patents.google.com

CN104579912B - Data pushing method and device - Google Patents

  • ️Tue Apr 07 2020

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 embodiment of the invention provides a data pushing method, which is applied to a network side server, wherein the network side server is used for maintaining the normal operation of instant messaging software, and as shown in figure 1, the method comprises the following steps:

step

101, obtaining a target user identifier, wherein the target user identifier is a user identifier with a login state being an online state.

And the network side server monitors and counts the online and offline states of all the user identifications through the user identification pool. The user identification pool can represent the login state of each user identification through an array of 0 and 1, wherein 0 represents offline and 1 represents online. In order to achieve the purpose that only a user who wants to be online sends push information, the user identification pool needs to be traversed through a round-robin training mode and the like, and then the user identification with the login state being the online state, namely the target user identification, is obtained. By controlling the time of polling, the time interval during which the push information can be sent is reduced. When the polling interval is 1 hour, push information will be received every 1 hour for online users.

And 102, acquiring attribute information of the target user identifier.

Attribute information of the target user identifier, which is used to indicate attribute information of the user represented by the target user identifier, for example: age, city, birthday, nature of work, hobbies, time to network entry, etc.

And 103, determining target push information from a preset push information set according to the operation records corresponding to the attribute information and the target user identification, wherein the relevance of the target push information, the operation records and the attribute information is high.

Wherein the operation record comprises at least one of the following records: web browsing records, consumption records, and time-frequency playing records.

The network side server records the operation record of the user identification besides the attribute information of the user identification. The targeted push information may be determined in three ways:

1. and determining target push information from a preset push information set according to the relevance of the attribute information and the operation record corresponding to the target user identification.

And determining a comparison user identifier with high correlation with the attribute information of the target user identifier, calculating a difference part between the operation record of the comparison user identifier and the operation record of the target user identifier, and determining target push information from the difference part.

2. And determining target push information from a preset push information set according to the relevance of the attribute information and the operation record corresponding to the target user identifier.

And finding out the content corresponding to each operation record corresponding to the target user identification, then finding out another content associated with the content, and determining push information corresponding to the another content from a preset push information set. One of the determined pieces of push information is selected as the target push information.

3. And determining the integral push proportion of each preset push message according to the proportion of the attribute information of the target user identification in each preset push message, and determining the push message with higher integral push proportion as the target push message.

Finding out the content corresponding to each preset push information in the preset push information set, acquiring the browsing record of all users for each content, finding out the proportion of the users in the browsing record, which is consistent with the attribute information of the target user newspaper, in the browsing record to obtain the push proportion corresponding to each item of attribute information, and then calculating the push proportion corresponding to each item of attribute information to obtain the integral push proportion of the preset push information for the target user identification. And determining the preset push information with the highest overall push proportion as target push information.

It should be noted that, the content described in the embodiment of the present invention may be, but is not limited to, any one of the following data: video data, news data, commodity data.

And

step

104, sending the target push information to the terminal logged with the target user identifier.

And determining a terminal logged with the target user identifier according to the login address of the target user identifier, and sending target push information to the terminal.

The data pushing method provided by the embodiment of the invention can acquire the target user identification of the online user, determines the target pushing information from the preset pushing information set according to the attribute information of the target user identification and the operation record corresponding to the target user identification, has higher relevance between the target pushing information and the operation record and the attribute information, and sends the target pushing information to the terminal logged with the target user identification. In the prior art, push information is sent to a terminal according to a random algorithm or a fixed sending relation, different push information adaptive to a user cannot be pushed according to different users, interference on normal use of the user is caused, and system resources are wasted. In the invention, the relevance of the target push information, the operation record and the attribute information is higher, so that the target push information possibly required by the user can be determined according to the operation record of the user and the attribute information of the user identification, the user can receive the interested push information with higher probability, and the waste of system resources is avoided.

During the use process, the following phenomena often occur: when a user logs in the instant messaging software on a computer, the user can pay attention to whether messages and other unprocessed information exist. After logging in for a period of time, the user leaves the computer for rest, the login state displayed in the network side server is still in an online state at the moment, but the user leaves the computer at the moment, and if the push information is sent to the user at the moment, the user cannot know the push information immediately, so that the instantaneity of the push information is reduced. To this end, an embodiment of the present invention further provides a data pushing method, which is a specific description of the method shown in fig. 1, and as shown in fig. 2,

step

101 is to obtain a target user identifier, where the target user identifier is a user identifier whose login state is an online state, and the method may be implemented in the following manner:

step

201, detecting the login state of each user identifier in the user identifier pool in real time.

