CN112668902A

CN112668902A - Resource allocation method and device, electronic equipment and storage medium

Info

Publication number: CN112668902A
Application number: CN202011628339.8A
Authority: CN
Inventors: 王超; 黄�俊; 王健鑫
Original assignee: Shanghai Zhongyuan Network Co ltd
Current assignee: Shanghai Zhongyuan Network Co ltd
Priority date: 2020-12-31
Filing date: 2020-12-31
Publication date: 2021-04-16
Anticipated expiration: 2040-12-31
Also published as: CN112668902B

Abstract

The application relates to a resource allocation method, a resource allocation device, an electronic device and a storage medium, wherein the resource allocation method comprises the following steps: determining a target access total amount of a target object reached in a target access time period; determining a target accumulated visit amount of each time slice based on the target visit time period and the target visit total amount; determining the current accumulated access amount of the target object from the starting time point of the target access time period to the ending time point of the current time slice; determining the target access amount of the next time slice based on the target accumulated access amount of the next time slice and the current accumulated access amount; and determining that the display bit meeting the target access amount of the next time slice is the target display bit of the next time slice. By the method and the device, the target display position corresponding to the next time slice of the target object can be directly determined according to the real-time access condition of the target object; the display position of the target object can be adjusted in real time; therefore, better guarantee is provided for the target object to reach the target access total amount in the target access time period.

Description

Resource allocation method and device, electronic equipment and storage medium

Technical Field

The present application relates to the field of data recall technologies, and in particular, to a resource allocation method and apparatus, an electronic device, and a storage medium.

Background

In an internet platform for providing content, when a certain content is newly put on shelf, in order to increase the access amount, the content newly put on shelf needs to be reserved, taking a video website as an example: when a certain video is newly put on shelf, the video needs to be protected and popularized, so that the video can obtain enough access, and more flow, user residence time and benefits are brought to a platform through the video.

The insurance policy is a policy for insurance of the recall source, namely: and simultaneously carrying out the policy of keeping the amount of a plurality of objects recalled by one or a plurality of label information. In addition, the reserve policy in the related art generally directly gives a weight to the ranking score of the videos that need to be reserved (the higher the ranking score is, the videos can be distributed to the display positions (positions in the page for displaying the content) with higher exposure), in a weighting manner, for example: the current ranking score of the video a is 1 score, and after twice weighting, the ranking score is 2 score, so that the video with the amount required to be saved is possibly distributed to the display position with higher exposure, therefore, although the method can perform weighting processing on the object needing to be weighted, as shown in the illustrated example, when the ranking score of the video is low, even if the weighting processing is performed, the score is lower than that of other objects, and therefore, the video is difficult to be distributed to the display position with higher exposure.

In view of the technical problems in the related art, no effective solution is provided at present.

Disclosure of Invention

In order to solve the technical problem or at least partially solve the technical problem, the present application provides a resource allocation method, apparatus, electronic device and storage medium.

In a first aspect, an embodiment of the present application provides a resource allocation method, including:

determining a target access total amount of a target object reached in a target access time period; wherein the target access time period consists of at least two time slices set consecutively;

determining a target accumulated visit amount of each time slice based on the target visit time period and the target visit total amount; the target accumulated visit volume of the time slice is the visit volume which needs to be reached by the target object from the starting time point of the target visit time period to the ending time point of the time slice;

determining the accumulated access volume of the target object from the starting time point of the target access time period to the ending time point of the current time slice as the current accumulated access volume;

determining the target access amount of the next time slice based on the target accumulated access amount of the next time slice and the current accumulated access amount; wherein, the target visit volume is the visit volume which needs to be reached by the target object in the next time slice;

and determining that the display position meeting the target access amount of the next time slice is the target display position of the next time slice, wherein the display position is a position resource used for displaying the target object on a display interface.

Optionally, as in the foregoing method, the determining the target cumulative visit amount for each time slice includes:

determining the total historical visit amount of all the display bits reached in the historical visit time period and the historical accumulated visit amount of each historical visit time slice; wherein the historical access time period consists of at least two continuously set historical access time slices;

determining the expected target visit total amount of all the display bits in the target visit time period and the expected accumulated visit amount of each time slice based on the corresponding relation between the historical visit time period and the target visit time period and the historical visit total amount; the expected accumulated access amount of the time slice is the access amount which can be expected to be reached by the display bit from the starting time point of the historical access time period to the ending time point of the time slice;

and determining the target accumulated visit amount of each time slice according to the corresponding relation between the target visit total amount and the expected accumulated visit amount of each time slice.

Optionally, as in the foregoing method, before determining that the presentation bits meeting the target access amount of the next time slice are the target presentation bits of the next time slice, the method further includes:

counting the visit volume of the target object in the current time slice, and determining the visit volume as the visit volume of the current time slice;

determining a target access amount of the current time slice based on the target accumulated access amount of each time slice;

and correcting the determined target visit amount of the next time slice based on the corresponding relation between the visit amount of the current time slice and the target visit amount of the current time slice.

Optionally, as in the foregoing method, after determining that the presentation bits that satisfy the target access amount of the next time slice are the target presentation bits of the next time slice, the method further includes:

determining a predicted display bit meeting the access amount of the current time slice in the next time slice based on the access amount of the current time slice;

obtaining a first ranking number of the target display position and a second ranking number of the predicted display position based on preset display position ranking information; the display bit ordering information is obtained by ordering the display bits according to the display bit access amount received by each display bit in a historical access time period and from high to low;

determining a position lifting value corresponding to the target object according to the first sequencing number and the second sequencing number;

when the position lifting value is smaller than or equal to a preset position lifting threshold value, obtaining the display position lifting information according to the position lifting value;

and when the position lifting value is greater than the position lifting threshold value, obtaining the display position lifting information according to the position lifting threshold value.

Optionally, as in the foregoing method, the method further includes:

determining display objects on each display position and the corresponding weights of the display objects;

determining a previous display position of the target display position according to the display position sorting information;

obtaining a target weight corresponding to the target object according to the first weight of the first display object on the target display position and the second weight of the second display object on the previous display position;

sorting the target object and all the display objects according to the target weight and the weight corresponding to the display object to obtain a first sorting result;

determining order information corresponding to the target object according to the first ordering result;

determining a real display bit to be allocated corresponding to the target object according to the sequence information and the display bit ordering information;

and allocating the target object to the real display bit to be allocated.

Optionally, as in the foregoing method, the determining, according to the order information and the display bit ordering information, a real display bit to be allocated corresponding to the target object includes:

sorting all the target objects according to the target weight corresponding to the target object to obtain a second sorting result;

screening all the target objects according to the second sorting result to obtain K-bit target objects to be adjusted before ranking, wherein K is an integer greater than or equal to 1;

and determining the real display bits to be distributed corresponding to the target objects to be adjusted according to the sequence information and the display bit ordering information.

Optionally, as in the foregoing method, the determining, according to the order information and the display bit ordering information, a real display bit to be allocated corresponding to the target object to be adjusted includes:

sorting the target objects and all the display objects according to the target weight of the target object to be adjusted and the weight corresponding to the display objects to obtain a third sorting result;

dividing the third sequencing result according to sequencing to obtain at least two continuous sequencing intervals;

inputting target weights and type information corresponding to the target objects to be adjusted respectively into a diversity model, determining the similarity between the target objects to be adjusted by the diversity model based on the type information, reordering the target objects to be adjusted in each ordering interval according to the target weights and the similarity between the target objects to be adjusted, and obtaining a fourth ordering result; in each reordered sorting interval, the number of target objects to be adjusted corresponding to the same type of information is less than or equal to a preset value;

and determining the real display position to be allocated corresponding to the target object to be adjusted according to the fourth sequencing result.

Optionally, as in the foregoing method, the method further includes:

determining historical access information of a target terminal;

determining corresponding content label information according to the historical access information;

determining object tag information of an object to be recalled;

screening all candidate objects according to the object label information to obtain a first object set comprising the object to be recalled; the number of the objects to be recalled in the first object set meets a first preset number requirement;

screening the first object set according to the content tag information to obtain a second object set;

selecting the target object from the second set of objects.

Optionally, as in the foregoing method, the method further includes:

determining historical access information of a target terminal;

determining target uploader information corresponding to the object to be recalled;

screening all candidate objects according to the target uploader information to obtain a third object set; the number of the candidate objects in the third object set meets a second preset number requirement;

screening the third object set according to the content tag information to obtain a fourth object set;

selecting the target object from the fourth set of objects.

In a second aspect, an embodiment of the present application provides a resource allocation apparatus, including:

the target object module is used for determining the total target access amount of the target object in the target access time period; wherein the target access time period consists of at least two time slices set consecutively;

the target accumulated visit amount module is used for determining the target accumulated visit amount of each time slice based on the target visit time period and the target visit total amount; the target accumulated visit volume of the time slice is the visit volume which needs to be reached by the target object from the starting time point of the target visit time period to the ending time point of the time slice;

the current accumulated access volume module is used for determining the accumulated access volume of the target object as the current accumulated access volume from the starting time point of the target access time period to the ending time point of the current time slice;

the target access amount module is used for determining the target access amount of the next time slice based on the target accumulated access amount of the next time slice and the current accumulated access amount; wherein, the target visit volume is the visit volume which needs to be reached by the target object in the next time slice;

and the target display position module is used for determining that the display position meeting the target access amount of the next time slice is the target display position of the next time slice, wherein the display position is a position resource used for displaying the target object on the display interface.

In a third aspect, an embodiment of the present application provides an electronic device, including: the system comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus;

the memory is used for storing a computer program;

the processor is configured to implement the processing method according to any one of the preceding claims when executing the computer program.

In a fourth aspect, an embodiment of the present application provides a storage medium including a stored program, where the program is run to perform the method steps of any one of the preceding claims.

Compared with the prior art, the technical scheme provided by the embodiment of the application has the following advantages:

according to the method provided by the embodiment of the application, the real-time access condition of a single target object is analyzed by carrying out the capacity preservation on the single target object, so that the target display position corresponding to the target object at the next moment can be obtained through real-time analysis; thereby ensuring that the target object is allocated a display bit which is more favorable for reaching the target access amount; therefore, better guarantee is provided for the target object to reach the target access total amount in the target access time period.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.

Fig. 1 is a flowchart of a resource allocation method according to an embodiment of the present application;

fig. 2 is a flowchart of a resource allocation method according to another embodiment of the present application;

fig. 3 is a flowchart of a resource allocation method according to another embodiment of the present application;

FIG. 4 is a flowchart of a resource allocation method in an application example of the present application;

fig. 5 is a block diagram of a data processing apparatus according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all 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 application.

Fig. 1 is a resource allocation method according to an embodiment of the present application, including the following steps S1 to S5:

s1, determining a target access total amount of a target object reached in a target access time period; wherein the target access period is composed of at least two time slices set consecutively.

Specifically, the target object may be content that needs to be secured (guaranteed to reach a preset access amount), for example: videos or articles that need to be held, etc.

The target access time period may be the total time period that the target object needs to be secured.

The target access amount may be a preset total amount of access to the target object that needs to be achieved within the target access time period.

The time slice is at least two time periods obtained by splitting a target access time period; the dividing mode may be preset fixed time length division, for example, the time length of each time slice is fixedly set to be 5 minutes or 15 minutes; the adjustment may be made according to the accuracy or the period of the access point control. In an alternative implementation manner, the start time of a time slice may be used as the identifier of the time slice, for example: when a time slice corresponds to a time period of 12:00:00 to 12:14:59, the time slice can be recorded as 12:00: 00.

For example: in the next week that the movie needs to be shown, the movie needs to reach an access amount of fifty million, in this scene, the target object is the movie to be shown, the target access time period is the next week that the movie is shown, the total target access amount is fifty million, and the division of the time slices can be adjusted according to the accuracy of the access amount, namely, the time slice corresponding to the 0 o 'clock to the 4 o' clock early morning field, the time slice corresponding to the 4 o 'clock to the 9 o' clock morning field, and the time slice corresponding to each hour in the 10 o 'clock to the 24 o' clock gold field.

Or, on the twenty-one day, the access amount of a set top box of a certain model reaches ten thousand, in this scenario, the target object is the set top box of a certain model, the target access time period is on the twenty-one day, the total target access amount is ten thousand, and the time slice division can be directly divided according to the time slice corresponding to each hour.

S2, determining the target accumulated visit amount of each time slice based on the target visit time period and the target visit total amount; the target accumulated access amount of the time slice is the access amount which the target object needs to reach from the starting time point of the target access time period to the ending time point of the time slice.

Specifically, after a target access time period and a target access amount are obtained, the unit time access amount of each time slice can be obtained in an average distribution mode, the sequence relation of each time slice is further determined through the time sequence relation among the time slices, and then the access amounts among the time slices and other previous time slices are accumulated according to the sequence relation to obtain a target accumulated access amount; or, based on the visit amount of a certain historical time period corresponding to the target visit time period, calculating to obtain the accumulated visit amount of each time slice in the historical time period, then determining the corresponding relation between the time slices in the historical time period and the time slices in the target visit time period, and determining the target accumulated visit amount of each time slice.

Specifically, the target accumulated access amount may correspond to a time slice, and may be accumulated from the beginning of the target access time period, and the access amount required to receive access is preset until the time slice ends.

And S3, determining the accumulated access volume of the target object as the current accumulated access volume from the starting time point of the target access time period to the ending time point of the current time slice.

Specifically, the starting time point of the target access time period may be a starting time point of video saving for the target object.

The current time slice may be a time slice corresponding to the current time point, and generally, the current accumulated access amount may be counted at an end time point of the current time slice, or may be counted after the end time point to obtain the current accumulated access amount. And takes it as the current accumulated access volume. By way of example: when the time period corresponding to the current time slice is 12:00:00 to 12:14:59, and the target access time period is 00:00:00 to 23:59:59, the current accumulated access amount may be the accumulated access amount received between 00:00:00 and 12:14: 59.

S4, determining the target access amount of the next time slice based on the target accumulated access amount of the next time slice and the current accumulated access amount; the target visit volume is the visit volume that the target object needs to reach in the next time slice.

Specifically, the next time slice may be a time slice adjacent to and subsequent in time sequence to the current time slice. By way of example: when the time period corresponding to the current time slice is 12:00:00 to 12:14:59, the time period corresponding to the next time slice is 12:15:00 to 12:29: 59.

The target accumulated access amount may be accumulated from a starting time point of the target access time period, and the target object presets a total access amount to be obtained when the next time slice is finished.

And the target access amount of the next time slice is the difference value between the target accumulated access amount and the current accumulated access amount.

And S5, determining that the display position meeting the target access amount of the next time slice is the target display position of the next time slice, wherein the display position is a position resource for displaying the object on the display interface.

Specifically, the display interface is an interface for displaying an object and is composed of a plurality of display positions; the display interface can be a display interface on the client side or a display interface on other terminals.

The display position can be a position for displaying an object on a display interface; generally, the amount of access received at different locations varies from location to location within the same unit of time. The target display position is a position for displaying the target object on the display interface.

And for the access amount which can be received by any presentation bit in the next time slice, the prediction can be obtained based on the historical access amount which is received by the presentation bit in the historical time slice corresponding to the next time slice.

Therefore, the target presentation bit meeting the target access amount of the next time slice can be selected from the access amounts of the presentation bits in the next time slice.

In an alternative implementation manner, the target presentation bits may be selected from presentation bits whose next time slice has an access amount greater than or equal to a target access amount of the next time slice.

Specifically, the target access amount of the next time slice needs to be satisfied by the selected target display bits, so that the access amount in the next time slice provided by the target display bits can satisfy the target access amount of the next time slice.

For example, the relationship between different display positions and exposure (access amount) in each time slice can be obtained based on historical data, and then the display position pos meeting the minimum display amount requirement of the target access amount at the next moment can be selected and obtained according to the access amount in the next time slice of each display position_mSetting a display position pos_mThe corresponding access amount in the next time slice is imp_{pos_m}Then p is_{tn}＝imp'_{goal,tn}/imp_{pos_m}∈(0,1]I.e. target access quantity imp 'of the next time slice'_{goal,tn}And the access quantity imp in the next time slice of the presentation bit_{posm}The ratio therebetween needs to be less than or equal to 1.

In summary, by using the method in this embodiment, the target display position corresponding to the next time slice of the target object can be determined in real time according to the access condition; thereby ensuring that the target object is allocated a display bit which is more favorable for reaching the target access amount; therefore, better guarantee is provided for the target object to reach the target access amount in the target access time period.

As shown in fig. 2, in some embodiments, the step S2 of determining the target accumulated visit volume per time slice, as the method described above, includes the steps S21 and S22 as follows:

step S21, determining the total historical visit amount reached by all the display positions in the historical visit time period and the historical accumulated visit amount of each historical visit time slice; wherein the historical access time period is composed of at least two continuously set historical access time slices.

Specifically, the historical access time period may be a time period for defining the historical data that needs to be acquired.

The historical access total is the sum of the total number of accesses received by each presentation bit during the historical access time period.

The historical access time slice may be a plurality of time slices obtained by splitting the historical access time period.

Step S22, based on the corresponding relation between the historical visit time quantum and the target visit time quantum and the historical visit total, determining the expected target visit total of all the display positions in the target visit time quantum and the expected accumulated visit amount of each time slice; the expected accumulated access amount of the time slice is the access amount which can be expected to be reached by the display bit from the starting time point of the historical access time period to the ending time point of the time slice.

Specifically, the expected accumulated access amount may be an access amount that the display bit is expected to reach from a starting time point of the target access time period to an ending time point of the time slice; the expected target total visit amount may be predicted according to the historical total visit amount, for example, when the historical total visit amount is obtained from 7 days of historical data, and the target visit time period is 1 day, the duration proportion is 7, and the guaranteed total visit amount may be obtained by summing up the 7 days of historical data to obtain the historical total visit amount, and then dividing by 7 to obtain the expected target visit amount.

Similarly, the expected cumulative visit amount corresponding to each time slice can be obtained by calculating the sum of the expected cumulative visit amounts of the same time slice on different days and then dividing the sum by the time length proportion, for example, when the expected cumulative visit amount of the time slice 12:00:00 to 12:14:59 in No. 12/08 of 2020 is required to be obtained: expected cumulative visits X1, X2, X3, X4, X5, X6, and X7 for time slices 12:00:00 to 12:14:59 in No. 01, No. 02, No. 03, No. 04, No. 05, No. 06, and No. 07, respectively, of 12 months in 2020; then, the average was calculated for X1, X2, X3, X4, X5, X6, and X7.

Optionally, the target access time period may be further split to obtain each time slice through the sequence characterization as described below, so as to obtain the expected accumulated access amount of all the display bits.

{imp_{goal,t0}，imp_{goal,t1}，...，imp_{goal,t|T|}}. Wherein, T0, T1, … and T | can be the starting time of each time slice, and the time length of each time slice is the same and is continuous with each other.

Step S23, determining a target accumulated visit amount for each time slice according to the corresponding relationship between the target visit amount and the expected accumulated visit amount for each time slice.

Specifically, the correspondence between the target access amount and the expected target access amount may be a proportional relationship between the target access amount and the expected target access amount, and since, in general, for a certain website, the total access amount and the access amount of each display bit therein change according to a certain rule with time, in order to facilitate management of the access amounts of all objects, the rule also needs to exist in the change situation of the access amounts of the target objects. Thus, a target cumulative access amount for a time slice is determined based on what may be a proportional relationship and the expected cumulative access amount for each time slice.

By the method in the embodiment, the access amount of the target object can be adjusted in equal proportion along with the change of the expected accumulated access amount of all the display bits, so that all the objects can be uniformly and orderly controlled.

In some embodiments, before the step S5 determines that the presentation bit satisfying the target access amount of the next time slice is the target presentation bit of the next time slice, the method further includes the following steps a1 to A3:

step A1, counting the visit amount of the target object in the current time slice, and determining the visit amount as the visit amount of the current time slice.

Step A2, based on the target accumulated visit amount of each time slice, determines the target visit amount of the current time slice.

Step a3, based on the correspondence between the visit volume of the current time slice and the target visit volume of the current time slice, corrects the determined target visit volume of the next time slice.

Specifically, the access amount of the current time slice may be the access amount received by the target object from the starting time point of the current time slice to the ending time point of the current time slice.

Since the target accumulated access amount of each time slice is determined in the foregoing step, the target access amount of the current time slice can be obtained by subtracting the target accumulated access amount of the current time slice from the target accumulated access amount of the next time slice.

By subtracting the target access amount of the current time slice from the access amount of the current time slice, the difference between the access amount of the current time slice and the target access amount can be determined, and whether the right is required to be lifted or not can be determined.

Since the target visit amount for the next time slice is predetermined, it can be corrected based on the difference value so that the target object can reach the accumulated visit amount for the next time slice in the next time slice.

By the method in the embodiment, the accumulated access amount of the next time slice can be quickly determined, and whether the adjustment of the display position is needed to be carried out in an adjustment mode can be judged accordingly, so that the real-time performance and the accuracy of the display adjustment can be effectively improved.

As shown in fig. 3, in some embodiments, as the aforementioned method, the step S5 further includes steps S61 to S65 after determining that the presentation bit meeting the target access amount of the next time slice is the target presentation bit of the next time slice:

step S61, determining the predicted display position meeting the access amount of the current time slice in the next time slice based on the access amount of the current time slice;

specifically, in order to guarantee the real-time performance of the access amount adjustment, the duration of one time slice may be a short time period such as 5 min. The predicted access amount of the next time slice can be predicted according to the real access amount of the current time slice. In an alternative implementation, the actual access amount may be used as the expected access amount.

Since the expected accumulated visit volume of each presentation bit in the next time slice is known, the expected target visit volume of each presentation bit in the next time slice is the predicted visit volume reached by the presentation bit in the next time slice; therefore, according to the numerical relationship between the expected access amount and the expected target access amount of each presentation bit in the next time slice (generally, the expected access amount needs to be smaller than the expected target access amount), the presentation bit to which the target object needs to be allocated when the expected access amount is reached is selected from the presentation bits, and the presentation bit is marked as the expected presentation bit.

S62, obtaining a first ranking number of a target display position and a second ranking number of a predicted display position based on preset display position ranking information; the display bit ordering information is obtained by ordering the display bits according to the total historical access amount received by each display bit in the historical access time period from high to low.

Specifically, the total historical access amount received by the presentation bit is the access amount received by each presentation bit in the historical access time period, and a corresponding relationship between the total historical access amount received by the presentation bit and the presentation bit can be established.

The display position sequencing information can be adjusted in real time according to historical accumulated access of each display position in different time slices; however, in general, when the access amount is greater than a certain preset value, the magnitude relationship between the historical accumulated access amounts of the respective display bits is basically kept consistent in the whole target access time period, and therefore the display bit ordering information is generally not changed in the whole target access time period.

The first ranking number is the ranking of the target access amount of the target presentation bit in the next time slice, among the expected target access amounts of all presentation bits.

The second rank number is the rank of the expected target access amount of the presentation bits in the next time slice among the expected target access amounts of all presentation bits.

And S63, determining a position lifting value corresponding to the target object according to the first ranking number and the second ranking number.

Specifically, the sorting information may be embodied by a numerical value, and therefore, the position-raising value may be obtained by a difference between a numerical value corresponding to the first sorting number and a numerical value corresponding to the second sorting number.

And S64, when the position lifting value is smaller than or equal to a preset position lifting threshold value, obtaining display position lifting information according to the position lifting value.

Specifically, the position lifting threshold may be set in advance, and when the position lifting value is smaller than or equal to the position lifting threshold, the real lifting value included in the display bit lifting information may be consistent with the position lifting value.

The presentation bit lifting information is used to represent information for lifting the right when the target object is distributed, and may be, in general, information for adjusting the presentation bit to which the target object is distributed.

In general, each presentation bit may be marked by a number, for example: marking the display bit with the highest visit quantity as 1, marking the display bit with the second visit quantity as 2, and then marking sequentially according to the visit quantity of each display bit.

Thus, after the expected display bits and the target display bits are obtained, the display bit boost information may be calculated by the difference between the numbers of the markers (i.e., the position boost value) used to characterize the expected display bits and the target display bits.

Because the logs written with the information related to the display bits are generally stored in a specific system, and meanwhile, in order to realize real-time adjustment of the display bits according to the access amount, the application has already accessed the system related to the access amount (to obtain the access amount information), and if another set of system is accessed, the data processing task is large, therefore, by adopting the method of the embodiment, the predicted display bits are obtained only through the real access amount, and then the obtained display bit promotion information can effectively simplify the calculation and accelerate the task.

And S65, when the position lifting value is larger than the position lifting threshold value, obtaining display position lifting information according to the position lifting threshold value.

Specifically, when the position promotion value is greater than the position promotion threshold, the real promotion value included in the display bit promotion information may be consistent with the position promotion threshold.

In an alternative implementation manner, for the target object v, the information for saving the target object v may be a saving (v, δ pos, p)_{tn}) Wherein δ pos is a position-increasing value, and δ pos is max (0, pos)_m-pos_k)；pos_mMay be the first rank number, pos, of the target display bit_kMay be a second rank number of the expected display bits, as can be seen from the previous embodiment, p_{tn}And the ratio of the target access amount of the target object in the next time slice to the target access amount of the target presentation bit in the next time slice is obtained.

Optionally, the position promotion threshold may be a value such as 200, and by the method in this embodiment, it may be prevented that poor content is distributed to the top, which may cause display bits with high access amount to be occupied by poor content, and further affect the overall consumption condition of the content on the distribution platform.

In some embodiments, the method as described above, further comprising steps P1 to P7 as described below:

step P1, determining display objects on all display positions in the current time slice and the weight of the display objects; and the weight of the display object is inversely related to the sorting number of the display bit where the display object is located.

Specifically, the display object is the content displayed on the display position by the current time slice; generally, a presentation object is an object that is presented on a partial presentation position in addition to a target object. Further, the presentation object may have content that does not require a guarantee amount.

In general, the larger the access quantity (obtained by prediction according to historical information, namely, the access quantity of the display position received by the display position in a certain historical access time period) which is expected to be obtained by the display position in a target time period is, the smaller the sequence number of the display position is; for example, when there are 100 display bits, the display bit with rank number 1 is expected to obtain the largest amount of accesses in the target time period, and the display bit with rank number 100 is expected to obtain the smallest amount of accesses in the target time period.

The weight may be a score obtained by evaluating each display object according to a target access amount of each display object, and generally, the weights corresponding to different display objects are different from each other, so as to avoid a situation that different display objects are distributed to the same display position. When the weight of the display object is larger, the display object is allocated to the display position with the smaller sorting number for display, so that the display object can obtain more access.

And P2, determining a previous display bit of the target display bit according to the display bit sorting information.

Specifically, the display bits are sorted according to the sequence of the display bit access amount from large to small in the display bit sorting information; the presentation bit access amount of the previous presentation bit is greater than the presentation bit access amount of the target presentation bit.

Step P3, obtaining the weight of the target object according to the weight of the first display object and the weight of the second display object, wherein the first display object is a display object on a target display position in the current time slice; the second display object is a display object on a previous display position in the current time slice.

Specifically, the weight of the target object may be an average of the weight of the first display object and the weight of the second display object, and thus, the weight of the target object may be guaranteed to be higher than the weight of the first display object (for example, if the target display bit is 40 bits, the target object is given with an average score of 39 bits and 40 bits of display objects, which guarantees that the 40 th bit is inserted).

And P4, sequencing the target object and all the display objects from large to small according to the weight to obtain a first sequencing result.

Specifically, the target objects and the display objects may be sorted from large to small according to the numerical values corresponding to the weights of the target objects and the numerical values of the weights of the display objects, and a first sorting result is obtained, where the sorting result may be a display list obtained by sorting in general.

Step P5. determines first order information corresponding to the target object according to the first order result.

Specifically, after the first sorting result is obtained, the order of the target object in the sorting result can be determined, and then order information is obtained.

And P6, selecting the display bit matched with the first order information according to the display bit ordering information as a real display bit to be distributed corresponding to the target object.

Specifically, the first order information is obtained by ordering the target object and the display object, and the display bit ordering information is obtained by ordering each display bit, so that the display bits corresponding to the target object can be obtained according to the display bits corresponding to the target object by determining the display bits corresponding to the target object according to the first order information. Therefore, the display bits with the ranking number consistent with the first-order information can be used as the real display bits to be allocated, for example: and when the first order information of the target object is 10, selecting the display bit with the ranking number of 10 as the real display bit to be allocated in all the display bits.

And P7, distributing the target object to the real display position to be distributed.

After the real display bits to be distributed are obtained, the target object can be distributed to the real display bits to be distributed by obtaining information such as links or interfaces of the real display bits to be distributed, so that the purpose of saving the target object is achieved.

In some embodiments, as in the foregoing method, when there are multiple target objects, the step P6 selects, according to the display bit ordering information, a display bit matching the first ordering information as a real display bit to be allocated corresponding to the target object, as described in the following steps P611 to P613:

and P611, sorting all the target objects from large to small according to the weight to obtain a second sorting result.

Specifically, the target objects may be sorted according to the target weight and the descending order to obtain a second sorting result; further, the second ranking result is a ranking result obtained by ranking only the target objects according to the target weight.

And P612, screening all the target objects to be adjusted to obtain the first K target objects to be adjusted according to the second sorting result, wherein K is an integer greater than or equal to 1.

Specifically, the target object to be adjusted is the target object ranked K top in the second sorting result; and after the second sorting result is obtained, the modified target weight can be reserved for the first K target objects to be adjusted, and the target weights of other target objects are restored to the original weights before modification.

And P613, selecting the display bits matched with the first order information of the first K target objects to be adjusted according to the display bit ordering information as the real display bits to be distributed corresponding to the target objects to be adjusted.

Specifically, after the target object to be adjusted is determined, the real display bits to be allocated corresponding to the target object to be adjusted may be determined according to the first order information obtained in step P4 and the display bit order information obtained in step P1.

Therefore, corresponding to any one of the K target objects to be adjusted, the display bits with the ranking number consistent with the first ranking information of the target object to be adjusted can be used as the real display bits to be allocated to the target object to be adjusted.

By the method in the embodiment, the display positions of the preset number of target objects can be adjusted, and the situation that excessive target objects needing to be subjected to volume preservation enter the final recommendation result (namely the display positions with high access volume) can be avoided.

In some embodiments, as the aforementioned method, the step P6 obtaining the target weight corresponding to the target object according to the first weight of the first display object at the target display position and the second weight of the second display object at the previous display position of the target display position includes the following steps P21 to P24:

and P21, determining a current sequencing result currently used for displaying the target object and the display object.

Specifically, the current sorting result may not be the sorting result before the flow related to the steps P1 to P6 is performed on the target object.

And P22, sequencing the display bit sequencing information corresponding to each target display bit from high to low to determine the weight calculation order corresponding to each target object.

Specifically, in this embodiment, there are at least two target objects, and each target object corresponds to one target display position, so there are at least two target display positions.

The weight calculation order includes: the order of the target weights is calculated for each target object.

When a plurality of target display positions exist, the position of each target object may be influenced after the position of each target object is adjusted; for example, when target object A, B, C is present, and the corresponding target display bits are 15, 20, and 30, respectively; if the position of C is calculated, inserting the target object C between 29 and 30 bits in the current sequencing result to obtain the purpose of corresponding target display position 30; thereafter, when the original corresponding display bit of the target object B is after 30 bits, and when the target object B is inserted between 19 and 20 bits in the current sorting result after the processing of C is completed, C is shifted back by one bit again at the adjusted position, resulting in a corresponding target display bit 31.

In view of this, if the target weight corresponding to the target object with the front target display position is calculated first, the above situation can be effectively avoided. Therefore, the display positions corresponding to the target display positions are sorted from high to low according to the sorting information of the display positions corresponding to the target display positions, so that the position relation among the target display positions can be obtained, and the weight calculation order corresponding to each target object is obtained according to the position relation.

And P23, according to the weight calculation sequence, sequentially obtaining the target weight corresponding to each target object according to the first weight and the second weight corresponding to each target object.

Specifically, each target object corresponds to one target display bit, so that each target object has a first weight and a second weight corresponding to each target object, and the target weight corresponding to each target object can be calculated.

After the weight calculation order is obtained, the target weight calculation is carried out on the target object with the most front target display position, then the target weight calculation is carried out on the target display position ranked second, and the target weights of other target objects are calculated according to the sequence.

And P24, updating the current sequencing result according to the target weight of the target object and the weights corresponding to all the display objects.

That is, after the target weight of each target object is obtained, the current sorting result is updated according to the target weight and the weights corresponding to all the display objects, and each update is performed on the basis of the previous update, so as to obtain the influence of each target weight on the current sorting result.

By the method in the embodiment, the target weight can accurately correspond to the specific target display position after being acquired, and the situation that the corresponding target display position changes due to the fact that other target objects are inserted into the front end can be avoided.

In some embodiments, as in the foregoing method, the step P6 selects the presentation bit matching the first order information according to the presentation bit ordering information, and the step P521 to P524 as the real to-be-allocated presentation bit corresponding to the target object includes the following steps:

step P521, acquiring type information of the target object and all display objects;

specifically, the type information of the target object and the display object may be obtained by obtaining description information or a label of the target object and the display object. For example: when the target object and the presentation object are videos, the object types may be: drama, animation, movie, etc., and may also be: drama, suspense, science fiction, and the like.

Step P522, inputting the weight and type information of the target object and all the display objects into a diversity model, determining the similarity between adjacent objects by the diversity model based on the type information, and reordering the target object and all the display objects according to the weight and the similarity to obtain a third ordering result; in the third sorting result, the similarity between the adjacent objects meets a preset condition: the similarity between the previous object and the next object is less than the similarity between the previous object and each of the rest of the objects, and/or the similarity between the adjacent objects is less than or equal to a similarity threshold.

In one implementation, the diversity model may be a DPP (deterministic Point Process) model. After the weight and type information of the target object and all the displayed objects are input into the DPP model, the DPP model may calculate the similarity between adjacent objects based on the type information, that is, the similarity between all the objects (the target object and the displayed object); in this embodiment, one of the conditions for triggering the rearrangement operation may be: the objects (which can be target objects and presentation objects) with the same type information in the same sorting interval (for example, 5) comprise two or more times. In this case, when the same object includes two or more objects to be adjusted, the ranking result of the objects to be adjusted with higher weight may be adjusted, and the value corresponding to the target weight of the objects to be adjusted is lifted, so that the corresponding target visit amount can be satisfied after the objects to be adjusted are adjusted, and the target weight of the low objects to be adjusted is kept unchanged; when the same object comprises an object to be adjusted and at least one display object, the sorting result of the display objects can be adjusted, the value corresponding to the target weight of the display objects is reduced, and the low target weight of the object to be adjusted is kept unchanged, so that the target access amount of the object to be adjusted in the sorting interval is not reduced after the display objects are adjusted.

And, when in the reordered adjacent objects, the similarity between the adjacent objects is higher than a preset value (for example, when the number of objects with the same type information in the adjacent objects is less than or equal to a preset value (for example, one), it means that the diversity of the object types is high, and therefore, the target object and/or the presentation object do not need to be adjusted); a third ordering result may then be obtained,

step P523, determining second order information corresponding to the target object according to the third ordering result;

and after the third sorting result is obtained, recording the order information of each target object as second order information.

And P524, selecting the display bit matched with the second order information according to the display bit ordering information as a real display bit to be distributed corresponding to the target object.

Specifically, because the diversity requirement between the adjacent objects of the third sorting result is met and the target access amount of each target object is also guaranteed, the real to-be-allocated display position of each target object to be adjusted is directly obtained according to the third sorting result.

By the method in the embodiment, on the basis of realizing the access amount preservation of the target object, the adjacent objects meet the diversity requirement, so that the content types received by the user can be richer after the user receives and pushes the content, and the use experience of the user is effectively improved.

In some embodiments, as in the previous method, the method further comprises steps Q11-Q13 as follows:

step Q11, acquiring an object tag of a target client historical access object and a preset recall object tag; wherein the object tag comprises a content tag and/or a source tag; .

Specifically, the target client may be a client to which the target object is distributed, and the historical access object may include: one or more objects that have been accessed in the platform by the target client.

The content tag may be a set of one or more tag information used to characterize the type of the target object. When the target object is a video, the content tags may include, but are not limited to: genre (e.g., movie, tv show, etc.), subject matter, age, etc.

The source tag may be an uploader tag used to characterize the content, such as: the ID of the uploader, or the characteristics of the uploader (such as age, sex, hobby and the like).

The preset recall object label can be a label corresponding to a capacity-preserving object needing capacity preservation; the preset recall object tag may be set based on characteristics of the inventory object, such as: upload time, actor name, and movie name, etc. Furthermore, through the preset recall object label, the guarantee amount object needing to be guaranteed can be directly selected, so that the probability that the guarantee amount object is pushed to the user is increased.

By way of example: when a target client A initiates an access request to a video website, historical access objects (such as video A, video B and the like) of the target client A to the video website are obtained, and each historical access object can be shown in a list form. And then determining the content label and/or the source label corresponding to the video A and the video B based on the labels of the video A and the video B.

Step Q12, screening the recall objects meeting the recall conditions in the object set to form a recall object set; wherein the recall condition is as follows: the similarity between the object tag of the recall object and the object tag of the history access object is greater than or equal to a first threshold, and the similarity between the object tag of the recall object and the preset recall object tag is greater than or equal to a second threshold.

Specifically, the object set may be a data set storing all objects, for example: one or more servers storing all videos in a certain video website.

Since the recall condition is: the similarity between the object tag of the recall object and the object tag of the history access object is greater than or equal to a first threshold, and the similarity between the object tag of the recall object and the preset recall object tag is greater than or equal to a second threshold. Therefore, by recalling the object whose similarity between the object tag of the recall object and the object tag of the history access object is greater than or equal to the first threshold, the recall object in which the user of the target client is interested can be acquired; through the similarity between the object tag of the recall object and the preset recall object tag being greater than or equal to the second threshold, the reserve object needing the reserve can be obtained; the two are combined to obtain the object which is interesting to the user and needs to be recalled for the amount preservation.

And step Q13, selecting the target object from the recall object set.

Specifically, after the recall object set is obtained, all objects in the recall object set may be used as target objects, or a preset number of target objects may be selected from the recall object set according to a preset number requirement.

Generally, the method of this embodiment may be applied to a process of roughly arranging content, and by using object tag information to recall, it may be ensured that an object to be recalled may be recalled by a certain recall source (tag for recall) certainly. By the method, the guarantee amount object needing the guarantee amount can be selected and obtained, meanwhile, the guarantee amount object can be ensured to be interested by the user, and the guarantee amount effect can be effectively improved.

As shown in fig. 4, the corresponding application of the method in any of the foregoing embodiments in video recall is as follows:

1. determining a video data set needing to be subjected to the capacity guarantee (namely, a collection of capacity guarantee videos needing to be subjected to the capacity guarantee operation comprises the target object), establishing an independent recall source corresponding to the capacity guarantee videos, independently recalling the capacity guarantee videos through the independent recall source, and performing rough arrangement and right withdrawal on the recalled capacity guarantee videos; the independent recall can be recalled in an object tag mode, and can also be recalled in a video uploader (source tag) recall mode.

2. And after the coarse ranking result is obtained, further screening the coarse ranking result according to the historical access information of the user to obtain a fine ranking result.

3. In a daily-level (one day is taken as a unit) guarantee task, the relationship between the display amount (the access amount) and the fine ranking (namely, the display position) and the change curve of the flow in one day (24 hours) along with the time are obtained through 7 calendar history data (optionally, a fine log (a log record of the result of all videos entering a fine model in one request) and a pingback log (a log record of behaviors such as display/click/play and the like of N videos really recommended to a user by a client) can be connected through event _ id, and the two logs are combined to further calculate the relationship between the fine position and the display according to the above mentioned mode).

4. Because the operation guarantee video data set can be updated in real time, the real-time data I of the guarantee video can be obtained, and a real-time task pair is needed; in addition, because the pingback is a streaming log, a user can obtain a pingback log every time the user has one action, and therefore, the real-time data II of the user action is obtained by obtaining the pingback log in real time. When calculation is carried out at each time T, the guaranteed video data set can update the guaranteed video data in real time, so that which videos need to be guaranteed at present are found from the operation guaranteed video data set; then, the real exposure of the last time slice of the reserve videos is obtained according to the pingback log, and the display positions pos corresponding to the next time slice of each reserve video are obtained according to the 24-hour flow curve of the large disk_k(ii) a And then calculating according to the operation guarantee video data set to obtain the target access amount of the next time slice, and obtaining the minimum display position pos meeting the display amount at the next time according to the 24-hour flow curve of the large disk_mFinally, δ pos and p are unified to each video that needs to be guaranteed.

5. And obtaining a real-time guarantee index of the guarantee video according to the real-time access data and the change curve.

6. According to the real-time guarantee index of the guarantee video and the relation between the display amount and the fine ranking (namely the weight of each video), and through a DPP diversity algorithm, the real-time fine ranking weight-lifting of the guarantee video is obtained (namely the real display position to be distributed is determined).

7. And pushing the guarantee video to the user through the engine according to the real-time fine-ranking and right-giving scheme of the guarantee video.

As shown in fig. 5, according to an embodiment of another aspect of the present application, there is also provided a data processing apparatus including:

the target object module 1 is used for determining the total target access amount of the target object in the target access time period; wherein the target access time period consists of at least two time slices which are continuously set;

the target accumulated visit amount module 2 is used for determining the target accumulated visit amount of each time slice based on the target visit time period and the target visit total amount; the target accumulated visit volume of the time slice is the visit volume which is required to be reached by the target object from the starting time point of the target visit time period to the ending time point of the time slice;

the current accumulated access volume module 3 is used for determining the accumulated access volume of the target object as the current accumulated access volume from the starting time point of the target access time period to the ending time point of the current time slice;

the target visit volume module 4 is used for determining the target visit volume of the next time slice based on the target accumulated visit volume of the next time slice and the current accumulated visit volume; the target visit volume is the visit volume which the target object needs to reach in the next time slice;

and the target display position module 5 is configured to determine that a display position meeting the target access amount of the next time slice is a target display position of the next time slice, where the display position is a position resource used for displaying the target object on the display interface.

Specifically, the specific process of implementing the functions of each module in the apparatus according to the embodiment of the present invention may refer to the related description in the method embodiment, and is not described herein again.

In some embodiments, the target accumulated visitation amount module comprises:

the historical information submodule is used for determining the total historical visit amount of all the display positions within the historical visit time period and the historical accumulated visit amount of each historical visit time slice; the historical access time period consists of at least two continuously set historical access time slices;

the expected accumulated access quantum module is used for determining the expected target access total amount of all the display bits in the target access time period and the expected accumulated access amount of each time slice based on the corresponding relation between the historical access time period and the target access time period and the historical access total amount; the expected accumulated access amount of the time slice is the access amount which can be reached by the display bit in the period from the starting time point of the historical access time period to the ending time point of the time slice;

and the target cumulative access quantum module is used for determining the target cumulative access amount of each time slice according to the corresponding relation between the target access total amount and the expected cumulative access amount of each time slice.

In some embodiments, the apparatus as in the previous paragraph, further comprising:

the statistical module is used for counting the visit volume of the target object in the current time slice and determining the visit volume as the visit volume of the current time slice;

the target visit amount module is used for determining the target visit amount of the current time slice based on the target accumulated visit amount of each time slice;

and the correcting module is used for correcting the determined target access amount of the next time slice based on the corresponding relation between the access amount of the current time slice and the target access amount of the current time slice.

the expected display bit module is used for determining an expected display bit meeting the access quantity of the current time slice in the next time slice based on the access quantity of the current time slice;

the sequencing number module is used for obtaining a first sequencing number of the target display position and a second sequencing number of the predicted display position based on preset display position sequencing information; the display bit ordering information is obtained by ordering each display bit from high to low according to the display bit access amount received by each display bit in the historical access time period;

the lifting value module is used for determining a position lifting value corresponding to the target object according to the first ranking number and the second ranking number;

the first lifting information module is used for obtaining display position lifting information according to the position lifting value when the position lifting value is smaller than or equal to a preset position lifting threshold value;

and the second lifting information module is used for obtaining the display position lifting information according to the position lifting threshold value when the position lifting value is greater than the position lifting threshold value.

the weight determining module is used for determining the display objects on all the display positions and the weights of the display objects in the current time slice; the weight of the display object is negatively related to the sequence number of the display position where the display object is located;

the previous display position determining module is used for determining a previous display position of the target display position according to the display position sorting information;

the weight calculation module is used for obtaining the weight of the target object according to the weight of the first display object and the weight of the second display object, wherein the first display object is a display object on a target display position in the current time slice; the second display object is a display object on a previous display position in the current time slice;

the sequencing module is used for sequencing the target object and all the display objects from large to small according to the weight to obtain a first sequencing result;

the first sequencing information module is used for determining first sequencing information corresponding to the target object according to the first sequencing result;

the real display bit module to be allocated is used for selecting the display bit matched with the first order information according to the display bit ordering information as the real display bit to be allocated corresponding to the target object;

and the allocation module is used for allocating the target object to the real display position to be allocated.

In some embodiments, as the foregoing apparatus, when there are a plurality of target objects, the module of real to-be-allocated presentation positions includes:

the second sequencing result submodule is used for sequencing all the target objects from large to small according to the weight to obtain a second sequencing result;

the screening submodule is used for screening all the target objects to obtain the first K target objects to be adjusted according to the second sorting result, wherein K is an integer greater than or equal to 1;

and the real display bit to be allocated submodule is used for selecting the display bits matched with the first sequence information of the first K target objects to be adjusted according to the display bit sequencing sub information and taking the display bits as the real display bits to be allocated corresponding to the target objects to be adjusted.

In some embodiments, as the aforementioned apparatus, the module of real to-be-allocated show bits includes:

the type information submodule is used for acquiring the type information of the target object and all the display objects;

the diversity submodule is used for inputting the weight and the type information of the target object and all the display objects into the diversity model, determining the similarity between adjacent objects by the diversity model based on the type information, and reordering the target object and all the display objects according to the weight and the similarity to obtain a third ordering result; in the third sorting result, the similarity between the adjacent objects meets a preset condition: the similarity between the previous object and the next object is less than the similarity between the previous object and each of the rest objects, and/or the similarity between the adjacent objects is less than or equal to a similarity threshold value;

the second order information submodule is used for determining second order information corresponding to the target object according to the third ordering result;

and the selection submodule is used for selecting the display bit matched with the second-order information according to the display bit ordering information as the real display bit to be distributed corresponding to the target object.

In some embodiments, the apparatus further comprises:

the system comprises a tag acquisition module, a target retrieval module and a retrieval module, wherein the tag acquisition module is used for acquiring an object tag of a historical access object of a target client and presetting a recall object tag; wherein the object tag comprises a content tag and/or a source tag;

the recall object set module is used for screening recall objects meeting recall conditions in the object set to form a recall object set; wherein the recall condition is as follows: the similarity between the object tag of the recall object and the object tag of the historical access object is greater than or equal to a first threshold, and the similarity between the object tag of the recall object and the preset recall object tag is greater than or equal to a second threshold;

and the target object selection module is used for selecting a target object from the recall object set.

According to another embodiment of the present application, there is also provided an electronic apparatus including: as shown in fig. 6, the electronic device may include: the system comprises a processor 1501, a communication interface 1502, a memory 1503 and a communication bus 1504, wherein the processor 1501, the communication interface 1502 and the memory 1503 complete communication with each other through the communication bus 1504.

A memory 1503 for storing a computer program;

the processor 1501 is configured to implement the steps of the above-described method embodiments when executing the program stored in the memory 1503.

The bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.

The communication interface is used for communication between the electronic equipment and other equipment.

The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.

The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component.

The embodiment of the present application further provides a storage medium, where the storage medium includes a stored program, and the program executes the method steps of the foregoing method embodiment when running.

It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

The foregoing are merely exemplary embodiments of the present invention, which enable those skilled in the art to understand or practice the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for resource allocation, comprising:

determining the accumulated access volume of the target object as the current accumulated access volume from the starting time point of the target access time period to the ending time point of the current time slice;

and determining that the display position meeting the target access amount of the next time slice is the target display position of the next time slice, wherein the display position is a position resource used for displaying an object on a display interface.

2. The method of claim 1, wherein said determining a target cumulative access amount for each of said time slices comprises:

3. The method of claim 1, wherein prior to determining that the presentation bits that satisfy the target amount of access for the next time slice are target presentation bits for the next time slice, the method further comprises:

4. The method of claim 3, further comprising, after the determining that the presentation bits that satisfy the target amount of access for the next time slice are target presentation bits for the next time slice:

obtaining a first ranking number of the target display position and a second ranking number of the predicted display position based on preset display position ranking information; the display bit ordering information is obtained by ordering the display bits according to the display bit access amount received by each display bit in a historical access time period from high to low;

determining a position lifting value corresponding to the target object according to the first ranking number and the second ranking number;

5. The method of claim 4, further comprising:

determining the display objects on all the display positions in the current time slice and the weight of the display objects; the weight of the display object is negatively related to the sequence number of the display position where the display object is located;

obtaining the weight of the target object according to the weight of the first display object and the weight of the second display object, wherein the first display object is a display object on the target display position in the current time slice; the second display object is a display object on the previous display position in the current time slice;

sorting the target object and all the display objects from large to small according to the weight to obtain a first sorting result;

determining first sequencing information corresponding to the target object according to the first sequencing result;

selecting a display bit matched with the first order information according to the display bit ordering information as a real display bit to be allocated corresponding to the target object;

and allocating the target object to the real display bit to be allocated.

6. The method according to claim 5, wherein when there are a plurality of target objects, selecting, according to the display bit ordering information, a display bit matching with the first ordering information as a real display bit to be allocated corresponding to the target object includes:

sorting all the target objects from large to small according to the weight to obtain a second sorting result;

according to the second sorting result, screening all the target objects to obtain first K target objects to be adjusted, wherein K is an integer greater than or equal to 1;

and selecting the display bits matched with the first order information of the first K target objects to be adjusted according to the display bit ordering information as the real display bits to be distributed corresponding to the target objects to be adjusted.

7. The method according to claim 5, wherein the selecting, according to the display bit ordering information, the display bit matching with the first ordering information as the real display bit to be allocated corresponding to the target object comprises:

acquiring the type information of the target object and all display objects;

inputting the weight and type information of the target object and all display objects into a diversity model, determining the similarity between adjacent objects by the diversity model based on the type information, and reordering the target object and all display objects according to the weight and the similarity to obtain a third ordering result; wherein, in the third sorting result, the similarity between adjacent objects satisfies a preset condition: the similarity between the previous object and the next object is less than the similarity between the previous object and each of the rest objects, and/or the similarity between the adjacent objects is less than or equal to a similarity threshold value;

determining second order information corresponding to the target object according to the third ordering result;

and selecting the display bit matched with the second order information as the real display bit to be distributed corresponding to the target object according to the display bit sorting information.

8. The method of claim 1, further comprising:

acquiring an object tag of a historical access object of a target client and a preset recall object tag; wherein the object tag comprises a content tag and/or a source tag;

in the object set, screening recall objects meeting recall conditions to form a recall object set; wherein the recall condition is: the similarity between the object tag of the recall object and the object tag of the historical access object is greater than or equal to a first threshold, and the similarity between the object tag of the recall object and the preset recall object tag is greater than or equal to a second threshold;

selecting the target object from the recall object set.

9. A data processing apparatus, comprising:

10. An electronic device, comprising: the system comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus;

the memory is used for storing a computer program;

the processor, when executing the computer program, implementing the method steps of any of claims 1-8.

11. A storage medium, characterized in that the storage medium comprises a stored program, wherein the program is operative to perform the method steps of any of the preceding claims 1-8.