Data delivery method and apparatus, device and medium
Through the file system-based data delivery method, using feature data and adjustment strategies to optimize data delivery, the problem of poor performance in existing systems when processing large amounts of data is solved, and efficient and accurate data delivery is achieved.
Patent Information
- Application Number
- PCT/CN2024/089784
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-05
- Filing Date
- 2024-04-25
- Publication Date
- 2024-11-14
AI Technical Summary
When existing data delivery systems process large amounts of data to be delivered, their performance is not ideal. They require a large number of remote calls, resulting in large time overhead and may lead to incomplete data, making it difficult to achieve the expected delivery effect.
Using a file system-based data delivery method, by obtaining the correlation between the characteristic data of the data and the delivery target, determining the adjustment action based on the predetermined adjustment strategy, optimizing the data delivery task to match the delivery target, reducing the number of remote calls, and improving data transmission efficiency and delivery performance.
Through the file system, data management is reduced, the number of remote calls is improved, the efficiency of data delivery tasks is improved, time waiting and fault retransmission problems are reduced, and more efficient data delivery is achieved, ensuring that the delivery results meet the expected goals.
Smart Images

Figure CN2024089784_14112024_PF_FP_ABST
Abstract
Description
Data delivery method, device, equipment and medium
[0001] This application claims priority to the Chinese invention patent application entitled “Data delivery method, device, equipment and medium” and application number 2023104989981, filed on May 5, 2023, the entire contents of which are incorporated by reference into this application. Technical Field
[0002] Exemplary implementations of the present disclosure generally relate to data delivery, and more particularly, to methods, apparatuses, devices, and computer-readable storage media for data delivery tasks. Background Art
[0003] With the development of computer technology, a variety of applications have been developed, and various types of data can be delivered to various applications in order to provide users of the applications with multi-faceted information. For example, data about a certain object (for example, a certain application, product, etc.) can be delivered to different types of applications (for example, video applications, short video applications, news applications, etc.). The amount of data to be delivered is huge, usually reaching hundreds of thousands or even millions. At this time, the delivery system needs to obtain and process each piece of data to be delivered one by one, and then set the corresponding delivery method. At this time, how to manage these data to be delivered in a more convenient and effective way to obtain the expected delivery effect has become a research hotspot and difficulty.
[0004] Summary of the Invention
[0005] In a first aspect of the present disclosure, a data delivery method is provided. In this method, a delivery target associated with a data delivery task is received. The delivery target describes a desired delivery result obtained by delivering multiple data items. Feature data for the data delivery task is obtained using a file system. The feature data describes an association between attribute data of the multiple data items and the delivery target. An adjustment action for adjusting the data delivery task is determined based on a predetermined adjustment strategy and the feature data. The adjustment action ensures that the delivery result of the adjusted data delivery task matches the delivery target.
[0006] In a second aspect of the present disclosure, a data delivery device is provided. The device includes: a receiving module configured to receive a delivery target associated with a data delivery task, the delivery target describing a delivery result expected to be obtained by delivering multiple data; an acquisition module configured to utilize a file system to acquire feature data of the data delivery task, the feature data describing an association between attribute data of the multiple data and the delivery target; and a determination module configured to determine an adjustment action for adjusting the data delivery task based on a predetermined adjustment strategy and the feature data, the adjustment action ensuring that the delivery result of the adjusted data delivery task matches the delivery target.
[0007] In a third aspect of the present disclosure, an electronic device is provided. The electronic device includes: at least one processing unit; and at least one memory, the at least one memory being coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the electronic device to perform the method according to the first aspect of the present disclosure.
[0008] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the processor implements the method according to the first aspect of the present disclosure.
[0009] It should be understood that the content described in this summary section is not intended to limit the key features or important features of the implementation of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] The above and other features, advantages and aspects of various implementations of the present disclosure will become more apparent hereinafter with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:
[0011] FIG1 shows a block diagram of an application environment according to an exemplary implementation of the present disclosure;
[0012] FIG2 shows a block diagram of data delivery according to some implementations of the present disclosure;
[0013] FIG3 shows a block diagram of performing data delivery based on a file system according to some implementations of the present disclosure;
[0014] FIG4 illustrates a block diagram of the architecture of a feature service according to some implementations of the present disclosure;
[0015] FIG5 illustrates a block diagram of determining an adjustment action according to some implementations of the present disclosure;
[0016] FIG6 is a block diagram illustrating a data structure of a sub-action in an adjustment action according to some implementations of the present disclosure;
[0017] FIG7 shows a block diagram of a subsequent processing procedure performed for an adjustment action according to some implementations of the present disclosure;
[0018] FIG8 shows a block diagram of priority control in an adjustment action according to some implementations of the present disclosure;
[0019] FIG9 shows a flow chart of a data delivery method according to some implementations of the present disclosure;
[0020] FIG10 shows a block diagram of a data delivery device according to some implementations of the present disclosure; and
[0021] FIG11 shows a block diagram of a device capable of implementing various implementations of the present disclosure. DETAILED DESCRIPTION
[0022] The following describes implementations of the present disclosure in more detail with reference to the accompanying drawings. Although certain implementations of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the implementations described herein. Rather, these implementations are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and implementations of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.
[0023] In the description of the implementation of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "an implementation" or "the implementation" should be understood as "at least one implementation". The term "some implementations" should be understood as "at least some implementations". The following may also include other explicit and implicit definitions. As used herein, the term "model" can represent the association relationship between various data. For example, the above-mentioned association relationship can be obtained based on a variety of technical solutions currently known and / or to be developed in the future.
[0024] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) must comply with the requirements of relevant laws, regulations and relevant provisions.
[0025] It is understandable that before using the technical solutions disclosed in the various embodiments of this disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved in this disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.
[0026] For example, in response to a user's active request, a prompt message is sent to the user to clearly inform the user that the operation requested will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the electronic device, application, server, storage medium, or other software or hardware that performs the operations of the disclosed technical solution based on the prompt message.
[0027] As an optional but non-limiting implementation, in response to receiving a user's active request, a prompt message may be sent to the user, for example, in the form of a pop-up window, in which the prompt message may be presented in text form. Furthermore, the pop-up window may also include a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device.
[0028] It is understandable that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of the present disclosure.
[0029] As used herein, the term "in response to" refers to a state in which a corresponding event occurs or a condition is satisfied. It will be understood that the timing of executing a subsequent action executed in response to the event or condition is not necessarily strongly correlated with the time when the event occurs or the condition is satisfied. For example, in some cases, a subsequent action may be executed immediately upon the occurrence of the event or the satisfaction of the condition; in other cases, the subsequent action may be executed some time after the occurrence of the event or the satisfaction of the condition.
[0030] Sample Environment
[0031] The data delivery system can deliver multiple types of data to various applications. The amount of data to be delivered is enormous, and the data delivery system needs to acquire and process each piece of data one by one, and then set up a corresponding delivery method. Referring to FIG1 , an application environment according to an exemplary implementation of the present disclosure is described, which shows a block diagram of an application environment 100 according to an exemplary implementation of the present disclosure. As shown in FIG1 , the delivery system 110 can process a large amount of related data 130 stored in different data sources 120.
[0032] In today's development environment, microservices are a popular approach to application development, and existing delivery systems 110 are often developed based on a microservices architecture. Microservices advocate for breaking a single application into a set of small services that coordinate and collaborate with each other to achieve the application's functionality. Each service runs in its own independent process, communicating with each other using lightweight communication mechanisms. Each service is built around a specific business and can be independently deployed in various application environments.
[0033] Accordingly, the delivery system 110 developed based on the microservice architecture usually avoids a unified, centralized service management mechanism as much as possible, which results in unsatisfactory performance in obtaining and processing the data to be delivered. At this time, the delivery system 110 needs to obtain relevant data 130 via remote calls 140. The amount of data to be delivered may be very large. For example, the delivery system 110 may need to process millions of data, which results in a single remote call being limited by the size limit of the network packet. At this time, it may be necessary to execute tens of thousands of remote calls (for example, each remote call obtains 100 data) to obtain the data to be delivered. On the one hand, a large number of remote calls lead to increased time overhead, and on the other hand, failure of remote calls may result in incomplete data. At this time, how to manage these data to be delivered and execute data delivery, and then achieve the expected delivery goals, has become a research hotspot and difficulty.
[0034] Overview of Data Delivery
[0035] In order to at least partially address the deficiencies in the prior art, a data delivery method is proposed according to an exemplary implementation of the present disclosure. In the context of the present disclosure, instead of a microservice architecture that can only process a relatively small amount of data, a technical solution for performing data delivery tasks based on a file system is provided. In summary, a large amount of data to be delivered and other associated data can be stored in a file at a certain address in the file system. When subsequently analyzing and processing these data, the address can be used as a unique identifier to access various data, thereby obtaining the data and then performing the corresponding data delivery task.
[0036] An overview of an exemplary implementation of the present disclosure is provided with reference to FIG2 , which illustrates a block diagram 200 of data delivery according to some implementations of the present disclosure. As shown in FIG2 , a delivery target 212 associated with a data delivery task may be received. The data delivery task may specify multiple data items to be delivered, such as data ranges for multiple data items to be retrieved from a certain location, or multiple data items associated with a certain account, etc. Delivery target 212 may describe the desired delivery result achieved by delivering the multiple data items 210.
[0037] For the sake of ease of description, in the context of the present disclosure, information push (e.g., advertising push, news push, etc.) will be used as a specific example of a data delivery task to describe the specific steps of an exemplary implementation method of the present disclosure. At this time, the data can be the information being pushed. For example, when it is desired to promote a certain application, the data can be news, introductions, download addresses, etc. about the application. The data delivery task at this time can, for example, specify the delivery of data on a certain (certain) channel (e.g., application, television, radio, etc.), the delivery of data to certain users, the delivery of data in a certain time period, etc. The delivery target 212 can, for example, describe the expected delivery results, for example, how many new users are expected to be obtained by delivering data, how many click-through rates are expected to be achieved, and so on.
[0038] As shown in FIG2 , a file system 270 can be provided to facilitate file-based access to various data. Specifically, a data service 230 associated with a data delivery task can be utilized to acquire feature data 220 associated with multiple data 210 and a delivery target 212. Feature data 220 can describe the relationship between the attribute data of the multiple data 210 and the delivery target 212, and feature data 220 can be stored at a certain address 222 (e.g., referred to as a first address) in the file system 270. In subsequent operations, feature data 220 can be accessed based on this address 222 to execute the corresponding data delivery task.
[0039] According to an exemplary implementation of the present disclosure, a predetermined adjustment strategy 250 may be provided. The adjustment strategy 250 may specify how the data delivery process is expected to be adjusted, for example, reducing the data delivery cost, increasing the user click-through rate, and so on. An adjustment action 260 for adjusting the data delivery task may be determined based on the predetermined adjustment strategy 250 and the feature data 222, and the adjustment action 260 makes the delivery result of the adjusted data delivery task match the delivery target 212. Specifically, the unified access interface 242 provided by the feature service 240 may be used to determine the adjustment action 260, thereby managing different data delivery tasks in a unified manner.
[0040] Using the exemplary implementation of the present disclosure, delivery system 110 does not need to rely on a large number of remote calls 140 to process each piece of data to be delivered. Instead, it can obtain all required data in a unified manner through file system 270 and then perform subsequent data delivery tasks. In this way, problems such as long waiting times and retransmissions caused by a large number of remote calls can be avoided, and the efficiency of data delivery tasks can be greatly improved.
[0041] Detailed process of data delivery
[0042] An overview of an exemplary implementation according to the present disclosure has been described with reference to FIG2 , and below, more details of data delivery are described with reference to FIG3 to 8 . FIG3 shows a block diagram 300 for performing data delivery based on a file system according to some implementations of the present disclosure. Here, the data service 230 can obtain relevant data 130 from one or more locations defined by the data ranges of multiple data 210. For example, when the data to be delivered comes from different data providers, the relevant data 130 can be pulled from multiple data sources across multiple geographical locations. The data service 230 can provide a unified service interface for pulling relevant data from outside the delivery system 110, and then generate attribute data 310 related to later data analysis and definition of data delivery tasks.
