CN106982250B - Information pushing method and device - Google Patents
Information pushing method and device Download PDFInfo
- Publication number
- CN106982250B CN106982250B CN201710125117.6A CN201710125117A CN106982250B CN 106982250 B CN106982250 B CN 106982250B CN 201710125117 A CN201710125117 A CN 201710125117A CN 106982250 B CN106982250 B CN 106982250B
- Authority
- CN
- China
- Prior art keywords
- users
- probability threshold
- target
- probability
- pushed
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Transfer Between Computers (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The present disclosure provides an information pushing method and apparatus, the method including: acquiring a first probability threshold according to the target number of the pushed users; determining a target terminal corresponding to a user to be pushed according to the first probability threshold; and executing push operation aiming at the target terminal. According to the method and the device, the first probability threshold value is obtained according to the target number of the pushed users, the corresponding pushing operation is executed, the probability threshold value is determined and the corresponding pushing operation is executed after the click probability calculation of all the users is completed, and therefore the real-time performance of the pushed information is improved. Meanwhile, the first probability threshold value is updated through the target number of the pushed users, so that the accuracy of the probability threshold value can be improved, and the actual pushing number of all users is closer to the preset pushing number.
Description
Technical Field
The present application relates to the field of communications technologies, and in particular, to an information pushing method and apparatus.
Background
With the continuous development of science and technology and the continuous improvement of living standard, intelligent terminal is more and more popularized. The intelligent terminal has an increasingly large influence on the daily life of people. And personalized pushing can be carried out on the user through the intelligent terminal. For example, news information may be pushed to the user, and so on.
At present, in order to perform personalized push to users, it is necessary to calculate click probabilities of target information to be pushed respectively by all users, then sort the target information according to a sequence of click probabilities from large to small, and push the information to users in the front of the ranking. However, since the personalized recommendation can be performed only after the click probabilities of all users are calculated, the calculation amount is very large, and the real-time performance of pushing is seriously affected.
Disclosure of Invention
In order to overcome the problems in the related art, the present disclosure provides an information pushing method and apparatus.
According to a first aspect of the embodiments of the present disclosure, there is provided an information pushing method, the method including:
acquiring a first probability threshold according to the target number of the pushed users;
determining a target terminal corresponding to a user to be pushed according to the first probability threshold;
and executing push operation aiming at the target terminal.
Optionally, all users are divided into K parts; the determining, according to the first probability threshold, a target terminal corresponding to a user to be pushed includes:
acquiring a K part which does not execute the push operation from the K part according to the sequence from 1 to K, wherein the K belongs to [1, K ];
and determining a target terminal corresponding to the user to be pushed from the kth part of users according to the first probability threshold.
Optionally, the target number is the number of pushed users in all users, or the target number is the number of pushed users in the (k-1) th part.
Optionally, K e (1, K ], the obtaining the first probability threshold according to the target number of the pushed users includes:
acquiring the target number of pushed users;
when the target number is larger than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is greater than the second probability threshold;
when the target number is less than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is smaller than the second probability threshold;
wherein, the N is0Is the desired number of pushed users.
Optionally, if the target number is the number of pushed users in all users,
or
If the target number is the number of pushed users in section k-1,
or
Where M represents the number of all users, MiIndicates the number of i-th partial users, Mk-1Representing the number of users of part k-1 and N representing the number of scheduled pushes among all users.
Optionally, when the target number is greater than N0By the formula h1A second probability threshold of the reconfiguration history;
when the target number is less than N0By the formula h1A second probability threshold of h/a reconfiguration history;
wherein h is1Represents the first probability threshold, h represents the second probability threshold, a represents a preset adjustment factor, a > 1.
Optionally, if k is 1, the obtaining a first probability threshold according to the target number of the pushed users includes:
randomly extracting users with a preset proportion from all users to obtain a sampling user group;
calculating a first click probability of each user in the sampling user group aiming at target information;
determining a first probability threshold based on the first click probability, the predetermined proportion, and a predetermined number of pushes among all users.
Optionally, the determining a first probability threshold according to the first click probability, the predetermined ratio, and a predetermined number of pushes among all users includes:
sequencing the first click probabilities in a descending order to obtain a probability set;
determining a first probability of click that is ranked (N x R) in the set of probabilities as a first probability threshold;
where N represents the predetermined number of pushes among all users and R represents the predetermined ratio.
Optionally, the determining, according to the first probability threshold, a target terminal corresponding to the user to be pushed from the kth user includes:
aiming at each user in the kth part of users, acquiring a second click probability of the user on target information;
and determining the terminal of the user with the second click probability being greater than or equal to the first probability threshold value as a target terminal corresponding to the user to be pushed.
According to a second aspect of the embodiments of the present disclosure, there is provided an information pushing apparatus including:
the threshold value acquisition module is configured to acquire a first probability threshold value according to the target number of the pushed users;
the terminal determining module is configured to determine a target terminal corresponding to a user to be pushed according to the first probability threshold;
a push operation module configured to execute a push operation for the target terminal.
Optionally, all users are divided into K parts; the terminal determination module includes:
the obtaining submodule is configured to obtain a K-th part without executing the pushing operation from the K parts according to the sequence from 1 to K, and the K belongs to [1, K ];
and the terminal determining submodule is configured to determine a target terminal corresponding to the user to be pushed from the kth part of users according to the first probability threshold.
Optionally, the target number is the number of pushed users in all users, or the target number is the number of pushed users in the (k-1) th part.
Optionally, K ∈ (1, K ], the threshold obtaining module includes:
the quantity acquisition submodule is configured to acquire the target quantity of the pushed users;
a threshold reconfiguration sub-module configured to reconfigure the number of the targets when the target number is greater than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is greater than the second probability threshold; when the target number is less than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is smaller than the second probability threshold; wherein, the N is0Is the desired number of pushed users.
Optionally, if the target number is the number of pushed users in all users,
or
If the target number is the number of pushed users in section k-1,
or
Where M represents the number of all users, MiIndicates the number of i-th partial users, Mk-1Representing the number of users of part k-1 and N representing the number of scheduled pushes among all users.
Optionally, the threshold reconfiguration sub-module is specifically configured to:
when the target number is larger than N0By the formula h1A second probability threshold of the reconfiguration history;
when the target number is less than N0By the formula h1A second probability threshold of h/a reconfiguration history;
wherein h is1Represents the first probability threshold, h represents the second probability threshold, a represents a preset adjustment factor, a > 1.
Optionally, if k is equal to 1, the threshold obtaining module includes:
the user group sampling sub-module is configured to randomly extract users with a preset proportion from all the users to obtain a sampling user group;
a probability calculation sub-module configured to calculate a first click probability for each user in the sampling user group with respect to target information;
a threshold determination submodule configured to determine a first probability threshold in dependence on the first click probability, the predetermined proportion and a predetermined number of pushes among all users.
Optionally, the threshold determining sub-module is specifically configured to:
sequencing the first click probabilities in a descending order to obtain a probability set;
determining a first probability of click that is ranked (N x R) in the set of probabilities as a first probability threshold;
where N represents the predetermined number of pushes among all users and R represents the predetermined ratio.
Optionally, the terminal determining submodule is specifically configured to:
aiming at each user in the kth part of users, acquiring a second click probability of the user on target information;
and determining the terminal of the user with the second click probability being greater than or equal to the first probability threshold value as a target terminal corresponding to the user to be pushed.
According to a third aspect of the embodiments of the present disclosure, there is provided an information pushing apparatus including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to:
acquiring a first probability threshold according to the target number of the pushed users;
determining a target terminal corresponding to a user to be pushed according to the first probability threshold;
and executing push operation aiming at the target terminal.
The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects:
according to the method and the device, the first probability threshold value is obtained according to the target number of the pushed users, the corresponding pushing operation is executed, the probability threshold value is determined and the corresponding pushing operation is executed after the click probability calculation of all the users is completed, and therefore the real-time performance of the pushed information is improved. Meanwhile, the click probability distribution of the users to the target information in all the users is uneven, so that the accuracy of the probability threshold value can be improved by updating the first probability threshold value according to the target number of the pushed users, and the actual push number in all the users is closer to the preset push number.
The method divides all users into K parts, obtains a first probability threshold value according to the target number of the pushed users, and executes pushing judgment and corresponding processing on the current group of users by using the first probability threshold value. After all the users in the current group are processed, the probability threshold value can be updated according to the number of the pushed users as the probability threshold value in the next group of information pushing judgment due to the change of the number of the pushed users, so that the probability threshold value is dynamically updated, the accuracy of the probability threshold value is improved, and the actual pushing number in all the users is closer to the preset pushing number.
According to the method and the device, the number of pushed users in all users is used as an influence factor for determining the first probability threshold, and the accuracy of determining the first probability threshold can be improved.
The method compares the target of the pushed user in the processed users with the expected number of the pushed users, and updates the probability threshold according to the comparison result, so that the updated probability threshold is obtained, and the probability threshold is updated rapidly.
The probability threshold is increased by multiplying the probability threshold by the preset adjustment factor larger than 1, and is reduced by dividing the probability threshold by the preset adjustment factor larger than 1, so that the probability threshold is updated quickly.
The present disclosure randomly extracts a predetermined proportion of users from all users, calculates the first click probability of only a part of users, and determines the first probability threshold using the first click probability, which can reduce the amount of calculation. In addition, since the first probability threshold has a high correlation with the target information, the first probability threshold obtained in this way has a higher accuracy than the directly specified threshold.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.
Fig. 1 is a schematic diagram illustrating an information push scenario according to an exemplary embodiment of the present disclosure.
Fig. 2 is a flowchart illustrating an information pushing method according to an exemplary embodiment of the present disclosure.
Fig. 3 is a flow chart illustrating another information pushing method according to an example embodiment of the present disclosure.
Fig. 4 is a block diagram illustrating an information pushing device according to an exemplary embodiment of the present disclosure.
Fig. 5 is a block diagram illustrating another information pushing device according to an example embodiment of the present disclosure.
Fig. 6 is a block diagram illustrating another information pushing device according to an example embodiment of the present disclosure.
Fig. 7 is a block diagram illustrating another information pushing device according to an example embodiment of the present disclosure.
Fig. 8 is a block diagram illustrating an apparatus for information push according to an exemplary embodiment of the present disclosure.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
The terminology used in the present disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present disclosure. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.
As shown in fig. 1, fig. 1 is a schematic diagram illustrating an information pushing scenario according to an exemplary embodiment of the present disclosure. The scenario may include a server and a plurality of user terminals. The server may include a server, a server cluster, a cloud platform, or the like. The user terminal may be an electronic device such as a computer, a smart phone, a tablet computer, a PDA (Personal digital assistant), a multimedia player, and a wearable device. The server can push target information to user terminals of different users, for example, a millet browser can be installed in the user terminal, and a server corresponding to the millet browser can push news information to the user terminal installed with the millet browser, so that a user can conveniently and effectively obtain the news information.
In order to improve the click rate of the target information and avoid inconvenience brought to the user by pushing the target information which is not interested by the user, personalized recommendation can be performed for different users. In the personalized push process, the push can be performed to a plurality of people most interested in the target information. In the related art, in order to realize personalized push, the click probabilities of target information to be pushed by all users are calculated firstly, then the target information is ranked according to the sequence of the click probabilities from large to small, a probability threshold value is determined according to the ranked click probabilities and the designated number of pushed people, and the information is pushed to the users with the click probabilities larger than the probability threshold value.
All users are all possible objects for information push, for example, the users may be users who register a target application program, and the target application program is a program that can perform information push, for example, the target application program may be a millet browser. The amount of all users is often large, for example, there are tens of millions of users, if the click probabilities of all users are calculated, personalized recommendation can be performed, the calculated amount is very large, the time consumed for calculating the click probabilities of all users is very long, and the real-time performance of pushing is seriously affected.
In order to avoid the defect of poor real-time performance of pushed information in the related technology, the method and the device for pushing the information acquire the first probability threshold according to the target number of the pushed users, determine the target terminal corresponding to the user to be pushed according to the first probability threshold, and execute pushing operation aiming at the target terminal. Therefore, the first probability threshold value is obtained according to the target number of the pushed users and corresponding pushing operation is executed, the probability threshold value is determined without waiting until the click probability calculation of all the users is completed, and the corresponding pushing operation is executed, so that the real-time performance of pushing information is improved.
Fig. 2 is a flowchart illustrating an information pushing method according to an exemplary embodiment of the present disclosure, which may be used in a server, and includes the following steps:
in step 201, a first probability threshold is obtained according to the target number of pushed users.
In step 202, a target terminal corresponding to the user to be pushed is determined according to the first probability threshold.
In step 203, a push operation is performed for the target terminal.
In the embodiment of the present disclosure, the pushed user is a user who has performed a pushing operation among the processed users, and the pushing operation may be an operation of pushing target information. The target information is information to be pushed, for example, the target information may be information such as sports news, scientific news, entertainment news, and the like.
The first probability threshold may be a quantile used to determine whether the user is a target user who needs to perform target information pushing, and the user to be pushed may be screened out by using the first probability threshold, so as to push target information to a target terminal of the user to be pushed.
The first probability threshold may be an initial threshold when the number of processed users is zero. And carrying out pushing judgment and corresponding processing by using the initial threshold value. After a part of users are processed, in order to avoid a large difference between the final pushed user number and the preset user number, the probability threshold may be adjusted according to the target number of pushed users in the processed users and the expected number of pushed users, where the adjustment is performed to make the actual pushed number in all users the same as the preset pushed number or make the difference within a specified range.
The processed users are users who have already carried out pushing judgment and corresponding processing, and the processed users comprise pushed users and non-pushed users; an unprocessed user refers to a user who has not performed push determination and corresponding processing. The push determination and corresponding processing may include step 202 and step 203. The target number of pushed users is the number of users who are actually pushed target information among the processed users. The desired number of pushed users is the number of users who desire pushed target information among the processed users. The target number and the expected number are compared, and the probability threshold value can be adjusted according to the comparison result.
Based on this, the present disclosure may factor in the target number of pushed users as a first probability threshold. Specifically, the target number of the pushed users and the expected number of the pushed users may be compared, and the first probability threshold may be obtained according to the comparison result.
After the first probability threshold is determined, for each unprocessed user, the click probability of the unprocessed user on the target information can be obtained; and if the click probability is greater than or equal to the first probability threshold, determining the terminal of the unprocessed user as a target terminal corresponding to the user to be pushed, and executing corresponding pushing operation aiming at the target terminal.
Regarding the click probability, the click probability is the probability that the user views the target information, and the user often views the target information by clicking on the target information, and therefore may be referred to as the view probability or the click probability. Different users have different interest degrees in the target information, so that the click probability is different. For example, a logistic regression model may be obtained through learning based on the information features and the user features, and the probability of clicking on the target information by the user may be calculated by using the logistic regression model. It is to be understood that other manners in the related art may also be adopted for calculating the click probability of the user on the target information, and are not limited herein.
After the click probability of the unprocessed user on the target information is obtained, the click probability can be compared with the first probability threshold, and whether the unprocessed user is the user to be pushed or not is judged according to the comparison result, so that the target terminal corresponding to the user to be pushed is obtained. The target terminal may be terminal identification information. Specifically, if the click probability is greater than or equal to a first probability threshold, determining the terminal of the user as a target terminal, and pushing target information to the target terminal; and if the click probability is smaller than the first probability threshold, judging that the terminal of the user is not the target terminal, and not carrying out pushing processing. This processing procedure may be referred to as push determination and corresponding processing in this embodiment. The push determination and corresponding processing can be performed for each unprocessed user.
It can be understood that, the first probability threshold may be used to perform the pushing judgment and corresponding processing on each unprocessed user, or after performing the pushing judgment and corresponding processing on a part of unprocessed users, the probability threshold may be obtained again according to the target number of the pushed users, and the newly obtained probability threshold is used to perform the pushing judgment and corresponding processing on the remaining unprocessed users.
According to the embodiment, the first probability threshold value is obtained according to the target number of the pushed users and the corresponding pushing operation is executed, the probability threshold value is determined and the corresponding pushing operation is executed after the click probability calculation of all the users is completed, and therefore the real-time performance of information pushing is improved. Meanwhile, the click probability distribution of the users to the target information in all the users is uneven, so that the accuracy of the probability threshold value can be improved by updating the first probability threshold value according to the target number of the pushed users, and the actual push number in all the users is closer to the preset push number.
In an optional implementation manner, all users to be processed may be grouped, after performing push judgment and corresponding processing on each user in a group according to a probability threshold, performing an update operation on the probability threshold according to a target number of the pushed users (step 201), and then performing push judgment and corresponding processing on all users in a next group according to an updated probability threshold (step 202 and step 203).
For example, all users may be divided into K sections, also called K groups. The division mode can be equal division or unequal division. For example, the groups are 20%, 30%, and are set according to the demand.
In one example, the K-th part, K e [1, K ], from which no push operation is performed may be obtained in the order from 1 to K. And determining a target terminal corresponding to the user to be pushed from the kth part of users according to the first probability threshold.
Here, a group/portion in which the push judgment and the corresponding processing are performed is referred to as a group/portion in which the push operation has been performed, and a group/portion in which the push judgment and the corresponding processing are not performed is referred to as a group/portion in which the push operation is not performed. This embodiment divides all users into K parts, and steps 201 to 203 may be performed for each part of users in sequence until all users have completed processing.
And as can be seen, all users are divided into K groups in advance, a first probability threshold value is obtained according to the target number of the pushed users, and pushing judgment and corresponding processing are performed on the users in the current group by using the first probability threshold value. After all the users in the current group are processed, the probability threshold value can be updated according to the number of the pushed users as the probability threshold value in the next group of information pushing judgment due to the change of the number of the pushed users, so that the probability threshold value is dynamically updated, the accuracy of the probability threshold value is improved, and the actual pushing number in all the users is closer to the preset pushing number.
In one example, the target number of pushed users may be the number of pushed users among all users. If the current group is the kth group, the target number of the pushed users is the number of the actual pushed users in the previous k-1 group, and the expected number of the pushed users is the number of the expected pushed users in the previous k-1 group.
The embodiment takes the number of pushed users in all users as the influence factor for determining the first probability threshold, so that the accuracy of determining the first probability threshold can be improved.
In another example, the target number of pushed users may be the number of pushed users in section k-1. If the current group is the kth group, the target number of the pushed users is the actual number of the pushed users in the kth-1 group, and the expected number of the pushed users is the expected number of the pushed users in the kth-1 group.
The embodiment takes the number of the pushed users in the last group of users as the influence factor for determining the first probability threshold, and realizes the updating of the probability threshold.
If k is 1, indicating that the number of processed users is zero, the first probability threshold may be an initial threshold.
In one example, the initial threshold may be a pre-specified probability threshold, and different types of target information may also be set with different click probability thresholds, which may be flexibly set.
Therefore, the initial threshold value is obtained in a preset mode, and the efficiency of obtaining the initial threshold value can be improved.
In another example, in order to improve the correlation between the initial threshold and the target information, the initial threshold may be determined based on the probability of clicking on the target information by the users in the sampled user group, specifically, if k is 1, the obtaining a first probability threshold according to the target number of the pushed users includes:
and randomly extracting a predetermined proportion of users from all users to obtain a sampling user group.
And calculating a first click probability of each user in the sampling user group aiming at target information.
Determining a first probability threshold based on the first click probability, the predetermined proportion, and a predetermined number of pushes among all users.
The predetermined ratio may be a preset sampling ratio, and may be flexibly configured according to the total number of users of all users. In one example, the predetermined ratio may be 1%, and if the total number of users is large, the predetermined ratio may be 0.1%. The extraction user may be an extraction user identifier, for example, a user account number of a preset proportion is extracted from all user names.
Therefore, the calculation amount can be reduced by randomly extracting a predetermined proportion of users from all users, calculating the first click probability of only part of the users, and determining the first probability threshold by using the first click probability. In addition, since the first probability threshold has a high correlation with the target information, the first probability threshold obtained in this way has a higher accuracy than the directly specified threshold.
The first probability threshold is determined according to the first click probability, the predetermined proportion and the predetermined push number of all users, so that when the click probability is filtered by using the first probability threshold, the number of click probabilities which are greater than or equal to the first probability threshold in the click probabilities of all users on the target information can be the same as the predetermined push number or the gap is within a specified range. The specified range is a previously set allowable error range.
One of the determination methods is described below.
Determining a first probability threshold based on the first click probability, the predetermined proportion, and a predetermined number of pushes among all users, comprising:
and sequencing the first click probabilities in a descending order to obtain a probability set.
Determining a first probability of click that is ranked (N R) in the set of probabilities as a first probability threshold.
Where N represents the predetermined number of pushes among all users and R represents the predetermined ratio.
The preset pushing quantity in all the users is the total number of the users which are expected to be screened from all the users and need to push the target information, the preset pushing quantity is multiplied by a preset proportion, the number of the users expected to be pushed in the sampling user group can be determined, and as the first click probability of each user in the sampling user group aiming at the target information is sequenced, the first click probability of the (N × R) th rank in the probability set can be determined as a first probability threshold.
It can be seen that the first probability threshold can be obtained quickly by sorting the first click probabilities in descending order to obtain a probability set, and determining the first click probability ranked (N × R) in the probability set as the first probability threshold.
If K ∈ (1, K ], indicating that the number of processed users is not zero, the probability threshold may be updated based on the target number of pushed users among the processed users and the desired number of pushed users.
In one example, the obtaining the first probability threshold according to the target number of the pushed users includes:
and acquiring the target number of the pushed users.
When the target number is larger than N0And reconfiguring a second probability threshold of the history to obtain a first probability threshold, wherein the first probability threshold is greater than the second probability threshold.
When the target number is less than N0And reconfiguring a second probability threshold of the history to obtain a first probability threshold, wherein the first probability threshold is smaller than the second probability threshold.
Wherein, the N is0Is the desired number of pushed users. The first probability threshold and the second probability threshold are both probability thresholds, and in order to distinguish between the probability thresholds before and after updating, the probability threshold before updating is referred to as the second probability threshold, and the probability threshold after updating is referred to as the first probability threshold.
It can be seen that in this embodiment, if the target number is greater than the desired number, this indicates that redundant users are screened out using the probability threshold, and therefore the probability threshold is adjusted upward. For example, a second probability threshold in the history is updated to a first probability threshold, and the first probability threshold is greater than the second probability threshold. If the target number is less than the desired number, it indicates that the number of users screened out using the probability threshold is insufficient, and therefore the click probability threshold is turned down. For example, a second probability threshold in the history is updated to a first probability threshold, and the first probability threshold is less than the second probability threshold. If the target number is equal to the expected number, which indicates that the number of users screened out by using the click probability threshold is appropriate, the probability threshold does not need to be adjusted, and the first probability threshold is the same as the second probability threshold.
The adjustment amplitude of the probability threshold can be determined by a preset adjustment factor. The adjustment factor for increasing and the adjustment factor for decreasing may be the same or different, and the following description will take the case where the two adjustment factors are the same.
In one example, when the target number is greater than N0By the formula h1A second probability threshold of the reconfiguration history; when the target number is less than N0By the formula h1A second probability threshold of ha reconfiguration history; wherein h is1Represents the first probability threshold, h represents the second probability threshold, a represents a preset adjustment factor, a > 1.
The preset adjusting factor is an adjusting speed control factor, and the larger the factor is, the faster the adjusting speed is, but the more unstable the adjusting speed is, so that the preset adjusting factor can be configured according to requirements. In one example, the preset adjustment factor may be a fixed value that is set in advance. For example, a is 1.1. In another example, the preset adjustment factor may be flexibly configured according to the gap between the target number and the desired number. For example, the larger the gap, the larger the preset adjustment factor, and the smaller the gap, the smaller the preset adjustment factor.
It can be seen that the probability threshold is increased by multiplying the probability threshold by the preset adjustment factor greater than 1, and the probability threshold is reduced by dividing the probability threshold by the preset adjustment factor greater than 1, so as to realize the quick update of the probability threshold.
In another example, the second probability threshold of the history may be reconfigured in other ways, for example, when the target number is greater than N0By the formula h1A second probability threshold of h + a reconfiguration history, when the target number is less than N0By the formula h1H-a reconfiguration history. For other reconfiguration partiesThe description of the formula is omitted here.
Regarding the desired number of pushed users, the desired number is the number of users who desire the pushed target information among the processed users. Several methods of determining the desired amount are listed below.
In one example, if the target number is the number of pushed users in all users, the desired number is the number of desired pushed users in all users.
At this time, the process of the present invention,
n represents the predetermined number of pushes among all users.
It can be seen that this embodiment directly multiplies the predetermined push number of all users by the processed user ratio to quickly obtain the desired number.
In particular, if all users are equally divided into K parts, the formula is adopted
Obtaining the desired number may improve the accuracy of determining the desired number, and thus the accuracy of determining the first probability threshold.
In another example, if the target number is the number of pushed users among all users, the desired number is the number of desired pushed users among the users used.
At this time, the process of the present invention,
m represents the number of all users, MiIndicating the number of users in the ith part and N indicating the number of scheduled pushes among all users.
Since there may be a case that the K part is not equally divided, the ratio of the number of users in each group to the total number of users is determined first
Determining the expected number of users in the group based on the ratio
Then adding the expected number of the processed groups of users
The expected number of pushed users among the processed users can be obtained.
Therefore, the expected quantity obtained by the formula is high in accuracy, and the accuracy of determining the first probability threshold is improved.
In one example, if the target number is the number of pushed users in section k-1, the desired number is the number of desired pushed users in section k-1.
At this time, the process of the present invention,
n represents the predetermined number of pushes among all users.
The present embodiment directly divides the predetermined push number among all users by K, and obtains the desired number quickly. In particular, if all users are equally divided into K parts, the formula is adopted
Obtaining the desired number may improve the accuracy of determining the desired number, and thus the accuracy of determining the first probability threshold.
In another example, if the target number is the number of pushed users in section k-1, the desired number is the number of desired pushed users in section k-1.
At this time, the process of the present invention,
m represents the number of all users, Mk-1Representing the number of users of part k-1 and N representing the number of scheduled pushes among all users.
Since there may be a case that all users are divided into K parts by using an unequal division manner, a ratio of the number of users in the K-1 th part to the total number of users is determined first
Determining the expected number of users in the group based on the ratio
To obtain N0。
Therefore, the expected quantity obtained by the formula is high in accuracy, and the accuracy of determining the first probability threshold is improved.
After the first probability threshold is obtained, aiming at each user in the kth part of users, obtaining a second click probability of the user on the target information; and determining the terminal of the user with the second click probability being greater than or equal to the first probability threshold value as a target terminal corresponding to the user to be pushed, and pushing target information to the target terminal.
The various technical features in the above embodiments can be arbitrarily combined, so long as there is no conflict or contradiction between the combinations of the features, but the combination is limited by the space and is not described one by one, and therefore, any combination of the various technical features in the above embodiments also belongs to the scope disclosed in the present specification.
The present disclosure exemplifies one of the combinations. As shown in fig. 3, fig. 3 is a flowchart of another information pushing method shown in the present disclosure according to an exemplary embodiment, the method includes:
in step 301, a predetermined percentage of users are randomly selected from all users to obtain a sampled user group.
In step 302, a first click probability for each user in the sampled user group with respect to the target information is calculated.
In step 303, a first probability threshold is determined based on the first click probability, the predetermined proportion and a predetermined number of pushes among all users.
In step 304, all users are equally divided into K groups.
In step 305, according to the sequence from 1 to K, a kth group which does not execute the pushing operation is obtained from the K group, a target terminal corresponding to the user to be pushed is determined from the kth group of users according to the first probability threshold, and target information is pushed to the target terminal.
Where K is 1, the first probability threshold used may be the threshold determined in step 303, and K is (1, K), the first probability threshold used is obtained based on the target number and the expected number of pushed users in the previous K-1 group of users that have performed the push operation.
When the target terminal corresponding to the user to be pushed is determined from the kth group of users according to the first probability threshold, a second click probability of the user on target information can be obtained for each user in the kth group of users, and if the second click probability is greater than or equal to the first probability threshold, the terminal of the user is determined as the target terminal of the user to be pushed, and the target information is pushed to the target terminal. And executing the pushing judgment and the corresponding processing on each user of the kth group in sequence until all the users of the kth group complete the processing. As can be seen, for each user, the click probability determination is complete, and the corresponding push processing is also complete.
In step 306, if K ≠ K, after performing push judgment and corresponding processing on all users in the kth group according to the first probability threshold, comparing the target number and the expected number of the pushed users in the previous K-1 group of users that have performed push operation, updating the first probability threshold according to the comparison result, making K equal to K +1, and returning to step 305, so as to perform push judgment and corresponding processing on the users in the next group by using the updated probability threshold.
In step 307, if K is equal to K, the push process ends.
According to the embodiment, the probability threshold is estimated based on the sampling user group, and the probability threshold is updated after each group of users finishes processing, so that the defect that the number of the finally pushed users is greatly different from the number of the preset users is overcome. Meanwhile, pushing judgment and corresponding pushing processing are carried out on each user in each group, and pushing instantaneity is improved.
For convenience of understanding, the following description is given with reference to a specific application example. Assume that the total number of users of all users is M, the predetermined number of push N among all users, and the predetermined ratio is 1%. Randomly selecting 1% of users from all users, obtaining a sampling user group with M/100 number of users, calculating the click probability of each user in the sampling user group to target information, sequencing the click probabilities of all users in the sampling user group according to the size to obtain a probability set, and determining the click probability of the ranking No. N/100 in the probability set as a probability threshold value h.
All users are divided into K equal parts. Aiming at the users in the first group, determining the click probability of the current user on the target information; and comparing the click probability with a click probability threshold h, judging whether the target information is pushed to the user terminal of the current user according to the comparison result, and executing corresponding pushing processing. After all users in the first group carry out pushing judgment according to the probability threshold value h and carry out corresponding processing, the number n of users actually pushed in the first group is determined1And
make a comparison if
The probability threshold is updated to h1H 1.1, if
The probability threshold is updated to h1H/1.1, and a probability threshold h for all users in the second group1And (5) carrying out pushing judgment and carrying out corresponding processing, and so on. When the calculation of the 1 to K-1 groups of users is completed, the number of the users actually pushed is n1～K-1If, if
The probability threshold is updated to hK-1＝hK-21.1, if
The probability threshold is updated to hK-1＝hK-21.1 and according to a probability threshold h for all users in the Kth groupK-1And carrying out pushing judgment and corresponding processing.
Corresponding to the embodiment of the information pushing method, the disclosure also provides an embodiment of an information pushing device and a server applied by the information pushing device.
As shown in fig. 4, fig. 4 is a block diagram of an information pushing apparatus shown in the present disclosure according to an exemplary embodiment, the apparatus including: a threshold obtaining module 41, a terminal determining module 42 and a pushing operation module 43.
Wherein, the threshold obtaining module 41 is configured to obtain the first probability threshold according to the target number of the pushed users.
And a terminal determining module 42 configured to determine a target terminal corresponding to the user to be pushed according to the first probability threshold.
A push operation module 43 configured to perform a push operation for the target terminal.
According to the embodiment, the first probability threshold value is obtained according to the target number of the pushed users and the corresponding pushing operation is executed, the probability threshold value is determined and the corresponding pushing operation is executed without waiting until the click probability calculation of all the users is completed, and therefore the real-time performance of information pushing is improved. Meanwhile, the click probability distribution of the users to the target information in all the users is uneven, so that the accuracy of the probability threshold value can be improved by updating the first probability threshold value according to the target number of the pushed users, and the actual push number in all the users is closer to the preset push number.
As shown in fig. 5, fig. 5 is a block diagram of another information pushing apparatus shown in the present disclosure according to an exemplary embodiment, which is based on the foregoing embodiment shown in fig. 4, and all users are divided into K parts; the terminal determining module 42 includes: an acquisition sub-module 421 and a terminal determination sub-module 422.
Wherein, the obtaining submodule 421 is configured to obtain the kth part, K e [1, K ], where the push operation is not performed from the K part in the order from 1 to K.
A terminal determining submodule 422 configured to determine, according to the first probability threshold, a target terminal corresponding to the user to be pushed from the kth user.
It can be seen from the above embodiments that all users are divided into K parts, the first probability threshold is obtained according to the target number of the pushed users, and the first probability threshold is used to perform pushing judgment and corresponding processing on the current group of users. After all the users in the current group are processed, the probability threshold value can be updated according to the number of the pushed users as the probability threshold value in the next group of information pushing judgment due to the change of the number of the pushed users, so that the probability threshold value is dynamically updated, the accuracy of the probability threshold value is improved, and the actual pushing number in all the users is closer to the preset pushing number.
In an optional implementation manner, the target number is the number of pushed users in all users, or the target number is the number of pushed users in the k-1 th part.
It can be seen from the above embodiments that, by using the number of pushed users in all users as an influencing factor for determining the first probability threshold, the accuracy of determining the first probability threshold can be improved.
As shown in fig. 6, fig. 6 is a block diagram of another information pushing apparatus shown in the present disclosure according to an exemplary embodiment, which is based on the aforementioned embodiment shown in fig. 4, where K e (1, K), and the threshold obtaining module 41 includes a number obtaining sub-module 411 and a threshold reconfiguring sub-module 412.
Wherein, the number obtaining sub-module 411 is configured to obtain the target number of the pushed users.
A threshold reconfiguration sub-module 412 configured to reconfigure the number of targets when the target number is greater than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is greater than the second probability threshold; when the target number is less than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is smaller than the second probability threshold; wherein, the N is0Is the desired number of pushed users.
It can be seen from the above embodiments that the target of the pushed user among the processed users is compared with the expected number of the pushed users, and the probability threshold is updated according to the comparison result, so that the updated probability threshold is obtained, and the probability threshold is rapidly updated.
In an alternative implementation, if the target number is the number of pushed users out of all users,
or
Where M represents the number of all users, MiIndicating the number of users in the ith part and N indicating the number of scheduled pushes among all users.
In an alternative implementation, if the target number is the number of pushed users in section k-1,
or
Where M represents the number of all users, Mk-1Representing the number of users of part k-1 and N representing the number of scheduled pushes among all users.
In an optional implementation manner, the threshold reconfiguration sub-module 412 is specifically configured to:
when the target number is larger than N0By the formula h1A second probability threshold of the reconfiguration history.
When the target number is less than N0By the formula h1H/a second probability threshold of reconfiguration history.
Wherein h is1Represents the first probability threshold, h represents the second probability threshold, a represents a preset adjustment factor, a > 1.
As can be seen from the above embodiments, the probability threshold is increased by multiplying the probability threshold by the preset adjustment factor greater than 1, and the probability threshold is decreased by dividing the probability threshold by the preset adjustment factor greater than 1, so as to achieve fast update of the probability threshold.
As shown in fig. 7, fig. 7 is a block diagram of another information pushing apparatus shown in the present disclosure according to an exemplary embodiment, on the basis of the foregoing embodiment shown in fig. 4, if k is 1, the threshold obtaining module 41 includes: a user group sampling sub-module 413, a probability calculation sub-module 414, and a threshold determination sub-module 415.
The user group sampling sub-module 413 is configured to randomly extract a predetermined proportion of users from all users to obtain a sampled user group.
A probability calculation submodule 414 configured to calculate a first click probability for each user in the sampled user group with respect to the target information.
A threshold determination submodule 415 configured to determine a first probability threshold based on the first click probability, the predetermined proportion, and a predetermined number of pushes among all users.
As can be seen from the above embodiments, it is possible to reduce the amount of calculation by randomly extracting a predetermined proportion of users from all users, calculating the first click probabilities of only a part of the users, and determining the first probability threshold using the first click probabilities. In addition, since the first probability threshold has a high correlation with the target information, the first probability threshold obtained in this way has a higher accuracy than the directly specified threshold.
In an optional implementation manner, the threshold determination sub-module 415 is specifically configured to:
and sequencing the first click probabilities in a descending order to obtain a probability set.
Determining a first probability of click that is ranked (N R) in the set of probabilities as a first probability threshold.
Where N represents the predetermined number of pushes among all users and R represents the predetermined ratio.
In an optional implementation manner, the terminal determining sub-module 422 is specifically configured to:
and acquiring a second click probability of the user on the target information for each user in the kth part of users.
And determining the terminal of the user with the second click probability being greater than or equal to the first probability threshold value as a target terminal corresponding to the user to be pushed.
Correspondingly, the disclosure also provides an information pushing device, which comprises a processor; a memory for storing processor-executable instructions; wherein the processor is configured to:
and acquiring a first probability threshold according to the target number of the pushed users.
And determining a target terminal corresponding to the user to be pushed according to the first probability threshold.
And executing push operation aiming at the target terminal.
The specific details of the implementation process of the functions and actions of each module in the device are referred to the implementation process of the corresponding step in the method, and are not described herein again.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, wherein the modules described as separate parts may or may not be physically separate, and the parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. One of ordinary skill in the art can understand and implement it without inventive effort.
As shown in fig. 8, fig. 8 is a block diagram illustrating an apparatus 800 for pushing information according to an exemplary embodiment.
For example, the apparatus 800 may be provided as a server device. Referring to FIG. 8, system 800 includes a processing component 822, which further includes one or more processors and memory resources, represented by memory 832, for storing instructions, such as applications, that are executable by processing component 822. The application programs stored in memory 832 may include one or more modules that each correspond to a set of instructions. Further, the processing component 822 is configured to execute instructions to perform the above-described information push method.
The system 800 may also include a power component 826 configured to perform power management of the system 800, a wired or wireless network interface 850 configured to connect the system 800 to a network, and an input/output (I/O) interface 858. The system 800 may operate based on an operating system stored in the memory 832.
Wherein the instructions in the memory 832, when executed by the processing component 822, enable the system 800 to perform a method of pushing information, comprising:
and acquiring a first probability threshold according to the target number of the pushed users.
And determining a target terminal corresponding to the user to be pushed according to the first probability threshold.
And executing push operation aiming at the target terminal.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
The above description is only exemplary of the present disclosure and should not be taken as limiting the disclosure, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.
Claims (20)
1. An information pushing method, characterized in that the method comprises:
obtaining a first probability threshold according to a comparison result of a target number of pushed users and an expected number, wherein the target number is the number of users who actually push target information, and the expected number is the number of users who expect to push target information;
determining a target terminal corresponding to a user to be pushed according to the first probability threshold;
and executing push operation aiming at the target terminal.
2. The method of claim 1, wherein all users are divided into K parts;
the determining, according to the first probability threshold, a target terminal corresponding to a user to be pushed includes:
acquiring a K part which does not execute the push operation from the K part according to the sequence from 1 to K, wherein the K belongs to [1, K ];
and determining a target terminal corresponding to the user to be pushed from the kth part of users according to the first probability threshold.
3. The method of claim 2, wherein the target number is a number of pushed users in all users or a number of pushed users in a k-1 part.
4. The method of claim 2, wherein K e (1, K ], the obtaining the first probability threshold according to the comparison of the target number of pushed users and the expected number comprises:
acquiring the target number of pushed users;
when the target number is larger than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is greater than the second probability threshold;
when the target number is less than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is smaller than the second probability threshold;
wherein, the N is0Is the desired number of pushed users.
5. The method of claim 4, wherein if the target number is the number of pushed users of all users,
or
If the target number is the number of pushed users in section k-1,
or
Where M represents the number of all users, MiIndicates the number of i-th partial users, Mk-1Representing the number of users of part k-1 and N representing the number of scheduled pushes among all users.
6. The method of claim 4,
when the target number is larger than N0By the formula h1A second probability threshold of the reconfiguration history;
when the target number is less than N0By the formula h1A second probability threshold of h/a reconfiguration history;
wherein h is1Represents the first probability threshold, h represents the second probability threshold, a represents a preset adjustment factor, a > 1.
7. The method of claim 2, further comprising:
if k is 1, randomly extracting a predetermined proportion of users from all users to obtain a sampling user group;
calculating a first click probability of each user in the sampling user group aiming at target information;
determining a first probability threshold based on the first click probability, the predetermined proportion, and a predetermined number of pushes among all users.
8. The method of claim 7, wherein determining a first probability threshold based on the first click probability, the predetermined proportion, and a predetermined number of pushes among all users comprises:
sequencing the first click probabilities in a descending order to obtain a probability set;
determining a first probability of click that is ranked (N x R) in the set of probabilities as a first probability threshold;
where N represents the predetermined number of pushes among all users and R represents the predetermined ratio.
9. The method according to claim 2, wherein the determining, according to the first probability threshold, the target terminal corresponding to the user to be pushed from the kth user includes:
aiming at each user in the kth part of users, acquiring a second click probability of the user on target information;
and determining the terminal of the user with the second click probability being greater than or equal to the first probability threshold value as a target terminal corresponding to the user to be pushed.
10. An information pushing apparatus, characterized in that the apparatus comprises:
a threshold value obtaining module configured to obtain a first probability threshold value according to a comparison result of a target number of pushed users and an expected number, wherein the target number is the number of users who actually push target information, and the expected number is the number of users who expect to push target information;
the terminal determining module is configured to determine a target terminal corresponding to a user to be pushed according to the first probability threshold;
a push operation module configured to execute a push operation for the target terminal.
11. The apparatus of claim 10, wherein all users are divided into K parts; the terminal determination module includes:
the obtaining submodule is configured to obtain a K-th part without executing the pushing operation from the K parts according to the sequence from 1 to K, and the K belongs to [1, K ];
and the terminal determining submodule is configured to determine a target terminal corresponding to the user to be pushed from the kth part of users according to the first probability threshold.
12. The apparatus of claim 11, wherein the target number is a number of pushed users in all users or a number of pushed users in part k-1.
13. The apparatus of claim 11, wherein K e (1, K ], said threshold acquisition module comprises:
the quantity acquisition submodule is configured to acquire the target quantity of the pushed users;
a threshold reconfiguration sub-module configured to reconfigure the number of the targets when the target number is greater than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is greater than the second probability threshold; when the target number is less than N0Reconfiguring a historical second probability threshold to obtain a first probability threshold, wherein the first probability threshold is smaller than the second probability threshold; wherein, the N is0Is the desired number of pushed users.
14. The apparatus of claim 13, wherein if the target number is the number of pushed users of all users,
or
If the target number is the number of pushed users in section k-1,
or
Where M represents the number of all users, MiIndicates the number of i-th partial users, Mk-1Representing the number of users of part k-1 and N representing the number of scheduled pushes among all users.
15. The apparatus according to claim 13, wherein the threshold reconfiguration sub-module is specifically configured to:
when the target number is larger than N0By the formula h1A second probability threshold of the reconfiguration history;
when the target number is less than N0By the formula h1A second probability threshold of h/a reconfiguration history;
wherein h is1Represents the first probability threshold, h represents the second probability threshold, a represents a preset adjustment factor, a > 1.
16. The apparatus of claim 11, wherein if k is 1, the threshold acquisition module comprises:
the user group sampling sub-module is configured to randomly extract users with a preset proportion from all the users to obtain a sampling user group;
a probability calculation sub-module configured to calculate a first click probability for each user in the sampling user group with respect to target information;
a threshold determination submodule configured to determine a first probability threshold in dependence on the first click probability, the predetermined proportion and a predetermined number of pushes among all users.
17. The apparatus according to claim 16, wherein the threshold determination submodule is specifically configured to:
sequencing the first click probabilities in a descending order to obtain a probability set;
determining a first probability of click that is ranked (N x R) in the set of probabilities as a first probability threshold;
where N represents the predetermined number of pushes among all users and R represents the predetermined ratio.
18. The apparatus according to claim 11, wherein the terminal determination submodule is specifically configured to:
aiming at each user in the kth part of users, acquiring a second click probability of the user on target information;
and determining the terminal of the user with the second click probability being greater than or equal to the first probability threshold value as a target terminal corresponding to the user to be pushed.
19. An information pushing apparatus, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to:
obtaining a first probability threshold according to a comparison result of a target number of pushed users and an expected number, wherein the target number is the number of users who actually push target information, and the expected number is the number of users who expect to push target information;
determining a target terminal corresponding to a user to be pushed according to the first probability threshold;
and executing push operation aiming at the target terminal.
20. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710125117.6A CN106982250B (en) | 2017-03-03 | 2017-03-03 | Information pushing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710125117.6A CN106982250B (en) | 2017-03-03 | 2017-03-03 | Information pushing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106982250A CN106982250A (en) | 2017-07-25 |
| CN106982250B true CN106982250B (en) | 2020-06-02 |
Family
ID=59338729
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710125117.6A Active CN106982250B (en) | 2017-03-03 | 2017-03-03 | Information pushing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106982250B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108171267B (en) * | 2017-12-28 | 2022-03-22 | 阿里巴巴（中国）有限公司 | User group division method and device and message pushing method and device |
| CN111131359B (en) * | 2018-10-31 | 2022-06-07 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating information |
| CN111273939B (en) * | 2020-01-20 | 2023-11-10 | Oppo（重庆）智能科技有限公司 | Information processing method, information processing device and terminal equipment |
| CN112328947B (en) * | 2020-12-31 | 2021-04-16 | 腾讯科技（深圳）有限公司 | Distribution method and device of to-be-recommended media information and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9466077B2 (en) * | 2009-06-30 | 2016-10-11 | Google Inc. | Propagating promotional information on a social network |
| CN105335491B (en) * | 2015-10-20 | 2018-11-09 | 杭州东信北邮信息技术有限公司 | Behavior is clicked come to the method and system of user's Recommended Books based on user |
| CN105915438B (en) * | 2016-04-15 | 2019-02-19 | 北京奇虎科技有限公司 | Information push method, apparatus and system |
-
2017
- 2017-03-03 CN CN201710125117.6A patent/CN106982250B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106982250A (en) | 2017-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109902849B (en) | User behavior prediction method and device, and behavior prediction model training method and device | |
| CN106982250B (en) | Information pushing method and device | |
| CN106651542B (en) | Article recommendation method and device | |
| CN109547546B (en) | Scheduling method of request task and scheduling center server | |
| CN109388674B (en) | Data processing method, device, equipment and readable storage medium | |
| CN109344314B (en) | Data processing method and device and server | |
| CN107968842B (en) | News pushing method, device and equipment based on distributed system | |
| CN110413867B (en) | Method and system for content recommendation | |
| CN110827924B (en) | Clustering method and device for gene expression data, computer equipment and storage medium | |
| CN111914159B (en) | Information recommendation method and terminal | |
| CN109815406B (en) | Data processing and information recommendation method and device | |
| WO2022048289A1 (en) | Content screening method and device | |
| CN112104505A (en) | Application recommendation method and device, server and computer-readable storage medium | |
| CN108667877B (en) | Method and device for determining recommendation information, computer equipment and storage medium | |
| CN112085058A (en) | Object combination recall method and device, electronic equipment and storage medium | |
| CN112148942B (en) | Business index data classification method and device based on data clustering | |
| CN110737458A (en) | code updating method and related device | |
| CN113742578A (en) | Data recommendation method and device, electronic equipment and storage medium | |
| CN114221991A (en) | Big data-based session recommendation feedback processing method and deep learning service system | |
| CN112614010A (en) | Load prediction method and device, storage medium and electronic device | |
| CN106651408B (en) | Data analysis method and device | |
| CN114880363A (en) | Data center flow prediction system, training method and prediction method | |
| CN116339932A (en) | Resource scheduling method, device and server | |
| CN114036391A (en) | Data pushing method and device, electronic equipment and storage medium | |
| CN114519593A (en) | Resource recall model updating method and device, 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 |