CN111353825A - Message transmission method and device - Google Patents

Message transmission method and device Download PDF

Info

Publication number
CN111353825A
CN111353825A CN202010161611.XA CN202010161611A CN111353825A CN 111353825 A CN111353825 A CN 111353825A CN 202010161611 A CN202010161611 A CN 202010161611A CN 111353825 A CN111353825 A CN 111353825A
Authority
CN
China
Prior art keywords
user
behavior
preset
probability
target object
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010161611.XA
Other languages
Chinese (zh)
Other versions
CN111353825B (en
Inventor
周江
向柳
卢剑锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Aibee Technology Co Ltd
Original Assignee
Beijing Aibee Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Aibee Technology Co Ltd filed Critical Beijing Aibee Technology Co Ltd
Priority to CN202010161611.XA priority Critical patent/CN111353825B/en
Publication of CN111353825A publication Critical patent/CN111353825A/en
Application granted granted Critical
Publication of CN111353825B publication Critical patent/CN111353825B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0202—Market predictions or forecasting for commercial activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Game Theory and Decision Science (AREA)
  • Computing Systems (AREA)
  • General Business, Economics & Management (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Marketing (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Probability & Statistics with Applications (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a message transmission method and a device, which can obtain user identity information and behavior information of a user aiming at various behaviors of a target object; obtaining a user characteristic vector according to the user identity information and the behavior information; predicting the probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current moment according to the user feature vector; predicting behavior data of a user performing a second preset type of behavior on the target object within a second preset time period after the current time according to the user feature vector, wherein the second preset time period is the same as or different from the first preset time period; screening the users according to the predicted probability and the predicted behavior data; and sending a preset message to at least one screened user. The invention sends the message to the screened user in a targeted manner, thereby avoiding the problem of large operation burden of the service system caused by pushing any message to all users.