[0043] In this way, delivery system 110 does not need to trigger a large number of remote calls based on microservices. Instead, it can simply send the data ranges of multiple data to be delivered to data service 230 to obtain the desired data. In this case, only two remote calls (i.e., the request process and the receive process) are required between delivery system 110 and data service 230 to complete the data acquisition process. Compared with existing architectures that require a large number of remote calls, the exemplary implementation of the present disclosure can utilize a mature file system to manage data processes, greatly improving data transmission efficiency, thereby reducing data latency and improving data delivery performance.
[0044] As shown in FIG3 , a data service 230 can be provided to query the relevant data 130 of the multiple data being delivered based on the data range, thereby generating corresponding attribute data 310. Furthermore, the attribute data 310 can be stored in a file at a certain address 312 (e.g., referred to as the second address) in the file system 320. According to an exemplary implementation of the present disclosure, the attribute data 310 can describe various aspects of information related to the data delivery task, such as but not limited to: data identifiers, account identifiers, data group identifiers, delivery channels, data locations, operating systems, bids, budgets, time, consumption, conversions, retention, user interactions, etc. associated with the multiple data.
[0045] Specifically, the data identifier associated with multiple data can uniquely identify each data, the account identifier can indicate the account to which the data belongs (for example, the account of the data provider); the data group identifier can indicate the location of the data group where the data is located, the delivery channel can indicate in which channels the data is delivered, the data location can indicate the location where the data is delivered (for example, the homepage, the secondary page, etc.), the operating system can indicate the operating system used by the user browsing the delivered data, the bid can indicate the relevant bid for the delivered data, the budget can indicate the relevant budget for the delivered data, the time can indicate the time when the data is delivered, the consumption can indicate the various aspects of the expenditure on the delivered data, the conversion can indicate the conversion generated by the delivered data, the retention can indicate the retention generated by the delivered data, and the user interaction can indicate the further interaction of the user with the data after browsing the delivered data (for example, user clicks, user registrations, etc.). Assuming that there are N data, and these data belong to two accounts 1 and 2 respectively, the data structure of the attribute data 310 at this time is as shown in Table 1 below. The "other attributes..." in Table 1 can represent columns of one or more attributes described above.
[0046] Table 1 Example of attribute data
[0047] According to an exemplary implementation of the present disclosure, attribute data 310 associated with a data delivery task may be stored in a file system 320 (e.g., a centralized file system and / or a distributed file system). The attribute data 310 may provide data support for defining the data delivery task. For example, based on one or more attributes, it may be determined whether the current data delivery method can meet the expected delivery target 212. If the delivery target 212 cannot be met, the delivery method of certain data among the multiple data may be adjusted accordingly, thereby ensuring that the adjusted data delivery task meets the delivery target 212. In this way, corresponding adjustment actions may be determined based on the collected real data, thereby improving the efficiency of the data delivery task.
[0048] According to an exemplary implementation of the present disclosure, delivery target 212 can be added to attribute data 310 to obtain corresponding feature data 220. Assuming that the delivery targets of two accounts are "Cost 25" and "Cost 30," respectively, a new column "Cost Target" can be added to Table 1, and each data row can be stored in a corresponding bucket in file system 320 (as shown in the last column "Bucket"). In this case, the data structure of feature data 220 is, for example, as shown in Table 2 below.
[0049] Table 2 Example of characteristic data
[0050] According to an exemplary implementation of the present disclosure, the data service 230 may store the feature data 220 at the address 222 in the file system 320. At this time, the portion of the feature data 220 corresponding to "cost 25" may be stored in the bucket 330, and the portion of the feature data 220 corresponding to "cost 30" may be stored in the bucket 332. In this way, data corresponding to different delivery targets may be indexed based on the buckets, thereby improving the efficiency of later data processing. It will be understood that although Table 2 shows a case where each account has only one delivery target, alternatively and / or additionally, multiple data of an account may correspond to multiple delivery targets, in which case the data associated with each delivery target may be stored in the corresponding bucket.
[0051] According to an exemplary implementation of the present disclosure, in the process of executing a data delivery task, feature data 220 can be periodically acquired at predetermined time intervals (e.g., 10 minutes or other time intervals) in order to dynamically adjust data delivery in real time. At this time, a corresponding global identifier can be set for each feature data 220 so as to uniquely identify the feature data 220 acquired at different time points in the file system 320. It will be understood that although FIG3 schematically illustrates that the attribute data 310 and the feature data 220 are stored in the same file system 320, alternatively and / or additionally, the attribute data 310 and the feature data 220 can be stored in different file systems. In this way, it is convenient to provide more flexibility for data storage.
[0052] According to an exemplary implementation of the present disclosure, a feature service 240 can be utilized to perform data delivery in a unified manner. See FIG4 for more details on the feature service. FIG4 shows a block diagram 400 of the architecture of the feature service 240 according to some implementations of the present disclosure. As shown in FIG4, the feature service 240 can provide a unified format 410 for storing feature data 220 associated with different data delivery tasks. The feature data 220 can be stored at an address 222 in a designated storage medium 420 (e.g., a storage device in the file system 320), thereby facilitating access to the feature data 220 at a globally unique address 222.
[0053] Furthermore, the feature service 240 can provide a unified access interface 242 to the delivery system 110, which can receive input parameters 430 set in a predetermined format and output output parameters 432 defined in a predetermined format, thereby determining the corresponding adjustment action 260. Using the exemplary implementation of the present disclosure, the delivery system 110 does not have to call the adjustment strategy 250 to process each data obtained by remote call one by one, but can treat all the data to be processed as a whole and process all the data in a unified manner via the access interface 242. In this way, data processing efficiency can be improved and potential time delays can be reduced. According to an exemplary implementation of the present disclosure, the input parameters 430 can be defined based on the data structure shown in Table 3 below.
[0054] Table 3 Examples of input parameters
[0055] Based on the input parameters shown in Table 3, the feature service 240 can read the feature data 22 at the address 222 in the file system 270. Further, the feature service 240 can select one or more data to be adjusted from the read feature data 220 based on the predetermined adjustment strategy 250. Here, each of the multiple data can be processed one by one, and the data to be processed can be selected based on whether the data meets the delivery target 212. Specifically, for each target data in the multiple data, the delivery result associated with the delivery target 212 can be determined based on the portion of the attribute data 310 corresponding to the target data. In other words, each data item in the feature data 220 (that is, each row of data in Table 2) can be processed one by one to determine whether the delivery result associated with the row of data matches the delivery target 212.
