A content processing method, apparatus, medium, and device

Abstract

The application discloses a content processing method, device, medium and equipment, comprising: obtaining delivery effect data, obtaining the identification of a target user and delivery effect slice data according to the delivery effect data; obtaining operation behavior data, obtaining first behavior data and second behavior data according to the operation behavior data and the identification of the target user; generating a first behavior strategy by analyzing the first behavior data; training a machine learning model according to the second behavior data and the delivery effect slice data to obtain a second behavior learning model, wherein the second behavior learning model represents a second behavior strategy; creating content based on the first behavior strategy, or adjusting content or adjusting the content delivery state based on the second behavior strategy. The application takes excellent operators as learning objects, analyzes their behavior patterns, and uses the behavior strategy obtained by learning analysis to effectively improve the content processing effect and the content delivery effect.

Description

Technical Field

[0001] This application relates to the field of machine learning, specifically to a content processing method, apparatus, medium, and device. Background Technology

[0002] Artificial Intelligence (AI) is a comprehensive technology within computer science that studies the design principles and implementation methods of various intelligent machines, enabling them to possess perception, reasoning, and decision-making capabilities. AI technology is a multidisciplinary field encompassing a wide range of areas, including natural language processing, machine learning, and deep learning. With technological advancements, AI will be applied in more fields and play an increasingly important role.

[0003] In content delivery applications, traditional methods emphasize the design of the delivery process, while automated delivery systems based on machine learning models are also being used. Taking advertising as an example, in the internet advertising field, there are generally two methods of ad delivery: one is for human advertisers to place ads; the other is for machines to place ads instead of human advertisers. Automated delivery by machines can undoubtedly improve delivery efficiency, but it also faces the challenge of ensuring delivery effectiveness. Summary of the Invention

[0004] To improve the effectiveness of content delivery, this application provides a content processing method, apparatus, medium, and device. The technical solution is as follows:

[0005] Firstly, this application provides a content processing method, the method comprising:

[0006] Obtain campaign performance data, and based on the campaign performance data, obtain the target user identifier and campaign performance slice data;

[0007] Obtain operation behavior data, and based on the operation behavior data and the target user's identifier, obtain first behavior data and second behavior data, wherein the first behavior data describes the target user's content creation behavior, and the second behavior data describes the target user's content adjustment behavior or the target user's content delivery status adjustment behavior.

[0008] Based on the analysis of the first behavioral data, a first behavioral strategy is generated;

[0009] The machine learning model is trained based on the second behavior data and the delivery effect slice data to obtain the second behavior learning model, which represents the second behavior strategy.

[0010] Content is created based on the first behavioral strategy, or content is adjusted or the content delivery status is adjusted based on the second behavioral strategy.

[0011] Secondly, this application provides a content processing apparatus, the apparatus comprising:

[0012] The first acquisition and processing module is used to acquire the delivery performance data and obtain the target user's identifier and delivery performance slice data based on the delivery performance data.

[0013] The second acquisition and processing module is used to acquire operation behavior data, and obtain first behavior data and second behavior data based on the operation behavior data and the identifier of the target user, wherein the first behavior data describes the target user's content creation behavior, and the second behavior data describes the target user's content adjustment behavior or the target user's content delivery status adjustment behavior.

[0014] The first behavior strategy generation module is used to analyze the first behavior data and generate a first behavior strategy.

[0015] The second behavior strategy generation module is used to train the machine learning model based on the second behavior data and the delivery effect slice data to obtain a second behavior learning model, wherein the second behavior learning model represents the second behavior strategy.

[0016] The content processing module is used to create content based on the first behavior strategy, or to adjust the content or the content delivery status based on the second behavior strategy.

[0017] Thirdly, this application provides a computer-readable storage medium storing at least one instruction or at least one program, which is loaded and executed by a processor to implement a content processing method as described in the first aspect.

[0018] Fourthly, this application provides a computer device including a processor and a memory, wherein the memory stores at least one instruction or at least one program, the at least one instruction or at least one program being loaded and executed by the processor to implement a content processing method as described in the first aspect.

[0019] Fifthly, the present invention provides a computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform a content processing method provided in the first aspect above.

[0020] The content processing method, apparatus, device, and storage medium provided in this application have the following technical advantages:

[0021] The solution provided in this application starts with campaign performance data, analyzes top performers, and learns their creation and adjustment behaviors. Specifically, statistical methods are used to identify behavioral patterns tending to occur in top performers' creation behavior, which are then applied to content creation to improve the overall quality of the content and achieve better content delivery. Simultaneously, in terms of content management, machine learning methods are used to build a learning model of the adjustment behaviors of top performers. Inputting the current performance of the delivered content into the trained learning model outputs corresponding adjustment actions, thereby achieving efficient content management. Applied to the advertising field, the solution provided in this application not only offers automated ad creation capabilities but also provides the ability to modify promotion plans, ads, and creatives, as well as management capabilities for ad targeting, ad activation / deactivation, and ad price adjustments, improving ad creation efficiency and ad delivery effectiveness.

[0022] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

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

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

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

[0026] Figure 3 This is a flowchart illustrating a content processing method applied to the field of advertising delivery, provided in an embodiment of this application.

[0027] Figure 4 This is a flowchart illustrating a process for obtaining target pitcher data based on delivery performance data, provided in an embodiment of this application.

[0028] Figure 5This is a schematic diagram of a process for obtaining first target pitcher action data and second target pitcher action data based on operation behavior data and target pitcher data, provided in an embodiment of this application;

[0029] Figure 6 This is a schematic diagram illustrating a process by which a pitcher adjusts their strategy based on the effectiveness of advertising, as provided in an embodiment of this application.

[0030] Figure 7 This is a schematic diagram of the structure of a deep neural network model provided in an embodiment of this application;

[0031] Figure 8 This is a flowchart illustrating another content processing method applied to the field of advertising delivery provided in this application embodiment;

[0032] Figure 9 This is a schematic diagram of a content processing device provided in an embodiment of this application.

[0033] Figure 10 This is a schematic diagram of the hardware structure of a device for implementing a content processing method provided in an embodiment of this application. Detailed Implementation

[0034] Artificial Intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to have perception, reasoning, and decision-making capabilities. AI technology is a comprehensive discipline involving a wide range of fields, encompassing both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operating / interactive systems, and mechatronics.

[0035] The solutions provided in this application involve technologies such as deep learning (DL) in artificial intelligence.

[0036] Deep learning (DL) is a major research direction in the field of machine learning (ML), bringing it closer to its original goal—artificial intelligence. Deep learning learns the inherent patterns and hierarchical representations of sample data; the information gained during this learning process greatly aids in interpreting data such as text, images, and sound. Its ultimate goal is to enable machines to possess analytical and learning capabilities like humans, capable of recognizing data such as text, images, and sound. Deep learning is a complex machine learning algorithm that has achieved results in speech and image recognition far exceeding previous related technologies. Deep learning has yielded significant achievements in search technology, data mining, machine learning, machine translation, natural language processing, multimedia learning, speech recognition, recommendation and personalization technologies, and other related fields. Deep learning enables machines to mimic human activities such as sight, hearing, and thought, solving many complex pattern recognition problems and significantly advancing artificial intelligence-related technologies.

[0037] The solutions provided in this application can be deployed in the cloud, and also involve cloud technologies.

[0038] Cloud technology refers to a hosting technology that unifies hardware, software, and network resources within a wide area network (WAN) or local area network (LAN) to achieve data computation, storage, processing, and sharing. It can also be understood as a general term for network technologies, information technologies, integration technologies, management platform technologies, and application technologies based on cloud computing business models. These technologies can form resource pools, allowing for on-demand use and flexibility. Backend services of cloud computing systems require substantial computing and storage resources, such as video websites, image websites, and many portal websites. With the rapid development and application of the internet industry, every item may have its own identification mark in the future, requiring transmission to backend systems for logical processing. Data at different levels will be processed separately, and various industry data require robust system support; therefore, cloud technology relies on cloud computing as its foundation. Cloud computing is a computing model that distributes computing tasks across a resource pool composed of numerous computers, enabling various application systems to obtain computing power, storage space, and information services as needed. The network providing these resources is called the "cloud." From the user's perspective, resources in the "cloud" are infinitely scalable, readily available, and can be used on demand, expanded at any time, and paid for based on usage. As a provider of fundamental cloud computing capabilities, a cloud resource pool platform, often referred to as a cloud platform or Infrastructure as a Service (IaaS), is established. This platform deploys various types of virtual resources within the resource pool for external customers to choose from. The cloud resource pool primarily includes: computing devices (which can be virtualized machines containing operating systems), storage devices, and network devices.

[0039] In content delivery applications, traditional methods emphasize the design of the delivery process, while automated delivery systems based on machine learning models are also being used. Taking advertising as an example, in the internet advertising field, there are generally two methods of ad delivery: one is for human advertisers to place ads; the other is for machines to place ads instead of human advertisers. Automated delivery by machines can undoubtedly improve delivery efficiency, but it also faces the challenge of ensuring delivery effectiveness.

[0040] To improve the effectiveness of content delivery, embodiments of this application provide a content processing method, apparatus, medium, and device. The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0041] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0042] To facilitate understanding of the technical solutions and their effects described in the embodiments of this application, the relevant technical terms are explained in the embodiments of this application:

[0043] Advertiser: The business entity that places advertisements.

[0044] Advertising agent: The person who places the advertisement, the specific person who operates the advertiser's account, usually serving one or more advertisers.

[0045] eCPM: Effective Cost Per Mille, refers to the advertising revenue generated per thousand impressions. An impression can be a webpage, an ad unit, or even a single ad. By default, eCPM refers to revenue per thousand pageviews. eCPM is merely a parameter used to reflect profitability and does not represent revenue itself.

[0046] oCPA: Optimized Cost per Action; it's a smart, automatic bidding strategy for performance advertising. Advertisers can select specific optimization goals (such as activation, order placement, or form booking) and provide their desired average cost per conversion. The system uses machine learning to estimate the conversion value of each impression based on the conversion data reported by the advertiser, automatically bids, and charges per click.

[0047] Please see Figure 1 This is a schematic diagram illustrating the implementation environment of a content processing method provided in this application embodiment, such as... Figure 1 As shown, the implementation environment may include at least client 01 and server 02.

[0048] Specifically, the client 01 may include devices such as smartphones, desktop computers, tablets, laptops, digital assistants, smart wearable devices, monitoring devices, and voice interaction devices. It may also include software running on the device, such as web pages provided to users by service providers, or applications provided by those service providers. Specifically, the client 01 can be used to record behavioral data and send it to the server 02.

[0049] Specifically, the server 02 can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. The server 02 may include network communication units, processors, and memory, etc. The terminal and server can be directly or indirectly connected via wired or wireless communication, which is not limited herein. Specifically, the server 02 can be used to store delivery effect data and operational behavior data, and to analyze the delivery effect data and operational behavior data to generate creation behavior strategies and adjustment behavior strategies for the creation, adjustment, and delivery status of specific content.

[0050] This application embodiment can also be implemented using cloud technology. Cloud technology refers to a hosting technology that unifies hardware, software, and network resources within a wide area network (WAN) or local area network (LAN) to achieve data computation, storage, processing, and sharing. It can also be understood as a general term for network technologies, information technologies, integration technologies, management platform technologies, and application technologies based on cloud computing business models. Cloud technology requires cloud computing as its support. Cloud computing is a computing model that distributes computing tasks across a resource pool composed of a large number of computers, enabling various application systems to obtain computing power, storage space, and information services as needed. The network providing these resources is called the "cloud." Specifically, the server 02 and the database are located in the cloud. The server 02 can be a physical machine or a virtualized machine.

[0051] The following describes a content processing method provided in this application. Figure 2 This is a flowchart illustrating a content processing method provided in an embodiment of this specification. This specification provides the operational steps of the method described in the embodiments or flowchart, but based on conventional or non-inventive labor, more or fewer operational steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order. In actual system or server product execution, the method can be executed sequentially according to the embodiments or accompanying drawings, or in parallel (e.g., in a parallel processor or multi-threaded processing environment). Please refer to... Figure 2 The content processing method provided in the embodiments of this specification may include the following steps:

[0052] S100: Obtain campaign performance data, and based on the campaign performance data, obtain the target user identifier and campaign performance slice data.

[0053] In the embodiments of this specification, the performance data refers to the performance data of historical content. The types of historical content include, but are not limited to, multimedia streams such as text, images, and videos. For example, if an advertisement is to be placed, the content is an advertisement, which can be displayed in the form of video or text and images. It is understood that the performance data is time-series data, containing information from the creation and execution of historical content to its termination, such as the content information of the historical content, the operator information of the historical content, the user information to which the historical content belongs, and the click-through rate / conversion rate / views data of the historical content. It is understood that the user refers to the operator of the historical content, and the target user refers to the operator among the users who meet the preset conditions.

[0054] For example, in the current internet advertising field, there are two main methods of ad placement: one is for advertisers to place ads, which relies on their experience and personal abilities; the other is for machines to place ads instead of advertisers. Automated ad placement, where machines replace advertisers, can improve the efficiency of ad placement, but it also faces the challenge of ensuring the effectiveness of the placement. By using accumulated ad placement performance data and advertiser behavior data for learning, the effectiveness of automated ad placement can be improved.

[0055] In one embodiment of this specification, specifically, as shown... Figure 4 As shown, obtaining the target user's identifier based on the campaign performance data may include the following steps:

[0056] S101: Based on the delivery performance data, aggregate, process, and analyze to obtain user characteristic data.

[0057] For example, refer to Figure 3 Taking advertising placement as an example, the placement performance data is aggregated and processed from the perspective of the advertiser to obtain placement performance data for each advertiser. Then, based on the placement performance data of each advertiser, multi-dimensional advertiser profile data is obtained, which is the object feature data. The advertiser profile data is used to characterize the advertiser's features, and may include, but is not limited to, account attributes, industry attributes, customer importance attributes, monetization ability attributes, creative value attributes, oCPA strategy attributes, and advertising performance attributes. Specifically, account attributes include the advertiser's identity, creation time, and the identity of their affiliated agency; industry attributes represent the industry information described by the advertiser, including specific information such as primary, secondary, and tertiary industries; clients are the advertisers served by the advertiser, and client importance attributes include the advertiser's budget and ranking, average daily spending and ranking; monetization ability attributes include the advertiser's average daily eCPM (effective cost per mile, advertising revenue earned per thousand impressions); creative value attributes include the advertiser's use of creatives and the number of times each type of creative was deployed; oCPA strategy attributes include the advertiser's use of oCPA (Optimized Cost per Action) capabilities; and advertising performance attributes include the number of clicks, impressions, spending, achievement rate, and conversion rate of the ads deployed by the advertiser.

[0058] S103: Calculate based on preset measurement indicators and user feature data, sort and filter according to the calculation results to obtain target users, and obtain the identifier of the target users.

[0059] In the embodiments of this specification, the quality of the analysis samples directly affects the quality of the final analysis results. Therefore, it is necessary to score and rank each user through user feature data, and select relatively excellent users as target users. Using target users as positive samples, behavioral strategies can be obtained by learning the operational behavior of target users, which can further improve the effectiveness of content delivery.

[0060] For example, refer to Figure 3 Taking advertising placement as an example, we identify relatively outstanding pitchers and use them as analysis samples to find good behavioral patterns. Based on pitcher profile data, we select multiple dimensions as metrics for outstanding pitchers, assigning different weights to different dimensions to calculate a score for each pitcher. These scores are then used to rank pitchers, and outstanding pitchers are selected according to preset criteria. The calculation formula is as follows:

[0061]

[0062] in,

[0063]

[0064] D i w represents the pitcher's performance index value in the i-th dimension. i Let be the weight of the i-th dimension.

[0065] In one embodiment of this specification, the campaign performance data is exemplarily divided into hourly segments to obtain hourly-level campaign performance data.

[0066] In the embodiments described in this specification, it is necessary to filter the operational behavior data to obtain the target user's behavior data. Therefore, it can be extracted from the operational behavior data based on the target user's identity identifier. In another feasible implementation, when it is necessary to further filter the target pitcher's operational behavior data in addition to extracting the target user's operational behavior data, the target user's identifier may also include other dimensions of the target user's feature data.

[0067] S200: Obtain operation behavior data, and based on the operation behavior data and the identifier of the target user, obtain first behavior data and second behavior data, wherein the first behavior data describes the target user's content creation behavior, and the second behavior data describes the target user's content adjustment behavior or the target user's content delivery status adjustment behavior.

[0068] In the embodiments of this specification, there are multiple categories of content processing behaviors. The embodiments of this specification use the first behavior and the second behavior for learning. The first behavior is the user's content creation behavior, and the second behavior is the user's content adjustment behavior or the user's content delivery status adjustment behavior.

[0069] In one embodiment of this specification, reference is made to Figure 3 Taking advertising placement as an example, all operations performed by advertisers or placement agents in the entire advertising placement system are recorded, forming raw data of operational behavior. This raw data can be categorized into many types of operational behaviors. The embodiments in this specification mainly involve two categories: ad creation behavior data and ad modification behavior data. Ad creation behavior includes the creation of promotion plans, ad groups, ad creatives, and ad targeting. Ad modification behavior includes, but is not limited to, modifying promotion plans, ad groups, ad creatives, audience targeting, adjusting ad prices, and starting / stopping ads. A promotion plan can contain one or more ad groups, an ad group can contain one or more ads, and an ad can contain one or more creatives. By classifying and learning operational behaviors, behavioral strategies for each type of behavior are obtained. This not only improves the overall level during content creation, thereby achieving better placement results, but also allows for targeted optimization and adjustments during content management, further enhancing the effectiveness of content placement.

[0070] In one feasible implementation, specifically, such as Figure 5 As shown, obtaining the first behavior data and the second behavior data based on the operation behavior data and the target user's identifier may include the following steps:

[0071] S201: Obtain target operation behavior data from the operation behavior data based on the identifier of the target user.

[0072] In one embodiment of this specification, the target user's operation behavior data is extracted from the operation behavior data based on the target user's identifier as target operation behavior data.

[0073] S203: Based on the operation object and operation content in the target operation behavior data, classify the target operation behavior data into behavior categories to obtain first behavior data and second behavior data containing behavior category labels.

[0074] For example, taking the application scenario of advertising placement as an example, the operation behavior classification aggregator can not only divide the target operation behavior data into first behavior data and second behavior data, but also further subdivide it into different behavior categories. For example, the first-level category label can be creation behavior or adjustment behavior, the second-level category label can be promotion plan creation behavior or promotion plan adjustment behavior, and the third-level category label can be ad creation behavior, ad targeting adjustment behavior, ad start and stop behavior, etc. This specification does not make specific limitations on this.

[0075] In one feasible implementation, specifically, obtaining the first behavior data and the second behavior data based on the operation behavior data and the target user's identifier may further include the following steps:

[0076] S205: Perform data structure transformation, data repair, or abnormal data filtering on the first behavior data and the second behavior data to obtain first behavior standard data and second behavior standard data.

[0077] Understandably, different analysis methods were used for the first and second behaviors, and the data structure would differ depending on the method. Therefore, the behavioral data can be transformed into a data structure that facilitates subsequent analysis. Furthermore, multiple backups of the behavioral data are maintained. When an anomaly occurs in any data point, the backup data can be used for verification, selecting the data with higher confidence levels for analysis, and using the standardized data from both the first and second behaviors as the basis for subsequent analysis.

[0078] S300: Analyze the first behavior data to generate a first behavior strategy.

[0079] In the embodiments of this specification, statistical analysis methods are used to analyze the first behavioral data of the target user to find the behavioral patterns that the target user tends to, and finally obtain the first behavioral strategy.

[0080] In one feasible implementation, specifically, the step of analyzing the first behavioral data to generate a first behavioral strategy may include the following steps:

[0081] S301: Divide the first behavior data into industry, first behavior object or target user dimensions to obtain first behavior segmentation data.

[0082] For example, taking advertising application scenarios as an example, there are significant differences in ad creation behavior between different industries. While secondary industries within the same primary industry show high similarity, some significant differences also exist. Besides segmenting by industry, we can also divide by the first action (creation behavior) object, such as promotion plans, ads, and creatives. We can further subdivide the first action object, such as images and text within creatives. Furthermore, we can also segment behavioral data based on the ranking of top advertisers. For example, we can divide the data into creation behavior data for the top 20% of top advertisers and creation behavior data for the bottom 80% of top advertisers, thus allowing us to compare the behavioral differences between top and bottom advertisers.

[0083] S303: Perform statistical analysis on the data segmented by the first behavior to generate a first behavior strategy.

[0084] For example, taking advertising application scenarios as an example, the first-behavior segmentation data obtained by segmenting based on industry dimensions can reveal differences in creation behavior between industries. Based on the data segmented according to the creation behavior object dimension, statistical analysis methods can be used to obtain the creation behavior characteristics of different creation objects within the same industry. In application, based on the advertiser's most basic needs information, such as the advertiser's industry attributes, promotion plans, ad groups, creatives, etc., can be created step by step according to the first-behavior strategy.

[0085] S400: The machine learning model is trained based on the second behavior data and the delivery effect slice data to obtain a second behavior learning model, which represents the second behavior strategy.

[0086] For example, taking advertising placement as an application scenario, the key to becoming a top-performing advertiser lies in their ability to make timely judgments and adjustments based on the advertising's performance to maximize exposure, clicks, or conversions. For instance... Figure 6 As shown, when pitchers see a decline in ad performance, they modify their promotion plans, ad placements, or creatives based on personal experience and placement rules, thus improving ad performance. This process can be summarized as: receiving an input (ad status) and generating an output (action). This is precisely the behavioral pattern that machine learning models aim to learn—what kind of output is obtained under what input. Secondary behavioral data from skilled pitchers serves as the training dataset for the model. After a period of learning, the model can acquire a large number of behavioral patterns. In application, the input is the current ad performance status, and the output is the action taken on the ad, thereby achieving intelligent ad monitoring.

[0087] In one feasible implementation, step S400 may specifically include the following steps:

[0088] S401: A training dataset is obtained based on the second behavioral data and the delivery effect slice data, wherein the second behavioral data and the delivery effect slice data are time series data.

[0089] S403: Using a supervised learning model, the machine learning model is trained using the training dataset to obtain a second behavior learning model.

[0090] Optionally, the machine learning model employs deep neural networks (DNNs), such as... Figure 7 As shown, a deep neural network consists of an input layer, multiple hidden layers, and an output layer, and uses a supervised learning model to train the model. It is understandable that, referring to... Figure 3 The resulting second behavior learning model (i.e., the modification behavior learning model) represents the second behavior strategy (i.e., the monitoring strategy). When applied, the current delivery effect of the content is input into the model to obtain the corresponding second behavior, including modification of the content or adjustment of the content delivery status, thus realizing the automated management of advertising.

[0091] S500: Create content based on the first behavior strategy, or adjust content or adjust the content delivery status based on the second behavior strategy.

[0092] In an automated delivery application scenario provided in the embodiments of this specification, step S500 may specifically include the following steps:

[0093] S501: Obtain the user's content delivery requirements, generate target content based on the content delivery requirements and the first behavior strategy, and deliver the target content.

[0094] Understandably, the first behavioral strategy can be used to automatically create content for ad placement. Based on the user's content placement needs, such as the advertiser's budget, placement metrics, and industry attributes, the first behavioral strategy automatically generates content for ad placement, realizing the creation function in intelligent automated ad placement, improving placement efficiency, and achieving the placement of higher-quality content. Figure 8 As shown, in the automated ad delivery system, ads are created based on the creation strategy, also known as the first action strategy, and are delivered and played according to the automatically created ad targeting.

[0095] S503: Obtain the delivery effect data of the target content, generate a second behavior target strategy for the target content based on the second behavior learning model and the delivery effect data of the target content, and adjust the target content or the delivery status of the target content according to the second behavior target strategy.

[0096] In the embodiments of this specification, the second behavioral strategy (second behavioral learning model) can be used to modify the target content or adjust the delivery status of the target content. The current delivery effect of the target content is input into the second behavioral learning model, and the trained second behavioral learning model will output a second behavioral target strategy for the target content. This enables the monitoring and management function in intelligent automated delivery, effectively improving content delivery performance and controlling cost deviations. Figure 8 As shown, in the automated advertising delivery system, based on the monitoring strategy, also known as the second behavior strategy (which can be represented by the second behavior learning model in the embodiments of this specification), a scheduled task is executed to monitor the advertising and update the advertising based on the queried advertising delivery effect.

[0097] In another feasible implementation, the content delivery needs or delivery content of the first user can also be obtained, and first guidance opinions can be obtained based on the first behavior strategy to guide the first user in creating content. That is, the first behavior strategy is applied to the content delivery recommendation system to provide guidance to users when creating delivery content.

[0098] In another feasible implementation, the effectiveness data of the content delivered to the second user can also be obtained. Based on the second behavioral learning model, second guidance suggestions can be obtained for the content delivered to the second user to guide the second user in adjusting the content or the content delivery status. That is, the second behavioral strategy (second behavioral learning model) is applied to the content delivery management system to provide guidance for users when managing the delivered content.

[0099] In another feasible implementation, taking the application scenario of advertising as an example, a pitcher training system can also be built based on the first behavior strategy and the second behavior strategy (second behavior learning model) to guide pitchers in reverse.

[0100] This application embodiment also provides a content processing device 900, such as... Figure 9 As shown, the device may include:

[0101] The first acquisition and processing module 910 is used to acquire delivery effect data and obtain the target user's identifier and delivery effect slice data based on the delivery effect data.

[0102] The second acquisition and processing module 920 is used to acquire operation behavior data, and obtain first behavior data and second behavior data based on the operation behavior data and the identifier of the target user, wherein the first behavior data describes the target user's content creation behavior, and the second behavior data describes the target user's content adjustment behavior or the target user's content delivery status adjustment behavior.

[0103] The first behavior strategy generation module 930 is used to analyze the first behavior data and generate a first behavior strategy.

[0104] The second behavior strategy generation module 940 is used to train the machine learning model based on the second behavior data and the delivery effect slice data to obtain a second behavior learning model, wherein the second behavior learning model represents the second behavior strategy.

[0105] The content processing module 950 is used to create content based on the first behavior strategy, or to adjust the content or the content delivery status based on the second behavior strategy.

[0106] In one embodiment of this specification, the content processing module 950 may include:

[0107] The content automatic generation unit is used to obtain the user's content delivery requirements, generate target content based on the content delivery requirements and the first behavior strategy, and deliver the target content.

[0108] The content automatic management unit is used to acquire the delivery effect data of the target content, generate a second behavior target strategy for the target content based on the second behavior learning model and the delivery effect data of the target content, and adjust the target content or the delivery status of the target content according to the second behavior target strategy.

[0109] It should be noted that the apparatus provided in the above embodiments is only illustrated by the division of the above functional modules when implementing its functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0110] This application provides a computer device including a processor and a memory. The memory stores at least one instruction or at least one program, which is loaded and executed by the processor to implement a content processing method as provided in the above method embodiments.

[0111] Figure 10 A schematic diagram of the hardware structure of a device for implementing a content processing method provided in an embodiment of this application is shown. This device may constitute or include the apparatus or system provided in the embodiment of this application. Figure 10As shown, device 10 may include one or more processors 1002 (shown as 1002a, 1002b, ..., 1002n in the figure) 1002 (processor 1002 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 1004 for storing data, and a transmission device 1006 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 10 The structure shown is for illustrative purposes only and does not limit the structure of the electronic device described above. For example, device 10 may also include a... Figure 10 The more or fewer components shown, or having the same Figure 10 The different configurations shown.

[0112] It should be noted that the aforementioned one or more processors 1002 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be wholly or partially integrated into any other element within device 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0113] The memory 1004 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the method described in the embodiments of this application. The processor 1002 executes various functional applications and data processing by running the software programs and modules stored in the memory 1004, thereby implementing the above-described content processing method. The memory 1004 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1004 may further include memory remotely located relative to the processor 1002, and these remote memories can be connected to the device 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0114] The transmission device 1006 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of device 10. In one example, the transmission device 1006 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 1006 may be a radio frequency (RF) module used for wireless communication with the Internet.

[0115] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows a user to interact with the user interface of device 10 (or a mobile device).

[0116] This application embodiment also provides a computer-readable storage medium, which can be disposed in a server to store at least one instruction or at least one program related to implementing a content processing method in the method embodiment. The at least one instruction or at least one program is loaded and executed by the processor to implement a content processing method provided in the above method embodiment.

[0117] Optionally, in this embodiment, the storage medium may be located at at least one of the multiple network servers in a computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0118] This invention also provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional embodiments described above.

[0119] As can be seen from the embodiments of the content processing method, apparatus, medium, and equipment provided in this application, the solution provided by this application starts from the data on the effectiveness of ad placement, takes excellent operators as the analysis object, and learns their creation and adjustment behaviors. Specifically, statistical methods are used to find the behavioral patterns tending to be adopted by excellent operators from their creation behaviors, and then these patterns are applied to content creation to improve the overall quality of the content and achieve better content generation and placement. Simultaneously, in terms of content management, machine learning methods are used to build a learning model of the adjustment behaviors of excellent operators. Inputting the current effect of the placed content into the trained learning model will output the corresponding adjustment operations, thereby achieving automated content management. Applied to the field of advertising placement, the solution provided by this application not only provides automated ad creation capabilities, but also provides the ability to modify promotion plans, ads, creatives, etc., and the management capabilities for ad targeting, ad start / stop, and ad price adjustments, thereby improving the efficiency of ad creation and the effectiveness of ad placement.

[0120] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0121] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0122] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0123] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A content processing method, characterized in that, The method includes: Acquire campaign performance data, and based on the campaign performance data, obtain the target user's identifier and campaign performance slice data; the target user is the content delivery party. The system acquires operation behavior data, and obtains first behavior data and second behavior data based on the operation behavior data and the identifier of the target user. The first behavior data describes the target user's content creation behavior, and the second behavior data describes the target user's content adjustment behavior or the target user's content delivery status adjustment behavior. The content indicates the multimedia resources used for delivery, and the content delivery status indicates the delivery operation for the content. Based on the analysis of the first behavioral data, a first behavioral strategy is generated; The machine learning model is trained based on the second behavior data and the delivery effect slice data to obtain the second behavior learning model, which represents the second behavior strategy. Obtain the user's content delivery requirements, generate target content based on the content delivery requirements and the first behavior strategy, and deliver the target content. Obtain the current delivery performance data of the target content; Based on the second behavior learning model, a second behavior target strategy for the target content is generated according to the current delivery effect data of the target content; Adjust the target content or the delivery status of the target content according to the second behavioral target strategy.

2. The content processing method according to claim 1, characterized in that, The step of obtaining the target user's identifier based on the delivery performance data includes: User characteristic data is obtained by aggregating, processing, and analyzing the aforementioned campaign performance data; Calculations are performed based on preset metrics and user characteristic data. Target users are obtained by sorting and filtering the calculation results, and the identifiers of the target users are acquired.

3. The content processing method according to claim 1, characterized in that, The step of obtaining the first behavior data and the second behavior data based on the operation behavior data and the target user's identifier includes: Based on the identifier of the target user, target operation behavior data is obtained from the operation behavior data; Based on the operation object and operation content in the target operation behavior data, the target operation behavior data is classified into behaviors to obtain first behavior data and second behavior data containing behavior category labels.

4. The content processing method according to claim 1, characterized in that, The step of obtaining the first behavior data and the second behavior data based on the operation behavior data and the target user's identifier further includes: The first behavioral data and the second behavioral data are transformed by data structure conversion, data repair or abnormal data filtering to obtain the first behavioral standard data and the second behavioral standard data.

5. The content processing method according to claim 1, characterized in that, The step of analyzing the first behavioral data to generate the first behavioral strategy includes: The first behavior data is divided into industry, first behavior object, or target user dimensions to obtain first behavior segmentation data; Statistical analysis is performed on the data segmented based on the first behavior to generate the first behavior strategy.

6. The content processing method according to claim 1, characterized in that, The step of training the machine learning model based on the second behavior data and the delivery effect slice data yields a second behavior learning model, which represents the second behavior strategy as follows: A training dataset is obtained based on the second behavioral data and the delivery effect slice data, wherein the second behavioral data and the delivery effect slice data are time series data. A supervised learning model is adopted, and the machine learning model is trained using the training dataset to obtain the second behavior learning model.

7. A content processing apparatus, characterized in that, The device includes: The first acquisition and processing module is used to acquire delivery performance data, and obtain the target user's identifier and delivery performance slice data based on the delivery performance data; the target user is the content delivery party. The second acquisition and processing module is used to acquire operation behavior data, and obtain first behavior data and second behavior data based on the operation behavior data and the identifier of the target user, wherein the first behavior data describes the target user's content creation behavior, and the second behavior data describes the target user's content adjustment behavior or the target user's content delivery status adjustment behavior. The first behavior strategy generation module is used to analyze the first behavior data and generate a first behavior strategy. The second behavior strategy generation module is used to train the machine learning model based on the second behavior data and the delivery effect slice data to obtain a second behavior learning model, wherein the second behavior learning model represents the second behavior strategy. The content processing module is used to obtain the user's content delivery requirements, generate target content based on the content delivery requirements and the first behavior strategy, and deliver the target content; obtain the current delivery effect data of the target content, generate a second behavior target strategy for the target content based on the second behavior learning model and the current delivery effect data of the target content, and adjust the target content or the delivery status of the target content according to the second behavior target strategy.

8. A computer-readable machine storage medium, characterized in that, The computer-readable storage medium stores at least one instruction or at least one program, which is loaded and executed by a processor to implement a content processing method as described in any one of claims 1 to 6.

9. A computer device, characterized in that, The device includes a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or at least one program being loaded by the processor and executed as a content processing method as described in any one of claims 1 to 6.

Images