Description

Message transmission method and device
Technical Field
The present invention relates to the field of information technologies, and in particular, to a method and an apparatus for transmitting a message.
Background
With the continuous development of information technology, service providers can push various service messages to users, so that the users can know and use the services provided by the service providers in time. For example: the video service provider may push a message of a certain thermoprint update to the user so that the user may view the updated content of the thermoprint after obtaining the message.
At present, for a certain service message, a service provider can push the service message to all users, and when the number of users is large, the technical problem that the service system is crashed due to too large operation burden of the service system of the service provider is easily caused when the service message is pushed to all users, so that the normal experience of the users is influenced.
Disclosure of Invention
In view of the above problems, the present invention provides a message transmission method and apparatus for overcoming the above problems or at least partially solving the above problems, and the technical solution is as follows:
a method of message transmission, comprising:
acquiring user identity information and behavior information of a plurality of behaviors of a user aiming at a target object;
obtaining a user characteristic vector according to the user identity information and the behavior information;
predicting the probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current moment according to the user feature vector;
predicting behavior data of a user performing a second preset type of behavior on the target object within a second preset time period after the current time according to the user feature vector, wherein the second preset time period is the same as or different from the first preset time period;
screening users according to the predicted probability and the predicted behavior data;
and sending a preset message to at least one screened user.
Optionally, the obtaining of the user identity information and the behavior information of the user for multiple behaviors of the target object includes:
the method comprises the steps of obtaining user identity information of a target user and behavior information of the target user for multiple behaviors of the target object, wherein the target user is a user who does not conduct a third preset type of behavior for the target object within a third preset time period before the current moment, and the third preset type is the same as the second preset type, or the third preset type is the same as the first preset type.
Optionally, the method further includes:
and predicting the probability of the second preset type of behavior aiming at the target object after the user receives the preset message according to the user feature vector.
Optionally, predicting, according to the user feature vector, a probability that the user does not perform a first preset type of behavior for the target object within a first preset time period after the current time includes:
and inputting the user feature vector into a pre-trained first behavior probability prediction model, and obtaining the probability that the user does not perform a first preset type of behavior for the target object within a first preset time period after the current time, which is predicted by the first behavior probability prediction model.
Optionally, the predicting, according to the user feature vector, behavior data of a second preset type of behavior performed by the user for the target object in a second preset time period after the current time includes:
and inputting the user characteristic vector into a pre-trained behavior data prediction model, and obtaining behavior data of a second preset type of behavior of the user predicted by the data prediction model for the target object in a second preset time period after the current time.
Optionally, the predicting, according to the user feature vector, the probability that the user performs the second preset type of behavior on the target object after receiving the preset message includes:
and inputting the user feature vector into a second pre-trained behavior prediction probability model, and obtaining the probability of the second preset type of behavior for the target object after the user predicted by the second behavior prediction probability model receives the preset message.
Optionally, the predicting, according to the user feature vector, the probability that the user performs the second preset type of behavior on the target object after receiving the preset message includes:
inputting the user characteristic vector into a pre-trained message click prediction model, and obtaining the click probability of clicking after a user predicted by the message click prediction model receives the preset message;
inputting the user characteristic vector into a pre-trained message feedback prediction model to obtain the feedback probability of the user predicted by the message feedback prediction model for feedback after receiving the preset message;
and determining the probability of the second preset type of behavior aiming at the target object after the user receives the preset message according to the click probability and the feedback probability.
A message transmission apparatus, comprising: an information obtaining unit, a user characteristic vector obtaining unit, a first probability obtaining unit, a behavior data obtaining unit, a user screening unit and a message sending unit,
the information obtaining unit is used for obtaining user identity information and behavior information of a plurality of behaviors of the user aiming at the target object;
the user characteristic vector obtaining unit is used for obtaining a user characteristic vector according to the user identity information and the behavior information;
the first probability obtaining unit is used for predicting the probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current time according to the user feature vector;
the behavior data obtaining unit is used for predicting behavior data of a second preset type of behavior of the user aiming at the target object in a second preset time period after the current time according to the user feature vector, wherein the second preset time period is the same as or different from the first preset time period;
the user screening unit is used for screening users according to the predicted probability and the predicted behavior data;
and the message sending unit is used for sending a preset message to at least one screened user.
Optionally, the information obtaining unit is specifically configured to obtain user identity information of a target user and behavior information of multiple behaviors of the target user for a target object, where the target user is a user who does not perform a third preset type of behavior for the target object within a third preset time period before a current time, where the third preset type is the same as the second preset type, or the third preset type is the same as the first preset type.
Optionally, the apparatus may further include: a second probability obtaining unit for obtaining a second probability,
the second probability obtaining unit is configured to predict, according to the user feature vector, a probability that the user performs the second preset type of behavior on the target object after receiving the preset message.
By means of the technical scheme, the message transmission method and the message transmission device can obtain the user identity information and the behavior information of the user aiming at various behaviors of the target object; obtaining a user characteristic vector according to the user identity information and the behavior information; predicting the probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current moment according to the user feature vector; predicting behavior data of a user performing a second preset type of behavior on the target object within a second preset time period after the current time according to the user feature vector, wherein the second preset time period is the same as or different from the first preset time period; screening users according to the predicted probability and the predicted behavior data; and sending a preset message to at least one screened user. The invention sends the message to the screened user in a targeted manner, thereby avoiding the problem of large operation burden of the service system caused by pushing any message to all users.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a flowchart illustrating a message transmission method according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating another message transmission method according to an embodiment of the present invention;
fig. 3 is a flowchart illustrating another message transmission method according to an embodiment of the present invention;
fig. 4 is a flowchart illustrating another message transmission method according to an embodiment of the present invention;
fig. 5 is a flowchart illustrating another message transmission method according to an embodiment of the present invention;
fig. 6 is a flowchart illustrating another message transmission method according to an embodiment of the present invention;
fig. 7 is a flowchart illustrating another message transmission method according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram illustrating a message transmission apparatus according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
As shown in fig. 1, a message transmission method provided in an embodiment of the present invention may include:
s100, obtaining user identity information and behavior information of a plurality of behaviors of the user aiming at the target object.
The user identity information may include morphological information, physiological information, and other information of the user. For example: the user identity information may include information such as the user's age, sex, height, weight, blood pressure, and body temperature. The target object may be a service provided by a service provider. For example: video, music, and novels. The behavior information may include a series of operations performed by the user with respect to the target object. For example: click, play, subscribe, screen capture, download, etc.
It can be understood that, according to different application scenarios in the embodiments of the present invention, information contained in the user identity information, the target object, and the behavior information may be different. For example: in an optional application scenario in the embodiment of the present invention, the target object may be a certain store, and the behavior information of the user for multiple behaviors of the target object may include: how many days the user has traded in the store for the first time, how many days the user has traded in the store for the last time, the total number of orders the user has made in the store, etc.
The user identity information, the target object and the behavior information in the embodiment of the present invention may be determined according to the actual needs of the service provider, and the embodiment of the present invention is not further limited herein.
S200, obtaining a user characteristic vector according to the user identity information and the behavior information.
Specifically, the embodiment of the invention can use the user identity information and the behavior information as the user characteristic vector. The embodiment of the invention can sort the user identity information and the behavior information according to a specific sequence to obtain the user characteristic vector. For example, when the user identity information includes age and gender and the behavior information includes play and subscription, the user feature vector may be "gender, age, subscription, play".
S300, predicting the probability that the user does not perform the first preset type of behavior on the target object within a first preset time period after the current time according to the user feature vector.
The first preset time period may be set according to the requirement of the service provider. The first preset type of behavior may be set according to the needs of the service provider. For example: if the video service provider needs to predict the probability that the user does not play a tv series within one week after the current time, the embodiment of the present invention may set the first preset time to seven days, and set the first preset type to play. The first preset type of behavior may be a behavior that a user uses a service provided by a service provider, for example: the user uses a video service provided by a video service provider, the video service comprising: video information browsing, video playing, video collecting, video downloading, video uploading and the like. If the user does not use the service provided by the service provider within a first preset time period after the current time, for the service provider, the user has run away within the first preset time period after the current time, and therefore, the probability that the user does not perform the first preset type of behavior on the target object within the first preset time period after the current time may be understood as: the user churn probability.
Optionally, based on the method shown in fig. 1, as shown in fig. 2, in another message transmission method provided in the embodiment of the present invention, step S300 may include:
s310, inputting the user feature vector into a pre-trained first behavior probability prediction model, and obtaining the probability that the first behavior probability prediction model predicts that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current time.
Wherein the first behavior probability prediction model may be a Deep Neural Network (DNN) model.
The training process of the first behavior probability prediction model in the embodiment of the present invention may include:
obtaining a user characteristic training vector marked with a behavior identifier, wherein the behavior identifier comprises a first preset type of behavior which is not performed by a user for a target object within a first preset time period, or a first preset type of behavior which is performed by the user for the target object within the first preset time period;
and performing machine learning on the user feature training vector to obtain a first behavior probability prediction model, wherein the input of the first behavior probability prediction model is the user feature vector, and the output of the first behavior probability prediction model is the probability that the user does not perform a first preset type of behavior on the target object within a first preset time period. It is understood that the probability is a predictive value.
It should be noted that, in an actual situation, since there may be a difference between the user feature vector and the user feature training vector, it is difficult for the trained first behavior probability prediction model to directly determine whether the user has not performed the first preset type of behavior with respect to the target object within the first preset time period according to the user feature vector, and the probability that the user has not performed the first preset type of behavior with respect to the target object within the first preset time period may be predicted. For example: according to the embodiment of the invention, the behavior of the user, which is not of the first preset type, aiming at the target object in the first preset time period is taken as 1, the behavior of the user, which is of the first preset type, aiming at the target object in the first preset time period is taken as 0, and the probability that the first behavior probability prediction model predicts that the user does not conduct the behavior of the first preset type aiming at the target object in the first preset time period after the current moment is possibly between 0 and 1 because the training vector of the user and the feature vector of the user possibly have difference. For example: the probability that the user corresponding to the user feature vector does not perform the first preset type of behavior on the target object within the first preset time period is 0.73, and the probability that the user corresponding to the user feature vector performs the first preset type of behavior on the target object within the first preset time period is 0.27. It can be understood that, for the same user feature vector, the sum of the probability that the user corresponding to the user feature vector does not perform the first preset type of behavior on the target object within the first preset time period and the probability that the user performs the first preset type of behavior on the target object within the first preset time period is 1.
Optionally, a first threshold may be set in the embodiment of the present invention, and when it is predicted that the probability that the user performs the first preset type of behavior for the target object within a first preset time period after the current time is greater than the first threshold according to the user feature vector, it may be determined that the user performs the first preset type of behavior for the target object within the first preset time period after the current time, and conversely, it may be determined that the user performs the first preset type of behavior for the target object within the first preset time period after the current time.
S400, predicting behavior data of a user performing a second preset type of behavior on the target object in a second preset time period after the current time according to the user feature vector, wherein the second preset time period is the same as or different from the first preset time period.
The second preset time period may be set according to the requirement of the service provider. Generally, the first preset time period is the same as the second preset time period. The second preset type of behavior may be set according to the needs of the service provider. For example: if the video service provider needs to predict the number of times that the user downloads a certain tv show within one week after the current time, the embodiment of the present invention may set the second preset time to seven days, and set the second preset type to download. The second preset type of behavior may be a behavior of the user using a certain service provided by the service provider, and the service may be a service that the service provider most desires to use by the user. For example: video services provided for video service providers include: video information browsing, video playing, video collecting, video downloading, video uploading and the like. The video service provider most expects the service used by the user to be video playing, and the second preset type of behavior may be a behavior of the user using the video playing service provided by the video service provider, such as: the video is viewed on a video website. Since the user uses the service which the service provider most expects the user to use, the user can be understood as having a certain value to the service provider, and the behavior data of the user performing the second preset type of behavior for the target object in the second preset time period after the current time can be understood as: the value of the user.
Optionally, based on the method shown in fig. 1, as shown in fig. 3, in another message transmission method provided in the embodiment of the present invention, step S400 may include:
s410, inputting the user feature vector into a pre-trained behavior data prediction model, and obtaining behavior data of a second preset type of behavior of the user aiming at the target object in a second preset time period after the current time predicted by the data prediction model.
The behavior data prediction model may be a Deep Neural Network (DNN) model.
The training process of the behavior data prediction model in the embodiment of the invention may include:
obtaining a user characteristic training vector marked with behavior data, wherein the behavior data is data of a second preset type of behavior of a user aiming at a target object in a second preset time period;
and performing machine learning on the user characteristic training vector to obtain a behavior data prediction model, wherein the input of the behavior data prediction model is the user characteristic vector, and the output of the behavior data prediction model is the behavior data of a second preset type of behavior of the user on the target object in a second preset time period.
It should be noted that, in an actual situation, since there may be a difference between the user feature vector and the user feature training vector, it is difficult for the trained behavior data prediction model to directly determine, according to the user feature vector, the behavior data of the user performing the second preset type of behavior on the target object within the second preset time period, and the behavior data of the user performing the second preset type of behavior on the target object within the second preset time period may be predicted. For example: the behavior data corresponding to the user feature training vector with the highest similarity to the user feature vector for training the behavior data prediction model is 5.3, and the behavior data predicted by the behavior data prediction model on the user feature vector may be 5.38.
Optionally, according to different application scenarios, the embodiment of the present invention may reserve integers for the behavior data according to a preset rule. The preset rule may be to round the data after the decimal point. For example: when the behavior data prediction model predicts that the behavior data of the video downloaded by the user in the second preset time period after the current time is 5.78, the behavior data prediction model may output that the behavior data of the video downloaded by the user in the second preset time period after the current time is 6.
It should be understood that fig. 1 shows only an alternative execution sequence of step S300 and step S400, and step S400 may be executed before step S300 or simultaneously with step S300, and the execution sequence of step S300 and step S400 is not limited herein in this embodiment of the present invention.
S500, screening the users according to the predicted probability and the predicted behavior data.
Specifically, the embodiment of the present invention may screen out the users whose predicted probability meets the preset probability condition and whose predicted behavior data meets the preset behavior data condition. The preset probability condition may be that the predicted probability is greater than a preset probability threshold. The preset behavior data condition may be that the predicted behavior data is greater than a preset behavior data threshold. The preset probability threshold and the preset behavior data threshold can be set according to the needs of the service provider. For example: when the preset probability threshold is 0.5 and the preset behavior data threshold is 7, the embodiment of the invention can screen out the users of which the predicted probability is greater than 0.5 and the predicted behavior data is greater than 7.
S600, sending a preset message to at least one screened user.
Wherein the preset message may be content that can be presented on the user mobile device. Specifically, the preset message may be content related to the target object. For example: television drama update, medicine getting reminding, member overdue reminding and the like.
The message transmission method provided by the embodiment of the invention can obtain the user identity information and the behavior information of the user aiming at various behaviors of the target object; obtaining a user characteristic vector according to the user identity information and the behavior information; predicting the probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current moment according to the user feature vector; predicting behavior data of a user performing a second preset type of behavior on the target object within a second preset time period after the current time according to the user feature vector, wherein the second preset time period is the same as or different from the first preset time period; screening users according to the predicted probability and the predicted behavior data; and sending a preset message to at least one screened user. The embodiment of the invention enables users with requirements to obtain the message by sending the message to the screened users in a targeted manner, thereby avoiding the problem of large operation burden of a service system caused by pushing any message to all users.
It can be understood that, in the message transmission method provided in the embodiment of the present invention, the user is screened according to the user identity information and the behavior information for the target object, and the preset message is sent to the screened user, so that the user who really has a demand for the target object can obtain the preset message, thereby avoiding message harassment caused by the user who has no demand for the target object or has a weak demand, and further improving the user experience. In order to intuitively understand the beneficial effects of the message transmission method provided by the embodiment of the present invention, an optional scenario of the embodiment of the present invention is described herein: aiming at a certain TV play which updates a set every week, the embodiment of the invention can screen out the user interested in the TV play to carry out message pushing aiming at the behavior information of whether the user subscribes to the TV play, clicks to play, completely watches the feature content, watches the highlights of the TV play and the like, thereby saving the resources of a service system and avoiding message harassment to other users while realizing accurate message pushing.
In an optional application scenario of the embodiment of the present invention, a marketplace may use the probability predicted in step S300 as a user loss probability, evaluate the user value by using the behavior data predicted in step S400, and further screen out users worth sending coupons according to the user loss probability and the user value, and send coupons to the users.
Optionally, based on the method shown in fig. 1, as shown in fig. 4, in another message transmission method provided in the embodiment of the present invention, step S100 may include:
s110, obtaining user identity information of a target user and behavior information of the target user for multiple behaviors of the target user on a target object, wherein the target user is a user who does not perform a third preset type of behavior on the target object within a third preset time period before the current moment, and the third preset type is the same as the second preset type, or the third preset type is the same as the first preset type.
It can be understood that, in order to make the execution results of steps S300 to S400 more accurate, it is necessary to use, as the user feature vector, a behavior of the target user that is not of the third preset type for the target object within a third preset time period before the current time. For ease of understanding, the description is made herein by way of example: if it is assumed that the embodiment of the present invention sets the third preset time period to 14 days and the third preset type to play according to the service provider requirement, then the embodiment of the present invention may obtain behavior information that the user does not perform a play behavior for the tv series within 14 days, and use the behavior information as a user feature vector. According to the embodiment of the invention, the target user does not perform the third preset type of behavior aiming at the target object in the third preset time period before the current moment as the user target characteristic vector, so that the execution results from the subsequent step S300 to the step S400 are more accurate.
Further, in order to prevent a large number of users from performing a second preset type of behavior on a target object after receiving a preset message to cause a sudden increase in access flow of a short-time service system in a short time, which may cause a service system crash, an embodiment of the present invention may further provide another message transmission method, as shown in fig. 5, where the method may further include:
s700, predicting the probability of the second preset type of behavior aiming at the target object after the user receives the preset message according to the user feature vector.
Specifically, as shown in fig. 6, in another message transmission method provided in the embodiment of the present invention, step S700 may include:
and S710, inputting the user feature vector into a second pre-trained behavior prediction probability model, and obtaining the probability that the second behavior prediction probability model predicts the behavior of the second preset type aiming at the target object after the user receives the preset message.