[0056] It will be understood that for different delivery targets 212, different attributes may be involved in determining the delivery results. For example, in the example where the delivery target 212 is "cost 25", for each row of data, the attributes "consumption" and "conversion" may be selected from the feature data 220. Further, the actual cost incurred in executing the current data delivery task may be determined based on the quotient of "consumption" and "conversion". Furthermore, it may be determined whether the data in the row is data to be processed based on a comparison of the actual cost with the delivery target 212 corresponding to the data row (e.g., "cost 25" or "cost 30"). If the difference between the two meets a predetermined condition (e.g., greater than a predetermined ratio (e.g., 20% or other ratio) or a predetermined value (e.g., 5 or other value)), the target data is identified as data to be adjusted.
[0057] For example, when the delivery goal 212 is "retention" for a certain time period, the "retention" for the corresponding time period in the attribute data 310 can be divided by "conversion" to obtain the retention rate for the corresponding time period (e.g., next-day retention, three-day retention, etc.). In this way, the required attributes can be selected from the feature data 220, thereby more accurately determining the data to be processed that is not conducive to achieving the delivery goal 212.
[0058] It will be understood that while the above description only uses "cost" and "retention" as examples of specific delivery objectives, alternatively and / or additionally, the delivery objectives may include at least one of the following: cost, clicks, click-through rate, number of new users per day, and user retention. For example, the clicks, click-through rate, and number of new users per day generated by the current data may be determined based on user interactions in the attribute data. In this way, data providers can be supported in defining different delivery objectives, thereby achieving data promotion and / or other purposes.
[0059] Once the data to be processed has been determined, an adjustment method for adjusting the data to be adjusted can be determined based on a predetermined adjustment strategy 250. It will be appreciated that this adjustment method can make the adjusted data delivery task more consistent with the desired delivery target 212. The adjustment strategy 250 can be implemented based on a variety of currently known and / or future developed methods. For example, the adjustment strategy 250 can be implemented based on machine learning techniques and / or other methods. In the context of this disclosure, the details of the adjustment strategy 250 are not of concern, and the adjustment strategy 250 can be understood as being implemented in a black-box manner.
[0060] According to an exemplary implementation of the present disclosure, the adjustment policy 250 may include one or more policy cards. For example, the function of one policy card may be to reduce the cost of data delivery, the function of another policy card may be to improve the retention rate, etc. Alternatively and / or additionally, based on the implementation of the policy card, a single policy card may also implement multiple functions, which is not limited in the context of the present disclosure.
[0061] Furthermore, adjustment sub-actions (e.g., simply referred to as sub-actions) corresponding to the data to be adjusted in adjustment action 260 can be generated based on the data to be adjusted and the adjustment method determined using adjustment strategy 250. In this case, adjustment action 260 generated by feature service 240 can include adjustment sub-actions for one or more of the multiple data. In this way, a corresponding adjustment sub-action can be determined for each data to be adjusted, thereby improving the accuracy of adjustment action 260, so that the adjusted data delivery task is more consistent with delivery target 212.
[0062] 5 , which illustrates a block diagram 500 for determining an adjustment action 260 according to some implementations of the present disclosure. As shown in FIG5 , feature data 220 may include multiple data items, such as data items 510, 512, 514, 516, ..., and 518 (corresponding to the respective data rows shown in Table 2 above). Feature data 220 may be stored at address 222, and feature service 240 may access feature data 220 and adjustment strategy 250 via input parameters as shown in Table 3 above, thereby obtaining corresponding adjustment actions 260.
[0063] Here, adjustment action 260 may include multiple sub-actions 520, 522, ..., and 524, and each sub-action corresponds to data to be adjusted from the multiple data. For example, as shown at 530, sub-action 520 is used to adjust the data placement associated with data item 510; as shown at line 532, sub-action 522 is used to adjust the data placement associated with data item 514; ...; and as shown at line 534, sub-action 524 is used to adjust the data placement associated with data item 518, and so on.
[0064] According to an exemplary implementation of the present disclosure, the adjustment method can adjust at least any one of the following: the operation type of the data to be adjusted, the operation level, the bid information, the budget information, the status information, and the like. Specifically, the operation type can define any one of the following types: for example, adjusting the advertisement bid, the advertisement status, the advertisement budget, the delivery period, and the like. The operation level can define the data in which level the adjustment action will affect, for example, a certain advertisement can be adjusted, the advertisement of a certain advertiser can be adjusted, the advertisement of a certain account can be adjusted, and so on. The bid information is valid when the operation type is "adjust advertisement bid", and can represent the adjusted bid information. The budget information is valid when the operation type is "adjust advertisement budget", and can represent the adjusted budget information. The status information is valid when the operation type is "adjust advertisement status", and can represent the adjusted status information.
[0065] In this way, the adjustment method for each piece of data to be adjusted can be defined in a more flexible and unified manner, thereby enabling the adjustment action to adjust the data delivery task in a direction that better matches the delivery target 212. According to an exemplary implementation of the present disclosure, output parameters 432 can be defined based on the data structure shown in Table 4 below, thereby defining each adjustment sub-action.
[0066] Table 4 Example of output parameters
[0067] Furthermore, the feature service 240 can be called and the determined adjustment action 260 can be output using the predetermined output parameters 432 in the access interface 242. It will be understood that Table 4 below is merely an example of output parameters, and that the output parameters 432 can include more, fewer, or different parameters. For example, the output parameters 432 can further include an "output destination" to determine where the adjustment action 260 is output. For example, the "output destination" can point to the entry point of the execution module for executing the adjustment action 260. In this way, the adjustment action 260 can be executed in a timely manner, so that the result of the data delivery task is more consistent with the delivery target.
[0068] According to an exemplary implementation of the present disclosure, the "end marker" shown in Table 4 can indicate whether all data has been processed completely, that is, whether the current sub-action is the last sub-action in the adjustment action 260. When processing each data item in the feature data 220 one by one, if it is determined that all data in the multiple data have been processed, an end marker can be added to the adjustment sub-action at the end of the adjustment action 260.
[0069] FIG6 shows a block diagram 600 of the data structure of sub-actions in adjustment action 260 according to some implementations of the present disclosure. As shown in FIG6 , each sub-action can include a corresponding "end marker." For example, sub-action 520 can include end marker 610, sub-action 522 can include end marker 612, ..., and sub-action 524 can include end marker 614, and so on. The default value of each end marker can be "false." After all data in feature data 220 has been processed, end marker 614 in the last sub-action 524 can be set to "true." In this way, the last sub-action in adjustment action 260 can be accurately indicated.
[0070] According to an exemplary implementation of the present disclosure, in order to improve the accuracy of the adjustment action 260, a further optimization process may be performed on the generated adjustment action 260. FIG7 shows a block diagram 700 of a subsequent processing process performed on the adjustment action 260 according to some implementations of the present disclosure. As shown in FIG7 , an optimization process 710 may be performed on the generated adjustment action 260, and the optimization process 710 may include a frequency control 712 and a priority control 714. First, the relevant details of the frequency control 712 are described. It will be understood that adjusting the delivery method of a certain data too frequently may make it difficult to obtain the long-term impact of the adjustment action on the data delivery task, and may introduce unstable factors into the data delivery task, so the historical adjustment frequency of a certain data may be considered when generating the adjustment action 260.
[0071] According to an exemplary implementation of the present disclosure, the adjustment history within a past historical time period (for example, several hours, or a day, etc.) can be considered. For example, a predetermined frequency condition can be set (for example, whether it is higher than 5 times a day, or other frequencies). If it is determined that a certain data has been frequently adjusted in the past (for example, the adjustment frequency is higher than 5 times a day), then the data may no longer be adjusted during the current round of adjustment; if it is determined that the adjustment frequency of a certain data in the past is not higher than 5 times a day), then during the current round of adjustment, a corresponding adjustment method can be further generated for the data, and the delivery result obtained according to the adjustment method matches the delivery target. In this way, the generated adjustment action 260 can adjust the data delivery task towards the delivery target 212 in a more accurate manner, thereby improving the performance of the delivery system 110.
[0072] According to an exemplary implementation of the present disclosure, delivery target 212 may include multiple adjustment targets, which may result in conflicts between the multiple adjustment targets. Figure 8 shows a block diagram 800 of priority control in adjustment actions according to some implementations of the present disclosure. As shown in Figure 8, delivery target 212 includes two adjustment targets 810 and 812. Adjustment target 810 is intended to reduce the cost of data delivery, and adjustment target 812 is intended to improve user retention.
[0073] In FIG8 , sub-actions 520, 522, ..., and 524 can be generated based on adjustment target 810, and sub-actions 820, 822, ..., and 824 can be generated based on adjustment target 812. In this case, sub-actions associated with different adjustment targets may involve the same data, and the adjustment operations for the same data may also be the same. For example, sub-action 520 and sub-action 824 involve the same data (e.g., data item 510) among multiple data. If sub-actions 520 and 824 involve the same operation type (e.g., adjusting the delivery period, or other types), then sub-action 824, which is located later, can be deleted.
[0074] It will be understood that since adjustment target 810 is located before policy card 812, it can be considered that cost control has a higher priority than retention rate control. At this time, sub-action 520 generated based on adjustment target 810 is located before sub-action 824 generated based on adjustment target 812, and thus the priority of sub-action 520 is higher than the priority of sub-action 824. Using the exemplary implementation of the present disclosure, deleting sub-action 824 can ensure that the data delivery process is adjusted in a direction that meets the higher priority. In this way, the data delivery task can be preferentially adjusted in the direction of cost control.
[0075] The process of determining adjustment action 260 has been described above. After determining adjustment action 260, adjustment action 260 can be further executed to adjust the data delivery task in a direction that is more consistent with delivery target 212. Further referring to FIG7 , execution process 720 can be called based on adjustment action 260 to execute the various sub-actions defined in adjustment action 260. It will be understood that it takes a certain amount of time for feature service 240 to traverse each data item in feature data 220 and generate the corresponding sub-actions. During this time, it can continuously check whether there is an end marker in adjustment action 260 to determine whether processing of the current feature data 220 has been completed.
[0076] According to an exemplary implementation of the present disclosure, if the end marker is not detected in the adjustment action 260, the end marker can be detected again in the adjustment action 260. Specifically, the end marker can be periodically detected at a predetermined time interval (e.g., 1 minute or other time interval). Alternatively and / or additionally, the feature service 240 can send an interrupt message to start the execution process 720 when all the multiple data are processed.
[0077] According to an exemplary implementation of the present disclosure, if an end marker is detected in adjustment action 260, the data delivery task can be executed based on the adjustment action. In other words, each sub-action in adjustment action 260 can be executed to complete the adjustment process. In this way, all of the multiple data being delivered can be processed in a holistic manner, thereby preventing one-sided adjustments caused by incomplete adjustment actions. As a result, the adjusted data delivery task can be more consistent with delivery target 212.
[0078] According to an exemplary implementation of the present disclosure, a predetermined waiting time length (for example, 5 minutes, or other time length) can be set for each round of the adjustment process. If the end mark is not detected within the predetermined time length, the method described above can be ended, that is, the execution process 720 shown in Figure 7 is terminated. Here, the waiting time length can be set based on the time interval for calling the adjustment process. For example, the waiting time length can be set to 50% (or other proportions) of the calling time interval. Utilizing the exemplary implementation of the present disclosure, corresponding adjustment actions can be generated and executed based on the latest acquired feature data. In this way, the real-time performance of the adjustment process can be improved, and the data delivery task can be continuously adjusted based on the feature data collected in real time, so that the adjusted data delivery task can be more in line with the delivery target 212.
[0079] According to an exemplary implementation of the present disclosure, a recording process 730 may be performed to record various historical parameters related to the adjustment process. Further, the recorded historical parameters may be used to guide subsequent adjustment processes.
[0080] According to an exemplary implementation of the present disclosure, the adjustment process described above can be performed again after a predetermined time interval to obtain another adjustment action for adjusting the data delivery task. Using the exemplary implementation of the present disclosure, new adjustment actions can be continuously generated over time, thereby making the adjusted data delivery task more consistent with the delivery target 212.
[0081] The various steps of data delivery have been described with reference to the accompanying drawings. The overall process for executing a data delivery task using the methods described above will be described below. Specifically, a user (e.g., an administrator for optimizing data delivery) can configure basic information for the management task, such as defining the data range, the adjustment strategy 250 to be used, the delivery target 212, the adjustment frequency, and so on. In this case, the delivery system 110 can run at fixed time intervals (e.g., every 10 minutes), that is, executing a round of the adjustment process every 10 minutes.
[0082] During each adjustment process, the data service 230 can pull advertising attribute data (such as bids, budgets, status, advertising positions, advertising channels, etc.) based on the data range. Here, the traditional method of multiple remote calls is abandoned. Instead, the pull conditions are informed to the data service 230, and the data service 230 queries all relevant attribute data at one time, and uploads it to a distributed file system (such as Hadoop Distributed File System, abbreviated as HDFS) in the form of a data file, and then informs the delivery system 110 of the file address in the form of a callback. Subsequently, the delivery system 110 receives the callback from the data service 230 and obtains the corresponding data through the file address. In this way, two remote calls (request and obtain) can be used to download the corresponding data, thereby completing large-scale data exchange. Furthermore, the attribute data file can be processed secondary, and a feature data file can be generated, which is then uploaded to a distributed file system (for example, HDFS).
[0083] Furthermore, the feature service 240 can obtain the feature data 220 and the adjustment strategy 250 through a unified input parameter 430, and then generate the corresponding adjustment action 260 through a unified output parameter 432. Specifically, the corresponding one or more adjustment sub-actions can be output through a software development kit (abbreviated as SDK), and an end mark can be written in the last sub-action. It will be understood that the delivery system 110 determines whether all the data has been processed based on the end mark. The delivery system 110 will not wait indefinitely, but will abandon the current round of adjustments when the maximum waiting time is reached. After obtaining the complete adjustment action 260, the delivery system 110 can merge the results according to the order of the policy configuration, and the results of the same operation type of the previous policy will overwrite the results of the subsequent policy. Further, the delivery system 110 can deliver the corresponding data to the delivery platform, thereby completing this round of adjustment process.
[0084] Compared to existing microservice-based delivery systems, the present disclosure proposes a large-file-based data exchange method, which can complete data exchange with just two remote call requests. This method improves the efficiency of delivery system 110, reduces the possibility of incomplete data due to call failures, and alleviates the pressure on backend storage systems.
[0085] Example Process
[0086] FIG9 shows a flow chart of a data delivery method 900 according to some implementations of the present disclosure. As shown in FIG9 , at box 910, a delivery target associated with a data delivery task is received, and the delivery target describes the delivery result expected to be obtained by delivering multiple data. At box 920, feature data of the data delivery task is obtained using a file system, and the feature data describes the association relationship between the attribute data of multiple data and the delivery target. At box 930, an adjustment action for adjusting the data delivery task is determined based on a predetermined adjustment strategy and the feature data, and the adjustment action makes the delivery result of the adjusted data delivery task match the delivery target.
[0087] According to an exemplary implementation of the present disclosure, the method 900 further includes: querying attribute data of the plurality of data according to data ranges of the plurality of data; and storing the attribute data in a file system.
[0088] According to an exemplary implementation of the present disclosure, acquiring the characteristic data includes: adding a delivery target to the attribute data at the second address to obtain the characteristic data.
[0089] According to an exemplary implementation of the present disclosure, the feature data is stored at a first address in the file system in a predetermined unified format, and the method 900 further includes: reading the feature data from the first address in the file system using predetermined input parameters in an access interface of the file system.
[0090] According to an exemplary implementation of the present disclosure, the method 900 further includes: outputting the adjustment action using a predetermined output parameter in an access interface of the file system.
[0091] According to an exemplary implementation of the present disclosure, determining an adjustment action includes: based on a predetermined adjustment strategy, selecting data to be adjusted from feature data, and the delivery result of the data to be adjusted does not meet the delivery target; based on the predetermined adjustment strategy, determining an adjustment method for adjusting the data to be adjusted; and based on the data to be adjusted and the adjustment method, generating an adjustment sub-action in the adjustment action corresponding to the data to be adjusted.
[0092] According to an exemplary implementation of the present disclosure, selecting data to be adjusted includes: determining, for target data among multiple data, a delivery result associated with the delivery target based on a portion of attribute data corresponding to the target data; and identifying the target data as data to be adjusted in response to determining that a difference between the delivery result and the delivery target satisfies a predetermined condition.
[0093] According to an exemplary implementation of the present disclosure, determining the adjustment method includes: in response to determining that the historical adjustment frequency of the data to be adjusted meets a predetermined frequency condition, determining the adjustment method, and matching the delivery result obtained according to the adjustment method with the delivery target.
[0094] According to an exemplary implementation of the present disclosure, determining the adjustment method includes determining at least any one of the following: an operation type, an operation level, bid information, budget information, and status information for adjusting the data to be adjusted.
[0095] According to an exemplary implementation of the present disclosure, the method 900 further includes: in response to determining that the first adjustment sub-action and the second adjustment sub-action in the adjustment action involve the same data among multiple data, and the first adjustment sub-action and the second adjustment sub-action have the same operation type, deleting the second adjustment sub-action, and the second adjustment sub-action is located after the first adjustment sub-action.
[0096] According to an exemplary implementation of the present disclosure, the method 900 further includes: in response to determining that all of the plurality of data are processed, adding an end marker to an adjustment sub-action at the end of the adjustment action.
[0097] According to an exemplary implementation of the present disclosure, the method 900 further includes: in response to detecting an end marker in the adjustment action, executing a data delivery task based on the adjustment action.
[0098] According to an exemplary implementation of the present disclosure, the method 900 further includes: in response to not detecting the end marker during the adjustment action, detecting the end marker again during the adjustment action.
[0099] According to an exemplary implementation of the present disclosure, the method 900 further includes: in response to not detecting the end marker within a predetermined time period, ending the method.
[0100] According to an exemplary implementation of the present disclosure, the method 900 further includes: after a predetermined time interval, executing the method again to obtain another adjustment action for adjusting the data delivery task.
[0101] According to an exemplary implementation of the present disclosure, attribute data includes at least any one of the following: data identification, account identification, data group identification, delivery channel, data location, operating system, bid, budget, time, consumption, conversion, retention, and user interaction associated with multiple data.
[0102] According to an exemplary implementation of the present disclosure, the delivery target includes at least any one of the following: cost, click volume, click-through rate, number of new users per day, and user retention.
[0103] Example devices and equipment
[0104] Figure 10 shows a block diagram of a data delivery device 1000 according to some implementations of the present disclosure. The device 1000 includes: a receiving module 1010 configured to receive a delivery target associated with a data delivery task, the delivery target describing a delivery result expected to be obtained by delivering multiple data; an acquisition module 1020 configured to use a file system to acquire feature data of the data delivery task, the feature data describing an association between attribute data of the multiple data and the delivery target; and a determination module 1030 configured to determine an adjustment action for adjusting the data delivery task based on a predetermined adjustment policy and the feature data, the adjustment action ensuring that the delivery result of the adjusted data delivery task matches the delivery target.
[0105] According to an exemplary implementation of the present disclosure, the apparatus 1000 further includes: a query module configured to query attribute data of multiple data according to data ranges of the multiple data using a data service; and a storage module configured to store the attribute data in a file system.
[0106] According to an exemplary implementation of the present disclosure, the acquisition module 1020 includes: an adding module configured to add a delivery target to the attribute data at the second address to obtain feature data.
[0107] According to an exemplary implementation of the present disclosure, the characteristic data is stored at a first address in a file system in a predetermined unified format, and the device 1000 further includes: an input module configured to read the characteristic data from the first address in the file system using predetermined input parameters in an access interface of the file system.
[0108] According to an exemplary implementation of the present disclosure, the apparatus 1000 further includes: an output module configured to output the adjustment action using predetermined output parameters in the access interface of the file system.
[0109] According to an exemplary implementation of the present disclosure, the determination module 1030 includes: a selection module, configured to select data to be adjusted from the feature data based on a predetermined adjustment strategy, and the delivery result of the data to be adjusted does not meet the delivery target; an adjustment method determination module, configured to determine an adjustment method for adjusting the data to be adjusted based on the predetermined adjustment strategy; and a generation module, configured to generate an adjustment sub-action corresponding to the data to be adjusted in the adjustment action based on the data to be adjusted and the adjustment method.
[0110] According to an exemplary implementation of the present disclosure, the selection module includes: a result determination module, configured to determine, for target data among multiple data, a delivery result associated with a delivery target based on a portion of attribute data corresponding to the target data; and an identification module, configured to identify the target data as data to be adjusted in response to determining that a difference between the delivery result and the delivery target satisfies a predetermined condition.
[0111] According to an exemplary implementation of the present disclosure, the adjustment method determination module includes: a frequency-based determination module, which is configured to determine the adjustment method in response to determining that the historical adjustment frequency of the data to be adjusted meets a predetermined frequency condition, and the delivery result obtained according to the adjustment method matches the delivery target.
[0112] According to an exemplary implementation of the present disclosure, the adjustment method determination module is configured to determine at least any one of the following: an operation type, an operation level, bid information, budget information, and status information for adjusting the data to be adjusted.
[0113] According to an exemplary implementation of the present disclosure, the device 1000 further includes: a deletion module, configured to delete the second adjustment sub-action in response to determining that the first adjustment sub-action and the second adjustment sub-action in the adjustment action involve the same data among multiple data, and the first adjustment sub-action and the second adjustment sub-action have the same operation type, and the second adjustment sub-action is located after the first adjustment sub-action.
[0114] According to an exemplary implementation of the present disclosure, the apparatus 1000 further includes: an end marker setting module configured to add an end marker to an adjustment sub-action at the end of the adjustment action in response to determining that all of the multiple data are processed.
[0115] According to an exemplary implementation of the present disclosure, the apparatus 1000 further includes: an execution module configured to execute a data delivery task based on the adjustment action in response to detecting an end marker in the adjustment action.
[0116] According to an exemplary implementation of the present disclosure, the apparatus 1000 further includes: a waiting module configured to detect an end marker again in the adjustment action in response to not detecting the end marker in the adjustment action.
[0117] According to an exemplary implementation of the present disclosure, the apparatus 1000 further includes: an ending module configured to stop calling the apparatus in response to not detecting an end marker within a predetermined time period.
[0118] According to an exemplary implementation of the present disclosure, the apparatus 1000 further includes: a periodic execution module configured to execute the method again after a predetermined time interval to obtain another adjustment action for adjusting the data delivery task.
[0119] According to an exemplary implementation of the present disclosure, attribute data includes at least any one of the following: data identification, account identification, data group identification, delivery channel, data location, operating system, bid, budget, time, consumption, conversion, retention, and user interaction associated with multiple data.
[0120] According to an exemplary implementation of the present disclosure, the delivery target includes at least any one of the following: cost, click volume, click-through rate, number of new users per day, and user retention.
[0121] FIG11 shows a block diagram of a device 1100 capable of implementing various implementations of the present disclosure. It should be understood that the computing device 1100 shown in FIG11 is merely exemplary and should not be construed as limiting the functionality and scope of the implementations described herein. The computing device 1100 shown in FIG11 can be used to implement the methods described above.
[0122] As shown in FIG11 , computing device 1100 is in the form of a general-purpose computing device. Components of computing device 1100 may include, but are not limited to, one or more processors or processing units 1110, memory 1120, storage device 1130, one or more communication units 1140, one or more input devices 1150, and one or more output devices 1160. Processing unit 1110 may be a real or virtual processor and is capable of performing various processes according to a program stored in memory 1120. In a multi-processor system, multiple processing units execute computer-executable instructions in parallel to increase the parallel processing capabilities of computing device 1100.
[0123] The computing device 1100 typically includes a plurality of computer storage media. Such media can be any available media accessible to the computing device 1100, including but not limited to volatile and non-volatile media, removable and non-removable media. The memory 1120 can be a volatile memory (e.g., registers, cache, random access memory (RAM)), a non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. The storage device 1130 can be a removable or non-removable medium and can include a machine-readable medium, such as a flash drive, a disk, or any other medium that can be used to store information and / or data (e.g., training data for training) and can be accessed within the computing device 1100.
[0124] The computing device 1100 may further include additional removable / non-removable, volatile / non-volatile storage media. Although not shown in FIG. 11 , a disk drive for reading from or writing to a removable, non-volatile disk (e.g., a "floppy disk") and an optical drive for reading from or writing to a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to a bus (not shown) by one or more data media interfaces. The memory 1120 may include a computer program product 1125 having one or more program modules configured to perform various methods or actions of various implementations of the present disclosure.
[0125] The communication unit 1140 enables communication with other computing devices via a communication medium. Additionally, the functionality of the components of the computing device 1100 can be implemented as a single computing cluster or multiple computing machines that can communicate via a communication connection. Thus, the computing device 1100 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or other network nodes.
[0126] Input device 1150 may be one or more input devices, such as a mouse, keyboard, or trackball. Output device 1160 may be one or more output devices, such as a display, a speaker, or a printer. Computing device 1100 may also communicate with one or more external devices (not shown) via communication unit 1140, as needed, such as storage devices, display devices, or the like, with one or more devices that allow a user to interact with computing device 1100, or with any device that allows computing device 1100 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface (not shown).
[0127] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which computer-executable instructions are stored, wherein the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is provided, on which a computer program is stored, and when the program is executed by a processor, the method described above is implemented.
[0128] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, apparatuses, devices, and computer program products implemented according to the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.
[0129] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, such that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.
[0130] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0131] The flow charts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple implementations of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a part for a module, program segment or instruction, and a part for a module, program segment or instruction comprises one or more executable instructions for realizing the logical function of the specification. In some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the function or action of the specification, or can be realized by a combination of special hardware and computer instructions.
[0132] While various implementations of the present disclosure have been described above, the foregoing description is intended to be illustrative, not exhaustive, and not limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is selected to best explain the principles of the implementations, their practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the various implementations disclosed herein.
Claims
1. A data delivery method, comprising: Receiving a delivery target associated with the data delivery task, wherein the delivery target describes a delivery result expected to be obtained by delivering a plurality of data; Acquiring characteristic data of the data delivery task by using a file system, wherein the characteristic data describes an association relationship between attribute data of the plurality of data and the delivery target; and An adjustment action for adjusting the data delivery task is determined based on a predetermined adjustment strategy and the characteristic data, wherein the adjustment action makes the delivery result of the adjusted data delivery task match the delivery target.
2. The method according to claim 1, further comprising: querying the attribute data of the plurality of data according to the data range of the plurality of data; as well as The attribute data is stored in the file system.
3. The method according to claim 2, wherein obtaining the characteristic data comprises: The delivery target is added to the attribute data in the file system to obtain the characteristic data.
4. The method according to claim 1, wherein the characteristic data is stored at a first address in the file system in a predetermined unified format, and the method further comprises: The characteristic data is read from the first address in the file system using a predetermined input parameter in the access interface of the file system.
5. The method according to claim 1, further comprising: The adjustment action is output using a predetermined output parameter in the access interface of the file system.
6. The method of claim 1 , wherein determining the adjustment action comprises: Based on the predetermined adjustment strategy, selecting data to be adjusted from the feature data, wherein a delivery result of the data to be adjusted does not meet the delivery target; Based on the predetermined adjustment strategy, determining an adjustment method for adjusting the data to be adjusted; as well as Based on the data to be adjusted and the adjustment method, the corresponding The adjustment sub-action should be applied to the data to be adjusted.
7. The method according to claim 6, wherein selecting the data to be adjusted comprises: For target data among the plurality of data, Determining a delivery result associated with the delivery target based on a portion of the attribute data corresponding to the target data; as well as In response to determining that the difference between the delivery result and the delivery target meets a predetermined condition, the target data is identified as the data to be adjusted.
8. The method according to claim 6, wherein determining the adjustment manner comprises: In response to determining that the historical adjustment frequency of the data to be adjusted meets a predetermined frequency condition, the adjustment method is determined, and the delivery result obtained according to the adjustment method matches the delivery target.
9. The method according to claim 7, wherein determining the adjustment method comprises determining at least any one of the following: an operation type, an operation level, bid information, budget information, and status information used to adjust the data to be adjusted.
10. The method according to claim 9, further comprising: In response to determining that the first adjustment sub-action and the second adjustment sub-action in the adjustment action involve the same data among the multiple data, and the first adjustment sub-action and the second adjustment sub-action have the same operation type, the second adjustment sub-action is deleted, and the second adjustment sub-action is located after the first adjustment sub-action.
11. The method according to claim 7, further comprising: In response to determining that all of the plurality of data are processed, an end marker is added to an adjustment sub-action at the end of the adjustment action.
12. The method according to claim 11, further comprising: In response to detecting the end marker in the adjustment action, the data delivery task is executed based on the adjustment action.
13. The method according to claim 12, further comprising: In response to not detecting the end marker in the adjustment action, detecting the end marker again in the adjustment action.
14. The method according to claim 13, further comprising: In response to not detecting the end marker within a predetermined length of time, the method ends.
15. The method according to claim 1, further comprising: After a predetermined time interval, the method is executed again to obtain another adjustment action for adjusting the data delivery task.
16. The method according to claim 1, wherein the attribute data includes at least any one of the following: data identification, account identification, data group identification, delivery channel, data location, operating system, bid, budget, time, consumption, conversion, retention, and user interaction associated with the multiple data.
17. The method according to claim 1, wherein the delivery target includes at least any one of the following: cost, click volume, click rate, number of new users per day, and user retention.
18. A data delivery device, comprising: A receiving module is configured to receive a delivery target associated with the data delivery task, wherein the delivery target describes a delivery result expected to be obtained by delivering a plurality of data; An acquisition module is configured to acquire characteristic data of the data delivery task by using a file system, wherein the characteristic data describes an association relationship between the attribute data of the plurality of data and the delivery target; as well as The determination module is configured to determine an adjustment action for adjusting the data delivery task based on a predetermined adjustment strategy and the characteristic data, wherein the adjustment action makes the delivery result of the adjusted data delivery task match the delivery target.
19. An electronic device comprising: at least one processing unit; as well as At least one memory, the at least one memory being coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions causing the electronic device to perform the method according to any one of claims 1 to 17 when executed by the at least one processing unit.
20. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the processor to implement the method according to any one of claims 1 to 17.
Citation Information
Patent Citations
Digital advertising input method based on deep machine learning and combination optimization
CN108830660A
Media information processing method and device, and storage medium
CN111626779A
Regulation and control method and device for delivery data and electronic equipment
CN115358768A
Data delivery method and device, equipment and medium
CN116645151A
System and method to facilitate targeted advertising
US20090106082A1