Step

202, when the login state of the user identifier in the user identifier pool is converted from the offline state to the online state, acquiring the user identifier of which the login state is converted from the offline state to the online state, wherein the user identifier is a target user identifier.

By detecting the login state of each user identifier, the user identifier of which the login state is converted from the offline state to the online state can be obtained. The login state is converted from the off-line state to the on-line state to show that the user just logs in, and at the moment, information is sent to the detected user identifier, so that the purpose of sending push messages to the user just logging in the instant chat software can be achieved.

The data push method provided by the embodiment of the invention can send push information to the user who just logs in, so that the reading proportion of the push information is improved.

As a specific description of the method shown in fig. 1,

step

103 is to determine target push information from a preset push information set according to an operation record corresponding to attribute information and a target user identifier, where the target push information has a high association with the operation record and the attribute information and can be implemented in any one of the following three ways:

1. determining target push information from a preset push information set according to the relevance of the attribute information and the operation record corresponding to the target user identification;

or 2, determining target push information from a preset push information set according to the relevance of the attribute information and the operation record corresponding to the target user identifier;

or 3, determining the overall push proportion of each preset push message according to the proportion of the attribute information of the target user identifier in each preset push message, and determining the push message with higher overall push proportion as the target push message.

As shown in fig. 3, the mode 1 can be implemented by the following steps:

step

301, searching a user identification list for a comparison user identification having similar attribute information with the target user identification.

The user identification list contains all user identifications including the user identification of the online user and the user identification of the offline user. The attribute information having similar attribute information is that the attribute values of some parts in the attribute information are the same.

And extracting one attribute information of the target user identifier, and searching a comparison user identifier which is the same as the extracted attribute information in the user identifier list. For example: the extracted attribute information is the age, and the age attribute value of the target user identifier is 27. In the list of user identifications at least one comparison user identification with an age attribute value of 27 is looked up.

Further, a plurality of attribute information of the target user identifier can be obtained, and a comparison user identifier which is the same as the extracted plurality of attribute information is searched in the user identifier list. For example: the extracted attribute information is age and gender, the age attribute value of the target user identifier is 27, and the gender is male. In the user identification list, at least one comparison user identification with the age attribute value of 27 and the gender of male is searched.

And 302, acquiring an operation record set corresponding to the user identification.

And according to the Internet access records recorded in the network side server, an operation record set consisting of at least one operation record corresponding to each comparison user identifier can be obtained.

Step

303, searching for at least one operation record which is not included in the operation record set corresponding to the target user identifier from the operation record set.

And comparing the operation record set obtained by comparing the user identification with the operation record set corresponding to the target user identification to obtain at least one operation record which is not contained in the operation record set corresponding to the target user identification. At least one operation record which is not included in the operation record set corresponding to the target user identification is the operation record which may be concerned by the user.

Step

304, selecting a target operation record from the at least one operation record.

Since only one push message is displayed at a time when the push message is displayed on the terminal, it is necessary to select a target operation record from at least one operation record. Since each operation record in the at least one operation record may be focused on by the user, a specific selection manner is not limited herein, for example: a random selection mode, an operation record with the highest occurrence rate and the like.

And 305, selecting target push information from a preset push information set according to the content corresponding to the target operation record.

The content corresponding to the target operation record is the main content to be represented by the push information. For example: and if the target operation record is the shopping record, the purchase object is the content corresponding to the target operation record. And if the target operation record is a video browsing record, the browsed video data is the content corresponding to the target operation record. And if the target operation record is a news browsing record, the theme or the type of the news is the content corresponding to the target operation record.

The method shown in fig. 3 is further explained below by means of an application scenario:

as shown in fig. 4, the user identifier C is a target user identifier, the age attribute value of the target user identifier is 27 gender attribute values, and the corresponding operation record is a shopping record, specifically, a purchased commodity C. The user identifier list comprises a user identifier A and a user identifier B, wherein: the age attribute value of the user identifier a is 27, the gender attribute value is female, the corresponding operation records are purchase commodity a and commodity C, the age attribute value of the user identifier B is 29, the gender attribute value is female, and the corresponding operation records are purchase commodity a and commodity C. At this time, the user identifier having the same gender attribute value and age attribute value is used as the comparison user identifier. Since the user identifier a and the user identifier C have a gender attribute value and an age attribute value, the user identifier a is determined to be a comparison user identifier (the user identifier a is similar to the target user identifier C). The operation record of the user identifier a is the purchase of the article a and the article C, and the operation record of the user identifier C is the purchase of the article C, so that the operation record of the purchase of the article a is determined as the target operation record. And determining preset push information for describing the commodity A as target push information.

In addition to determining the target push information according to the correlation between the attribute information of the target user identifier and the attribute information of the comparison user identifier, the target push information may also be determined according to the correlation between the push information corresponding to the operation record corresponding to the target user identifier and the preset push information in the preset push information set, as shown in fig. 5, and the method 2 may be implemented by the following steps:

and 501, acquiring an operation record set corresponding to the target user identifier.

Step

502, in a preset push information set, at least one first comparison push information corresponding to each operation record in an operation record set is searched.

And determining the first contrast push information according to the content corresponding to the operation record. The attribute information in the main content of the first contrast push information is identical to the content corresponding to the operation record.

Step

503, determining at least one second contrast push information having similar attributes to each found first contrast push information in the preset push information set.

The partial attribute information in the main body content of the second comparative push information coincides with the partial attribute information of the main body content of the first comparative push information. For example: the content corresponding to the operation record is a mobile phone A, the manufacturer attribute of the mobile phone A is a manufacturer B, and the price attribute is 2000. And finding out the push information with the manufacturer attribute of manufacturer B and the price attribute of 2000 from the preset push set as first comparison push information. In order to improve accuracy, the types of attribute information may be increased, for example, a mobile phone system attribute value, a screen size attribute value, and the like may be referred to in addition to the vendor attribute value and the price attribute value.

Step

504, at least one third comparison push information which is not included in the operation record set is found out from the at least one second comparison push information.

Each second comparison push message is a message that the user pays attention to, and therefore, the manner for finding out the at least one third comparison push message that is not included in the operation record set from the at least one second comparison push message may be, but is not limited to, a random selection manner, and the like.

Step

505, selecting a target push information from the at least one third comparative push information.

The method shown in fig. 5 is further explained below by means of an application scenario:

as shown in fig. 6, the user identifier C is a target user identifier, the corresponding operation record is a shopping record, and the specific content is a purchased commodity C. According to the content of the purchased commodity C, the first comparative push information having the same attribute information as the commodity C can be determined from the preset push information list, and the main content of the push information is the commodity C. And searching at least one second comparison push information which is the same as the manufacturer attribute value and the price attribute value of the commodity C in the preset push information list, and if the main content of the second preset push information obtained at this time is the commodity B and the commodity C, selecting one of two comparison push information of which the main content is the commodity B and the commodity C respectively as third comparison push information, for example, the main content is the third comparison push information of the commodity B. And sending the third comparison push information to the terminal.

Further, the following further describes the method 3 by an application scenario:

the attribute information of the target user identifier includes gender attribute information, age attribute information, region attribute information and purchased commodity attribute information, which are male, 23 years old, Guangdong and commodity C respectively. And respectively calculating the proportion of the main content of the preset push information corresponding to each attribute information. If 2 pieces of preset push information are currently provided, namely the commodity A and the commodity B, the click rates of the commodity A on the male, the 23 year old, the Guangdong and the commodity C are respectively calculated, and the statistical results are shown in table 1:

TABLE 1

Attribute information Attribute value Click rate in Commodity A Click rate in item B
Sex For male 0.8718 0.1740
Age (age) 23 0.2416 0.908
Region of origin Guangdong (Chinese character of Guangdong) 0.413 0.553
Shopping records Commodity C 0.1 0.03

Bayesian formula states that: the probability of each independent event occurring simultaneously is proportional to the product of the probabilities of each event occurring. The click rates of the attribute information in the commodity A in the table 1 are multiplied to obtain the overall pushing proportion of 0.0087 of the commodity A, and similarly, the overall pushing proportion of 0.0026 of the commodity B is obtained, and since 0.0087 is greater than 0.0026, the target user identification is higher in relevance of the commodity A than the commodity B. And determining preset push information with the commodity A as the main content as target push information.

The data pushing method provided by the embodiment of the invention can determine the target pushing information through three modes, so that the pushing information has higher correlation with the operation record or attribute information of the user, and the pushing efficiency is improved.

As a specific description of the method shown in fig. 1,

step

104, sending the target push information to the terminal logged with the target user identifier, which may be implemented in the following manner:

if the quota amount of the target push information is larger than zero, sending the target push information to a terminal logged with the target user identifier, wherein the quota amount is used for the remaining times of sending the target push information;

and modifying the quota amount according to the sending amount of the target push information.

And if the quota amount of the target push information is equal to zero, canceling to send the target push information to the terminal logged with the target user identifier, and re-determining the target push information, wherein the re-determined target push information is different from the current target push information. And if the target push information cannot be determined again, canceling sending the target push information to the terminal.

The data pushing method provided by the embodiment of the invention can realize sending the pushing information according to the quota amount by modifying the quota amount, thereby ensuring that the same pushing information sent to each terminal does not exceed the scope of the quota amount.

As shown in fig. 7, before sending the target push information to the terminal logged with the target user identifier in

step

104, the method further includes:

step

701, acquiring the quantity of target push information received by a target user identifier within a preset time.

The preset time and the preset times can control the times of sending the target push information to the terminal by the network side server. The preset time is 1 day, 5 days and 7 days, and the preset times are 1 time and 3 times.

And step 702, if the number is larger than the preset number, canceling to send the target push information to the terminal logged with the target user identifier.

The data pushing method provided by the embodiment of the invention can cancel sending the target pushing information to the terminal logged with the target user identification when the number of the target pushing information received by the target user identification in the preset time is more than the preset times, thereby further reducing the interference on the normal operation of the user.

An embodiment of the present invention further provides a data pushing apparatus, where the apparatus is configured to execute the methods shown in fig. 1 to fig. 7, and as shown in fig. 8, the

apparatus

81 includes:

an obtaining

unit

811, configured to obtain a target user identifier, where the target user identifier is a user identifier whose login status is an online status;

the obtaining

unit

811 is further configured to obtain attribute information of the target user identifier;

a determining

unit

812, configured to determine, according to the attribute information obtained by the obtaining

unit

811 and an operation record corresponding to the target user identifier, target push information from a preset push information set, where a relevance between the target push information and the operation record and the attribute information is higher;

a sending

unit

813, configured to send the target push information determined by the determining

unit

812 to a terminal logged with the target user identifier.

Further, as shown in fig. 9, the obtaining

unit

811 includes:

a

detection subunit

8111, configured to detect, in real time, a login state of each user identifier in the user identifier pool;

an obtaining

subunit

8112, configured to, when the detecting

subunit

8111 detects that the login state of the user identifier in the user identifier pool is converted from the offline state to the online state, obtain the user identifier of which the login state is converted from the offline state to the online state, where the user identifier is a target user identifier.

Further, the determining

unit

812 includes:

a first determining

subunit

8121, configured to determine, according to the relevance of the attribute information and the operation record corresponding to the target user identifier, target push information from a preset push information set;

a second determining

subunit

8122, configured to determine, according to the association between the attribute information and the operation record corresponding to the target user identifier, target push information from a preset push information set;

a third determining

subunit

8123, configured to determine an overall pushing ratio of each preset pushing information according to a ratio of the attribute information of the target user identifier in each preset pushing information, and determine the pushing information with a higher overall pushing ratio as the target pushing information.

Further, the first determining

subunit

8121 is further configured to:

searching a user identification list for a comparison user identification with similar attribute information to the target user identification;

acquiring an operation record set corresponding to the comparison user identifier;

searching at least one operation record which is not contained in the operation record set corresponding to the target user identification from the operation record set;

selecting a target operation record from the at least one operation record;

and selecting target push information from a preset push information set according to the content corresponding to the target operation record.

Further, the second determining

subunit

8122 is further configured to:

acquiring an operation record set corresponding to the target user identifier;

searching at least one first comparison push information corresponding to each operation record in the operation record set in a preset push information set;

determining at least one second comparative push information with similar attributes to each of the searched first comparative push information in the preset push information set;

finding out at least one third comparison push information which is not contained in the operation record set from the at least one second comparison push information;

selecting a target push information from the at least one third comparative push information.

Further, the sending

unit

813 includes:

a sending subunit 8131, configured to send the target push information to a terminal logged with the target user identifier when a quota amount of the target push information is greater than zero, where the quota amount is used for the remaining number of times of sending the target push information;

a modifying subunit 8132, configured to modify the quota amount according to the sending amount of the target push information.

Further, the obtaining

unit

811 is further configured to obtain the number of the target push information received by the target user identifier within a preset time;

the modifying sub-unit 8132 is further configured to cancel sending the target push information to the terminal logged with the target user identifier when the number acquired by the acquiring

unit

811 is greater than a preset number of times.

It will be clear to those skilled in the art that, for convenience and simplicity of description, the foregoing division of the functional modules is merely used as an example, and in practical applications, the above function distribution may be performed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to perform all or part of the above described functions. For the specific working processes of the system, the apparatus and the unit described above, reference may be made to the corresponding processes in the foregoing method embodiments, and details are not described here again.

The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.