Wherein the second behavior prediction probability model may be a Click-Through-Rate (CTR) model based on a deep learning algorithm. The Deep learning algorithm may be a gradient boosting iterative decision tree (GBDT) algorithm, a Factorization Machine (FM) algorithm, a generalized Linear and Deep neural Network (Wide & Deep) algorithm, and the like.
The embodiment of the invention can obtain the second behavior prediction probability model by combining whether the user performs the second preset type of behaviors on the target object after receiving the message and the corresponding user characteristic training vector for machine learning. It should be noted that, in an actual situation, since there may be a difference between the user feature vector and the user feature training vector, it is difficult for the trained second behavior prediction probability model to directly determine whether the user performs the second preset type of behavior with respect to the target object after receiving the preset message, and the probability that the user performs the second preset type of behavior with respect to the target object after receiving the preset message may be predicted. For example: the embodiment of the invention can predict the probability of whether the user plays the TV play after receiving the preset message of the TV play update.
Optionally, as shown in fig. 7, in another message transmission method provided in the embodiment of the present invention, step S700 may include:
s720, inputting the user characteristic vector into a pre-trained message click prediction model, and obtaining the click probability of the message click prediction model for predicting the click of the user after receiving the preset message.
Wherein, the message Click prediction model can be a Click-Through-Rate (CTR) model based on a deep learning algorithm. The Deep learning algorithm may be a gradient boosting iterative Decision Tree (GBDT) algorithm, a Factorization Machine (FM) algorithm, a generalized Linear and Deep neural Network (Wide & Deep) algorithm, and the like.
The embodiment of the invention can obtain the message click prediction model by combining the fact that whether the user clicks after receiving the preset message and the corresponding user characteristic vector to perform machine learning. It should be noted that, in an actual situation, because there may be a difference between the user feature vector and the user feature training vector, it is difficult for the trained message click prediction model to directly determine whether the user clicks after receiving the preset message, and the click probability of clicking after receiving the preset message can be predicted.
And S730, inputting the user characteristic vector into a pre-trained message feedback prediction model, and obtaining the feedback probability of the message feedback prediction model for predicting that the user performs feedback after receiving the preset message.
The message feedback prediction model may be a Click-Through-Rate (CTR) model based on a deep learning algorithm. The Deep learning algorithm may be a gradient boosting iterative Decision Tree (GBDT) algorithm, a Factorization Machine (FM) algorithm, a generalized Linear and Deep neural Network (Wide & Deep) algorithm, and the like.
The embodiment of the invention can obtain the message feedback prediction model by combining the fact that whether the user performs feedback after receiving the preset message and the corresponding user characteristic training vector to perform machine learning. It should be noted that, in an actual situation, because there may be a difference between the user feature vector and the user feature training vector, it is difficult for the trained message feedback prediction model to directly determine whether the user performs feedback after receiving the preset message, and a feedback probability that the user performs feedback after receiving the preset message may be predicted.
And S740, determining the probability of the second preset type of behavior aiming at the target object after the user receives the preset message according to the click probability and the feedback probability.
Specifically, the click probability and the feedback probability may be multiplied to obtain a product, and the product is used as the probability of the second preset type of behavior performed on the target object after the user receives the preset message.
For ease of understanding, the description is made herein by way of example: the method and the device can obtain the probability that the user clicks the preset message after receiving the preset message updated by the TV play and the probability that the user feeds back after receiving the preset message, wherein the feedback can be that the user returns to the TV play page, and then the probability that the user plays the TV play is calculated according to the clicking probability and the feedback probability.
It can be understood that, because the invention can selectively send messages to the screened users, it can not only reduce the burden of the service system, but also bring profits to the service provider, and the following takes the example of the distribution of the coupons in the mall to calculate how the coupons are distributed, so as to obtain the maximum profits.
In an optional application scenario of the embodiment of the present invention, the marketplace may use the click probability in step S720 as a coupon rate for the user to receive the coupon after sending the coupon to the user, use the feedback probability in step S730 as a verification rate for the user to use the coupon after sending the coupon to the user, and finally determine the probability that the user consumes in the marketplace through the coupon rate and the verification rate.
According to the embodiment of the invention, the probability of the second preset type of behavior aiming at the target object after the user receives the preset message is predicted, so that the access flow of the service system can be estimated in advance, the service system is upgraded and modified in advance, and the service system is prevented from being crashed.
When the embodiment of the invention is applied to a scene of sending the coupons in a shopping mall, the preference degree of the user to different types of coupons and the income of the user from the coupons to the shopping mall are considered to be maximized in the limited various types of coupons. The embodiment of the invention can realize the optimal distribution of the coupons by the following formula:
Figure BDA0002405997850000131
the constraint conditions are as follows:
Figure BDA0002405997850000132
i∈I。
wherein,
Figure BDA0002405997850000133
is a screened user group, wherein the user group comprises the number u of users, I is the total number of coupons, I is one type of coupon, CiThe total number of coupons of one type,
Figure BDA0002405997850000134
to determine whether to send i-type coupons to user u,
Figure BDA0002405997850000135
for sending a benefit of i-type coupon to user u, wherein
Figure BDA0002405997850000136
The behavior data of the user performing the second preset type of behavior for the target object in the second preset time period after the current time may be predicted in step S400, or the probability of performing the second preset type of behavior for the target object after the user receives the preset message may be predicted in step S700.
Corresponding to the foregoing method embodiment, an embodiment of the present invention further provides a message transmission apparatus, whose structure is shown in fig. 8, and may include: the information obtaining unit 100, the user feature vector obtaining unit 200, the first probability obtaining unit 300, the behavior data obtaining unit 400, the user screening unit 500, and the message sending unit 600.
The information obtaining unit 100 is configured to obtain user identity information and behavior information of a plurality of behaviors of the user with respect to the target object.
The user identity information may include morphological information, physiological information, and other information of the user. For example: the user identity information may include information such as the user's age, sex, height, weight, blood pressure, and body temperature. The target object may be a service provided by a service provider. For example: video, music, and novels. The behavior information may include a series of operations performed by the user with respect to the target object. For example: click, play, subscribe, screen capture, download, etc.
The user feature vector obtaining unit 200 is configured to obtain a user feature vector according to the user identity information and the behavior information.
Specifically, the embodiment of the invention can use the user identity information and the behavior information as the user characteristic vector. The embodiment of the invention can sort the user identity information and the behavior information according to a specific sequence to obtain the user characteristic vector.
The first probability obtaining unit 300 is configured to predict, according to the user feature vector, a probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current time.
The first preset time period may be set according to the requirement of the service provider. The first preset type of behavior may be set according to the needs of the service provider. The first preset type of behavior may be a behavior in which a user uses a service provided by a service provider.
Optionally, the first probability obtaining unit 300 is specifically configured to input the user feature vector into a first behavior probability prediction model trained in advance, and obtain a probability that a user predicted by the first behavior probability prediction model does not perform a first preset type of behavior on the target object within a first preset time period after the current time.
Wherein the first behavior probability prediction model may be a Deep Neural Network (DNN) model.
Optionally, a first threshold may be set in the embodiment of the present invention, and when it is predicted that the probability that the user performs the first preset type of behavior for the target object within a first preset time period after the current time is greater than the first threshold according to the user feature vector, it may be determined that the user performs the first preset type of behavior for the target object within the first preset time period after the current time, and conversely, it may be determined that the user performs the first preset type of behavior for the target object within the first preset time period after the current time.
The behavior data obtaining unit 400 is configured to predict, according to the user feature vector, behavior data of a second preset type of behavior performed by the user for the target object within a second preset time period after the current time, where the second preset time period is the same as or different from the first preset time period.
The second preset time period may be set according to the requirement of the service provider. Generally, the first preset time period is the same as the second preset time period. The second preset type of behavior may be set according to the needs of the service provider.
The second preset type of behavior may be a behavior of the user using a certain service provided by the service provider, and the service may be a service that the service provider most desires to use by the user.
Optionally, the behavior data obtaining unit 400 is specifically configured to input the user feature vector into a pre-trained behavior data prediction model, and obtain behavior data of a second preset type of behavior performed by the user in a second preset time period after the current time by using the data prediction model, for the target object.
The behavior data prediction model may be a Deep Neural Network (DNN) model.
The user screening unit 500 is configured to screen users according to the predicted probability and the predicted behavior data.
Specifically, the embodiment of the present invention may screen out the users whose predicted probability meets the preset probability condition and whose predicted behavior data meets the preset behavior data condition. The preset probability condition may be that the predicted probability is greater than a preset probability threshold. The preset behavior data condition may be that the predicted behavior data is greater than a preset behavior data threshold. The preset probability threshold and the preset behavior data threshold can be set according to the needs of the service provider.
The message sending unit 600 is configured to send a preset message to at least one screened user.
Wherein the preset message may be content that can be presented on the user mobile device. Specifically, the preset message may be content related to the target object.
The message transmission device provided by the embodiment of the invention can obtain the identity information of a user and the behavior information of the user aiming at various behaviors of a target object; obtaining a user characteristic vector according to the user identity information and the behavior information; predicting the probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current moment according to the user feature vector; predicting behavior data of a user performing a second preset type of behavior on the target object within a second preset time period after the current time according to the user feature vector, wherein the second preset time period is the same as or different from the first preset time period; screening users according to the predicted probability and the predicted behavior data; and sending a preset message to at least one screened user. The embodiment of the invention enables users with requirements to obtain the message by sending the message to the screened users in a targeted manner, thereby avoiding the problem of large operation burden of a service system caused by pushing any message to all users.
Optionally, the information obtaining unit 100 is specifically configured to obtain user identity information of a target user and behavior information of multiple behaviors of the target user for a target object, where the target user is a user who does not perform a third preset type of behavior for the target object within a third preset time period before the current time, where the third preset type is the same as the second preset type, or the third preset type is the same as the first preset type.
Optionally, the message transmission apparatus provided in the embodiment of the present invention may further include: a second probability obtaining unit.
The second probability obtaining unit is configured to predict, according to the user feature vector, a probability that the user performs the second preset type of behavior on the target object after receiving the preset message.
Optionally, the second probability obtaining unit is specifically configured to input the user feature vector into a second behavior prediction probability model trained in advance, and obtain a probability that the user predicted by the second behavior prediction probability model performs the second preset type of behavior on the target object after receiving the preset message.
Wherein the second behavior prediction probability model may be a Click-Through-Rate (CTR) model based on a deep learning algorithm. The Deep learning algorithm may be a gradient boosting iterative decision tree (GBDT) algorithm, a Factorization Machine (FM) algorithm, a generalized Linear and Deep neural Network (Wide & Deep) algorithm, and the like.
Optionally, the second probability obtaining unit includes: the system comprises a click probability obtaining subunit, a feedback probability obtaining subunit and a second probability obtaining subunit.
And the click probability obtaining subunit is configured to input the user feature vector into a pre-trained message click prediction model, and obtain a click probability of clicking after a user predicted by the message click prediction model receives the preset message.
Wherein, the message Click prediction model can be a Click-Through-Rate (CTR) model based on a deep learning algorithm. The Deep learning algorithm may be a gradient boosting iterative Decision Tree (GBDT) algorithm, a Factorization Machine (FM) algorithm, a generalized Linear and Deep neural Network (Wide & Deep) algorithm, and the like.
And the feedback probability obtaining subunit is configured to input the user feature vector into a pre-trained message feedback prediction model, and obtain a feedback probability that a user predicted by the message feedback prediction model feeds back the preset message after receiving the preset message.
The message feedback prediction model may be a Click-Through-Rate (CTR) model based on a deep learning algorithm. The Deep learning algorithm may be a gradient boosting iterative Decision Tree (GBDT) algorithm, a Factorization Machine (FM) algorithm, a generalized Linear and Deep neural Network (Wide & Deep) algorithm, and the like.
And the second probability obtaining subunit is configured to determine, according to the click probability and the feedback probability, a probability that the user performs the second preset type of behavior on the target object after receiving the preset message.
Specifically, the second probability obtaining subunit may multiply the click probability and the feedback probability to obtain a product, and use the product as the probability of performing the second preset type of behavior on the target object after the user receives the preset message.
Optionally, another message transmission apparatus provided in the embodiment of the present invention may further include: and a distribution unit.
The distribution unit is used for executing an algorithm of optimal distribution of the coupons.
The message transmission device comprises a processor and a memory, wherein the information obtaining unit 100, the user feature vector obtaining unit 200, the first probability obtaining unit 300, the behavior data obtaining unit 400, the user screening unit 500, the message sending unit 600 and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can set one or more than one, and the messages are sent to the screened users in a targeted mode by adjusting the kernel parameters.
An embodiment of the present invention provides a storage medium on which a program is stored, the program implementing the message transmission method when executed by a processor.
The embodiment of the invention provides a processor, which is used for running a program, wherein the message transmission method is executed when the program runs.
The embodiment of the invention provides equipment, which comprises at least one processor, at least one memory and a bus, wherein the memory and the bus are connected with the processor; the processor and the memory complete mutual communication through a bus; the processor is used for calling the program instructions in the memory to execute the message transmission method. The device herein may be a server, a PC, a PAD, a mobile phone, etc.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device: (steps of method claims, independent + rights from).
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a device includes one or more processors (CPUs), memory, and a bus. The device may also include input/output interfaces, network interfaces, and the like.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip. The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims (10)

1. A method for message transmission, comprising:
acquiring user identity information and behavior information of a plurality of behaviors of a user aiming at a target object;
obtaining a user characteristic vector according to the user identity information and the behavior information;
predicting the probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current moment according to the user feature vector;
predicting behavior data of a user performing a second preset type of behavior on the target object within a second preset time period after the current time according to the user feature vector, wherein the second preset time period is the same as or different from the first preset time period;
screening users according to the predicted probability and the predicted behavior data;
and sending a preset message to at least one screened user.
2. The method of claim 1, wherein obtaining the user identity information and behavior information of the user for a plurality of behaviors of the target object comprises:
the method comprises the steps of obtaining user identity information of a target user and behavior information of the target user for multiple behaviors of the target object, wherein the target user is a user who does not conduct a third preset type of behavior for the target object within a third preset time period before the current moment, and the third preset type is the same as the second preset type, or the third preset type is the same as the first preset type.
3. The method of claim 1, further comprising:
and predicting the probability of the second preset type of behavior aiming at the target object after the user receives the preset message according to the user feature vector.
4. The method according to claim 1, wherein the predicting, according to the user feature vector, a probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current time includes:
and inputting the user feature vector into a pre-trained first behavior probability prediction model, and obtaining the probability that the user does not perform a first preset type of behavior for the target object within a first preset time period after the current time, which is predicted by the first behavior probability prediction model.
5. The method according to claim 1, wherein the predicting behavior data of the user performing a second preset type of behavior for the target object within a second preset time period after the current time according to the user feature vector comprises:
and inputting the user characteristic vector into a pre-trained behavior data prediction model, and obtaining behavior data of a second preset type of behavior of the user predicted by the data prediction model for the target object in a second preset time period after the current time.
6. The method according to claim 3, wherein predicting the probability that the user performs the second preset type of action on the target object after receiving the preset message according to the user feature vector comprises:
and inputting the user feature vector into a second pre-trained behavior prediction probability model, and obtaining the probability of the second preset type of behavior for the target object after the user predicted by the second behavior prediction probability model receives the preset message.
7. The method according to claim 3, wherein predicting the probability that the user performs the second preset type of action on the target object after receiving the preset message according to the user feature vector comprises:
inputting the user characteristic vector into a pre-trained message click prediction model, and obtaining the click probability of clicking after a user predicted by the message click prediction model receives the preset message;
inputting the user characteristic vector into a pre-trained message feedback prediction model to obtain the feedback probability of the user predicted by the message feedback prediction model for feedback after receiving the preset message;
and determining the probability of the second preset type of behavior aiming at the target object after the user receives the preset message according to the click probability and the feedback probability.
8. A message transmission apparatus, comprising: an information obtaining unit, a user characteristic vector obtaining unit, a first probability obtaining unit, a behavior data obtaining unit, a user screening unit and a message sending unit,
the information obtaining unit is used for obtaining user identity information and behavior information of a plurality of behaviors of the user aiming at the target object;
the user characteristic vector obtaining unit is used for obtaining a user characteristic vector according to the user identity information and the behavior information;
the first probability obtaining unit is used for predicting the probability that the user does not perform a first preset type of behavior on the target object within a first preset time period after the current time according to the user feature vector;
the behavior data obtaining unit is used for predicting behavior data of a second preset type of behavior of the user aiming at the target object in a second preset time period after the current time according to the user feature vector, wherein the second preset time period is the same as or different from the first preset time period;
the user screening unit is used for screening users according to the predicted probability and the predicted behavior data;
and the message sending unit is used for sending a preset message to at least one screened user.
9. The apparatus according to claim 8, wherein the information obtaining unit is specifically configured to obtain user identity information of a target user and behavior information of the target user for multiple behaviors of a target object, where the target user is a user who has not performed a third preset type of behavior on the target object within a third preset time period before a current time, and the third preset type is the same as the second preset type, or the third preset type is the same as the first preset type.
10. The apparatus of claim 8, further comprising: a second probability obtaining unit for obtaining a second probability,
the second probability obtaining unit is configured to predict, according to the user feature vector, a probability that the user performs the second preset type of behavior on the target object after receiving the preset message.
CN202010161611.XA 2020-03-10 2020-03-10 Message transmission method and device Active CN111353825B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010161611.XA CN111353825B (en) 2020-03-10 2020-03-10 Message transmission method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010161611.XA CN111353825B (en) 2020-03-10 2020-03-10 Message transmission method and device

Publications (2)

Publication Number Publication Date
CN111353825A true CN111353825A (en) 2020-06-30
CN111353825B CN111353825B (en) 2024-01-12

Family

ID=71197401

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010161611.XA Active CN111353825B (en) 2020-03-10 2020-03-10 Message transmission method and device

Country Status (1)

Country Link
CN (1) CN111353825B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114387023A (en) * 2022-01-12 2022-04-22 上海意略明数字科技股份有限公司 User data processing method and device, storage medium and computing equipment

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106649774A (en) * 2016-12-27 2017-05-10 北京百度网讯科技有限公司 Artificial intelligence-based object pushing method and apparatus
US20180005271A1 (en) * 2015-12-15 2018-01-04 Tencent Technology (Shenzhen) Company Limited Information processing method, server, and computer storage medium
CN107730284A (en) * 2016-08-10 2018-02-23 腾讯科技（深圳）有限公司 A kind of information-pushing method and device
CN108334887A (en) * 2017-01-19 2018-07-27 腾讯科技（深圳）有限公司 A kind of user's choosing method and device
CN108520324A (en) * 2018-04-13 2018-09-11 北京京东金融科技控股有限公司 Method and apparatus for generating information
CN108551462A (en) * 2018-02-28 2018-09-18 阿里巴巴集团控股有限公司 A kind of information-pushing method, device and equipment
CN109446431A (en) * 2018-12-10 2019-03-08 网易传媒科技（北京）有限公司 For the method, apparatus of information recommendation, medium and calculate equipment
US20190080363A1 (en) * 2017-09-14 2019-03-14 Amadeus S.A.S. Methods and systems for intelligent adaptive bidding in an automated online exchange network
US20190102652A1 (en) * 2016-08-31 2019-04-04 Tencent Technology (Shenzhen) Company Limited Information pushing method, storage medium and server
CN109858970A (en) * 2019-02-02 2019-06-07 中国银行股份有限公司 A kind of user's behavior prediction method, apparatus and storage medium
CN110148013A (en) * 2019-04-22 2019-08-20 阿里巴巴集团控股有限公司 A kind of user tag distribution forecasting method, apparatus and system
CN110263242A (en) * 2019-01-04 2019-09-20 腾讯科技（深圳）有限公司 Content recommendation method, device, computer readable storage medium and computer equipment

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180005271A1 (en) * 2015-12-15 2018-01-04 Tencent Technology (Shenzhen) Company Limited Information processing method, server, and computer storage medium
CN107730284A (en) * 2016-08-10 2018-02-23 腾讯科技（深圳）有限公司 A kind of information-pushing method and device
US20190102652A1 (en) * 2016-08-31 2019-04-04 Tencent Technology (Shenzhen) Company Limited Information pushing method, storage medium and server
CN106649774A (en) * 2016-12-27 2017-05-10 北京百度网讯科技有限公司 Artificial intelligence-based object pushing method and apparatus
CN108334887A (en) * 2017-01-19 2018-07-27 腾讯科技（深圳）有限公司 A kind of user's choosing method and device
US20190080363A1 (en) * 2017-09-14 2019-03-14 Amadeus S.A.S. Methods and systems for intelligent adaptive bidding in an automated online exchange network
CN108551462A (en) * 2018-02-28 2018-09-18 阿里巴巴集团控股有限公司 A kind of information-pushing method, device and equipment
CN108520324A (en) * 2018-04-13 2018-09-11 北京京东金融科技控股有限公司 Method and apparatus for generating information
CN109446431A (en) * 2018-12-10 2019-03-08 网易传媒科技（北京）有限公司 For the method, apparatus of information recommendation, medium and calculate equipment
CN110263242A (en) * 2019-01-04 2019-09-20 腾讯科技（深圳）有限公司 Content recommendation method, device, computer readable storage medium and computer equipment
CN109858970A (en) * 2019-02-02 2019-06-07 中国银行股份有限公司 A kind of user's behavior prediction method, apparatus and storage medium
CN110148013A (en) * 2019-04-22 2019-08-20 阿里巴巴集团控股有限公司 A kind of user tag distribution forecasting method, apparatus and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
李怀强等: "移动大数据匿名双向身份认证方法仿真", 《计算机仿真》, no. 09, pages 1 - 9 *
段林侠;孙晓艳;王稚慧;: "面向安全应用消息传输的异构网络选择算法", 计算机科学与探索, no. 04, pages 90 - 102 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114387023A (en) * 2022-01-12 2022-04-22 上海意略明数字科技股份有限公司 User data processing method and device, storage medium and computing equipment

Also Published As

Publication number Publication date
CN111353825B (en) 2024-01-12

Similar Documents

Publication Publication Date Title
CN107885796B (en) Information recommendation method, device and equipment
Pielot et al. Beyond interruptibility: Predicting opportune moments to engage mobile phone users
CN109145210B (en) Information recommendation method and device and storage medium
US9691035B1 (en) Real-time updates to item recommendation models based on matrix factorization
CN111178970B (en) Advertisement putting method and device, electronic equipment and computer readable storage medium
US10719769B2 (en) Systems and methods for generating and communicating application recommendations at uninstall time
US9864951B1 (en) Randomized latent feature learning
US20190303994A1 (en) Recommendation System using Linear Stochastic Bandits and Confidence Interval Generation
CN112104505B (en) Application recommendation method, device, server and computer readable storage medium
CN109685545B (en) Method and device for predicting virtual network resources to be issued and electronic equipment
US20210192549A1 (en) Generating analytics tools using a personalized market share
US20200311487A1 (en) Utilizing a bayesian approach and multi-armed bandit algorithms to improve distribution timing of electronic communications
CN114116822A (en) Information push method, terminal and storage medium
KR20180137351A (en) Item recommendation method and apparatus based on user behavior
US20190066200A1 (en) Bid optimization
CN111353825A (en) Message transmission method and device
CN107729544B (en) Method and device for recommending applications
CN111260416A (en) Method and device for determining associated user of object
CN112449217A (en) Method and device for pushing video
GB2567929A (en) Digital experience targeting using bayesian approach
US11605042B1 (en) Artificial intelligence system for content presentation impact estimation
CN113592549A (en) Conversion rate prediction method and device and electronic equipment
CN112307319A (en) Page generation method and device
CN113626699B (en) Target user identification method, device, electronic equipment and storage medium
CN114979275B (en) Method for processing resource request, electronic equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant