CN116647601A - Method and system for dynamically adjusting request quantity of promotion content based on filling rate - Google Patents

Method and system for dynamically adjusting request quantity of promotion content based on filling rate Download PDF

Info

Publication number
CN116647601A
CN116647601A CN202310920166.4A CN202310920166A CN116647601A CN 116647601 A CN116647601 A CN 116647601A CN 202310920166 A CN202310920166 A CN 202310920166A CN 116647601 A CN116647601 A CN 116647601A
Authority
CN
China
Prior art keywords
dsp platform
filling rate
dsp
platform
real
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202310920166.4A
Other languages
Chinese (zh)
Other versions
CN116647601B (en
Inventor
甄建琦
周暾
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Chuangzhihui Technology Co ltd
Original Assignee
Beijing Chuangzhihui Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Chuangzhihui Technology Co ltd filed Critical Beijing Chuangzhihui Technology Co ltd
Priority to CN202310920166.4A priority Critical patent/CN116647601B/en
Publication of CN116647601A publication Critical patent/CN116647601A/en
Application granted granted Critical
Publication of CN116647601B publication Critical patent/CN116647601B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0277—Online advertisement
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/2866—Architectures; Arrangements
    • H04L67/30—Profiles
    • H04L67/306—User profiles
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Abstract

The disclosure provides a method and a system for dynamically adjusting promotion content request amount based on filling rate, wherein the method comprises the following steps: setting observation periods and attenuation coefficients of the DSP platforms, obtaining historical interface service conditions of at least one observation period of the DSP platforms, and adjusting the request quantity of the DSP platforms by combining the historical filling rate of the DSP platforms in the at least one observation period and the attenuation coefficients of the DSP platforms; acquiring real-time response data of any adjusted DSP platform, acquiring a real-time interface service condition of the DSP platform, updating a filling rate prediction model based on the real-time response data of the DSP platform, and calculating the real-time filling rate of the DSP platform; and adjusting the request quantity of the DSP platform based on the real-time interface service condition and the real-time filling rate of the DSP platform. According to the method and the device, the request quantity is dynamically adjusted according to the historical filling rate and the real-time filling rate of each DSP platform, so that the invalid request cost is reduced.

