CN111461778A - Advertisement pushing method and device - Google Patents
Advertisement pushing method and device Download PDFInfo
- Publication number
- CN111461778A CN111461778A CN202010242625.4A CN202010242625A CN111461778A CN 111461778 A CN111461778 A CN 111461778A CN 202010242625 A CN202010242625 A CN 202010242625A CN 111461778 A CN111461778 A CN 111461778A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- model
- time
- user
- reading
- 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
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
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a method and a device for pushing advertisements, which are used for acquiring advertisement information of an advertisement to be published and user information of a target user; processing advertisement information of the advertisement to be published and user information of a target user by utilizing a pre-trained push time model to obtain push time of the advertisement to be published corresponding to the target user; the push time model is a neural network model obtained by utilizing a plurality of training samples; each training sample comprises user information of a historical user, advertisement information of a piece of published advertisement effectively read by the historical user and actual reading time; and pushing the advertisement to be published to the target user in the advertisement to be published pushing time corresponding to the target user. The push time model provided by the scheme can predict the most possible time for the user to effectively read the advertisement to be released by combining the preferences of the user on different advertisements in different time periods in the past, and pushes the advertisement at the time, thereby further improving the advertisement push efficiency.
Description
Technical Field
The present invention relates to the field of advertisement technologies, and in particular, to a method and an apparatus for pushing advertisements.
Background
With the development of computer technology and internet technology, computers in various forms and smart phones are used more and more widely. Accordingly, it is also a common information promotion means to push advertisements to users when they use these electronic devices.
Currently, an advertisement push strategy for these electronic devices is generally that when a user accesses a specific network platform (such as a web page, client software, and various application programs) using a device, the user's preference for information is determined according to the type of information that the user has browsed in the past and part of personal information of the user, and then advertisement content related to the preferred information is pushed to the user, so that the user is ensured to have a higher probability of reading pushed advertisements, and an effect of improving advertisement push efficiency is achieved.
However, the existing method pushes the same type of advertisement to the user at any time, without considering that the user's preference may change at different time periods of the day, so the pushing efficiency of the method is still low.
Disclosure of Invention
In view of the problems in the prior art, the present application provides a method and an apparatus for advertisement delivery, and particularly provides a scheme for determining the delivery time of an advertisement by analyzing preferences of a user in different time periods, so as to further improve the efficiency of advertisement delivery.
A first aspect of the present application discloses an advertisement push method, including:
acquiring advertisement information of an advertisement to be issued and user information of a target user;
processing the advertisement information of the advertisement to be released and the user information of the target user by utilizing a pre-trained push time model to obtain the push time of the advertisement to be released corresponding to the target user; the push time model is a neural network model obtained by training a plurality of training samples; each training sample comprises user information of a historical user, and advertisement information and actual reading time of a piece of published advertisement effectively read by the historical user; the actual reading time refers to a time during which the historical user effectively reads the published advertisement;
and pushing the advertisement to be published to the target user in the advertisement to be published pushing time corresponding to the target user.
Optionally, the process of training the push time model includes:
constructing a plurality of training samples by utilizing a plurality of advertisement reading records in an advertisement reading record library;
acquiring a preset neural network model and determining initial model parameters of the neural network model to obtain an initial push model;
inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample;
calculating the model loss of the initial pushing model according to the pushing time of each training sample and the actual reading time of the training sample;
judging whether the model loss meets a preset model convergence condition or not;
if the model loss does not meet the model convergence condition, updating parameters of the initial pushing model according to the model loss, and returning to execute the step of inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample;
and if the model loss meets the model convergence condition, determining the initial push model as the push time model.
Optionally, each of the advertisement reading records includes: the system comprises a user identifier of a historical user, an advertisement identifier of a published advertisement, and starting time and ending time of reading the published advertisement by the historical user;
the method for constructing a plurality of training samples by utilizing a plurality of advertisement reading records in the advertisement reading record library comprises the following steps:
calculating the difference between the starting time of the advertisement reading record and the ending time of the advertisement reading record aiming at each advertisement reading record to obtain the reading duration of the advertisement reading record;
judging whether the reading time of each advertisement reading record is greater than a preset effective reading time or not;
for each advertisement reading record, if the reading time of the advertisement reading record is longer than the effective reading time, determining the advertisement reading record as an effective reading record;
and aiming at each effective reading record, combining the user information of the historical user of the effective reading record, the advertisement information of the published advertisement of the effective reading record and the starting time of the effective reading record to obtain a training sample corresponding to the effective reading record.
Optionally, the determining initial model parameters of the neural network model includes:
randomly generating a plurality of initial parameter individuals according to the structure of the neural network model; wherein each initial parameter individual comprises a plurality of parameters;
optimizing the initial parameter individuals by using a genetic algorithm to obtain optimal parameter individuals;
determining a plurality of parameters in the optimal parameter individual as initial model parameters of the neural network model.
Optionally, the user information at least includes: user occupation information, user family information and user preference information;
the advertisement information includes at least an advertisement type.
This application second aspect provides a device of advertisement propelling movement, includes:
the system comprises an acquisition unit, a display unit and a display unit, wherein the acquisition unit is used for acquiring advertisement information of an advertisement to be issued and user information of a target user;
the training unit is used for training a neural network model by utilizing a plurality of training samples to obtain a push time model; each training sample comprises user information of a historical user, and advertisement information and actual reading time of a piece of published advertisement effectively read by the historical user; the actual reading time refers to a time during which the historical user effectively reads the published advertisement;
the processing unit is used for processing the advertisement information of the advertisement to be issued and the user information of the target user by utilizing the push time model to obtain the push time of the advertisement to be issued corresponding to the target user;
and the pushing unit is used for pushing the advertisement to be issued to the target user in the advertisement to be issued pushing time corresponding to the target user.
Optionally, the training unit trains a neural network model by using a plurality of training samples, and when obtaining the push time model, the training unit is specifically configured to:
constructing a plurality of training samples by utilizing a plurality of advertisement reading records in an advertisement reading record library;
acquiring a preset neural network model and determining initial model parameters of the neural network model to obtain an initial push model;
inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample;
calculating the model loss of the initial pushing model according to the pushing time of each training sample and the actual reading time of the training sample;
judging whether the model loss meets a preset model convergence condition or not;
if the model loss does not meet the model convergence condition, updating parameters of the initial pushing model according to the model loss, and returning to execute the step of inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample;
and if the model loss meets the model convergence condition, determining the initial push model as the push time model.
Optionally, each of the advertisement reading records includes: the system comprises a user identifier of a historical user, an advertisement identifier of a published advertisement, and starting time and ending time of reading the published advertisement by the historical user;
when the training unit constructs a plurality of training samples by using a plurality of advertisement reading records in the advertisement reading record library, the training unit is specifically configured to:
calculating the difference between the starting time of the advertisement reading record and the ending time of the advertisement reading record aiming at each advertisement reading record to obtain the reading duration of the advertisement reading record;
judging whether the reading time of each advertisement reading record is greater than a preset effective reading time or not;
for each advertisement reading record, if the reading time of the advertisement reading record is longer than the effective reading time, determining the advertisement reading record as an effective reading record;
and aiming at each effective reading record, combining the user information of the historical user of the effective reading record, the advertisement information of the published advertisement of the effective reading record and the starting time of the effective reading record to obtain a training sample corresponding to the effective reading record.
Optionally, when the training unit determines the initial model parameters of the neural network model, the training unit is specifically configured to:
randomly generating a plurality of initial parameter individuals according to the structure of the neural network model; wherein each initial parameter individual comprises a plurality of parameters;
optimizing the initial parameter individuals by using a genetic algorithm to obtain optimal parameter individuals;
determining a plurality of parameters in the optimal parameter individual as initial model parameters of the neural network model.
Optionally, the user information at least includes: user occupation information, user family information and user preference information;
the advertisement information includes at least an advertisement type.
The application provides a method and a device for pushing advertisements, which are used for acquiring advertisement information of an advertisement to be published and user information of a target user; processing advertisement information of the advertisement to be published and user information of a target user by utilizing a pre-trained push time model to obtain push time of the advertisement to be published corresponding to the target user; the push time model is a neural network model obtained by utilizing a plurality of training samples; each training sample comprises user information of a historical user, advertisement information of a piece of published advertisement effectively read by the historical user and actual reading time; and pushing the advertisement to be published to the target user in the advertisement to be published pushing time corresponding to the target user. The push time model provided by the scheme can predict the most possible time for the user to effectively read the advertisement to be released by combining the preferences of the user on different advertisements in different time periods in the past, and pushes the advertisement at the time, thereby further improving the advertisement push efficiency.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a method for pushing an advertisement according to an embodiment of the present application;
fig. 2 is a flowchart of a method for training a push time model according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of an advertisement delivery apparatus according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The advertisement push efficiency can be understood as the ratio of the number of times that the user effectively reads the pushed advertisement to the total number of times of advertisement push in a period of time. For example, assuming that a certain advertisement presenter pushes an advertisement to a total of 100 users in january, each user has pushed 20 times, the total number of advertisement pushes in january is 2000, wherein the users effectively read 1000 times, and the advertisement push efficiency in january is 50%. After receiving the prompt message of advertisement push, the user enters the corresponding advertisement detail page and stays for a long enough time (specifically, the time may be greater than or equal to 15 seconds), and then the user is considered to have effectively read the advertisement.
The existing advertisement push method only considers what types of users may be interested in what types of advertisements, but does not consider that the preferences of the advertisement types of the users may also change with different time periods, so that the advertisement push efficiency is not high. In order to solve the problem, the application provides a method and a device for pushing advertisements, so as to improve the efficiency of pushing advertisements.
A first embodiment of the present application provides an advertisement push method, please refer to fig. 1, which specifically includes the following steps:
s101, obtaining advertisement information of the advertisement to be released and user information of a target user.
Optionally, the advertisement information includes at least an advertisement type. The advertisement type is a classification aiming at advertisement content, and particularly, for the advertisement provided by a bank, the advertisement type can comprise a financial product advertisement, a interest rate adjustment advertisement, a business process adjustment advertisement and the like.
In the expression form of the advertisement, the advertisement can be static characters and pictures, and can also be a video, wherein when the advertisement is a video, the advertisement information can also include the duration of the advertisement.
Optionally, the user information at least includes: user occupation information (for indicating the current occupation of the user), user family information (for indicating the family member status of the user), and user preference information (characterizing the personal preferences of the user, e.g., car, sports, reading, etc.).
Optionally, the user information may be obtained by collecting behavior data of the user on the internet and analyzing the behavior data, or may be actively collected by pushing a corresponding questionnaire to the user.
S102, processing the advertisement information of the advertisement to be released and the user information of the target user by using the push time model to obtain the push time of the advertisement to be released corresponding to the target user.
The push time model is a neural network model obtained by training a plurality of training samples; each training sample comprises user information of a historical user, advertisement information of a piece of published advertisement effectively read by the historical user and actual reading time; the actual reading time refers to the time of the historical user's effective reading of the published advertisement.
The push time of the advertisement to be published corresponding to the target user can be predicted by using a push time model, and the target user is most likely to be interested in the advertisement to be published, or the target user is most likely to be effective in reading the advertisement to be published.
S103, pushing the advertisement to be published to the target user in the advertisement to be published pushing time corresponding to the target user.
The specific implementation process of step S103 is to detect whether the target user logs in to the specified advertisement publishing platform through any electronic device in real time after determining the to-be-published advertisement push time corresponding to the target user. For example, if a certain bank needs to push the advertisement, whether the target user logs in an online bank or an online shopping mall of the bank through a computer or a mobile phone is detected in real time.
After detecting that the target user logs in the appointed platform (namely the target user is online), judging whether the current time is matched with the advertisement pushing time to be issued of the target user. The matching may be that the absolute value of the difference between the current time and the time for pushing the advertisement to be published of the target user is less than a certain threshold. For example, if the current time is 5 minutes before or after the to-be-published advertisement push time of the target user, the current time and the to-be-published advertisement push time of the target user may be considered to be matched, that is, if the to-be-published advertisement push time of the target user is 20:05, the current time and the to-be-published advertisement push time of the target user may be considered to be matched as long as the current time is within a time period of 20:00 to 20: 10.
And if the target user is online and the current time is matched with the advertisement pushing time to be issued of the target user, pushing the advertisement to be issued to the electronic equipment which is currently used for logging in the appointed platform by the target user.
The step of pushing the advertisement to be released to the electronic equipment means that prompt information of the advertisement to be released is sent to the electronic equipment, the electronic equipment displays the prompt information to a user after receiving the prompt information successfully, the user can select to click or not click the prompt information, after clicking the prompt information, the electronic equipment requests a detailed page of the pushed advertisement to be released to a server, and the obtained detailed page is displayed to the user for the user to read the advertisement.
The application provides an advertisement pushing method, which comprises the steps of obtaining advertisement information of an advertisement to be published and user information of a target user; processing advertisement information of the advertisement to be published and user information of a target user by utilizing a pre-trained push time model to obtain push time of the advertisement to be published corresponding to the target user; the push time model is a neural network model obtained by utilizing a plurality of training samples; each training sample comprises user information of a historical user, advertisement information of a piece of published advertisement effectively read by the historical user and actual reading time; and pushing the advertisement to be published to the target user in the advertisement to be published pushing time corresponding to the target user.
It can be understood that the push time model provided by the scheme is a push time model constructed by analyzing advertisement types which are interested by a plurality of users in different time periods in the past. Therefore, the push time model can accurately predict what type of advertisement is most likely to be interested in, or most likely to be effectively read, at what time by a user having a characteristic described by specific user information, and output the predicted time in the form of push time.
To sum up, the advertisement push scheme provided by the application can ensure that the pushed advertisement can be effectively read by the user with high probability when the advertisement is pushed to the user at the predicted push time, so that the push efficiency of the advertisement is effectively improved.
A second embodiment of the present application provides a training method for pushing a time model, please refer to fig. 2, which includes the following steps:
s201, constructing a plurality of training samples by using a plurality of advertisement reading records in the advertisement reading record library.
The advertisement reading record library is a database which is constructed by an advertisement delivery party (referring to a merchant needing to deliver advertisements to users, specifically, the advertisement delivery party can be a bank) by collecting behavior data when the users visit the relevant network platforms of the advertisement delivery party, and comprises a plurality of advertisement reading records.
An advertisement reading record comprising: the system comprises a user identifier of a historical user, an advertisement identifier of a published advertisement, and a starting time and an ending time of reading the published advertisement by the historical user.
If an advertisement has been previously pushed to a user, the user is marked as a historical user, and the advertisement that has been pushed becomes a published advertisement.
The construction process of the advertisement reading record library is described below by taking a bank as an example. The bank will provide web-page version of internet bank and related internet shopping mall on the personal computer, and will provide internet bank client on the mobile phone, the user can use the account to access the above network platform to obtain information after registering the internet bank account. For each user, the server of the bank can predict whether the current time is the idle time of the user according to the identity information of the user when the user accesses the network platform. For example, if the user identity is a business employee, then 18:00 to 22:00 on weekdays and 7:00 to 22:00 on holidays are predicted to be the free time of the user.
Alternatively, an electronic questionnaire may be pushed to the user, and information filled in by the user on the questionnaire is collected to determine the free time of the user.
And if the current time is the idle time of the user, selecting one or more advertisements from the plurality of advertisements stored in the server to push to the user.
After the server pushes the advertisement to the user, the electronic equipment of the user can pop up related prompt information to prompt that the user currently has the advertisement to push, and the user can select to click or not click the prompt information. After the user clicks the prompt information, the server pushes a detailed page of the advertisement to the user, the time when the user clicks the prompt information is recorded as the starting time, the user is marked to read the advertisement, and after the user quits the detailed page of the advertisement, the server records the time when the user quits as the ending time. Therefore, the user identifier (which may be a user account) of the user, the advertisement identifier (which may be a number configured for each advertisement in advance as an advertisement identifier) of the advertisement read by the user, and the start time and the end time constitute an advertisement reading record.
Based on the description of the advertisement reading record, the specific method for constructing the training sample by using the advertisement reading record comprises the following steps:
and calculating the difference between the starting time of the advertisement reading record and the ending time of the advertisement reading record aiming at each advertisement reading record to obtain the reading duration of the advertisement reading record.
And judging whether the reading time of the advertisement reading record is greater than the preset effective reading time or not aiming at each advertisement reading record.
The effective reading time period is a preset time period and may be set to 15 seconds, for example. After a user clicks a prompt to enter a detailed page of a corresponding advertisement, the advertisement may be found to be not interesting, and then the detailed page is immediately closed and the advertisement is not read. By setting the effective reading duration, the reading record of the advertisement which is interested by the user can be screened from the plurality of advertisement reading records, and the reading record of the advertisement which is not interested by the user is abandoned.
And aiming at each advertisement reading record, if the reading time length of the advertisement reading record is greater than the effective reading time length, determining the advertisement reading record as the effective reading record.
And finally, aiming at each effective reading record, combining the user information of the historical user of the effective reading record, the advertisement information of the published advertisement of the effective reading record and the starting time of the effective reading record to obtain a training sample corresponding to the effective reading record.
S202, obtaining a preset neural network model and determining initial model parameters of the neural network model to obtain an initial pushing model.
The preset neural network model may be a neural network model constructed based on any one of the existing neural network frameworks, and is not limited herein. For example, the predetermined neural network model may be a Back Propagation (BP) neural network model including an input layer, an output layer and a hidden layer.
After the architecture of the neural network model is determined, the number of model parameters involved in the neural network model and the relationship between the model parameters are determined. An initial push model can be obtained only by giving initial parameter values of each model parameter.
Specifically, the initial model parameters in step S202 may be determined by using a genetic algorithm based on the training samples obtained and the architecture of the neural network model determined.
S203, inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample.
The push time of a training sample is the time predicted by the initial push model that theoretically the user involved in the training sample will effectively read the corresponding advertisement (i.e. the reading time is longer than the effective reading time, in other words, the user may be considered to be interested in the advertisement).
If the training sample includes the user information of the user X and the advertisement information of the advertisement Y, the push time output by the initial push model is the predicted time that the user X may effectively read the advertisement Y, for example, the output push time is 21:00, which means that the initial push model predicts that the user X may effectively read the advertisement Y at 21: 00.
And S204, calculating the model loss of the initial push model according to the push time of each training sample and the actual reading time of the training sample.
Specifically, for each training sample, the pushing time of the training sample is subtracted from the actual reading time to obtain a difference of the training sample, and finally, the sum of squares of the differences of all the training samples is calculated to obtain the model loss of the initial pushing model.
And S205, judging whether the model loss meets a preset model convergence condition.
The model convergence condition may be that the model loss is less than a preset loss threshold. That is, if the current model loss is less than the preset loss threshold, the model loss is considered to satisfy the model convergence condition, step S207 is executed, and if the current model loss is greater than or equal to the preset loss threshold, the model loss is considered not to satisfy the model convergence condition, and step S206 is executed.
And S206, updating the parameters of the initial push model according to the model loss.
After step S206 is executed, the initial push model is updated, and an updated initial push model is generated.
When the execution of step S206 is completed, the process returns to step S203.
And S207, determining the initial push model as a push time model.
If the current model loss meets the model convergence condition, it indicates that, for a specific user and a specific type of advertisement, the time that is predicted by the initial push model to effectively read the advertisement by the user has a high probability, which is the time that the user actually effectively reads the advertisement, in other words, the initial push model can accurately predict at the moment when the specific user is most likely to effectively read the specific type of advertisement, and therefore, the initial push model at the moment can be determined as a push time model and used for predicting the push time of the advertisement.
Optionally, after the push time model is determined, a plurality of verification samples may be further constructed by using data in the advertisement reading record library, then, the user information and the advertisement information of each verification sample are input into the push time model determined in step S207, the push time model is verified by comparing the push time output by the push time model with the actual reading time of the verification sample, if the verification is passed, prediction is performed by using the push time model, and if the verification is not passed, training is performed again.
Specifically, the process of determining the initial model parameters of the neural network model by using the genetic algorithm in step S202 is as follows:
firstly, a plurality of initial parameter individuals are randomly generated according to the structure of a neural network model.
Each initial parameter individual comprises N parameters, the parameters can be real numbers generated randomly, and the value of N depends on the structure of a set neural network model.
After determining the structure of the neural network model, as described above, it is determined how many parameters the model contains. Assuming that the structure of the neural network model determined in step S202 involves 50 parameters, N is equal to 50, i.e. each initial parameter comprises 50 real numbers randomly determined from numerical values.
And optimizing the plurality of initial parameter individuals by using a genetic algorithm to obtain the optimal parameter individuals.
The optimization of the initial parameter individual specifically means:
and determining the genetic probability of each initial parameter individual based on the fitness of the initial parameter individual, wherein the lower the fitness is, the higher the genetic probability is, and simultaneously determining a uniform mutation probability. Exchanging every two initial parameter individuals based on the determined genetic probability, namely exchanging a part of parameters of one initial parameter individual with a part of parameters of another initial parameter individual to obtain two new parameter individuals, and simultaneously reserving the original initial parameter individuals. On the other hand, for each initial parameter individual, mutation is performed based on the above mutation probability, that is, the values of some parameters of the initial parameter individual are randomly adjusted. Through the process, a generation of parameter individuals can be obtained.
After the first-generation parameter individuals are obtained, the parameter individuals with overlarge fitness (larger than a certain threshold value) are deleted from all the parameter individuals (including the initial parameter individuals and the first-generation parameter individuals), and then the heredity and variation processes are repeated on the rest parameter individuals to obtain the second-generation parameter individuals. And by analogy, generating three-generation and four-generation parameter individuals until all the parameter individuals have the parameter individuals with the fitness smaller than a certain threshold value or the algebra larger than a certain threshold value. After the algorithm is ended, the parameter individual with the minimum fitness in all the parameter individuals is the optimal parameter individual.
In the present application, the fitness of a parameter individual is defined as the model loss of a neural network model constructed by using the parameter individual.
As described above, each initial parameter entity includes a number of parameters equal to the number of parameters required for the neural network model. Aiming at any initial parameter individual, after all parameters of the initial parameter individual are substituted into the structure of the determined neural network model, a complete neural network model can be obtained, then the model loss of the neural network model constructed by the parameter individual can be calculated according to the steps S203 and S204,
after the optimal parameter individuals are generated, each parameter included in the optimal parameter individuals is the initial parameter of the neural network model in the step S202.
With reference to fig. 3, the apparatus for pushing an advertisement according to the foregoing embodiments of the present application further includes:
the acquiring unit 301 is configured to acquire advertisement information of an advertisement to be published and user information of a target user.
The training unit 302 is configured to train a neural network model using a plurality of training samples to obtain a push time model.
Each training sample comprises user information of a historical user, advertisement information of a piece of published advertisement effectively read by the historical user and actual reading time; the actual reading time refers to the time of the historical user's effective reading of the published advertisement.
The processing unit 303 is configured to process the advertisement information of the advertisement to be published and the user information of the target user by using the push time model, so as to obtain the push time of the advertisement to be published corresponding to the target user.
The pushing unit 304 is configured to push the advertisement to be published to the target user in the advertisement to be published pushing time corresponding to the target user.
The training unit 302 is configured to train a neural network model by using a plurality of training samples, and when obtaining the push time model, is specifically configured to:
constructing a plurality of training samples by utilizing a plurality of advertisement reading records in an advertisement reading record library;
acquiring a preset neural network model and determining initial model parameters of the neural network model to obtain an initial push model;
inputting the user information and the advertisement information of each training sample into an initial pushing model to obtain the pushing time of each training sample;
calculating the model loss of the initial pushing model according to the pushing time of each training sample and the actual reading time of the training sample;
judging whether the model loss meets a preset model convergence condition or not;
if the model loss does not meet the model convergence condition, updating parameters of the initial pushing model according to the model loss, and returning to execute the step of inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample;
and if the model loss meets the model convergence condition, determining the initial push model as a push time model.
Specifically, each advertisement reading record comprises: the system comprises a user identifier of a historical user, an advertisement identifier of a published advertisement, and starting time and ending time of the historical user for reading the published advertisement;
when the training unit 302 constructs a plurality of training samples using the plurality of advertisement reading records in the advertisement reading record library, it is specifically configured to:
calculating the difference between the starting time of the advertisement reading record and the ending time of the advertisement reading record aiming at each advertisement reading record to obtain the reading duration of the advertisement reading record;
judging whether the reading time of each advertisement reading record is greater than the preset effective reading time or not;
for each advertisement reading record, if the reading time of the advertisement reading record is longer than the effective reading time, determining the advertisement reading record as the effective reading record;
and aiming at each effective reading record, combining the user information of the historical user of the effective reading record, the advertisement information of the published advertisement of the effective reading record and the starting time of the effective reading record to obtain a training sample corresponding to the effective reading record.
Optionally, when the training unit 302 determines the initial model parameters of the neural network model, it is specifically configured to:
randomly generating a plurality of initial parameter individuals according to the structure of the neural network model; wherein each initial parameter individual comprises a plurality of parameters;
optimizing a plurality of initial parameter individuals by using a genetic algorithm to obtain optimal parameter individuals;
and determining a plurality of parameters in the optimal parameter individuals as initial model parameters of the neural network model.
Optionally, the user information at least includes: user professional information, user family information and user preference information, and the advertisement information at least comprises advertisement types.
For the device for pushing an advertisement provided in this embodiment, specific working principles thereof may refer to the method for pushing an advertisement provided in any embodiment of the present application, and details are not described here again.
The application provides a device for pushing advertisements, wherein an acquisition unit 301 acquires advertisement information of an advertisement to be published and user information of a target user; the training unit 302 obtains a push time model by training a plurality of training samples; each training sample comprises user information of a historical user, advertisement information of a piece of published advertisement effectively read by the historical user and actual reading time. The processing unit 303 processes the advertisement information of the advertisement to be published and the user information of the target user by using a pre-trained push time model to obtain the push time of the advertisement to be published corresponding to the target user; the pushing unit 304 pushes the advertisement to be published to the target user at the advertisement to be published pushing time corresponding to the target user.
It can be understood that the push time model provided by the scheme is a push time model constructed by analyzing advertisement types which are interested by a plurality of users in different time periods in the past. Therefore, the push time model can accurately predict what type of advertisement is most likely to be interested in, or most likely to be effectively read, at what time by a user having a characteristic described by specific user information, and output the predicted time in the form of push time.
To sum up, the advertisement push scheme provided by the application can ensure that the pushed advertisement can be effectively read by the user with high probability when the advertisement is pushed to the user at the predicted push time, so that the push efficiency of the advertisement is effectively improved.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, 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 a process, method, article, or apparatus that comprises the element.
It should be noted that the terms "first", "second", and the like in the present invention are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.
Those skilled in the art can make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method of advertisement push, comprising:
acquiring advertisement information of an advertisement to be issued and user information of a target user;
processing the advertisement information of the advertisement to be released and the user information of the target user by utilizing a pre-trained push time model to obtain the push time of the advertisement to be released corresponding to the target user; the push time model is a neural network model obtained by training a plurality of training samples; each training sample comprises user information of a historical user, and advertisement information and actual reading time of a piece of published advertisement effectively read by the historical user; the actual reading time refers to a time during which the historical user effectively reads the published advertisement;
and pushing the advertisement to be published to the target user in the advertisement to be published pushing time corresponding to the target user.
2. The method of claim 1, wherein training the push time model comprises:
constructing a plurality of training samples by utilizing a plurality of advertisement reading records in an advertisement reading record library;
acquiring a preset neural network model and determining initial model parameters of the neural network model to obtain an initial push model;
inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample;
calculating the model loss of the initial pushing model according to the pushing time of each training sample and the actual reading time of the training sample;
judging whether the model loss meets a preset model convergence condition or not;
if the model loss does not meet the model convergence condition, updating parameters of the initial pushing model according to the model loss, and returning to execute the step of inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample;
and if the model loss meets the model convergence condition, determining the initial push model as the push time model.
3. The method of claim 2, wherein each of the advertisement reading records comprises: the system comprises a user identifier of a historical user, an advertisement identifier of a published advertisement, and starting time and ending time of reading the published advertisement by the historical user;
the method for constructing a plurality of training samples by utilizing a plurality of advertisement reading records in the advertisement reading record library comprises the following steps:
calculating the difference between the starting time of the advertisement reading record and the ending time of the advertisement reading record aiming at each advertisement reading record to obtain the reading duration of the advertisement reading record;
judging whether the reading time of each advertisement reading record is greater than a preset effective reading time or not;
for each advertisement reading record, if the reading time of the advertisement reading record is longer than the effective reading time, determining the advertisement reading record as an effective reading record;
and aiming at each effective reading record, combining the user information of the historical user of the effective reading record, the advertisement information of the published advertisement of the effective reading record and the starting time of the effective reading record to obtain a training sample corresponding to the effective reading record.
4. The method of claim 2, wherein the determining initial model parameters of the neural network model comprises:
randomly generating a plurality of initial parameter individuals according to the structure of the neural network model; wherein each initial parameter individual comprises a plurality of parameters;
optimizing the initial parameter individuals by using a genetic algorithm to obtain optimal parameter individuals;
determining a plurality of parameters in the optimal parameter individual as initial model parameters of the neural network model.
5. The method according to claim 1, wherein the user information comprises at least: user occupation information, user family information and user preference information;
the advertisement information includes at least an advertisement type.
6. An advertisement push device, comprising:
the system comprises an acquisition unit, a display unit and a display unit, wherein the acquisition unit is used for acquiring advertisement information of an advertisement to be issued and user information of a target user;
the training unit is used for training a neural network model by utilizing a plurality of training samples to obtain a push time model; each training sample comprises user information of a historical user, and advertisement information and actual reading time of a piece of published advertisement effectively read by the historical user; the actual reading time refers to a time during which the historical user effectively reads the published advertisement;
the processing unit is used for processing the advertisement information of the advertisement to be issued and the user information of the target user by utilizing the push time model to obtain the push time of the advertisement to be issued corresponding to the target user;
and the pushing unit is used for pushing the advertisement to be issued to the target user in the advertisement to be issued pushing time corresponding to the target user.
7. The apparatus according to claim 6, wherein the training unit is configured to train a neural network model using a plurality of training samples, and when obtaining the push time model, to:
constructing a plurality of training samples by utilizing a plurality of advertisement reading records in an advertisement reading record library;
acquiring a preset neural network model and determining initial model parameters of the neural network model to obtain an initial push model;
inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample;
calculating the model loss of the initial pushing model according to the pushing time of each training sample and the actual reading time of the training sample;
judging whether the model loss meets a preset model convergence condition or not;
if the model loss does not meet the model convergence condition, updating parameters of the initial pushing model according to the model loss, and returning to execute the step of inputting the user information and the advertisement information of each training sample into the initial pushing model to obtain the pushing time of each training sample;
and if the model loss meets the model convergence condition, determining the initial push model as the push time model.
8. The apparatus of claim 7, wherein each of said advertisement reading records comprises: the system comprises a user identifier of a historical user, an advertisement identifier of a published advertisement, and starting time and ending time of reading the published advertisement by the historical user;
when the training unit constructs a plurality of training samples by using a plurality of advertisement reading records in the advertisement reading record library, the training unit is specifically configured to:
calculating the difference between the starting time of the advertisement reading record and the ending time of the advertisement reading record aiming at each advertisement reading record to obtain the reading duration of the advertisement reading record;
judging whether the reading time of each advertisement reading record is greater than a preset effective reading time or not;
for each advertisement reading record, if the reading time of the advertisement reading record is longer than the effective reading time, determining the advertisement reading record as an effective reading record;
and aiming at each effective reading record, combining the user information of the historical user of the effective reading record, the advertisement information of the published advertisement of the effective reading record and the starting time of the effective reading record to obtain a training sample corresponding to the effective reading record.
9. The apparatus according to claim 7, wherein the training unit, when determining the initial model parameters of the neural network model, is specifically configured to:
randomly generating a plurality of initial parameter individuals according to the structure of the neural network model; wherein each initial parameter individual comprises a plurality of parameters;
optimizing the initial parameter individuals by using a genetic algorithm to obtain optimal parameter individuals;
determining a plurality of parameters in the optimal parameter individual as initial model parameters of the neural network model.
10. The apparatus of claim 6, wherein the user information comprises at least: user occupation information, user family information and user preference information;
the advertisement information includes at least an advertisement type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010242625.4A CN111461778B (en) | 2020-03-31 | 2020-03-31 | Advertisement pushing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010242625.4A CN111461778B (en) | 2020-03-31 | 2020-03-31 | Advertisement pushing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111461778A true CN111461778A (en) | 2020-07-28 |
| CN111461778B CN111461778B (en) | 2023-07-14 |
Family
ID=71678455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010242625.4A Active CN111461778B (en) | 2020-03-31 | 2020-03-31 | Advertisement pushing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111461778B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112399204A (en) * | 2020-11-03 | 2021-02-23 | 重庆农村商业银行股份有限公司 | Alternate type information pushing method, device, equipment and storage medium |
| CN113448528A (en) * | 2021-07-12 | 2021-09-28 | 中国银行股份有限公司 | Method and device for displaying background picture on self-service teller machine |
| CN113822696A (en) * | 2021-03-09 | 2021-12-21 | 北京沃东天骏信息技术有限公司 | Information display method, device, system, electronic equipment and medium |
| CN113869964A (en) * | 2021-11-30 | 2021-12-31 | 腾讯科技（深圳）有限公司 | Data processing method and related device |
| CN116228319A (en) * | 2023-05-04 | 2023-06-06 | 广州钛动科技股份有限公司 | Intelligent advertisement putting method and system based on buried points |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080250450A1 (en) * | 2007-04-06 | 2008-10-09 | Adisn, Inc. | Systems and methods for targeted advertising |
| CN105187863A (en) * | 2015-07-31 | 2015-12-23 | 小米科技有限责任公司 | Advertisement playing method and device |
| EP3057055A1 (en) * | 2013-10-08 | 2016-08-17 | Wides Limited | Advertisement informatin sharing system |
| US20170186043A1 (en) * | 2015-12-29 | 2017-06-29 | Le Holdings (Beijing) Co., Ltd. | Advertisement recommendation method and based on face recogntion and recognition and electronic device |
| CN107332879A (en) * | 2017-06-05 | 2017-11-07 | 广东欧珀移动通信有限公司 | Method, mobile terminal and storage medium that a kind of information is pushed |
| CN108229994A (en) * | 2016-12-21 | 2018-06-29 | 北京金山云网络技术有限公司 | A kind of information-pushing method and device |
-
2020
- 2020-03-31 CN CN202010242625.4A patent/CN111461778B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080250450A1 (en) * | 2007-04-06 | 2008-10-09 | Adisn, Inc. | Systems and methods for targeted advertising |
| EP3057055A1 (en) * | 2013-10-08 | 2016-08-17 | Wides Limited | Advertisement informatin sharing system |
| CN105187863A (en) * | 2015-07-31 | 2015-12-23 | 小米科技有限责任公司 | Advertisement playing method and device |
| US20170186043A1 (en) * | 2015-12-29 | 2017-06-29 | Le Holdings (Beijing) Co., Ltd. | Advertisement recommendation method and based on face recogntion and recognition and electronic device |
| CN108229994A (en) * | 2016-12-21 | 2018-06-29 | 北京金山云网络技术有限公司 | A kind of information-pushing method and device |
| CN107332879A (en) * | 2017-06-05 | 2017-11-07 | 广东欧珀移动通信有限公司 | Method, mobile terminal and storage medium that a kind of information is pushed |
Non-Patent Citations (1)
| Title |
|---|
| 陆枫 等: "基于细粒度标签的在线视频广告投放机制研究" * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112399204A (en) * | 2020-11-03 | 2021-02-23 | 重庆农村商业银行股份有限公司 | Alternate type information pushing method, device, equipment and storage medium |
| CN113822696A (en) * | 2021-03-09 | 2021-12-21 | 北京沃东天骏信息技术有限公司 | Information display method, device, system, electronic equipment and medium |
| CN113448528A (en) * | 2021-07-12 | 2021-09-28 | 中国银行股份有限公司 | Method and device for displaying background picture on self-service teller machine |
| CN113869964A (en) * | 2021-11-30 | 2021-12-31 | 腾讯科技（深圳）有限公司 | Data processing method and related device |
| CN116228319A (en) * | 2023-05-04 | 2023-06-06 | 广州钛动科技股份有限公司 | Intelligent advertisement putting method and system based on buried points |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111461778B (en) | 2023-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111461778B (en) | Advertisement pushing method and device | |
| US7594189B1 (en) | Systems and methods for statistically selecting content items to be used in a dynamically-generated display | |
| US8620746B2 (en) | Scoring quality of traffic to network sites | |
| US9245252B2 (en) | Method and system for determining on-line influence in social media | |
| TWI570645B (en) | Advertisement mehtod and apparatus using user analyzing platform and marketing platform based on cohort | |
| US9607273B2 (en) | Optimal time to post for maximum social engagement | |
| US8732015B1 (en) | Social media pricing engine | |
| US20080313040A1 (en) | Content distribution system including cost-per-engagement based advertising | |
| US10776816B2 (en) | System and method for building a targeted audience for an online advertising campaign | |
| WO2012048244A2 (en) | System and method for real-time advertising campaign adaptation | |
| US9911130B1 (en) | Attribution modeling using regression analysis | |
| CN104077711A (en) | Method and system for determining changes in brand awareness after exposure to on-line advertisements | |
| Jenamani et al. | A stochastic model of e-customer behavior | |
| US20100217668A1 (en) | Optimizing Delivery of Online Advertisements | |
| US11983089B2 (en) | Contribution incrementality machine learning models | |
| Roos et al. | The effect of links and excerpts on internet news consumption | |
| US20230316106A1 (en) | Method and apparatus for training content recommendation model, device, and storage medium | |
| CN111814062A (en) | Information pushing method and device, server and storage medium | |
| TWI657395B (en) | Opinion leader related network-based trading system, method, and storage medium | |
| CN111461827A (en) | Product evaluation information pushing method and device | |
| CN112819528A (en) | Crowd pack online method and device and electronic equipment | |
| CN109829593B (en) | Credit determining method and device for target object, storage medium and electronic device | |
| WO2011014422A2 (en) | System and method for generating a valuation of online users and websites from user activities | |
| CN113850416A (en) | Advertisement promotion cooperation object determining method and device | |
| CN114119078A (en) | Target resource determination method, device, electronic 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |