Media content delivery processing method, apparatus, device, medium, and program product

Abstract

The application discloses a media content delivery processing method, device, equipment, medium and program product. The method comprises the following steps: acquiring a plurality of first pull exposure time lengths, a first average preset delivery index and a plurality of first real delivery indexes corresponding to a plurality of content delivery exposures of a preset content delivery position in a first historical time period; clustering the plurality of first real delivery indexes based on a preset time interval and the plurality of first pull exposure time lengths, obtaining at least one first real delivery index corresponding to each pull exposure time length interval, and determining an average real delivery index corresponding to each pull exposure time length interval; fitting a preset index decay model based on the average real delivery index and the first average preset delivery index; and performing media content delivery processing on the preset content delivery position based on the preset index decay model. The technical scheme provided by the application can improve the rationality, effectiveness and delivery income of content delivery.

Description

Technical Field

[0001] This application relates to the field of Internet technology, and in particular to a media content delivery and processing method, apparatus, equipment, medium, and program product. Background Technology

[0002] With the development of internet technology, some businesses often use internet platforms to recommend media content to their target audience. Correspondingly, the platform selects suitable media content for display from a wide range of media based on various factors, such as the business's bid ECPM (effective cost per mille, the advertising revenue earned per thousand impressions), and retrieves it to local cache or temporary storage for later display. However, in practice, to avoid frequent exposure of media content affecting user experience, a media content exposure frequency control strategy is often used for pre-loaded media content delivery. This exposure frequency control strategy can extend the exposure time of some pre-loaded media content, increasing the risk of over-broadcasting. This directly affects the actual ECPM and other delivery metrics of the pre-loaded media content, causing a decline in performance and ultimately hindering the platform's effective and reasonable content delivery, leading to a decrease in advertising revenue. Summary of the Invention

[0003] This application provides a media content delivery processing method, apparatus, equipment, storage medium, and computer program product, which can...

[0004] On the one hand, this application provides a media content delivery and processing method, the method comprising:

[0005] Get multiple first pull exposure durations corresponding to multiple content delivery exposures within a first historical time period, a first average preset delivery index corresponding to the multiple content delivery exposures, and multiple first real delivery indices corresponding to the multiple content delivery exposures.

[0006] Based on a preset time interval and the plurality of first pull exposure durations, the plurality of first real delivery metrics are clustered to obtain at least one first real delivery metric corresponding to each of the plurality of pull exposure duration intervals.

[0007] Based on at least one first real delivery indicator corresponding to each of the multiple pull exposure duration intervals, determine the average real delivery indicator corresponding to each pull exposure duration interval.

[0008] Based on the average actual delivery index and the first average preset delivery index corresponding to each pull exposure duration interval, a preset index decay model is fitted. The preset index decay model is used to characterize the decay change from the preset delivery index to the actual delivery index and the correspondence between the pull exposure duration.

[0009] Based on the preset index decay model, media content is delivered to the preset content delivery position.

[0010] On the other hand, a media content delivery and processing device is provided, the device comprising:

[0011] The first data acquisition module is configured to acquire multiple first pull exposure durations corresponding to multiple content delivery exposures within a first historical time period, a first average preset delivery index corresponding to the multiple content delivery exposures, and multiple first real delivery indices corresponding to the multiple content delivery exposures.

[0012] The delivery metric clustering module is configured to perform clustering of the multiple first real delivery metrics based on a preset time interval and the multiple first pull exposure durations, to obtain at least one first real delivery metric corresponding to each of the multiple pull exposure duration intervals.

[0013] The first indicator determination module is configured to execute at least one first real delivery indicator based on each of the plurality of pull exposure duration intervals to determine the average real delivery indicator corresponding to each pull exposure duration interval.

[0014] The attenuation model fitting module is configured to perform a fitting of a preset index attenuation model based on the average actual delivery index and the first average preset delivery index corresponding to each pull exposure duration interval. The preset index attenuation model is used to characterize the attenuation change from the preset delivery index to the actual delivery index and the correspondence between the pull exposure duration.

[0015] The media content processing module is configured to perform media content delivery processing on the preset content delivery position based on the preset index decay model.

[0016] On the other hand, an electronic device is provided, including: a processor;

[0017] Memory used to store the processor's executable instructions;

[0018] The processor is configured to execute the instructions to implement the media content delivery processing method described in any of the preceding claims.

[0019] On the other hand, a computer-readable storage medium is provided, which, when the instructions in the storage medium are executed by the processor of an electronic device, enables the electronic device to perform any of the media content delivery processing methods described above.

[0020] On the other hand, a computer program product is provided, which includes a computer program that, when executed by an electronic device, implements the media content delivery processing method described in any of the above claims.

[0021] The media content delivery and processing method, apparatus, equipment, storage medium, and computer program product provided in this application have the following technical effects:

[0022] In the process of media content delivery for preset content placement positions, this application obtains multiple first pull-out exposure durations, first average preset delivery metrics, and multiple first actual delivery metrics corresponding to multiple content delivery exposures within a first historical time period. Based on preset time intervals and multiple first pull-out exposure durations, the multiple first actual delivery metrics are clustered to obtain at least one first actual delivery metric corresponding to each of the multiple pull-out exposure duration intervals. The average actual delivery metric corresponding to each pull-out exposure duration interval is determined. Then, based on the average actual delivery metric and the first average preset delivery metric corresponding to each pull-out exposure duration interval, a preset metric decay model is fitted to characterize the decay change from preset delivery metrics to actual delivery metrics and the correspondence between pull-out exposure durations. Combining the preset metric decay model with media content delivery processing for preset content placement positions can effectively reduce the risk of over-broadcasting of pre-loaded media content caused by extended pull-out exposure durations. This can also improve the rationality and effectiveness of content delivery on the delivery platform while increasing the platform's delivery revenue. Attached Figure Description

[0023] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the application environment of a media content delivery processing method provided in an embodiment of this application;

[0025] Figure 2 This is a flowchart illustrating a media content delivery processing method provided in an embodiment of this application;

[0026] Figure 3This is a flowchart illustrating a process for fitting a preset index decay model based on the average real delivery index and the first average preset delivery index corresponding to each pull exposure duration interval, provided in an embodiment of this application.

[0027] Figure 4 This is a schematic diagram of a process for media content delivery to a preset content delivery position based on a preset index decay model, provided in an embodiment of this application.

[0028] Figure 5 This is a schematic diagram illustrating the media content delivery and exposure process corresponding to a preset content delivery position, according to an exemplary embodiment.

[0029] Figure 6 This is a schematic diagram of a process for media content delivery to a preset content delivery position based on a preset index decay model, provided in an embodiment of this application.

[0030] Figure 7 This is a schematic diagram of the structure of a media content delivery and processing device provided in an embodiment of this application;

[0031] Figure 8 This is a block diagram of an electronic device for media content processing provided in an embodiment of this application. Detailed Implementation

[0032] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0033] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0034] In this application embodiment, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0035] Please see Figure 1 , Figure 1 This is a schematic diagram of the application environment of a media content delivery processing method provided in an embodiment of this application. The application environment may include at least a server 100 and a terminal 200.

[0036] In an optional embodiment, server 100 can be used for media content delivery processing. Server 100 can be an independent physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides cloud computing services.

[0037] In an optional embodiment, terminal 200 can provide services such as media content retrieval and exposure based on the server's media content processing capabilities. Specifically, terminal 200 can be, but is not limited to, electronic devices such as smartphones, desktop computers, tablets, laptops, smart speakers, digital assistants, augmented reality (AR) / virtual reality (VR) devices, smart wearable devices, in-vehicle terminals, and smart TVs; it can also be software running on the aforementioned electronic devices, such as applications and mini-programs. The operating system running on the electronic device in this embodiment can include, but is not limited to, Android, iOS, Linux, and Windows.

[0038] In addition, it should be noted that, Figure 1 The examples shown are merely one application environment for a media content delivery and processing method, and the embodiments in this specification are not limited to the above.

[0039] In the embodiments described in this specification, the server 100 and the terminal 200 can be directly or indirectly connected via wired or wireless communication, and this application does not impose any restrictions on this.

[0040] The following describes a media content delivery and processing method according to this application. Figure 2This is a flowchart illustrating a media content delivery processing method provided in an embodiment of this application. This specification provides method operation steps as shown in the embodiments or flowcharts, but based on conventional or non-inventive methods, more or fewer operation steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order. In actual system or server product execution, the methods can be executed sequentially according to the embodiments or drawings, or in parallel (e.g., in a parallel processor or multi-threaded processing environment). Specifically, as shown... Figure 2 As shown, the method may include:

[0041] S201: Obtain multiple first pull exposure durations, first average preset delivery metrics, and multiple first real delivery metrics corresponding to multiple content delivery exposures within the first historical time period for the preset content delivery position.

[0042] In one specific embodiment, the preset content placement position can be a media content display position used for media content placement and exposure; specifically, the preset content placement position can be a media content display position used for pre-loaded media content placement and exposure; correspondingly, in this embodiment, the media content can be pre-loaded media content, that is, media content cached on the local terminal device is retrieved in advance before triggering the media content exposure request. For example, taking the preset content placement position as the media content display position in the information display page published by the subscription account as an example, the pre-retrieval of media content can be triggered when the subscription account publishes information; taking the preset content placement position as the media content display position in the information flow browsing page as an example, the pre-retrieval of media content can be triggered before the user account triggers the information flow browsing page display.

[0043] In a specific embodiment, the first historical time period can be a pre-set historical time period, such as the most recent week, which can be set according to the actual application. Each content delivery exposure corresponds to a first pull-and-exposure duration; specifically, the first pull-and-exposure duration corresponding to any content delivery exposure can be the pull-and-exposure duration corresponding to that content delivery exposure (i.e., the pull-and-exposure duration of the media content corresponding to that content delivery exposure). Specifically, the pull-and-exposure duration can be the time interval from pulling the corresponding media content to its exposure. Specifically, the content pull-and-exposure time and content exposure time corresponding to each content delivery exposure can be obtained first; accordingly, the corresponding pull-and-exposure duration can be determined by combining the content exposure time and the content pull-and-exposure time.

[0044] In a specific embodiment, the first average preset delivery index can be the average of multiple first preset delivery indices corresponding to multiple content delivery exposures within a first historical time period; specifically, the sum of multiple first preset delivery indices can be divided by the number of exposures (the total number of content delivery exposures) to obtain the aforementioned first average preset delivery index. Specifically, each content delivery exposure corresponds to one first preset delivery index; specifically, the first preset delivery index corresponding to any content delivery exposure can be the preset delivery index corresponding to that content delivery exposure (i.e., the preset delivery index of the media content corresponding to that content delivery exposure); specifically, the preset delivery index can be an indicator used to measure the estimated revenue that the media content corresponding to the delivery exposure can bring; for example, the preset delivery index can be the estimated cost per thousand impressions (ECPM) when participating in media content bidding. Each content delivery exposure corresponds to a first real delivery metric; specifically, the first real delivery metric for any content delivery exposure can be the real delivery metric for that content delivery exposure (i.e., the real delivery metric for the media content corresponding to that content delivery exposure); specifically, the real delivery metric can be a metric used to measure the actual revenue that the media content corresponding to the delivery exposure can bring; for example, the real delivery metric can be the actual cost per thousand impressions (ECPM) when the media content is actually exposed.

[0045] S203: Based on a preset time interval and multiple first pull exposure durations, cluster multiple first real delivery metrics to obtain at least one first real delivery metric corresponding to each of the multiple pull exposure duration intervals.

[0046] In a specific embodiment, the preset time interval can be set according to actual application requirements. Specifically, the preset time interval can be a time length standard used to divide the exposure duration intervals corresponding to multiple first real delivery indicators. For example, the preset time interval is one minute. Correspondingly, the multiple exposure duration intervals can include (0,1], (1,2], (2,3]... and so on. The time unit corresponding to the exposure duration interval can be minutes. Correspondingly, the first real delivery indicator with a first exposure duration greater than 0 and less than or equal to 1 minute is at least one first real delivery indicator corresponding to the exposure duration interval (0,1]. The first real delivery indicator with a first exposure duration greater than 1 and less than or equal to 2 minutes is at least one first real delivery indicator corresponding to the exposure duration interval (1,2).

[0047] S205: Based on at least one first real delivery metric corresponding to each of the multiple pull exposure duration intervals, determine the average real delivery metric corresponding to each pull exposure duration interval.

[0048] In one specific embodiment, the sum of at least one first real delivery metric corresponding to each pull exposure duration interval can be divided by the number of exposures (total number of content delivery exposures) corresponding to the pull exposure duration interval to obtain the average real delivery metric corresponding to the pull exposure duration interval.

[0049] S207: Based on the average real delivery index and the first average preset delivery index corresponding to each pull exposure duration interval, fit the preset index decay model.

[0050] In a specific embodiment, the above-mentioned preset index decay model can be used to characterize the decay change from the preset delivery index to the actual delivery index and the correspondence between the pull exposure time.

[0051] In one specific embodiment, the aforementioned preset index decay model may include a first index decay model and a second index decay model, optionally, such as Figure 3 As shown, the preset index decay model fitted based on the average actual delivery index and the first average preset delivery index corresponding to each pull exposure duration interval can include:

[0052] S2071: Obtain the first preset decay function with the pull exposure duration as the independent variable, the second preset decay function with the pull exposure duration as the independent variable, and the preset duration threshold;

[0053] S2073: Based on a preset duration threshold, determine the average real delivery index corresponding to the first pull exposure duration interval adapted by the first preset decay function and the average real delivery index corresponding to the second pull exposure duration interval adapted by the second preset decay function from the average real delivery index corresponding to multiple pull exposure duration intervals.

[0054] S2075: Based on the average actual delivery index and the first average preset delivery index corresponding to the first pull exposure time interval, the fitting parameters in the first preset decay function are fitted to obtain the first index decay model.

[0055] S2077: Based on the average actual delivery index corresponding to the second pull exposure duration interval and the first average preset delivery index, the fitting parameters in the second preset decay function are fitted to obtain the second index decay model.

[0056] In a specific embodiment, the aforementioned preset duration threshold is the critical value of the pull exposure duration interval adapted by the first preset decay function and the second preset decay function, and the aforementioned preset duration threshold can be the preset exposure time interval corresponding to the preset content delivery position (the interval between the exposure times of two adjacent media content of the preset content delivery position).

[0057] In a specific embodiment, the first preset attenuation function and the second preset attenuation function can be determined in the following manner:

[0058] Obtain at least two preset decay functions and preset content placement positions for multiple content placement exposures in multiple third historical time periods, the third average preset placement index for multiple content placement exposures in each third historical time period, and the multiple fourth real placement index for multiple content placement exposures in each third historical time period.

[0059] Based on the multiple fourth real delivery indicators corresponding to each third historical time period and the third average preset delivery indicator corresponding to each third historical time period, the fitting parameters in each preset attenuation function are segmented and fitted to obtain at least two first initial attenuation models corresponding to the first pull exposure time interval and at least two second initial attenuation models corresponding to the second pull exposure time interval.

[0060] Determine the difference data of the first parameter in each first initial decay model over multiple third historical time periods and the difference data of the second parameter in each second initial decay model over multiple third historical time periods;

[0061] Determine the first fit difference data for each first initial decay model in multiple third historical time periods and the second fit difference data for each second initial decay model in multiple third historical time periods;

[0062] Based on the first parameter difference data and the first fitting difference data, a first index decay model is determined from at least two first initial decay models;

[0063] Based on the difference data of the second parameter and the difference data of the second fit, the second index decay model is determined from at least two second initial decay models.

[0064] In one specific embodiment, at least two preset decay functions can be pre-set in conjunction with actual applications, such as at least two of a preset exponential decay function, a preset Gaussian decay function, and a preset logarithmic decay function. The at least two preset decay functions use the exposure duration as the independent variable.

[0065] In one specific embodiment, the multiple third historical time periods can be different sub-time periods corresponding to the same duration within a pre-set fourth historical time period, such as multiple weeks within the last two months. Specifically, one of the multiple third historical time periods can be the same as or different from at least one of the first and second historical time periods. For details regarding the fourth pull exposure duration, the third average preset delivery metric, and the fourth actual delivery metric, please refer to the above descriptions, which will not be repeated here.

[0066] In one specific embodiment, the first exposure duration range can be a duration range in which the exposure duration is less than or equal to a preset duration threshold; the second exposure duration range can be a duration range in which the exposure duration is greater than the preset duration threshold.

[0067] In a specific embodiment, the fitting parameters in each preset attenuation function can be fitted according to the fourth real delivery index corresponding to the first pull exposure duration interval among the multiple fourth real delivery indicators corresponding to each third historical time period (i.e., the fourth real delivery index whose corresponding pull exposure duration interval is located in the first pull exposure duration interval) and the third average preset delivery index corresponding to each third historical time period, so as to obtain at least two first initial attenuation models corresponding to the first pull exposure duration interval.

[0068] In a specific embodiment, based on the fourth real delivery index corresponding to the second pull exposure duration interval (i.e., the fourth real delivery index whose corresponding pull exposure duration interval is located in the second pull exposure duration interval) among the multiple fourth real delivery indices corresponding to each third historical time period and the third average preset delivery index corresponding to each third historical time period, the fitting parameters in each preset decay function are fitted to obtain at least two second initial decay models corresponding to the second pull exposure duration interval.

[0069] In one specific embodiment, a preset fitting algorithm can be used for fitting processing. For example, the preset fitting algorithm can be a nonlinear least squares fitting algorithm. In one specific embodiment, during the fitting process of the fitting parameters in each preset decay function using the preset fitting algorithm, an expression can be constructed to show the correspondence between the decay change from the preset delivery index to the actual delivery index and the pull-out exposure time, based on the preset decay function:

[0070]

[0071] in, It can represent the average preset delivery index; α(t) can represent the preset decay function with the pull exposure duration t as the independent variable; ECPM costIt can represent the actual delivery index; further, it can combine multiple fourth actual delivery indices corresponding to each third historical time period and the third average preset delivery index corresponding to each third historical time period to perform piecewise fitting processing on the fitting parameters of the preset attenuation function in the above expression (fitting processing is performed separately for the first pull exposure time interval and the second pull exposure time interval), and determine the fitting parameters of the preset attenuation function in the above expression (i.e., determine the specific values ​​of the fitting parameters), so as to obtain at least two first initial attenuation models corresponding to the first pull exposure time interval and at least two second initial attenuation models corresponding to the second pull exposure time interval.

[0072] In a specific embodiment, the first parameter difference data can characterize the fluctuation of the fitted parameter in the corresponding first initial decay model over multiple third historical time periods; optionally, the first parameter difference data is positively correlated with the fluctuation of the fitted parameter in the corresponding first initial decay model over multiple third historical time periods; the second parameter difference data can characterize the fluctuation of the fitted parameter in the corresponding second initial decay model over multiple third historical time periods; optionally, the second parameter difference data is positively correlated with the fluctuation of the fitted parameter in the corresponding second initial decay model over multiple third historical time periods; specifically, the change curve of the fitted parameter over multiple third historical time periods can be determined, and the corresponding parameter difference data can be determined by combining the fluctuation amplitude of the change curve; optionally, the corresponding parameter difference data can also be determined by calculating the standard deviation of the fitted parameter over multiple third historical time periods, etc.

[0073] In a specific embodiment, the first fitting difference data can characterize the difference between the actual delivery index predicted by the corresponding first initial decay model in multiple third historical time periods and the corresponding fourth actual delivery index; optionally, the first fitting difference data can be positively correlated with the difference between the actual delivery index predicted by the corresponding first initial decay model in multiple third historical time periods and the corresponding fourth actual delivery index; the second fitting difference data corresponds to the difference between the actual delivery index predicted by the second initial decay model in multiple third historical time periods and the corresponding fourth actual delivery index; optionally, the second fitting difference data can be positively correlated with the difference between the actual delivery index predicted by the corresponding second initial decay model in multiple third historical time periods and the corresponding fourth actual delivery index; optionally, the corresponding fitting difference data can be determined by calculating the average residual between the predicted actual delivery index and the corresponding fourth actual delivery index in multiple third historical time periods.

[0074] In a specific embodiment, when the difference between the first fitting difference data and the difference between the predicted actual delivery index and the corresponding fourth actual delivery index of the corresponding first initial decay model in multiple third historical time periods are positively correlated, and the fluctuation of the first parameter difference data and the fitting parameters in the corresponding first initial decay model in multiple third historical time periods are positively correlated, in the process of determining the first index decay model from at least two first initial decay models based on the first parameter difference data and the first fitting difference data, the first initial decay model with the smallest first parameter difference data and the smallest first fitting difference data can be preferentially selected as the first index decay model; optionally, when the first parameter difference data are the same, the first initial decay model with the smallest first fitting difference data is preferentially selected as the first index decay model; when the first fitting difference data are the same, the first initial decay model with the smallest first parameter difference data is preferentially selected as the first index decay model. Different initial decay models correspond to the smallest first fitting difference data and the smallest first parameter difference data. The first fitting difference data and the first parameter difference data corresponding to each first initial decay model can be weighted and summed by combining the preset weights corresponding to the first fitting difference data and the first parameter difference data respectively, and the first initial decay model with the smallest weighted sum value is selected as the first index decay model. Specifically, the preset weights can be set according to the actual application requirements.

[0075] In a specific embodiment, when the difference between the first fitting difference data and the corresponding first initial decay model in predicted actual delivery indicators and corresponding fourth actual delivery indicators over multiple third historical time periods is negatively correlated, and the fluctuation of the first parameter difference data and the fitting parameters in the corresponding first initial decay model over multiple third historical time periods is negatively correlated, in the process of determining the first index decay model from at least two first initial decay models based on the first parameter difference data and the first fitting difference data, the first initial decay model with the largest first parameter difference data and the largest first fitting difference data can be preferentially selected as the first index decay model; optionally, when the first parameter difference data are the same, the first initial decay model with the largest first fitting difference data is preferentially selected as the first index decay model; when the first fitting difference data are the same, the first initial decay model with the largest parameter difference data is preferentially selected as the first index decay model. The largest first fitting difference data and the largest first parameter difference data correspond to different first initial decay models. The first fitting difference data and the first parameter difference data can be weighted and summed according to their respective preset weights. The first initial decay model with the largest weighted sum value is selected as the first index decay model. Specifically, the preset weights can be set according to the actual application requirements.

[0076] In one specific embodiment, based on the second parameter difference data and the second fitting difference data, the specific refinement of the second index decay model is determined from at least two second initial decay models. This can be referred to as the specific refinement of the first index decay model determined from at least two first initial decay models based on the first parameter difference data and the first fitting difference data, which will not be repeated here.

[0077] In the above embodiments, at least two preset attenuation functions are combined to perform segmented fitting of the attenuation value from the preset delivery index to the actual delivery index, with the preset exposure time interval of the preset content delivery position as the dividing point. In the segmented fitting process, multiple fourth pull exposure durations corresponding to multiple content delivery exposures of the preset content delivery position in multiple third historical time periods, the third average preset delivery index corresponding to multiple content delivery exposures in each third historical time period, and multiple fourth actual delivery indices corresponding to multiple content delivery exposures in each third historical time period are combined. For at least two first initial attenuation models and at least two second initial attenuation models after fitting, the difference in attenuation fitting parameters (parameter difference data) and fitting error (fitting difference data) are used to further screen out the final index attenuation model that is suitable for different pull exposure duration intervals. This can effectively improve the accuracy of the final index attenuation model in representing the relationship between the attenuation change from the preset delivery index to the actual delivery index and the pull exposure duration. This improves the rationality of subsequent delivery processing such as pulling and exposure frequency control of media content corresponding to the preset content delivery position based on the index attenuation model, and effectively improves the content delivery revenue of the platform.

[0078] In one specific embodiment, the average true delivery index corresponding to the first pull exposure time interval adapted by the first preset attenuation function can be the average true delivery index of the corresponding pull exposure time interval in the first pull exposure time interval; the average true delivery index corresponding to the second pull exposure time interval adapted by the second preset attenuation function can be the average true delivery index of the corresponding pull exposure time interval in the second pull exposure time interval.

[0079] In a specific embodiment, the details of steps 2075 and S2077 described above can be found in the relevant descriptions above, and will not be repeated here.

[0080] In a specific embodiment, the preset decay function corresponding to the first index decay model is a Gaussian decay function, and the preset decay function corresponding to the second index decay model is a logarithmic decay function. Specifically, the first index decay model and the second index decay model α(t) are as follows:

[0081]

[0082] Where a1, b1, c1, a2, b2, and c2 can be fitting parameters, and t0 can be a preset time threshold.

[0083] Accordingly, the decay model for the first indicator can be:

[0084]

[0085] The first index decay model can be:

[0086]

[0087] in, It can represent the average preset delivery index; α(t) can represent the preset decay function with the pull exposure duration t as the independent variable; ECPM cost It can represent the actual delivery metrics; a1, b1, c1, a2, b2, and c2 can represent the fitting parameters, and t0 can be a preset duration threshold.

[0088] In the above embodiments, during the process of fitting the preset index decay model by combining the average actual delivery index and the first average preset delivery index corresponding to each pull exposure duration interval, the preset exposure time interval corresponding to the preset content delivery position is used as the segmented critical value of the pull exposure duration interval. The fitting parameters in the first preset decay function and the second preset decay function with the pull exposure duration as the independent variable are subjected to segmented fitting processing. This can improve the accuracy of the fitted index decay model in representing the relationship between the decay change of the preset delivery index to the actual delivery index and the pull exposure duration. In turn, it can improve the rationality of subsequent delivery processing such as pulling and exposure frequency control of media content corresponding to the preset content delivery position based on the index decay model, and effectively improve the content delivery revenue of the platform.

[0089] In this embodiment, during the fitting of the aforementioned preset index decay model, besides using the ratio of the actual delivery index corresponding to each pull-out exposure duration interval divided by a preset historical time period and the average preset delivery index of the preset historical time period to characterize the decay from the preset delivery index to the actual delivery index, the following can also be used: the ratio of the actual delivery index corresponding to each pull-out exposure duration interval divided by a preset historical time period and the preset delivery index corresponding to each pull-out exposure duration interval divided by a preset historical time period; the difference between the preset delivery index corresponding to each pull-out exposure duration interval divided by a preset historical time period and the actual delivery index corresponding to each pull-out exposure duration interval divided by a preset historical time period; and the ratio of the difference between the preset delivery index corresponding to each pull-out exposure duration interval divided by a preset historical time period and the actual delivery index corresponding to each pull-out exposure duration interval divided by a preset historical time period, as a percentage of the preset delivery index corresponding to each pull-out exposure duration interval divided by a preset historical time period. The ratio of the target value is used to characterize the decay from the preset delivery index to the actual delivery index. However, analysis of actual data from different time periods reveals that when the decay is characterized by the ratio of the actual delivery index corresponding to each pull exposure duration interval divided by the preset historical time period to the average preset delivery index of the preset historical time period, the decay trends of different time periods are almost consistent. Furthermore, different decay patterns emerge before and after the exposure frequency control strategy's time limit. The latter three decay values ​​exhibit a more discrete distribution as the pull exposure duration increases. Therefore, this application uses the ratio of the actual delivery index corresponding to each pull exposure duration interval divided by the preset historical time period to the average preset delivery index of the preset historical time period to characterize the decay from the preset delivery index to the actual delivery index. This ratio serves as the input for subsequent decay fitting. Segmented fitting is performed using the preset exposure time interval, and the preset delivery index is the average preset delivery index of the preset historical time period. This effectively avoids the impact of preset delivery index fluctuations and improves the accuracy of the fitted preset index decay model in representing the decay change from the preset delivery index to the actual delivery index and the correspondence between the pull exposure duration and the actual delivery index.

[0090] S209: Based on the preset index decay model, perform media content delivery processing on the preset content delivery positions.

[0091] In an optional embodiment, taking the retrieval of media content corresponding to a preset content delivery position using a preset index decay model as an example, the aforementioned media content delivery process may include executing a response operation for the media content retrieval request for the preset content delivery position, such as... Figure 4 As shown, the media content delivery processing for preset content delivery positions based on the preset index decay model can include:

[0092] S401: In response to a media content retrieval request, obtain the second retrieval exposure duration of the media content to be retrieved corresponding to the preset content delivery position, the second preset delivery index of the media content to be retrieved, the third retrieval exposure duration of the retrieved media content corresponding to the preset content delivery position, and the third preset delivery index of the retrieved media content.

[0093] S403: Based on the preset index decay model, the second pull exposure duration, the second preset delivery index, the third pull exposure duration, and the third preset delivery index, predict the second real delivery index corresponding to the media content to be pulled and the third real delivery index corresponding to the media content already pulled, respectively.

[0094] S405: Based on the second and third real delivery metrics, determine the pull analysis results corresponding to the preset content delivery position. The pull analysis results are used to indicate whether pulling the media content to be pulled to update the already pulled media content corresponding to the preset content delivery position increases the delivery revenue corresponding to the preset content delivery position.

[0095] S407: Based on the pull analysis results, execute the response operation for the media content pull request.

[0096] In one specific embodiment, the media content retrieval request can be a request to re-retrieve pre-loaded media content; specifically, taking the media content display position in the information display page published by the subscription account as an example, the above-mentioned media content retrieval request can be triggered when the subscription account publishes information; taking the media content display position in the information flow browsing page as an example, the media content retrieval request can be triggered before the user account triggers the information flow browsing page display, for example, by triggering the corresponding application to start, or by a preset time.

[0097] In a specific embodiment, the second pull exposure duration can be the pull exposure duration corresponding to the media content to be pulled. Specifically, since the media content to be pulled has not yet been pulled, the second pull exposure duration can be the time length standard (preset time interval) of the pull exposure duration interval corresponding to the real indicator data during the fitting process of the preset indicator decay model. Specifically, the second preset delivery indicator can be the preset delivery indicator corresponding to the media content to be pulled. The third pull exposure duration can be the pull exposure duration corresponding to the current pull of the already pulled media content. Specifically, the already pulled media content can be the media content to be exposed that is cached locally on the terminal device. Specifically, the third preset delivery indicator can be the preset delivery indicator corresponding to the current pull of the already pulled media content.

[0098] In an optional embodiment, the preset index decay model includes a first index decay model and a second index decay model, and the method may further include:

[0099] Obtain a preset duration threshold, which is the critical value of the pull exposure duration range adapted by the first index decay model and the second index decay model, and the preset duration threshold is the preset exposure time interval corresponding to the preset content delivery position.

[0100] Based on the second pull exposure duration, the third pull exposure duration, and the preset duration threshold, the first target attenuation model corresponding to the media content to be pulled and the second target attenuation model corresponding to the media content already pulled are determined from the preset index attenuation model.

[0101] Accordingly, based on the preset indicator decay model, the second pull exposure duration, the second preset delivery indicator, the third pull exposure duration, and the third preset delivery indicator, the predicted second actual delivery indicator for the media content to be pulled and the third actual delivery indicator for the media content already pulled include:

[0102] The second actual delivery indicator is determined based on the first target attenuation model, the second pull exposure duration, and the second preset delivery indicator;

[0103] Based on the second target attenuation model, the third pull exposure duration, and the third preset delivery index, the third actual delivery index is determined.

[0104] In a specific embodiment, if the pull exposure duration corresponding to the media content to be pulled is less than or equal to a preset duration threshold, the first index attenuation model can be used as the first target attenuation model; if the pull exposure duration corresponding to the media content to be pulled is greater than the preset duration threshold, the second index attenuation model can be used as the first target attenuation model; correspondingly, if the pull exposure duration corresponding to the already pulled media content is less than or equal to the preset duration threshold, the first index attenuation model can be used as the second target attenuation model; if the pull exposure duration corresponding to the already pulled media content is greater than the preset duration threshold, the second index attenuation model can be used as the second target attenuation model.

[0105] In one specific embodiment, the second pull exposure duration and the second preset delivery index can be input into the first target attenuation model to obtain the second actual delivery index; the third pull exposure duration and the third preset delivery index can be input into the second target attenuation model to obtain the third actual delivery index.

[0106] In the above embodiments, the preset exposure time interval corresponding to the preset content placement position is used as the critical value of the pull exposure duration interval adapted to the first index decay model and the second index decay model. This allows for more targeted selection of the adapted index decay model to predict the actual placement index, improve the accuracy of the actual placement index, and further enhance the rationality and effectiveness of re-pulling pre-loaded media content while improving the placement platform's revenue.

[0107] In an optional embodiment, the above-mentioned pull analysis results for determining the preset content delivery position based on the second and third real delivery metrics include:

[0108] The difference between the second and third actual delivery metrics is used as the result of the pull analysis.

[0109] In one specific embodiment, if the difference between the second real delivery indicator and the third real delivery indicator is greater than zero, the pull analysis result used to instruct the pulling of media content to be pulled to update the already pulled media content corresponding to the preset content delivery position will increase the delivery revenue corresponding to the preset content delivery position; if the difference between the second real delivery indicator and the third real delivery indicator is less than or equal to zero, the pull analysis result used to instruct the pulling of media content to be pulled to update the already pulled media content corresponding to the preset content delivery position will not increase the delivery revenue corresponding to the preset content delivery position.

[0110] In the above embodiments, the difference between the second and third real delivery metrics is used as the result of the pull analysis. This allows for re-pulling of media content if it can improve delivery revenue, thus effectively ensuring the delivery revenue of the delivery platform.

[0111] In an optional embodiment, the above method may further include:

[0112] Obtain the predicted exposure metrics and preset pull consumption data corresponding to the preset content placement positions;

[0113] Accordingly, the pull analysis results for determining the preset content placement positions based on the second and third real delivery metrics can include:

[0114] Determine the difference in the delivery metrics between the second and third actual delivery metrics;

[0115] Based on the difference in delivery metrics and the predicted exposure metrics, the target pull consumption data is determined. The target pull consumption data represents the incremental delivery revenue brought by re-pulling the media content to be pulled.

[0116] The comparison between the target fetch consumption data and the preset fetch consumption data is used as the fetch analysis result.

[0117] In one specific embodiment, the aforementioned predicted exposure index represents the probability that media content corresponding to a preset content placement position will be exposed. Optionally, the content feature information of the media content exposed in the preset content placement position within the fifth historical time period and the account feature information of the corresponding exposure account can be input into the preset exposure index prediction model for exposure index prediction processing to obtain the aforementioned predicted exposure index. Specifically, the preset exposure index prediction model can be a pre-trained artificial intelligence model for predicting exposure indicators.

[0118] Furthermore, the above is only one way to determine the predicted exposure metric. In practical applications, other methods can be combined to determine the predicted exposure metric corresponding to the preset content placement position.

[0119] In a specific embodiment, the aforementioned preset retrieval consumption data can be the incremental lower limit of the delivery revenue corresponding to the re-retrieval of media content; specifically, it can be determined in advance based on the system resource consumption required for the system to retrieve media resources, and the incremental lower limit of the delivery revenue corresponding to the re-retrieval of media content is a value greater than zero, so as to ensure that the delivery revenue brought by the re-retrieved media content (media content to be retrieved) is greater than the delivery revenue brought by the already retrieved media content.

[0120] In one specific embodiment, the difference between the second and third actual delivery metrics can be the second actual delivery metric minus the third actual delivery metric. Specifically, the product of the delivery metric difference and the predicted exposure metric can be used as the target to pull consumption data.

[0121] In one specific embodiment, when the target pull consumption data is greater than the preset pull consumption data, a pull analysis result can be determined to indicate that the pull of media content to be pulled in order to update the already pulled media content corresponding to the preset content delivery position increases the delivery revenue corresponding to the preset content delivery position; when the target pull consumption data is less than or equal to the preset pull consumption data, a pull analysis result can be determined to indicate that the pull of media content to be pulled in order to update the already pulled media content corresponding to the preset content delivery position does not increase the delivery revenue corresponding to the preset content delivery position.

[0122] In the above embodiments, during the pull analysis process, which combines the actual delivery metrics corresponding to the already pulled media content and the media to be pulled, the target pull consumption data representing the increase in delivery revenue brought about by re-pulling the media content is determined by first combining the difference between the actual delivery metrics corresponding to the media to be pulled and the actual delivery metrics corresponding to the already pulled media content, and the predicted exposure metrics corresponding to the preset content delivery position. Then, the pull analysis results are determined by comparing the target pull consumption data with the lower limit of the increase in delivery revenue corresponding to the re-pulled media content (preset pull consumption data). This allows for a more comprehensive and accurate analysis of whether re-pulling media content can improve delivery revenue, thereby improving the rationality of platform media content pull and improving the platform's delivery revenue.

[0123] In an optional embodiment, the above-described response operation to the media content fetch request based on the fetch analysis results may include:

[0124] If the pull analysis results indicate that the media content to be pulled should be pulled in order to update the already pulled media content corresponding to the preset content placement position and increase the placement revenue corresponding to the preset content placement position, the media content to be pulled should be pulled, and the already pulled media content cached locally should be replaced based on the media content to be pulled.

[0125] In the above embodiments, when the pull analysis results indicate that the media content to be pulled should be pulled in order to update the already pulled media content corresponding to the preset content placement position and increase the placement revenue corresponding to the preset content placement position, the media content to be pulled should be pulled, and the already pulled media content cached locally should be replaced based on the media content to be pulled. This can improve the rationality of the platform's media content retrieval while effectively ensuring the placement revenue of the placement platform.

[0126] In an optional embodiment, if the pull analysis result indicates that pulling the media content to be pulled to update the pulled media content corresponding to the preset content delivery position does not increase the delivery revenue corresponding to the preset content delivery position, a non-pull response operation can be performed on the media content pull request.

[0127] In a specific embodiment, such as Figure 5As shown, Figure 5 This is a schematic diagram illustrating the media content delivery and exposure process corresponding to a preset content delivery position, according to an exemplary embodiment; specifically, in conjunction with... Figure 5 As can be seen, the log module can collect data such as historical fetch time, historical exposure time, historical preset delivery metrics, and historical actual delivery metrics by combining the media content delivery and exposure processing of user terminals. It can also calculate data such as historical fetch exposure duration, historical average preset delivery metrics, and historical actual delivery metrics based on the collected data, and provide data support for the preset metric decay model. Correspondingly, the preset metric decay model can predict the corresponding actual delivery metrics. Furthermore, the content fetch module can perform fetch analysis based on the actual delivery metrics predicted by the preset metric decay model. If the analysis indicates that media content needs to be fetched again (i.e., fetching the media content to be fetched to update the already fetched media content corresponding to the preset content delivery position to increase the delivery revenue corresponding to the preset content delivery position), it sends a media content fetch request to the media content server and returns the fetched media content to the user terminal. Furthermore, when the user triggers exposure, the content display module on the user terminal side can trigger the display of pre-fetched (pre-loaded) media content.

[0128] In the above embodiments, when a media content retrieval request is triggered for a preset content placement position, the second retrieval exposure duration, the second preset placement index of the media content to be retrieved corresponding to the preset content placement position, the third retrieval exposure duration, and the third preset placement index of the already retrieved media content corresponding to the preset content placement position are obtained. Based on the preset index decay model, the second retrieval exposure duration, the second preset placement index, the third retrieval exposure duration, and the third preset placement index, the second actual placement index and the third actual placement index corresponding to the media content to be retrieved are predicted respectively. This can effectively ensure the accuracy of the actual placement index. Furthermore, by combining the second and third actual placement indices, a retrieval analysis result is determined to indicate whether retrieving the media content to be retrieved to update the already retrieved media content corresponding to the preset content placement position increases the placement revenue of the preset content placement position. Response operations for the media content retrieval request are then executed. This can improve the rationality of media content retrieval on the platform while effectively ensuring the placement revenue of the placement platform.

[0129] In an optional embodiment, taking the control of content exposure frequency by combining a preset index attenuation model as an example, the above media content delivery process may include exposure frequency control processing of media content in preset content delivery positions, such as... Figure 6 As shown, the media content delivery processing for preset content delivery positions based on the preset index decay model can include:

[0130] S601: Obtain the second average preset delivery index corresponding to multiple historical delivery exposures of the preset content delivery position within the second historical time period and multiple fifth real delivery indices corresponding to multiple historical delivery exposures within the second historical time period.

[0131] S603: Determine the target real delivery indicator that is at a preset percentage among multiple fifth real delivery indicators;

[0132] S605: Determine the target exposure recycle time based on the preset index decay model, the second average preset delivery index, and the target actual delivery index;

[0133] S607: Based on the target exposure recycle time, perform exposure frequency control processing on media content in the preset content delivery position.

[0134] In one specific embodiment, the second historical time period can be set according to the actual application, such as the most recent week. Optionally, the second historical time period can be different from or the same as the first historical time period. Specifically, the second average preset delivery index can be the average of multiple preset delivery indexes corresponding to multiple content delivery exposures within the second historical time period.

[0135] In one specific embodiment, each content delivery exposure corresponds to a fifth true delivery metric; specifically, the fifth true delivery metric corresponding to any content delivery exposure can be the true delivery metric corresponding to that content delivery exposure.

[0136] In a specific embodiment, the preset percentile can be set according to the actual application. Specifically, the preset percentile can represent the ranking position of the target real delivery indicator when multiple fifth real delivery indicators are ranked based on the indicator size. For example, if the preset percentile is 50% and the target real delivery indicator is the median, the ranking position is the middle position. Specifically, it can be sorted in ascending order according to the indicator size or in descending order according to the indicator size.

[0137] In one specific embodiment, the preset index decay model includes a first index decay model and a second index decay model. Optionally, a third target decay model can be determined from the first and second index decay models by combining the exposure replay duration corresponding to the target actual delivery index. Specifically, if the exposure replay duration corresponding to the target actual delivery index is less than or equal to a preset duration threshold, the first index decay model can be used as the third target decay model; if the exposure replay duration corresponding to the target actual delivery index is greater than the preset duration threshold, the second index decay model can be used as the third target decay model. Further, the second average preset delivery index and the target actual delivery index are input into the third target decay model to obtain the target exposure replay duration.

[0138] In an optional embodiment, the above method may further include:

[0139] Get the content delivery feedback information corresponding to the current user account;

[0140] Accordingly, the above-mentioned exposure frequency control processing of media content in preset content placement positions based on target exposure return duration may include:

[0141] If the target exposure return duration is greater than the first preset threshold, and the content delivery feedback information indicates that the current user account is interested in the content delivered in the preset content delivery position, the preset exposure time interval corresponding to the preset content delivery position will be reduced.

[0142] or,

[0143] If the target exposure return time is less than or equal to the second preset threshold, and the content delivery feedback information indicates that the current user account is not interested in the content delivered in the preset content delivery position, increase the preset exposure time interval corresponding to the preset content delivery position.

[0144] In one specific embodiment, the content delivery feedback information can be used to indicate whether the current user account is interested in the content delivered to the preset content delivery position; optionally, it can be determined by combining the historical operation data of the current user account; the first preset threshold is greater than or equal to the second preset threshold.

[0145] In the above embodiments, during the content exposure frequency control process, the target exposure return duration for exposure frequency control is determined by combining the second average preset delivery index corresponding to multiple historical delivery exposures of the preset content delivery position within the second historical time period, the target real delivery index at the preset percentage position among multiple fifth real delivery indicators corresponding to multiple historical delivery exposures within the second historical time period, and the preset index decay model characterizing the decay change from the preset delivery index to the real delivery index and the corresponding relationship between the pull exposure duration. This effectively ensures the rationality of the exposure time interval setting. Furthermore, by combining the feedback information used to indicate the current user account's interest in the delivery content corresponding to the preset content delivery position for exposure frequency control, the rationality and effectiveness of content exposure frequency control can be improved, while also better meeting the content exposure needs of different users and enhancing the user experience.

[0146] Furthermore, it should be noted that the preset index decay model in this application embodiment can be updated periodically by combining the exposure data (exposure duration, average preset index, and actual index, etc.) within the most recent preset historical time period of the preset placement position. This improves the accuracy of the preset index decay model in representing the relationship between the decay change of the preset index to the actual index and the exposure duration, thereby improving the prediction accuracy of the actual index and the effectiveness and rationality of exposure frequency control.

[0147] As can be seen from the technical solutions provided in the embodiments of this specification above, in the process of media content delivery processing for preset content delivery positions, this specification obtains multiple first pull-out exposure durations, first average preset delivery indicators, and multiple first real delivery indicators corresponding to multiple content delivery exposures within a first historical time period for the preset content delivery position. Based on the preset time interval and the multiple first pull-out exposure durations, the multiple first real delivery indicators are clustered to obtain at least one first real delivery indicator corresponding to each of the multiple pull-out exposure duration intervals, and the average real delivery indicator corresponding to each pull-out exposure duration interval is determined. Then, based on the average real delivery indicator and the first average preset delivery indicator corresponding to each pull-out exposure duration interval, a preset indicator decay model is fitted to characterize the decay change from the preset delivery indicator to the real delivery indicator and the correspondence between the pull-out exposure duration. Combining the preset indicator decay model with the media content delivery processing for the preset content delivery position can effectively reduce the risk of over-broadcasting of pre-loaded media content caused by the extension of the pull-out exposure duration. In addition, it can also improve the rationality and effectiveness of content delivery on the delivery platform while increasing the delivery revenue of the delivery platform.

[0148] This application also provides a media content delivery and processing device, such as... Figure 7 As shown, the above-mentioned device includes:

[0149] The first data acquisition module 710 is configured to acquire multiple first pull exposure durations, first average preset delivery indicators, and multiple first real delivery indicators corresponding to multiple content delivery exposures within a first historical time period for a preset content delivery position.

[0150] The delivery metric clustering module 720 is configured to perform clustering of multiple first real delivery metrics based on a preset time interval and multiple first pull exposure durations, to obtain at least one first real delivery metric corresponding to each of the multiple pull exposure duration intervals.

[0151] The first indicator determination module 730 is configured to execute at least one first real delivery indicator based on each of the multiple pull exposure duration intervals to determine the average real delivery indicator corresponding to each pull exposure duration interval.

[0152] The attenuation model fitting module 740 is configured to perform a fitting of a preset index attenuation model based on the average real delivery index and the first average preset delivery index corresponding to each pull exposure duration interval. The preset index attenuation model is used to characterize the attenuation change from the preset delivery index to the real delivery index and the correspondence between the pull exposure duration.

[0153] The media content processing module 750 is configured to perform media content delivery processing on preset content delivery positions based on a preset index decay model.

[0154] In an optional embodiment, the media content delivery process includes performing a response operation to a media content fetch request for a preset content delivery location; the media content processing module 750 includes:

[0155] The first data acquisition unit is configured to respond to the media content retrieval request by acquiring the second retrieval exposure duration of the media content to be retrieved corresponding to the preset content delivery position, the second preset delivery index of the media content to be retrieved, the third retrieval exposure duration of the retrieved media content corresponding to the preset content delivery position, and the third preset delivery index of the retrieved media content.

[0156] The first indicator determination unit is configured to execute based on a preset indicator decay model, a second pull exposure duration, a second preset delivery indicator, a third pull exposure duration, and a third preset delivery indicator to predict the second real delivery indicator corresponding to the media content to be pulled and the third real delivery indicator corresponding to the media content already pulled, respectively.

[0157] The pull analysis unit is configured to perform pull analysis based on the second and third real delivery metrics to determine the pull analysis results corresponding to the preset content delivery position. The pull analysis results are used to indicate whether pulling the media content to be pulled to update the already pulled media content corresponding to the preset content delivery position increases the delivery revenue corresponding to the preset content delivery position.

[0158] The pull response unit is configured to perform a response operation based on the pull analysis results in response to the media content pull request.

[0159] In an optional embodiment, the above-described apparatus further includes:

[0160] The second data acquisition module is configured to acquire the predicted exposure index and the preset pull consumption data corresponding to the preset content placement position. The predicted exposure index represents the probability that the media content corresponding to the preset content placement position will be exposed; the preset pull consumption data is the lower limit of the incremental revenue of the media content to be re-pulled.

[0161] The pull analysis unit includes:

[0162] The unit for determining the difference between the delivery indicators is configured to determine the difference between the second and third actual delivery indicators.

[0163] The unit for determining consumption data is configured to determine the target consumption data based on the difference between the delivery metrics and the predicted exposure metrics. The target consumption data represents the incremental delivery revenue brought by re-fetching the media content to be fetched.

[0164] The first pull analysis result determination unit is configured to use the comparison result between the target pull consumption data and the preset pull consumption data as the pull analysis result.

[0165] In an optional embodiment, the preset index decay model includes a first index decay model and a second index decay model, and the above-mentioned device further includes:

[0166] The preset duration threshold acquisition module is configured to acquire a preset duration threshold. The preset duration threshold is the critical value of the pull exposure duration range adapted by the first index decay model and the second index decay model, and the preset duration threshold is the preset exposure time interval corresponding to the preset content delivery position.

[0167] The attenuation model determination module is configured to determine, based on the second pull exposure duration, the third pull exposure duration, and the preset duration threshold, the first target attenuation model corresponding to the media content to be pulled and the second target attenuation model corresponding to the media content already pulled from the preset index attenuation model.

[0168] The first indicator determination unit includes:

[0169] The second indicator determination unit is configured to determine the second actual delivery indicator based on the first target attenuation model, the second pull exposure duration, and the second preset delivery indicator.

[0170] The third indicator determination unit is configured to determine the third actual delivery indicator based on the second target attenuation model, the third pull exposure duration, and the third preset delivery indicator.

[0171] In an optional embodiment, the pull analysis unit includes:

[0172] The first pull analysis result determination unit is configured to use the difference between the second and third actual delivery metrics as the pull analysis result.

[0173] In an optional embodiment, the pull response unit includes:

[0174] The pull execution unit is configured to pull media content when the pull analysis result indicates that the media content to be pulled should be pulled in order to update the already pulled media content corresponding to the preset content delivery position and increase the delivery revenue corresponding to the preset content delivery position.

[0175] The content update unit is configured to replace the locally cached fetched media content with the media content to be fetched.

[0176] In an optional embodiment, the media content delivery processing includes exposure frequency control processing of media content at a preset content delivery position, and the media content processing module 750 includes:

[0177] The second data acquisition unit is configured to acquire the second average preset delivery index corresponding to multiple historical delivery exposures of the preset content delivery position within the second historical time period and multiple fifth real delivery indices corresponding to multiple historical delivery exposures within the second historical time period.

[0178] The fourth indicator determination unit is configured to determine the target real delivery indicator that is in a preset percentile among multiple fifth real delivery indicators;

[0179] The exposure recycle duration determination unit is configured to determine the target exposure recycle duration based on a preset index decay model, a second average preset delivery index, and a target actual delivery index.

[0180] The exposure frequency control processing unit is configured to perform exposure frequency control processing on media content in preset content delivery positions based on the target exposure reflow duration.

[0181] In an optional embodiment, the above-described apparatus further includes:

[0182] The content delivery feedback information acquisition module is configured to acquire the content delivery feedback information corresponding to the current user account;

[0183] The exposure frequency control processing unit includes:

[0184] The first frequency control processing unit is configured to reduce the preset exposure time interval corresponding to the preset content delivery position when the target exposure return duration is greater than the first preset threshold and the content delivery feedback information indicates that the current user account is interested in the delivery content corresponding to the preset content delivery position.

[0185] or,

[0186] The second frequency control processing unit is configured to increase the preset exposure time interval corresponding to the preset content delivery position when the target exposure return time is less than or equal to the second preset threshold and the content delivery feedback information indicates that the current user account is not interested in the delivery content corresponding to the preset content delivery position.

[0187] Among them, the first preset threshold is greater than or equal to the second preset threshold.

[0188] In an optional embodiment, the preset index decay model includes a first index decay model and a second index decay model, and the decay model fitting module 740 includes:

[0189] The third data acquisition unit is configured to execute the acquisition of a first preset decay function with the pull exposure duration as the independent variable, a second preset decay function with the pull exposure duration as the independent variable, and a preset duration threshold; the preset duration threshold is the critical value of the pull exposure duration interval adapted by the first preset decay function and the second preset decay function, and the preset duration threshold is the preset exposure time interval corresponding to the preset content delivery position;

[0190] The fifth indicator determination unit is configured to perform, based on a preset duration threshold, to determine, from multiple average actual delivery indicators corresponding to the first pull exposure duration interval adapted by the first preset decay function, the average actual delivery indicator corresponding to the second pull exposure duration interval adapted by the second preset decay function, respectively.

[0191] The first fitting processing unit is configured to perform fitting processing on the fitting parameters in the first preset decay function based on the average real delivery index and the first average preset delivery index corresponding to the first pull exposure time interval, so as to obtain the first index decay model.

[0192] The second fitting processing unit is configured to perform fitting processing on the fitting parameters in the second preset decay function based on the average real delivery index corresponding to the second pull exposure time interval and the first average preset delivery index, so as to obtain the second index decay model.

[0193] In an optional embodiment, the first preset attenuation function and the second preset attenuation function are determined using the following modules:

[0194] The third data acquisition module is configured to acquire at least two preset decay functions and preset content delivery positions for multiple content delivery exposures in multiple third historical time periods, the third average preset delivery index for multiple content delivery exposures in each third historical time period, and the multiple fourth real delivery index for multiple content delivery exposures in each third historical time period.

[0195] The segmented fitting processing module is configured to perform segmented fitting processing on the fitting parameters in each preset attenuation function based on multiple fourth real delivery indicators and the third average preset delivery indicator corresponding to each third historical time period, so as to obtain at least two first initial attenuation models corresponding to the first pull exposure time interval and at least two second initial attenuation models corresponding to the second pull exposure time interval.

[0196] The parameter difference data determination module is configured to perform the determination of first parameter difference data of the fitted parameters in each first initial decay model over multiple third historical time periods and second parameter difference data of the fitted parameters in each second initial decay model over multiple third historical time periods.

[0197] The fitting difference data determination module is configured to perform the determination of first fitting difference data for each first initial decay model in multiple third historical time periods and second fitting difference data for each second initial decay model in multiple third historical time periods;

[0198] The first attenuation model determination module is configured to perform a first index attenuation model determination from at least two first initial attenuation models based on first parameter difference data and first fitting difference data.

[0199] The second attenuation model determination module is configured to determine a second index attenuation model from at least two second initial attenuation models based on second parameter difference data and second fitted difference data.

[0200] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0201] Figure 8 This is a block diagram of an electronic device for media content processing provided in an embodiment of this application. The electronic device may be a server, and its internal structure diagram may be as follows. Figure 8 As shown, the electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a media content delivery processing method.

[0202] Those skilled in the art will understand that Figure 8The structure shown is merely a block diagram of a portion of the structure related to the present disclosure and does not constitute a limitation on the electronic device to which the present disclosure is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0203] In an exemplary embodiment, an electronic device is also provided, including: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the media content delivery processing method as described in the embodiments of this disclosure.

[0204] In an exemplary embodiment, a computer-readable storage medium is also provided, wherein when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the media content delivery processing method of the present disclosure embodiments.

[0205] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the media content delivery processing method provided in the various optional implementations described above.

[0206] It is understood that in the specific implementation of this application, user-related data is involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0207] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0208] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0209] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for media content delivery and processing, characterized in that, The method includes: Get multiple first pull exposure durations corresponding to multiple content delivery exposures within a first historical time period, a first average preset delivery index corresponding to the multiple content delivery exposures, and multiple first real delivery indices corresponding to the multiple content delivery exposures. Based on a preset time interval and the plurality of first pull exposure durations, the plurality of first real delivery metrics are clustered to obtain at least one first real delivery metric corresponding to each of the plurality of pull exposure duration intervals. Based on at least one first real delivery indicator corresponding to each of the multiple pull exposure duration intervals, determine the average real delivery indicator corresponding to each pull exposure duration interval. Based on the average actual delivery index and the first average preset delivery index corresponding to each pull exposure duration interval, a preset index decay model is fitted. The preset index decay model is used to characterize the decay change from the preset delivery index to the actual delivery index and the correspondence between the pull exposure duration. Based on the preset index decay model, media content is delivered to the preset content delivery position.

2. The method according to claim 1, characterized in that, The media content delivery process includes executing a response operation for the media content retrieval request for the preset content delivery position, and the media content delivery process for the preset content delivery position based on the preset index decay model includes: In response to the media content retrieval request, the system obtains the second retrieval exposure duration of the media content to be retrieved corresponding to the preset content delivery position, the second preset delivery index of the media content to be retrieved, the third retrieval exposure duration of the retrieved media content corresponding to the preset content delivery position, and the third preset delivery index of the retrieved media content. Based on the preset index decay model, the second pull exposure duration, the second preset delivery index, the third pull exposure duration, and the third preset delivery index, the second real delivery index corresponding to the media content to be pulled and the third real delivery index corresponding to the media content already pulled are predicted respectively. Based on the second and third real delivery metrics, the pull analysis results corresponding to the preset content delivery position are determined. The pull analysis results are used to indicate whether pulling the media content to be pulled to update the already pulled media content corresponding to the preset content delivery position increases the delivery revenue corresponding to the preset content delivery position. Based on the pull analysis results, execute the response operation for the media content pull request.

3. The method according to claim 2, characterized in that, The method further includes: Obtain the predicted exposure metric and the preset pull consumption data corresponding to the preset content placement position. The predicted exposure metric represents the probability that the media content corresponding to the preset content placement position will be exposed. The preset pull consumption data is the lower limit of the incremental revenue of the media content to be re-pulled. The step of determining the pull analysis results corresponding to the preset content delivery position based on the second and third real delivery metrics includes: Determine the difference in delivery metrics between the second and third actual delivery metrics; Based on the difference in the delivery metrics and the predicted exposure metrics, the target pull consumption data is determined, whereby the target pull consumption data represents the incremental delivery revenue brought about by re-pulling the media content to be pulled. The comparison result between the target pull consumption data and the preset pull consumption data is used as the pull analysis result.

4. The method according to claim 2, characterized in that, The preset index decay model includes a first index decay model and a second index decay model, and the method further includes: Obtain a preset duration threshold, wherein the preset duration threshold is the critical value of the pull exposure duration interval adapted by the first index decay model and the second index decay model, and the preset duration threshold is the preset exposure time interval corresponding to the preset content delivery position; Based on the second pull exposure duration, the third pull exposure duration, and the preset duration threshold, the first target attenuation model corresponding to the media content to be pulled and the second target attenuation model corresponding to the already pulled media content are determined from the preset index attenuation model; The step of predicting the second actual delivery indicator corresponding to the media content to be retrieved and the third actual delivery indicator corresponding to the media content already retrieved, based on the preset indicator decay model, the second pull exposure duration, the second preset delivery indicator, the third pull exposure duration, and the third preset delivery indicator, respectively, includes: The second actual delivery indicator is determined based on the first target attenuation model, the second pull exposure duration, and the second preset delivery indicator; The third actual delivery indicator is determined based on the second target attenuation model, the third pull exposure duration, and the third preset delivery indicator.

5. The method according to claim 2, characterized in that, The step of determining the pull analysis results corresponding to the preset content delivery position based on the second and third real delivery metrics includes: The difference between the second and third actual delivery metrics is used as the result of the pull analysis.

6. The method according to claim 2, characterized in that, The step of performing a response operation for the media content fetch request based on the fetch analysis results includes: When the pull analysis result indicates that the media content to be pulled should be pulled in order to update the already pulled media content corresponding to the preset content delivery position and increase the delivery revenue corresponding to the preset content delivery position, the media content to be pulled should be pulled, and the already pulled media content cached locally should be replaced based on the media content to be pulled.

7. The method according to claim 1, characterized in that, The media content delivery processing includes exposure frequency control processing of the media content in the preset content delivery position, and the media content delivery processing of the preset content delivery position based on the preset index attenuation model includes: Obtain the second average preset delivery index corresponding to multiple historical delivery exposures of the preset content delivery position within the second historical time period, and multiple fifth real delivery indices corresponding to multiple historical delivery exposures within the second historical time period. Determine the target real delivery indicator that is at a preset percentage among the plurality of fifth real delivery indicators; The target exposure recycle time is determined based on the preset index decay model, the second average preset delivery index, and the target actual delivery index. Based on the target exposure reflow duration, the media content in the preset content delivery position is subjected to exposure frequency control processing.

8. The method according to claim 7, characterized in that, The method further includes: Get the content delivery feedback information corresponding to the current user account; The exposure frequency control processing of the media content in the preset content delivery position based on the target exposure reflow duration includes: If the target exposure return duration is greater than a first preset threshold, and the content delivery feedback information indicates that the current user account is interested in the delivery content corresponding to the preset content delivery position, then the preset exposure time interval corresponding to the preset content delivery position is reduced. or, If the target exposure return time is less than or equal to the second preset threshold, and the content delivery feedback information indicates that the current user account is not interested in the delivery content corresponding to the preset content delivery position, the preset exposure time interval corresponding to the preset content delivery position shall be increased. Wherein, the first preset threshold is greater than or equal to the second preset threshold.

9. The method according to any one of claims 1 to 8, characterized in that, The preset index decay model includes a first index decay model and a second index decay model. The step of fitting the preset index decay model based on the average actual delivery index corresponding to each pull exposure duration interval and the first average preset delivery index includes: Obtain a first preset decay function with pull exposure duration as the independent variable, a second preset decay function with pull exposure duration as the independent variable, and a preset duration threshold; the preset duration threshold is the critical value of the pull exposure duration interval adapted by the first preset decay function and the second preset decay function, and the preset duration threshold is the preset exposure time interval corresponding to the preset content delivery position; Based on the preset duration threshold, the average real delivery index corresponding to the first pull exposure duration interval adapted by the first preset decay function and the average real delivery index corresponding to the second pull exposure duration interval adapted by the second preset decay function are determined from the average real delivery index corresponding to the multiple pull exposure duration intervals. Based on the average actual delivery index corresponding to the first pull exposure time interval and the first average preset delivery index, the fitting parameters in the first preset decay function are fitted to obtain the first index decay model. Based on the average actual delivery index corresponding to the second pull exposure duration interval and the first average preset delivery index, the fitting parameters in the second preset decay function are fitted to obtain the second index decay model.

10. The method according to claim 9, characterized in that, The first preset attenuation function and the second preset attenuation function are determined in the following manner: Obtain at least two preset decay functions and multiple fourth pull exposure durations corresponding to multiple content delivery exposures in multiple third historical time periods, a third average preset delivery index corresponding to multiple content delivery exposures in each third historical time period, and multiple fourth real delivery indices corresponding to multiple content delivery exposures in each third historical time period. Based on the multiple fourth real delivery indicators corresponding to each third historical time period and the third average preset delivery indicator corresponding to each third historical time period, the fitting parameters in each preset attenuation function are subjected to segmented fitting processing to obtain at least two first initial attenuation models corresponding to the first pull exposure time interval and at least two second initial attenuation models corresponding to the second pull exposure time interval. Determine the first parameter difference data of the fitting parameters in each first initial decay model within the plurality of third historical time periods and the second parameter difference data of the fitting parameters in each second initial decay model within the plurality of third historical time periods; Determine the first fitting difference data for each first initial decay model within the plurality of third historical time periods and the second fitting difference data for each second initial decay model within the plurality of third historical time periods; Based on the first parameter difference data and the first fitting difference data, the first index decay model is determined from the at least two first initial decay models; Based on the second parameter difference data and the second fitting difference data, the second index decay model is determined from the at least two second initial decay models.

11. A media content delivery and processing device, characterized in that, The device includes: The first data acquisition module is configured to acquire multiple first pull exposure durations corresponding to multiple content delivery exposures within a first historical time period, a first average preset delivery index corresponding to the multiple content delivery exposures, and multiple first real delivery indices corresponding to the multiple content delivery exposures. The delivery metric clustering module is configured to perform clustering of the multiple first real delivery metrics based on a preset time interval and the multiple first pull exposure durations, to obtain at least one first real delivery metric corresponding to each of the multiple pull exposure duration intervals. The first indicator determination module is configured to execute at least one first real delivery indicator based on each of the plurality of pull exposure duration intervals to determine the average real delivery indicator corresponding to each pull exposure duration interval. The attenuation model fitting module is configured to perform a fitting of a preset index attenuation model based on the average actual delivery index and the first average preset delivery index corresponding to each pull exposure duration interval. The preset index attenuation model is used to characterize the attenuation change from the preset delivery index to the actual delivery index and the correspondence between the pull exposure duration. The media content processing module is configured to perform media content delivery processing on the preset content delivery position based on the preset index decay model.

12. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the media content delivery processing method as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is able to perform the media content delivery processing method as described in any one of claims 1 to 10.

14. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by an electronic device, it implements the media content delivery processing method according to any one of claims 1 to 10.

Images