Description

Method and system for dynamically adjusting request quantity of promotion content based on filling rate
Technical Field
The disclosure relates to the technical field of Internet, in particular to a method and a system for dynamically adjusting popularization content request amount based on filling rate.
Background
In the field of programmed transaction, the ADX (AD Exchange) Platform needs to send requests of promotion contents to a plurality of DSP (Demand-Side Platform) platforms to bid so as to return the promotion contents suitable for the media Side or the user Side, but the DSP platforms may not have promotion contents suitable for the media Side or the user Side, or the DSP platforms cannot be temporarily released due to insufficient budget, which can result in far less return quantity of the DSP platforms than the request quantity, lower filling rate and full bandwidth in most of the time, thereby increasing bandwidth cost and affecting effective requests and responses of the DSP platforms. Therefore, in order to improve the filling rate of the DSP platform, improve the utilization rate of the request, and reduce the invalid request cost, it is necessary to observe and analyze the filling performance of each DSP platform, so as to dynamically adjust the amount of popularization content requests allocated based on the filling performance of each DSP platform.
Disclosure of Invention
The present disclosure provides a method and a system for dynamically adjusting request amount of promotion content based on filling rate, so as to solve the problems of high invalid request cost, low filling rate and the like in the prior art, and realize dynamic adjustment of request amount of each platform of each DSP.
In a first aspect, the present disclosure provides a method for dynamically adjusting a request amount of promotion content based on a filling rate, the method being applied to an ADX platform, the method comprising:
setting an observation period and an attenuation coefficient of each DSP platform;
collecting historical response data of at least one observation period of each DSP platform, extracting relevant characteristics of the filling rate, constructing a filling rate prediction model based on the relevant characteristics of the filling rate, and calculating the historical filling rate of each DSP platform in the at least one observation period;
acquiring the historical interface service condition of at least one observation period of each DSP platform, and adjusting the request quantity of each DSP platform by combining the historical filling rate of each DSP platform in at least one observation period and the attenuation coefficient of each DSP platform;
acquiring real-time response data of any adjusted DSP platform, acquiring a real-time interface service condition of the DSP platform, updating a filling rate prediction model based on the real-time response data of the DSP platform, and calculating the real-time filling rate of the DSP platform;
and adjusting the request quantity of the DSP platform based on the real-time interface service condition and the real-time filling rate of the DSP platform.
According to the method for dynamically adjusting the request quantity of promotion content based on the filling rate, the historical response data of at least one observation period of each DSP platform is collected, and the method comprises the following steps: analyzing the flow of the promotion content requests of different links in four dimensions of the DSP platform, promotion position, operating system and promotion form, and collecting response data of the promotion content requests of different links in at least one observation period of each DSP platform; wherein the response data includes: filling condition, response time and promotion information, wherein the promotion information comprises the type of promotion content, the size of promotion content, the material quality of promotion content, the delivery time of promotion content and the price of promotion content.
According to the method for dynamically adjusting the request quantity of the promotion content based on the filling rate, the observation period comprises a tolerance period and an adjustment period, and the attenuation coefficient comprises a first attenuation coefficient and a second attenuation coefficient; the method for acquiring the historical interface service condition of at least one observation period of each DSP platform and adjusting the request quantity of each DSP platform by combining the historical filling rate of each DSP platform in at least one observation period and the attenuation coefficient of each DSP platform comprises the following steps: based on historical response data of at least one observation period of any DSP platform, acquiring historical interface service conditions of at least one observation period of the DSP platform; judging whether the history interface service condition of the DSP platform is normal or not; if the history interface service condition of the DSP platform is normal, judging whether the history filling rate of the DSP platform in at least one observation period is lower than a preset filling rate, and if so, adjusting the request quantity of the DSP platform based on the attenuation coefficient of the DSP platform; and if the historical interface service condition of the DSP platform is abnormal, fusing the DSP platform.
According to the method for dynamically adjusting the request quantity of promotion content based on the filling rate, the method for adjusting the request quantity of the DSP platform based on the attenuation coefficient of the DSP platform comprises the following steps: determining a current attenuation coefficient based on the attenuation coefficient of the DSP platform; calculating the request attenuation amount of the DSP platform, wherein the request attenuation amount is the product of the initial request amount and the current attenuation coefficient; and adjusting the request quantity of the DSP platform to be the difference between the request attenuation quantity and the initial request quantity.
According to the method for dynamically adjusting the request quantity of promotion content based on the filling rate, the method for determining the current attenuation coefficient based on the attenuation coefficient of the DSP platform comprises the following steps: if the observation period is a tolerance period and the historical filling rate of the DSP platform is lower than a preset filling rate, determining that the current attenuation coefficient is a first attenuation coefficient and determining that the next observation period is an adjustment period; if the observation period is an adjustment period and the historical filling rate of the DSP platform is lower than a preset filling rate, judging whether the current attenuation coefficient of the DSP platform is lower than a preset attenuation ratio, if so, determining that the current attenuation coefficient is the sum of the current attenuation coefficient and the second attenuation coefficient, otherwise, keeping unchanged; if the observation period is an adjustment period and the historical filling rate of the DSP platform is not lower than the preset filling rate, resetting the current attenuation coefficient and determining the next observation period as a tolerance period.
According to the method for dynamically adjusting the request quantity of the promotion content based on the filling rate, which is provided by the present disclosure, whether the service condition of the historical interface of the DSP platform is normal is judged, which specifically comprises: judging whether a link of any promotion content request is available or not; and/or judging whether the response time of any promotion content request accords with the preset response time; and/or judging whether the promotion information of any promotion content request accords with the standard specification.
According to the method for dynamically adjusting the request quantity of the promotion content based on the filling rate, the request quantity of the DSP platform is adjusted based on the real-time interface service condition and the real-time filling rate of the DSP platform, and the method comprises the following steps: judging whether the real-time interface service condition of the DSP platform is recovered to be normal or not for the DSP platform with abnormal history interface service condition, if so, canceling the fusing treatment of the DSP platform, resetting the current attenuation coefficient, and determining the next observation period as a tolerance period; otherwise, continuing fusing the DSP platform; and judging whether the real-time interface service condition of the DSP platform is normal or not for the DSP platform with normal history interface service condition, if so, judging whether the real-time filling rate of the DSP platform is lower than the preset filling rate, adjusting the request amount of the DSP platform based on the attenuation coefficient of the DSP platform, and if not, fusing the DSP platform.
According to the method for dynamically adjusting the request quantity of promotion content based on the filling rate, the method for acquiring the real-time response data of any adjusted DSP platform, acquiring the real-time interface service condition of the DSP platform, updating the filling rate prediction model based on the real-time response data of the DSP platform, and calculating the real-time filling rate of the DSP platform comprises the following steps: distributing real-time promotion content requests to different links of each DSP platform according to the request amount adjusted by each DSP platform, and acquiring real-time response data of the promotion content requests under different links in the current observation period; determining the real-time interface service condition of any DSP platform based on the real-time response data of the current observation period of the DSP platform; based on real-time response data of a current observation period of any DSP platform, extracting associated features of the filling rate, updating a filling rate prediction model based on the associated features of the filling rate, and calculating the real-time filling rate of the DSP platform.
According to the method for dynamically adjusting the request quantity of the promotion content based on the filling rate, the associated characteristics of the filling rate comprise the size of the promotion content, the material quality of the promotion content and the delivery time of the promotion content; the constructing the filling rate prediction model based on the associated features of the filling rate comprises the following steps: calculating the weight of the associated feature of the filling rate by adopting a data analysis method; and constructing a filling rate prediction model based on the associated features of the filling rate, the weights of the associated features of the filling rate and the request quantity.
According to the system for dynamically adjusting the request quantity of promotion content based on the filling rate, which is provided by the present disclosure, the system is applied to an ADX platform, and the system comprises:
the setting module is used for setting the observation period and attenuation coefficient of each DSP platform;
the analysis module is used for collecting historical response data of at least one observation period of each DSP platform, extracting the associated features of the filling rate, constructing a filling rate prediction model based on the associated features of the filling rate, and calculating the historical filling rate of each DSP platform in the at least one observation period;
the first adjusting module is used for acquiring the historical interface service condition of at least one observation period of each DSP platform and adjusting the request quantity of each DSP platform by combining the historical filling rate of each DSP platform in at least one observation period and the attenuation coefficient of each DSP platform;
The backtracking module is used for acquiring the real-time response data of any adjusted DSP platform, acquiring the real-time interface service condition of the DSP platform, updating a filling rate prediction model based on the real-time response data of the DSP platform, and calculating the real-time filling rate of the DSP platform;
and the second adjusting module is used for adjusting the request quantity of the DSP platform based on the real-time interface service condition and the real-time filling rate of the DSP platform.
In summary, according to the method and system for dynamically adjusting the request amount of promotion content based on the filling rate, the request amount of each DSP platform is adjusted by setting the observation period and the attenuation coefficient of the DSP platform and combining the historical filling rate and the real-time filling rate of each DSP platform to reflect the change of the filling rate, so that the invalid request cost of each DSP platform is reduced, the request utilization rate of each DSP platform is improved, and the filling rate of each DSP platform is improved; the method has the advantages that response data are collected for promotion content requests of different links in four dimensions of a DSP platform, promotion positions, an operating system and promotion forms, so that more fine granularity observation and analysis of flow data are realized, the filling condition of the promotion content requests distributed by the ADX platform by the different links in each DSP platform can be accurately mastered, and the dynamic adjustment of the request quantity of each DSP platform is better realized; through deep combined analysis of historical response data and real-time response data, the filling rate change of popularization content requests of each DSP platform is accurately analyzed, and therefore real-time adjustment and accurate adjustment of the request quantity of each DSP platform are achieved.
Drawings
In order to more clearly illustrate the present disclosure or the prior art solutions, a brief description will be given below of the drawings that are needed in the embodiments or prior art descriptions, it being apparent that the drawings in the following description are some embodiments of the present disclosure and that other drawings may be obtained from these drawings without inventive effort to a person of ordinary skill in the art.
FIG. 1 is a schematic flow chart of a method for dynamically adjusting a request amount of promotion content based on a filling rate provided by the present disclosure;
FIG. 2 is a flow chart of a method for adjusting a request amount of the DSP platform based on an attenuation coefficient of the DSP platform provided by the present disclosure;
fig. 3 is a schematic diagram of a system structure for dynamically adjusting a request amount of promotion content based on a filling rate provided in the present disclosure.
Icon: 310-setting up a module; 320-an analysis module; 330-a first adjustment module; 340-backtracking module; 350-a second adjustment module.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the present disclosure more apparent, the technical solutions in the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the present disclosure, and it is apparent that the described embodiments are some, but not all, embodiments of the present disclosure. All other embodiments, which can be made by one of ordinary skill in the art without inventive effort, based on the embodiments in this disclosure are intended to be within the scope of this disclosure.
Fig. 1 is a schematic flow chart of a method for dynamically adjusting a request amount of promotion content based on a filling rate, which is provided by the present disclosure, and the method is applied to an ADX platform. Referring to fig. 1, the method comprises:
s11, setting the observation period and attenuation coefficient of each DSP platform.
Specifically, it can be understood that the ADX platform needs a plurality of DSP platforms to distribute the promotion content requests, and the promotion content and the quotation of each DSP platform are different, and the received request amount is also not too small, so in order to better adjust the request amount of each DSP platform, it is necessary to configure the observation period and the attenuation coefficient for each DSP platform separately. Meanwhile, since the filling performance of each DSP platform is also changed all the time, in order to adjust the request quantity according to the filling performance more accurately, different observation periods and attenuation coefficients are required to be set for each DSP platform, so that the request quantity distributed to each DSP platform by the ADX platform can be dynamically adjusted.
Wherein the observation period comprises a tolerance period and an adjustment period, and the attenuation coefficient comprises a first attenuation coefficient and a second attenuation coefficient; the tolerance period refers to an observation period with strong tolerance to filling performance, the adjustment period refers to an observation period with strong tolerance to filling performance and needs to adjust the request quantity according to filling performance, the tolerance period is longer than the adjustment period, the first attenuation coefficient is a request quantity attenuation coefficient corresponding to the tolerance period, the second attenuation coefficient is a request quantity attenuation coefficient corresponding to the adjustment period, and the first attenuation coefficient is larger than the second attenuation coefficient, so that the request quantity can be adjusted greatly in the tolerance period, and the request quantity can be adjusted frequently in a small amplitude in the adjustment period.
In some embodiments, the tolerance period may be set to 30min, the observation period may be set to 10min, the first attenuation coefficient may be set to 50%, and the second attenuation coefficient may be set to 10%, so that the requested amount may be attenuated by 50% when the observation period is the tolerance period, and 10% when the observation period is the adjustment period, every 10 min.
It should be noted that, the first attenuation coefficient and the second attenuation coefficient set for each DSP platform may be kept constant, or may be adjusted according to an actual request amount or an actual requirement, which is not limited in the disclosure. In addition, since there are multiple different links under each DSP platform, in some embodiments, the observation period and the attenuation coefficient may be set uniformly for the different links under each DSP platform, or may be set individually for the different link analysis under each DSP platform, which is not limited in this disclosure.
S12, collecting historical response data of at least one observation period of each DSP platform, extracting relevant characteristics of the filling rate, constructing a filling rate prediction model based on the relevant characteristics of the filling rate, and calculating the historical filling rate of each DSP platform in the at least one observation period.
Specifically, it can be understood that, after each DSP platform receives the promotion content request distributed by the ADX platform, the promotion content corresponding to the promotion content request is returned, that is, each DSP platform responds to the promotion content request, and specific response results include no return, invalid return and valid return, where no return refers to that a promotion party of the DSP platform does not have promotion content or budget release promotion content currently; the invalid return refers to that the DSP platform returns corresponding promotion content according to the promotion content request, and the corresponding promotion content does not meet the standard specification and needs to be filtered out, so that links of the promotion content request cannot be filled; the effective return means that the promotion content request of the DSP platform is filled with promotion content. Thus, to obtain a filled representation of each DSP platform for promotional content requests, it is necessary to collect response data to the requests for at least one observation period thereof.
Specifically, it is also understood that step S12 includes the steps of:
step S121, collecting historical response data of at least one observation period of each DSP platform.
Specifically, statistics is performed on historical data of filling performance of promotion content requests by each DSP platform, and analysis is performed from different links in four dimensions of the DSP platform, promotion bit, operating system and promotion form to obtain historical response data of a plurality of observation periods under the different links. For example, there are 4 DSP platforms, 5 promotion bits, 2 operating systems, and 3 promotion forms, at this time, under four dimensions of the DSP platforms, promotion bits, operating systems, and promotion forms, there are 4×5×2×3, and a total of 120 different links, and each promotion content request corresponds to one link, and one link may correspond to a plurality of promotion content requests, at this time, the filling performance of each DSP platform for the promotion content request is counted, that is, the filling performance of the promotion content request under different links of each DSP platform is counted.
In some embodiments, the collecting historical response data for at least one observation period for each DSP platform includes: analyzing the flow of the promotion content requests of different links in four dimensions of the DSP platform, promotion position, operating system and promotion form, and collecting response data of the promotion content requests of different links in at least one observation period of each DSP platform; wherein the response data includes: filling condition, response time and promotion information, wherein the promotion information comprises the type of promotion content, the size of promotion content, the material quality of promotion content, the delivery time of promotion content and the price of promotion content.
In the above embodiment, the DSP platform refers to a platform of multiple resource ends connected to the ADX platform, the promotion site is a location where the promotion end is used to deliver promotion content, the operating system is a system of a device of the promotion end, such as a common Android system, an iOS system, and the like, and the promotion form is a propagation form of promotion content, including a television medium, a paper medium, and the like. By collecting response data of promotion content requests of different links in four dimensions of a DSP platform, promotion bits, an operating system and promotion forms, the observation and analysis of flow data with finer granularity are realized, and compared with the flow analysis of the traditional level, the method and the device can more accurately grasp the filling condition of the promotion content requests distributed by different links to an ADX platform under each DSP platform, thereby better realizing the dynamic adjustment of the request quantity of each DSP platform.
And step S122, extracting the associated features of the filling rate, and constructing a filling rate prediction model based on the associated features of the filling rate.
The relevant characteristics of the filling rate comprise the size of the promotion content, the material quality of the promotion content and the delivery time of the promotion content;
specifically, for each feature in the response data, the filling condition is used for indicating whether the DSP platform returns the promotion content for the promotion content request, the response time is used for indicating the response time of the DSP platform for the promotion content request, the promotion information is used for indicating the type, size, material quality, delivery time, price and the like of the promotion content returned by the DSP platform for the promotion content request, the type of the promotion content is military, historic, entertainment and the like, the size of the promotion content is the layout size of the promotion content delivered at the promotion terminal equipment, the material quality is the image quality and the like of the promotion content, the delivery time is the duration of the promotion content needing to be delivered, and the price is the delivery price of the DSP platform for the promotion content. Therefore, in order to analyze the filling performance of each DSP platform, it is necessary to extract the collected historical response data, and obtain the relevant features affecting the filling rate, that is, the features related to the return amount corresponding to the request amount of the promotion content, including the size of the promotion content, the material quality of the promotion content, and the delivery time of the promotion content.
More specifically, the constructing the filling rate prediction model based on the associated features of the filling rate includes: calculating the weight of the associated feature of the filling rate by adopting a data analysis method; and constructing a filling rate prediction model based on the associated features of the filling rate, the weights of the associated features of the filling rate and the request quantity. That is, after extracting the associated features of the filling rate, the weights of the associated features of the filling rate are determined under different links respectively by adopting a statistical analysis method, so that whether the popularization content is returned or not is determined for each popularization content request, and a filling rate prediction model is constructed. The data analysis method may be a correlation analysis method, a regression analysis method, or other statistical analysis methods, which is not limited in this disclosure.
For example, the associated feature of the filling rate is denoted as A 1 、A 2 、A 3 The weight of each associated feature is determined to be w respectively through a data analysis method 1 、w 2 、w 3 Then further construct a filling rate prediction model under any link, e.g. R 1 /R 2 Wherein R is 1 Indicating the return amount of promotional content request under the link is based on A 1 、A 2 、A 3 Calculated, R 2 Representing the request amount under the link, R 1 The calculation formula of (2) needs to be determined according to the promotion content request under the link, and can be a multi-element linear equation or a multi-element nonlinear equation, which is not limited in the disclosure.
It should be noted that, the filling rate prediction model constructed in step S122 is constructed for each link under each DSP platform, and the filling rate prediction models between different links do not affect each other. The method is characterized in that a filling rate prediction model is constructed for the historical response data of each link under each DSP platform, the historical response data of the DSP platform is subjected to deep analysis, and the flow request cost and filling performance of each DSP platform can be accurately analyzed in finer dimensionality, so that the request quantity of each DSP platform can be better adjusted.
Step S123, calculating the historical filling rate of each DSP platform in at least one observation period.
Specifically, according to step S122, not only the filling rate prediction model under different links of each DSP platform but also the filling rate under different links of each DSP platform may be obtained. On this basis, considering that a plurality of links exist under each DSP platform, calculating the historical filling rate of each DSP platform in at least one observation period includes: and calculating the historical filling rate of different links of each DSP platform in at least one observation period.
S13, acquiring the historical interface service condition of at least one observation period of each DSP platform, and adjusting the request quantity of each DSP platform by combining the historical filling rate of each DSP platform in at least one observation period and the attenuation coefficient of each DSP platform.
Specifically, it can be understood that based on the historical response data of at least one observation period of each DSP platform, not only the historical filling rate of each DSP platform in at least one observation period, but also the historical interface service condition of at least one observation period of each DSP platform can be determined, and the interface of each DSP platform is an interface for connecting with the ADX platform, receiving the promotion content request and returning the promotion content, where the interface service condition can accurately reflect whether the promotion content requests of different links have service anomalies, such as anomalies caused by technical or network reasons, response time-out, and non-compliance of promotion information with standard specifications.
Specifically, it can be understood that the obtaining the historical interface service condition of at least one observation period of each DSP platform and adjusting the request amount of each DSP platform in combination with the historical filling rate of each DSP platform in at least one observation period and the attenuation coefficient of each DSP platform includes:
Step S131, based on the historical response data of at least one observation period of any DSP platform, the historical interface service condition of at least one observation period of the DSP platform is obtained.
Specifically, according to the filling condition, response time, popularization information and the like of the historical response data of at least one observation period of any DSP platform, the historical interface service condition of the DSP platform is determined, namely, the historical interface service conditions of different links of the DSP platform are determined.
And step S132, judging whether the history interface service condition of the DSP platform is normal.
In the step S132, it is substantially determined whether the interface of the DSP platform has abnormal service, if the service condition of the history interface is normal, step S133 is executed, otherwise step S134 is executed. Judging whether the history interface service condition of the DSP platform is normal or not, specifically comprising:
judging whether a link of any promotion content request is available or not; and/or the number of the groups of groups,
judging whether the response time of any promotion content request accords with the preset response time; and/or the number of the groups of groups,
and judging whether the promotion information of any promotion content request accords with the standard specification.
Specifically, the historical interface service condition of the DSP platform needs to meet three conditions of link availability, response time meeting preset response time and popularization information meeting standard specification at the same time, namely, the condition that link availability, response time meeting preset response time and popularization information meeting standard are met, and when one condition is not met, the historical interface service condition of the DSP platform is considered to be abnormal, and when one condition is not met, for example, the interface service abnormality is directly determined when the link is not available, response time and popularization time are not needed to be judged any more, so that the acceleration of judging the interface service condition of the DSP platform is facilitated.
More specifically, judging whether the link of any promotion content request is available or not refers to judging whether the link of any promotion content request has abnormal error reporting caused by technical or network reasons or not; judging whether the response time of any promotion content request accords with the preset response time, namely whether the response time is overtime, wherein the response time of each DSP platform to the promotion content request is controlled within 120ms, and the overtime response is marked as no response or no return; judging whether the promotion information of any promotion content request accords with the standard specification refers to judging whether the size, price and the like of the promotion information accord with the standard specification, for example, judging whether the data format of the promotion content accords with the format specification, judging whether the size of the promotion content meets the delivery size, judging whether the price of the promotion content is higher than a preset lower limit of the delivery price and the like.
And step S133, if the history interface service condition of the DSP platform is normal, judging whether the history filling rate of the DSP platform in at least one observation period is lower than a preset filling rate, and if so, adjusting the request quantity of the DSP platform based on the attenuation coefficient of the DSP platform.
Specifically, it can be understood that, based on the history filling performance of each DSP platform in the observation period, the request amount distributed to each DSP platform is adjusted, and when the history filling performance is lower than the preset filling rate, the request amount distributed to each DSP platform can be adjusted down, so that the invalid request is reduced, and the filling rate is improved. When analyzing the filling table of each DSP platform in the observation period, the method can judge whether the interface service condition is normal or not, and judge whether the filling rate is lower than the preset filling rate or not under the condition that the interface service is normal.
Specifically, it may be further understood that when the historical filling rate of the DSP platform in at least one observation period is lower than the preset filling rate, the request amount of the DSP platform needs to be adjusted according to the first attenuation coefficient and the second attenuation coefficient of the DSP platform, so that invalid requests are reduced, and the filling rate of the DSP platform is improved.
And step S134, if the history interface service condition of the DSP platform is abnormal, fusing the DSP platform.
The fusing process is to stop the transaction of the promotion content between the DSP platform and the ADX platform, and when the service condition of the historical interface of any DSP platform in a certain observation period is abnormal, the distribution of the promotion content request of the DSP platform can be stopped in the next period.
S14, acquiring the real-time response data of any adjusted DSP platform, acquiring the real-time interface service condition of the DSP platform, updating a filling rate prediction model based on the real-time response data of the DSP platform, and calculating the real-time filling rate of the DSP platform.
Specifically, it can be understood that step S13 is to determine the first adjustment policy of the request amount based on the historical filling rate of each DSP platform, and adjust the request amount distributed to each DSP platform at the current time according to the first adjustment policy, but when the request amount is adjusted according to the first adjustment policy at the current time, new popularization content requests and corresponding response data will also be generated under different links of each DSP platform, so that the request amount can be reduced only based on the historical filling performance adjustment request amount when the DSP platform budget is insufficient, but the request amount cannot be recovered in time when the DSP platform increases the budget, so that in order to realize real-time dynamic adjustment of the request amount of each DSP platform, the result of the first adjustment policy needs to be monitored and analyzed, the response data corresponding to the new popularization content requests is collected, and the real-time filling rate in the current observation period is calculated.
Specifically, it may be further understood that the acquiring the adjusted real-time response data of any DSP platform, obtaining a real-time interface service condition of the DSP platform, updating a filling rate prediction model based on the real-time response data of the DSP platform, and calculating the real-time filling rate of the DSP platform includes:
and step S141, distributing real-time promotion content requests to different links of each DSP platform according to the request amount adjusted by each DSP platform, and acquiring real-time response data of the promotion content requests under different links in the current observation period.
Specifically, the request amount and the current observation period which are distributed to the DSP are determined and adjusted according to the steps S12 and S13, and promotion content requests under different links of each DSP platform after the request amount is adjusted are monitored and analyzed, so that real-time response data of the current observation period are obtained.
Step S142, determining the real-time interface service condition of any DSP platform based on the real-time response data of the current observation period of the DSP platform.
Specifically, in the current observation period, the real-time interface service condition of any DSP platform is determined based on the filling condition, response time, popularization information and the like of the real-time response data of the popularization content request under different links of the DSP platform.
Step S143, extracting relevant characteristics of the filling rate based on real-time response data of a current observation period of any DSP platform, updating a filling rate prediction model based on the relevant characteristics of the filling rate, and calculating the real-time filling rate of the DSP platform.
Specifically, in the current observation period, based on real-time response data of a promotion content request under any link of any DSP platform, extracting relevant features of filling rates such as the size of promotion content, the material quality of promotion content, the delivery time of promotion content and the like, updating a filling rate prediction model constructed based on historical response data based on the relevant features of the filling rates, and calculating the real-time filling rate so as to facilitate an adjustment scheme of the request quantity of each DSP platform for the next observation period.
S15, the request quantity of the DSP platform is adjusted based on the real-time interface service condition and the real-time filling rate of the DSP platform.
Specifically, it can be understood that step S13 is to determine a first adjustment policy of the request amount based on the historical filling rate of each DSP platform, and steps S14 and S15 are to determine a second adjustment policy of the request amount based on the real-time filling rate of each DSP platform, so as to implement real-time dynamic adjustment of the request amount of each DSP platform.
Specifically, it may be further understood that the adjusting the request amount of the DSP platform based on the real-time interface service condition and the real-time filling rate of the DSP platform includes:
judging whether the real-time interface service condition of the DSP platform is recovered to be normal or not for the DSP platform with abnormal history interface service condition, if so, canceling the fusing treatment of the DSP platform, resetting the current attenuation coefficient, and determining the next observation period as a tolerance period; otherwise, continuing fusing the DSP platform;
and judging whether the real-time interface service condition of the DSP platform is normal or not for the DSP platform with normal history interface service condition, if so, judging whether the real-time filling rate of the DSP platform is lower than the preset filling rate, adjusting the request amount of the DSP platform based on the attenuation coefficient of the DSP platform, and if not, fusing the DSP platform.
The method for judging whether the real-time interface service condition of the DSP platform is normal may refer to the method for judging whether the history interface service condition of step S132 is normal, and the request amount for adjusting the DSP platform based on the attenuation coefficient of the DSP platform is the same as the request amount adjusting method of step S133, which is not described herein.
In some embodiments, for a DSP platform with abnormal interface service conditions, the source of the abnormal information may be analyzed according to the interface service conditions, different abnormal information may correspond to different error codes, if the source of the abnormal information is the response judgment of the ADX platform to the DSP, at this time, a modification request may be sent to the DSP platform according to report statistics, so as to solve the abnormal information caused by the response judgment of the ADX platform to the DSP, thereby enabling the interface service of the DSP platform to be restored.
According to the method for dynamically adjusting the request quantity of the promotion content based on the filling rate, the request quantity of each DSP platform is adjusted by setting the observation period and the attenuation coefficient of the DSP platform and combining the historical filling rate and the real-time filling rate of each DSP platform to reflect the filling rate change, so that the invalid request cost of each DSP platform is reduced, the request utilization rate of each DSP platform is improved, and the filling rate of each DSP platform is improved; the method has the advantages that response data are collected for promotion content requests of different links in four dimensions of a DSP platform, promotion positions, an operating system and promotion forms, so that more fine granularity observation and analysis of flow data are realized, the filling condition of the promotion content requests distributed by the ADX platform by the different links in each DSP platform can be accurately mastered, and the dynamic adjustment of the request quantity of each DSP platform is better realized; through deep combined analysis of historical response data and real-time response data, the filling rate change of popularization content requests of each DSP platform is accurately analyzed, and therefore real-time adjustment and accurate adjustment of the request quantity of each DSP platform are achieved.
FIG. 2 is a flow chart of a method for adjusting a request amount of the DSP platform based on an attenuation coefficient of the DSP platform provided by the present disclosure; referring to fig. 2, the adjusting the request amount of the DSP platform based on the attenuation coefficient of the DSP platform includes:
s21, determining a current attenuation coefficient based on the attenuation coefficient of the DSP platform;
wherein the determining the current attenuation coefficient based on the attenuation coefficient of the DSP platform includes:
step S211, if the observation period is a tolerant period and the historical filling rate of the DSP platform is lower than a preset filling rate, determining the current attenuation coefficient as a first attenuation coefficient and determining the next observation period as an adjustment period;
specifically, it may be understood that the current attenuation coefficient refers to an attenuation proportion of the DSP platform request amount determined based on that the historical filling rate is the current observation period, the determining that the next observation period is the adjustment period refers to the next period of the observation period corresponding to the historical response data, if the historical response data of two observation periods are collected, the first observation period is a tolerance period, and the historical filling rate of the first observation period is lower than the preset filling rate, the determining that the next observation period is the adjustment period, that is, the second observation period is the adjustment period.
Specifically, it can be further understood that when the historical filling rate of the DSP platform is not lower than the preset filling rate in the tolerant period, the observation is continued, the request amount is kept unchanged, and the next observation period is still the tolerant period.
Step S211, if the observation period is an adjustment period and the historical filling rate of the DSP platform is lower than the preset filling rate, judging whether the current attenuation coefficient of the DSP platform is lower than the preset attenuation ratio, if so, determining that the current attenuation coefficient is the sum of the current attenuation coefficient and the second attenuation coefficient, otherwise, keeping unchanged.
Specifically, the preset attenuation ratio is used for ensuring that the non-attenuation of the request quantity of each DSP is 0, that is, if the current attenuation coefficient is lower than the preset attenuation ratio, the current attenuation coefficient needs to be updated to continuously adjust the request quantity of the DSP platform, and if the current attenuation coefficient is not lower than the preset attenuation ratio, the request quantity of the DSP platform is indicated to be adjusted to be the lowest, that is, the current coefficient is not required to be continuously adjusted. The preset attenuation ratio may be adjusted according to the second attenuation coefficient, for example, the second attenuation coefficient is 10%, the preset attenuation ratio may be set to 90%, and the preset attenuation ratio may also be set according to the initial request amount of each DSP platform, which is not limited in the disclosure.
Step S211, if the observation period is an adjustment period and the historical filling rate of the DSP platform is not lower than the preset filling rate, resetting the current attenuation coefficient and determining the next observation period as a tolerance period.
Specifically, resetting the current attenuation coefficient refers to resetting the current attenuation coefficient to zero, that is, if the historical filling rate of the DSP platform is not lower than the preset filling rate in the adjustment period, the observation period needs to be restored from the adjustment period to the tolerance period, and the request amount of the DSP platform is restored to the initial request amount by resetting the current attenuation coefficient.
In some embodiments, when the historical filling rate of the DSP platform is not lower than the preset filling rate in the adjustment period, the current attenuation coefficient can be reset, the DSP platform can directly restore to the initial request amount, or the growth coefficient can be selectively set, the request amount is increased by gradually increasing the growth coefficient, so that the request amount is slowly restored to the request amount or the initial request amount meeting the preset filling rate, instead of directly restoring to the initial request amount in one step, and more invalid requests still exist after the DSP platform is directly restored to the initial request amount.
S22, calculating the request attenuation of the DSP platform, wherein the request attenuation is the product of the initial request amount and the current attenuation coefficient.
Specifically, it may be understood that the initial request amount is a request amount before the request amount adjustment is not performed by each DSP platform, and after determining the current attenuation coefficient according to step S21, a product of the initial request amount and the current attenuation coefficient may be calculated and used as the request attenuation amount of the DSP platform.
S23, adjusting the request quantity of the DSP platform to be the difference between the request attenuation quantity and the initial request quantity.
Specifically, it may be understood that the request amount adjustment of the DSP platform adjusts the distribution of the request amount in the observation period to which the current time belongs according to the history filling rate of each observation period of the history data, and the adjustment of the request amount is the initial request amount of each DSP platform minus the request attenuation amount calculated in step S22.
For example, by setting the observation period and the attenuation coefficient of the DSP platform 1 in step S11, the observation period includes a tolerance period of 20min and an adjustment period of 10min, the attenuation coefficient includes a first attenuation coefficient of 30% and a second attenuation coefficient of 10%, and by collecting historical response data obtained in 60min under different links of the DSP platform 1 in step S12, the adjustment of the request amount of each DSP platform according to steps S21-S23 mainly includes the following typical request amount adjustment:
(1) Within 60min, the historical filling rate within 20min of the first observation period 'tolerance period' accords with the preset filling rate, and then calculation and comparison of the historical filling rate within 20min of the next tolerance period are continued:
if the history filling rate of the tolerant period is not lower than the preset filling rate all the time, the observation period is kept as the tolerant period, and the request amount is kept unchanged;
if the historical filling rate of any tolerance period is lower than the preset filling rate, determining that the current attenuation coefficient is the first attenuation coefficient, adjusting the request quantity of the DSP platform to be the product of the initial request quantity and the current attenuation coefficient, and adjusting the observation period from the tolerance period to the adjustment period so as to continue calculation and comparison of the historical filling rate in the next adjustment period of 10 min.
(2) Within 60min, the historical filling rate in 20min of the first observation period 'tolerance period' is consistent with being lower than the preset filling rate, the request quantity of the DSP platform is adjusted to be the product of the initial request quantity and the current attenuation coefficient, and the observation period is adjusted from the tolerance period to the adjustment period, so that calculation and comparison of the historical filling rate in 10min of the next adjustment period are continued:
if the historical filling rate of the adjustment period is not lower than the preset filling rate, the observation period is readjusted to be a tolerance period, and the current attenuation coefficient is reset to restore the request quantity to the initial request quantity;
If the historical filling rate in the adjustment period is lower than the preset filling rate and the current attenuation coefficient is lower than the preset attenuation ratio, determining that the current attenuation coefficient is the sum of the current attenuation coefficient and the second attenuation coefficient, adjusting the request amount of the DSP platform to be the product of the initial request amount and the current attenuation coefficient, and continuing to calculate and compare the historical filling rate in the next adjustment period of 10 min;
if the historical filling rate of the adjustment period is lower than the preset filling rate and the current attenuation coefficient is not lower than the preset attenuation ratio, otherwise, keeping the current attenuation coefficient unchanged, and continuing to calculate and compare the historical filling rate in the next adjustment period of 10 min.
According to the method for adjusting the request quantity of the DSP platform based on the attenuation coefficient of the DSP platform, the current attenuation coefficient is determined through the attenuation coefficient of the DSP platform, and the request quantity of the DSP platform is determined, so that the change of the filling rate can be displayed in real time based on the historical filling rate and the real-time filling rate of each DSP platform, invalid requests are reduced, the filling rate is improved, and the request quantity of each DSP platform is dynamically adjusted according to the change of the filling rate; by setting different observation periods and attenuation coefficients, the request quantity of each DSP platform can be dynamically adjusted according to the change of the filling rate, various adjustment methods for the request quantity are realized, the request quantity can be greatly adjusted, the request quantity can be also slightly adjusted for a plurality of times in a short time, the request quantity of the DSP platform can be timely recovered when the popularization content of the DSP platform is recovered, and the cost, the bandwidth cost and the like of invalid requests are reduced while the accurate release of the popularization content is ensured.
Fig. 3 is a schematic diagram of a system structure for dynamically adjusting a request amount of promotion content based on a filling rate, where the system is applied to an ADX platform. Referring to fig. 3, the system includes: a setting module 310, an analyzing module 320, a first adjusting module 330, a backtracking module 340, and a second adjusting module 350.
A setting module 310, configured to set an observation period and an attenuation coefficient of each DSP platform;
the analysis module 320 is configured to collect historical response data of at least one observation period of each DSP platform, extract relevant features of the filling rate, construct a filling rate prediction model based on the relevant features of the filling rate, and calculate a historical filling rate of each DSP platform in the at least one observation period;
the first adjusting module 330 is configured to obtain a history interface service condition of at least one observation period of each DSP platform, and adjust a request amount of each DSP platform in combination with a history filling rate of each DSP platform in the at least one observation period and an attenuation coefficient of each DSP platform;
the backtracking module 340 is configured to collect real-time response data of any adjusted DSP platform, obtain a real-time interface service condition of the DSP platform, update a filling rate prediction model based on the real-time response data of the DSP platform, and calculate a real-time filling rate of the DSP platform;
And a second adjusting module 350, configured to adjust the request amount of the DSP platform based on the real-time interface service condition and the real-time filling rate of the DSP platform.
According to the system for dynamically adjusting the request quantity of the promotion content based on the filling rate, the observation period and the attenuation coefficient of the DSP platform are set through the setting module, and the request quantity of each DSP platform is adjusted by combining the historical filling rate and the real-time filling rate of each DSP platform to reflect the filling rate change, so that the invalid request cost of each DSP platform is reduced, the request utilization rate of each DSP platform is improved, and the filling rate of each DSP platform is improved; the analysis module is used for collecting response data of promotion content requests of different links in four dimensions of a DSP platform, promotion positions, an operating system and promotion forms, so that the observation and analysis of flow data with finer granularity are realized, the filling condition of the promotion content requests distributed by the ADX platform by the different links in each DSP platform can be accurately mastered, and the dynamic adjustment of the request quantity of each DSP platform is better realized; the first adjusting module and the second adjusting module are used for carrying out deep combined analysis on the historical response data and the real-time response data, and the filling rate change of the popularization content request of each DSP platform is accurately analyzed, so that real-time adjustment and accurate adjustment of the request quantity of each DSP platform are realized.
The above described embodiments of the control device are merely illustrative, wherein the "units" used as separate components description may be a combination of software and/or hardware implementing the predetermined functions, which may or may not be physically separate. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
Finally, it should be noted that: the above embodiments are merely for illustrating the technical solution of the present disclosure, and are not limiting thereof; although the present disclosure has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the disclosure, and are intended to be included in the scope of the present invention.

Claims (10)

1. A method for dynamically adjusting a request amount of promotion content based on a filling rate, wherein the method is applied to an ADX platform, the method comprising:
Setting an observation period and an attenuation coefficient of each DSP platform;
collecting historical response data of at least one observation period of each DSP platform, extracting relevant characteristics of the filling rate, constructing a filling rate prediction model based on the relevant characteristics of the filling rate, and calculating the historical filling rate of each DSP platform in the at least one observation period;
acquiring the historical interface service condition of at least one observation period of each DSP platform, and adjusting the request quantity of each DSP platform by combining the historical filling rate of each DSP platform in at least one observation period and the attenuation coefficient of each DSP platform;
acquiring real-time response data of any adjusted DSP platform, acquiring a real-time interface service condition of the DSP platform, updating a filling rate prediction model based on the real-time response data of the DSP platform, and calculating the real-time filling rate of the DSP platform;
and adjusting the request quantity of the DSP platform based on the real-time interface service condition and the real-time filling rate of the DSP platform.
2. The method for dynamically adjusting a request amount for promotional content based on a fill factor according to claim 1, wherein said collecting historical response data for at least one observation period for each DSP platform comprises:
analyzing the flow of the promotion content requests of different links in four dimensions of the DSP platform, promotion position, operating system and promotion form, and collecting response data of the promotion content requests of different links in at least one observation period of each DSP platform;
Wherein the response data includes: filling condition, response time and promotion information, wherein the promotion information comprises the type of promotion content, the size of promotion content, the material quality of promotion content, the delivery time of promotion content and the price of promotion content.
3. The method for dynamically adjusting a request amount for promotional content based on a fill factor according to claim 1, wherein said observation period comprises a tolerance period and an adjustment period, and wherein said attenuation coefficients comprise a first attenuation coefficient and a second attenuation coefficient; the method for acquiring the historical interface service condition of at least one observation period of each DSP platform and adjusting the request quantity of each DSP platform by combining the historical filling rate of each DSP platform in at least one observation period and the attenuation coefficient of each DSP platform comprises the following steps:
based on historical response data of at least one observation period of any DSP platform, acquiring historical interface service conditions of at least one observation period of the DSP platform;
judging whether the history interface service condition of the DSP platform is normal or not;
if the history interface service condition of the DSP platform is normal, judging whether the history filling rate of the DSP platform in at least one observation period is lower than a preset filling rate, and if so, adjusting the request quantity of the DSP platform based on the attenuation coefficient of the DSP platform;
And if the historical interface service condition of the DSP platform is abnormal, fusing the DSP platform.
4. The method for dynamically adjusting a request amount for promotional content based on a fill factor of claim 3 wherein said adjusting a request amount for said DSP platform based on an attenuation coefficient of said DSP platform comprises:
determining a current attenuation coefficient based on the attenuation coefficient of the DSP platform;
calculating the request attenuation amount of the DSP platform, wherein the request attenuation amount is the product of the initial request amount and the current attenuation coefficient;
and adjusting the request quantity of the DSP platform to be the difference between the request attenuation quantity and the initial request quantity.
5. The method for dynamically adjusting a requested amount of promotional content based upon a fill factor of claim 4 wherein said determining a current attenuation coefficient based upon an attenuation coefficient of said DSP platform comprises:
if the observation period is a tolerance period and the historical filling rate of the DSP platform is lower than a preset filling rate, determining that the current attenuation coefficient is a first attenuation coefficient and determining that the next observation period is an adjustment period;
if the observation period is an adjustment period and the historical filling rate of the DSP platform is lower than a preset filling rate, judging whether the current attenuation coefficient of the DSP platform is lower than a preset attenuation ratio, if so, determining that the current attenuation coefficient is the sum of the current attenuation coefficient and the second attenuation coefficient, otherwise, keeping unchanged;
If the observation period is an adjustment period and the historical filling rate of the DSP platform is not lower than the preset filling rate, resetting the current attenuation coefficient and determining the next observation period as a tolerance period.
6. The method for dynamically adjusting request amount of promotion content based on filling rate according to claim 3, wherein determining whether the historical interface service condition of the DSP platform is normal comprises:
judging whether a link of any promotion content request is available or not; and/or the number of the groups of groups,
judging whether the response time of any promotion content request accords with the preset response time; and/or the number of the groups of groups,
and judging whether the promotion information of any promotion content request accords with the standard specification.
7. The method for dynamically adjusting request amount of promotion content based on filling rate according to claim 3, wherein the adjusting request amount of the DSP platform based on real-time interface service condition and real-time filling rate of the DSP platform comprises:
judging whether the real-time interface service condition of the DSP platform is recovered to be normal or not for the DSP platform with abnormal history interface service condition, if so, canceling the fusing treatment of the DSP platform, resetting the current attenuation coefficient, and determining the next observation period as a tolerance period; otherwise, continuing fusing the DSP platform;
And judging whether the real-time interface service condition of the DSP platform is normal or not for the DSP platform with normal history interface service condition, if so, judging whether the real-time filling rate of the DSP platform is lower than the preset filling rate, adjusting the request amount of the DSP platform based on the attenuation coefficient of the DSP platform, and if not, fusing the DSP platform.
8. The method for dynamically adjusting request amount of promotion content based on filling rate according to claim 1, wherein the acquiring real-time response data of any adjusted DSP platform, acquiring real-time interface service conditions of the DSP platform, updating a filling rate prediction model based on the real-time response data of the DSP platform, and calculating the real-time filling rate of the DSP platform comprises:
distributing real-time promotion content requests to different links of each DSP platform according to the request amount adjusted by each DSP platform, and acquiring real-time response data of the promotion content requests under different links in the current observation period;
determining the real-time interface service condition of any DSP platform based on the real-time response data of the current observation period of the DSP platform;
based on real-time response data of a current observation period of any DSP platform, extracting associated features of the filling rate, updating a filling rate prediction model based on the associated features of the filling rate, and calculating the real-time filling rate of the DSP platform.
9. The method for dynamically adjusting the request amount of the promotion content based on the filling rate according to claim 1, wherein the associated features of the filling rate comprise the size of the promotion content, the material quality of the promotion content and the delivery time of the promotion content; the constructing the filling rate prediction model based on the associated features of the filling rate comprises the following steps:
calculating the weight of the associated feature of the filling rate by adopting a data analysis method;
and constructing a filling rate prediction model based on the associated features of the filling rate, the weights of the associated features of the filling rate and the request quantity.
10. A system for dynamically adjusting a request amount for promoted content based on a filling rate, the system being applied to an ADX platform, the system comprising:
the setting module is used for setting the observation period and attenuation coefficient of each DSP platform;
the analysis module is used for collecting historical response data of at least one observation period of each DSP platform, extracting the associated features of the filling rate, constructing a filling rate prediction model based on the associated features of the filling rate, and calculating the historical filling rate of each DSP platform in the at least one observation period;
the first adjusting module is used for acquiring the historical interface service condition of at least one observation period of each DSP platform and adjusting the request quantity of each DSP platform by combining the historical filling rate of each DSP platform in at least one observation period and the attenuation coefficient of each DSP platform;
The backtracking module is used for acquiring the real-time response data of any adjusted DSP platform, acquiring the real-time interface service condition of the DSP platform, updating a filling rate prediction model based on the real-time response data of the DSP platform, and calculating the real-time filling rate of the DSP platform;
and the second adjusting module is used for adjusting the request quantity of the DSP platform based on the real-time interface service condition and the real-time filling rate of the DSP platform.
CN202310920166.4A 2023-07-26 2023-07-26 Method and system for dynamically adjusting request quantity of promotion content based on filling rate Active CN116647601B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310920166.4A CN116647601B (en) 2023-07-26 2023-07-26 Method and system for dynamically adjusting request quantity of promotion content based on filling rate

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310920166.4A CN116647601B (en) 2023-07-26 2023-07-26 Method and system for dynamically adjusting request quantity of promotion content based on filling rate

Publications (2)

Publication Number Publication Date
CN116647601A true CN116647601A (en) 2023-08-25
CN116647601B CN116647601B (en) 2023-09-29

Family

ID=87625083

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310920166.4A Active CN116647601B (en) 2023-07-26 2023-07-26 Method and system for dynamically adjusting request quantity of promotion content based on filling rate

Country Status (1)

Country Link
CN (1) CN116647601B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120041816A1 (en) * 2010-07-19 2012-02-16 Yehuda Ari Buchalter Systems and methods for determining competitive market values of an ad impression
CN110782290A (en) * 2019-10-28 2020-02-11 苏州跃盟信息科技有限公司 Media file processing method and device, storage medium and processor
CN111052167A (en) * 2017-09-14 2020-04-21 艾玛迪斯简易股份公司 Method and system for intelligent adaptive bidding in automated online trading network
CN111401943A (en) * 2020-03-10 2020-07-10 厦门美图之家科技有限公司 Multi-source advertisement bidding system and method
CN115730972A (en) * 2022-11-30 2023-03-03 杭州网易云音乐科技有限公司 Method, device and equipment for dynamically setting guaranteed price
CN115953204A (en) * 2023-01-10 2023-04-11 湖南快乐阳光互动娱乐传媒有限公司 Flow request control method and device

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120041816A1 (en) * 2010-07-19 2012-02-16 Yehuda Ari Buchalter Systems and methods for determining competitive market values of an ad impression
CN111052167A (en) * 2017-09-14 2020-04-21 艾玛迪斯简易股份公司 Method and system for intelligent adaptive bidding in automated online trading network
CN110782290A (en) * 2019-10-28 2020-02-11 苏州跃盟信息科技有限公司 Media file processing method and device, storage medium and processor
CN111401943A (en) * 2020-03-10 2020-07-10 厦门美图之家科技有限公司 Multi-source advertisement bidding system and method
CN115730972A (en) * 2022-11-30 2023-03-03 杭州网易云音乐科技有限公司 Method, device and equipment for dynamically setting guaranteed price
CN115953204A (en) * 2023-01-10 2023-04-11 湖南快乐阳光互动娱乐传媒有限公司 Flow request control method and device

Also Published As

Publication number Publication date
CN116647601B (en) 2023-09-29

Similar Documents

Publication Publication Date Title
CN110971444B (en) Alarm management method, device, server and storage medium
CA2482370C (en) Systems, methods and computer program products for obtaining a best available reimbursement price in a pharmaceutical prescription transaction
US20100030597A1 (en) Internet Content Delivery Management System
CN103136623B (en) The online management method of progress payment application
CN101179360A (en) System and method for managing perceived response time
US20070299746A1 (en) Converged tool for simulation and adaptive operations combining it infrastructure performance, quality of experience, and finance parameters
US11689433B2 (en) Method and scoring node for estimating a user's quality of experience for a delivered service
CN106559265A (en) A kind of cloud service provider ability integration appraisal system and examining method
US20140279406A1 (en) Automated risk monitoring method and system
CN116647601B (en) Method and system for dynamically adjusting request quantity of promotion content based on filling rate
CN111949277A (en) Intelligent gray scale release production method and device, computer equipment and storage medium
US8068584B2 (en) System and method for selecting a profile for a digital subscriber line
CN115983918A (en) Advertisement pushing method, medium, device and computing equipment
CN110442455A (en) A kind of data processing method and device
CN112367540B (en) Method and device for monitoring live broadcast stream pushing number on line
US9197739B1 (en) System, method, and computer program for providing guaranteed quality of service
CN109040197A (en) A kind of intelligent gateway management system and method based on web SiteServer LBS
CN114971412A (en) Order distribution method, order distribution platform, order distribution system and storage medium
CN111429060B (en) Express bill payment management method and equipment
CN114048010A (en) Method, device, equipment and storage medium for controlling service timeout time
CN114143822A (en) Traffic management method, operation management platform, charging pile and storage medium
KR102346138B1 (en) System for trading media platform channel
CN111507819A (en) Dynamic risk quota management method
CN107909460B (en) Position synchronization method, device, data center and storage medium
CN114448838A (en) System reliability evaluation method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant