日互动股份有限公司 | Data processing system for acquiring target message flow |
CN114820080A (en) * | 2022-05-24 | 2022-07-29 | 上海钧正网络科技有限公司 | User grouping method, system, device and medium based on crowd circulation |
CN116434601A (en) * | 2023-04-25 | 2023-07-14 | 广州市单元信息科技有限公司 | Intelligent processing method and system for message pushing |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104917896A (en) * | 2015-06-12 | 2015-09-16 | 努比亚技术有限公司 | Data pushing method and terminal equipment |
CN105096158A (en) * | 2015-07-01 | 2015-11-25 | 北京奇虎科技有限公司 | Method and device for pushing information |
CN105701191A (en) * | 2016-01-08 | 2016-06-22 | 腾讯科技(深圳)有限公司 | Push information click rate estimation method and device |
CN105915438A (en) * | 2016-04-15 | 2016-08-31 | 北京奇虎科技有限公司 | Message pushing method, apparatus, and system |
CN107613022A (en) * | 2017-10-20 | 2018-01-19 | 广州优视网络科技有限公司 | Content delivery method, device and computer equipment |
-
2017
- 2017-11-20 CN CN201711160588.7A patent/CN107766580A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104917896A (en) * | 2015-06-12 | 2015-09-16 | 努比亚技术有限公司 | Data pushing method and terminal equipment |
CN105096158A (en) * | 2015-07-01 | 2015-11-25 | 北京奇虎科技有限公司 | Method and device for pushing information |
CN105701191A (en) * | 2016-01-08 | 2016-06-22 | 腾讯科技(深圳)有限公司 | Push information click rate estimation method and device |
CN105915438A (en) * | 2016-04-15 | 2016-08-31 | 北京奇虎科技有限公司 | Message pushing method, apparatus, and system |
CN107613022A (en) * | 2017-10-20 | 2018-01-19 | 广州优视网络科技有限公司 | Content delivery method, device and computer equipment |
Cited By (43)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109218390A (en) * | 2018-07-12 | 2019-01-15 | 北京比特智学科技有限公司 | User's screening technique and device |
CN109218390B (en) * | 2018-07-12 | 2021-09-10 | 北京比特智学科技有限公司 | User screening method and device |
CN108920717A (en) * | 2018-07-27 | 2018-11-30 | 百度在线网络技术(北京)有限公司 | For showing the method and device of information |
CN109167816A (en) * | 2018-08-03 | 2019-01-08 | 广州虎牙信息科技有限公司 | Information-pushing method, device, equipment and storage medium |
CN109241408A (en) * | 2018-08-14 | 2019-01-18 | 深圳开黑科技有限公司 | User's portrait processing method, equipment and the system for playing platform are accompanied based on game |
CN109218411B (en) * | 2018-08-22 | 2022-10-18 | 中国平安人寿保险股份有限公司 | Data processing method and device, computer readable storage medium and electronic equipment |
CN109218411A (en) * | 2018-08-22 | 2019-01-15 | 中国平安人寿保险股份有限公司 | Data processing method and device, computer readable storage medium, electronic equipment |
CN110875949A (en) * | 2018-09-04 | 2020-03-10 | 京东数字科技控股有限公司 | Method and device for pushing information |
CN109166012A (en) * | 2018-09-21 | 2019-01-08 | 苏州创旅天下信息技术有限公司 | The method and apparatus of classification and information push for stroke predetermined class user |
CN109166012B (en) * | 2018-09-21 | 2021-05-28 | 苏州创旅天下信息技术有限公司 | Method and device for classifying users in travel reservation class and pushing information |
CN111061979B (en) * | 2018-10-17 | 2023-04-21 | 北京字节跳动网络技术有限公司 | User tag pushing method and device, electronic equipment and medium |
CN111061979A (en) * | 2018-10-17 | 2020-04-24 | 北京字节跳动网络技术有限公司 | User label pushing method and device, electronic equipment and medium |
CN111292107A (en) * | 2018-12-06 | 2020-06-16 | 北京嘀嘀无限科技发展有限公司 | User data processing method and device, electronic equipment and storage medium |
CN109684554A (en) * | 2018-12-26 | 2019-04-26 | 腾讯科技(深圳)有限公司 | The determination method and news push method of the potential user of news |
CN109741113A (en) * | 2019-01-10 | 2019-05-10 | 博拉网络股份有限公司 | A kind of user's purchase intention prediction technique based on big data |
CN109726322A (en) * | 2019-01-28 | 2019-05-07 | 北京百度网讯科技有限公司 | Method and apparatus for pushed information |
CN110399550B (en) * | 2019-03-14 | 2023-08-15 | 腾讯科技(深圳)有限公司 | Information recommendation method and device |
CN110399550A (en) * | 2019-03-14 | 2019-11-01 | 腾讯科技(深圳)有限公司 | A kind of information recommendation method and device |
CN110135942A (en) * | 2019-04-12 | 2019-08-16 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus and computer readable storage medium |
CN110532467A (en) * | 2019-08-22 | 2019-12-03 | 上海易点时空网络有限公司 | Activity recommendation method and device, equipment, storage medium based on push model |
CN110532467B (en) * | 2019-08-22 | 2022-06-07 | 上海易点时空网络有限公司 | Activity recommendation method and device based on push model, equipment and storage medium |
CN110782291A (en) * | 2019-10-29 | 2020-02-11 | 加和(北京)信息科技有限公司 | Advertisement delivery user determination method and device, storage medium and electronic device |
CN111159561A (en) * | 2019-12-31 | 2020-05-15 | 青梧桐有限责任公司 | Method for constructing recommendation engine according to user behaviors and user portrait |
CN111090815A (en) * | 2019-12-31 | 2020-05-01 | 恩亿科(北京)数据科技有限公司 | Label generation method and device |
CN111860878A (en) * | 2019-12-31 | 2020-10-30 | 北京骑胜科技有限公司 | Repair reporting method for shared bicycle, server, electronic device and storage medium |
CN111754287A (en) * | 2020-02-14 | 2020-10-09 | 北京沃东天骏信息技术有限公司 | Article screening method, apparatus, device and storage medium |
CN111612366A (en) * | 2020-05-27 | 2020-09-01 | 中国联合网络通信集团有限公司 | Channel quality evaluation method and device, electronic equipment and storage medium |
CN111681050A (en) * | 2020-06-05 | 2020-09-18 | 深圳市卡牛科技有限公司 | Advertisement pushing method, device, equipment and storage medium |
CN111681050B (en) * | 2020-06-05 | 2023-08-29 | 深圳市卡牛科技有限公司 | Advertisement pushing method, device, equipment and storage medium |
CN112015975A (en) * | 2020-07-15 | 2020-12-01 | 北京淇瑀信息科技有限公司 | Financial user-oriented information pushing method and device based on Newton's cooling law |
CN112015975B (en) * | 2020-07-15 | 2023-11-14 | 北京淇瑀信息科技有限公司 | Information pushing method and device for financial users based on Newton's law of cooling |
CN112148994A (en) * | 2020-10-23 | 2020-12-29 | 腾讯科技(深圳)有限公司 | Information push effect evaluation method and device, electronic equipment and storage medium |
CN112148994B (en) * | 2020-10-23 | 2024-04-19 | 腾讯科技(深圳)有限公司 | Information push effect evaluation method and device, electronic equipment and storage medium |
CN112532748A (en) * | 2020-12-24 | 2021-03-19 | 北京百度网讯科技有限公司 | Message pushing method, device, equipment, medium and computer program product |
CN113110942A (en) * | 2021-03-26 | 2021-07-13 | 支付宝(杭州)信息技术有限公司 | Message distribution method and device, computer equipment and storage medium |
CN113761003A (en) * | 2021-04-15 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | User portrait data processing method and device, electronic equipment and readable storage medium |
CN113360768A (en) * | 2021-06-29 | 2021-09-07 | 平安信托有限责任公司 | Product recommendation method, device and equipment based on user portrait and storage medium |
CN113742571B (en) * | 2021-08-03 | 2024-04-26 | 大箴(杭州)科技有限公司 | Message pushing method and device based on big data and storage medium |
CN113742571A (en) * | 2021-08-03 | 2021-12-03 | 大箴(杭州)科技有限公司 | Message pushing method and device based on big data and storage medium |
CN114820080A (en) * | 2022-05-24 | 2022-07-29 | 上海钧正网络科技有限公司 | User grouping method, system, device and medium based on crowd circulation |
CN114707097A (en) * | 2022-05-31 | 2022-07-05 | 每日互动股份有限公司 | Data processing system for acquiring target message flow |
CN116434601B (en) * | 2023-04-25 | 2023-12-29 | 广州市单元信息科技有限公司 | Intelligent processing method and system for message pushing |
CN116434601A (en) * | 2023-04-25 | 2023-07-14 | 广州市单元信息科技有限公司 | Intelligent processing method and system for message pushing |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107766580A (en) | The method for pushing and device of message | |
CN110399550B (en) | Information recommendation method and device | |
US20210342745A1 (en) | Artificial intelligence model and data collection/development platform | |
CN109343857B (en) | Method, apparatus and storage medium for deploying machine learning model on line | |
CN111898023A (en) | Message pushing method and device, readable storage medium and computing equipment | |
CN107704396A (en) | The method of testing and device of application program | |
CN110413867B (en) | Method and system for content recommendation | |
CN110688598A (en) | Service parameter acquisition method and device, computer equipment and storage medium | |
CN109189935B (en) | APP propagation analysis method and system based on knowledge graph | |
CN113360800A (en) | Method and device for processing featureless data, computer equipment and storage medium | |
CN111177541B (en) | Data analysis method and device based on user tag generation time | |
CN115860836A (en) | E-commerce service pushing method and system based on user behavior big data analysis | |
CN113297287B (en) | Automatic user policy deployment method and device and electronic equipment | |
CN114091684A (en) | Method and device for enhancing interpretability of service result | |
CN116756576B (en) | Data processing method, model training method, electronic device and storage medium | |
CN111062736A (en) | Model training and clue sequencing method, device and equipment | |
CN113569162A (en) | Data processing method, device, equipment and storage medium | |
CN110209958B (en) | Thematic making method, device, terminal and storage medium | |
CN110275880B (en) | Data analysis method, device, server and readable storage medium | |
CN115563069B (en) | Data sharing processing method and system based on artificial intelligence and cloud platform | |
CN114944942B (en) | Data reporting method and device based on effective buried points, computer equipment and medium | |
US20220035992A1 (en) | Ai driven content correction built on personas | |
CN109902831B (en) | Service decision processing method and device | |
CN114203304A (en) | Information pushing method based on smart medical big data and smart medical cloud server | |
CN111522747A (en) | Application processing method, device, equipment and medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180306 |