Information processing method and apparatus, and computer-readable storage medium

1 . An information processing method, applied to a server, comprising:

counting feedback data of historical push information, the feedback data comprising at least exposure data and click data;

generating a first probability distribution of a click-through rate of each piece of push information in the historical push information based on the exposure data and the click data;

determining first predicted click-through rates of to-be-pushed push information according to the first probability distribution, and selecting a preset quantity of pieces of first push information from the to-be-pushed push information according to the first predicted click-through rates;

using a preset target click-through rate prediction model to obtain a target predicted click-through rate of each of the pieces of first push information; and

selecting target push information for pushing from the pieces of first push information according to the target predicted click-through rate, wherein selecting target push information comprises:

obtaining a target exposure data by summing the exposure data of each of the pieces of first push information;

determining whether the target exposure data is lower than a preset level;

in response to the target exposure data being lower than the preset level, selecting the target push information for pushing by a random selection process based on the target predicted click-through rate; and

in response to the target exposure data not being lower than the preset level, selecting the target push information for pushing based on a highest target predicted click-through rate.

2 . The information processing method according to claim 1 , wherein generating the first probability distribution comprises:

obtaining priori distribution information corresponding to a click-through rate of the historical push information, and generating a first probability distribution corresponding to the click-through rate of each piece of push information according to the priori distribution information;

counting click data and non-click data generated in response to each piece of push information being exposed, to generate posteriori distribution information corresponding to the click-through rate; and

adjusting the first probability distribution according to the posteriori distribution information corresponding to the click-through rate, to obtain the first probability distribution corresponding to the click-through rate of each piece of push information.

3 . The information processing method according to claim 1 , wherein selecting the preset quantity of pieces of first push information according to the first predicted click-through rates comprises:

selecting the preset quantity of pieces of first push information in descending order of the first predicted click-through rates.

4 . The information processing method according to claim 1 , wherein the feedback data further comprises conversion data, and the method further comprises:

before selecting the preset quantity of pieces of first push information according to the first predicted click-through rates, obtaining target conversion data of each piece of push information;

in response to detecting that the target conversion data is less than a first preset threshold, selecting the preset quantity of pieces of first push information according to the first predicted click-through rates;

in response to detecting that the target conversion data is not less than the first preset threshold, counting conversion data and non-conversion data generated in response to each piece of push information being clicked, to generate posteriori distribution information corresponding to a conversion rate;

generating a second probability distribution corresponding to a conversion rate of each piece of push information according to the posteriori distribution information corresponding to the conversion rate; and

selecting the preset quantity of pieces of first push information by combining the first probability distribution and the second probability distribution.

5 . The information processing method according to claim 4 , wherein selecting the preset quantity of pieces of first push information by combining the first probability distribution and the second probability distribution comprises:

obtaining a predicted click-through rate of each piece of push information according to the first probability distribution;

obtaining a predicted conversion rate of each piece of push information according to the second probability distribution;

combining the predicted click-through rate and the predicted conversion rate of each piece of push information to obtain a combination rate; and

selecting the preset quantity of pieces of first push information in descending order of combination rates.

6 . The information processing method according to claim 4 , wherein the feedback data further comprises virtual expanse data, and the method further comprises:

calculating virtual cost data of each piece of push information according to the virtual expanse data and the conversion data; and

freezing push information whose virtual cost data is greater than preset virtual data.

7 . The information processing method according to claim 1 , wherein selecting the target push information comprises:

in response to detecting that the target exposure data is less than a second preset threshold, normalizing the target predicted click-through rate to obtain target predicted vector information of a preset quantity of dimensions;

dividing probability intervals based on the target predicted vector information, randomly accessing the probability intervals, and determining push information corresponding to an accessed probability interval as the target push information for pushing; and

in response to detecting that the target exposure data is not less than the second preset threshold, determining push information with the highest target predicted click-through rate as the target push information for pushing.

8 . The information processing method according to claim 1 , further comprising:

acquiring update information of the push information according to a preset cycle; and

performing an update operation on the push information according to the update information.

9 . An information processing apparatus, comprising:

a memory storing a plurality of instructions; and

a processor configured to execute the plurality of instructions, wherein upon execution of the plurality of instructions, the processor is configured to:

count feedback data of historical push information, the feedback data comprising at least exposure data and click data;

generate a first probability distribution corresponding to a click-through rate of each piece of push information in the historical push information based on the exposure data and the click data;

determine first predicted click-through rates of to-be-pushed push information according to the first probability distribution, select a preset quantity of pieces of first push information from the to-be-pushed push information according to the first predicted click-through rates;

use a preset target click-through rate prediction model to obtain a target predicted click-through rate of each of the pieces of first push information; and

select target push information for pushing from the pieces of first push information according to the target predicted click-through rate, wherein the processor is further configured to:

obtain a target exposure data by summing the exposure data of each of the pieces of first push information;

determine whether the target exposure data is lower than a preset level;

in response to the target exposure data being lower than the preset level, select the target push information for pushing by a random selection process based on the target predicted click-through rate; and

in response to the target exposure data not being lower than the preset level, selecting the target push information for pushing based on a highest target predicted click-through rate.

10 . The information processing apparatus according to claim 9 , wherein the processor, in order to generate the first probability distribution, is configured to execute the plurality of instructions to:

obtain priori distribution information corresponding to a click-through rate of the historical push information, and generate a first probability distribution corresponding to the click-through rate of each piece of push information according to the priori distribution information;

count click data and non-click data generated in response to each piece of push information being exposed, to generate posteriori distribution information corresponding to the click-through rate; and

adjust the first probability distribution according to the posteriori distribution information corresponding to the click-through rate, to obtain the first probability distribution corresponding to the click-through rate of each piece of push information.

11 . The information processing apparatus according to claim 9 , wherein the processor, in order to select the preset quantity of pieces of first push according to the first predicted click-through rates, is configured to execute the plurality of instructions to:

select the preset quantity of pieces of first push information in descending order of the first predicted click-through rates.

12 . The information processing apparatus according to claim 9 , wherein the feedback data further comprises conversion data, and wherein the processor, upon execution of the plurality of instructions, is further configured to:

obtain target conversion data of each piece of push information;

in response to detecting that the target conversion data is less than a first preset threshold, select the preset quantity of pieces of first push information according to the first predicted click-through rates; and

in response to detecting that the target conversion data is not less than the first preset threshold, count conversion data and non-conversion data generated in response to each piece of push information being clicked, to generate posteriori distribution information corresponding to a conversion rate;

generate a second probability distribution corresponding to a conversion rate of each piece of push information according to the posteriori distribution information corresponding to the conversion rate; and

select the preset quantity of pieces of first push information by combining the first probability distribution and the second probability distribution.

13 . The information processing apparatus according to claim 12 , wherein the processor, in order to select the preset quantity of pieces of first push information by combining the first probability distribution and the second probability distribution, is configured to:

obtain a predicted click-through rate of each piece of push information according to the first probability distribution;

obtain a predicted conversion rate of each piece of push information according to the second probability distribution;

combine the predicted click-through rate and the predicted conversion rate of each piece of push information to obtain a combination rate; and

select the preset quantity of pieces of first push information in descending order of combination rates.

14 . The information processing apparatus according to claim 12 , wherein the feedback data further comprises virtual expanse data, and wherein the processor, upon execution of the plurality of instructions, is further configured to:

calculate virtual cost data of each piece of push information according to the virtual expanse data and the conversion data; and

freeze at least one piece of the push information whose virtual cost data is greater than preset virtual data.

15 . The information processing apparatus according to claim 9 , wherein the processor, in order to select the target push information for pushing, is configured to execute the plurality of instructions to:

in response to detecting that the target exposure data is less than a second preset threshold, normalize the target predicted click-through rate to obtain target predicted vector information of a preset quantity of dimensions;

divide probability intervals based on the target predicted vector information, randomly access the probability intervals, and determine push information corresponding to an accessed probability interval as the target push information for pushing; and

in response to detecting that the target exposure data is not less than the second preset threshold, determine push information with the highest target predicted click-through rate as the target push information for pushing.

16 . The information processing apparatus according to claim 9 , wherein the processor, upon execution of the plurality of instructions, is further configured to:

acquire update information of the push information according to a preset cycle; and

perform an update operation on the push information according to the update information.

17 . A non-transitory computer-readable storage medium storing a plurality of instructions, the instructions configured to be executed by a processor, and upon execution by the processor, cause the processor to:

count feedback data of historical push information, the feedback data comprising at least exposure data and click data;

generate a first probability distribution corresponding to a click-through rate of each piece of push information in the historical push information based on the exposure data and the click data;

determine first predicted click-through rates of to-be-pushed push information according to the first probability distribution, select a preset quantity of pieces of first push information from the to-be-pushed push information according to the first predicted click-through rates;

use a preset target click-through rate prediction model to obtain a target predicted click-through rate of each of the pieces of first push information; and

select target push information for pushing from the pieces of first push information according to the target predicted click-through rate, wherein the instructions further cause the processor to:

obtain a target exposure data by summing the exposure data of each of the pieces of first push information;

determine whether the target exposure data is lower than a preset level;

in response to the target exposure data being lower than the preset level, select the target push information for pushing by a random selection process based on the target predicted click-through rate; and

in response to the target exposure data not being lower than the preset level, selecting the target push information for pushing based on a highest target predicted click-through rate.

18 . The non-transitory computer-readable storage medium of claim 17 , wherein the instructions, in order to cause the processor to generate the first probability distribution, is configured to cause the processor to:

obtain priori distribution information corresponding to a click-through rate of the historical push information, and generate a first probability distribution corresponding to the click-through rate of each piece of push information according to the priori distribution information;

count click data and non-click data generated in response to each piece of push information being exposed, to generate posteriori distribution information corresponding to the click-through rate; and

adjust the first probability distribution according to the posteriori distribution information corresponding to the click-through rate, to obtain the first probability distribution corresponding to the click-through rate of each piece of push information.