Data processing method and related apparatus

Abstract

A data processing method, which can be applied to the field of artificial intelligence. The method comprises: acquiring a first prompt, which includes description information of an advertisement and attribute information of a medium, wherein the medium is a publishing platform of the advertisement, and the first prompt instructs to generate a publishing requirement of the advertisement on the medium according to the description information and the attribute information of the medium; according to the first prompt and by means of a language model, obtaining the publishing requirement, which is on the medium; and on the basis of receiving an advertisement access request of a user on the medium and according to the relationship between the publishing requirement and intent information of the user, determining whether to publish the advertisement to the user. In the present application, a language model is guided by means of a first prompt, such that an advertiser can be aided in efficiently and precisely constructing a scenario-granularity advertisement publishing task (i.e., a publishing requirement) when having no background knowledge about the media.

Description

A data processing method and related device

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on July 20, 2023, with application number 202310896986.4 and invention name “A data processing method and related device”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of artificial intelligence, and in particular to a data processing method and related devices. Background Art

[0003] Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a branch of computer science that seeks to understand the essence of intelligence and develop new intelligent machines that can respond in a manner similar to human intelligence. AI also studies the design principles and implementation methods of various intelligent machines, enabling them to possess the capabilities of perception, reasoning, and decision-making.

[0004] Machine learning systems, including personalized recommendation systems, train the parameters of machine learning models based on input data and labels using optimization methods such as gradient descent. Once the model parameters converge, the model can be used to predict unknown data. For example, in a personalized recommendation system, click-through rate prediction involves input data including user and product attributes. Predicting personalized recommendations based on user preferences plays a crucial role in improving the accuracy of the recommendation system.

[0005] With the release of large language models (LLMs) such as ChatGPT, many manufacturers have begun researching their applications in recommendation tasks. ChatGPT is a large language model that uses reinforcement learning with human feedback to align its generative capabilities with human intent. This approach enables ChatGPT to achieve comparable performance to human conversations.

[0006] In the ad delivery system, advertisers can set ad delivery tasks based on their needs within a demand-side platform (DSP). Tasks are anchored by user group tags and keywords, and advertisers select keywords that apply to their tasks. However, due to limited knowledge, advertisers may not select the most beneficial keywords that best reflect their needs, resulting in low ad delivery accuracy.

[0007] Summary of the Invention

[0008] This application provides a data processing method that can help advertisers efficiently and accurately construct scene-granular advertising delivery tasks without any media background knowledge.

[0009] In a first aspect, the present application provides a data processing method, comprising: obtaining a first prompt, the first prompt including description information of an advertisement and attribute information of a media; the media being a delivery platform for the advertisement, the first prompt indicating generation of delivery requirements for the advertisement on the media based on the description information and the attribute information of the media; obtaining delivery requirements on the media through a language model based on the first prompt; and determining whether to deliver the advertisement to the user based on a relationship between the delivery requirements and the user's intention information based on a received advertisement access request from the user on the media.

[0010] By guiding the language model through the first prompt, advertisers can efficiently and accurately build scene-specific advertising delivery tasks without any media background knowledge, and can adjust advertising delivery tasks in an interactive manner.

[0011] In one possible implementation, the delivery demand generated by the language model includes multiple slots and a description corresponding to each slot, wherein each slot corresponds to a demand; specifically, the multiple slots can be specified in the prompt, and the first prompt can be used to guide the language model to generate a description of the delivery demand corresponding to each slot, that is, the first prompt can specifically indicate to generate multiple slots and a description corresponding to each slot based on the description information and the attribute information of the media, and each slot corresponds to a delivery demand.

[0012] In one possible implementation, obtaining the delivery demand on the media through a language model according to the first prompt includes: obtaining candidate delivery demand on the media through a language model according to the first prompt; receiving modification information for the candidate delivery demand, obtaining the delivery demand on the media through a language model according to a second prompt, and the second prompt instructing to modify the candidate delivery demand according to the modification information.

[0013] Determining delivery tasks is often not a one-time process. After the language model determines delivery requirements based on the first prompt, advertisers can adjust the language model's results, and the language model can regenerate delivery requirements based on these adjustments. For example, tasks assigned by the language model can be adjusted and regenerated based on other interactive methods such as conversations.

[0014] Furthermore, media attribute information typically includes descriptive information about the media and may also include attribute information about the user who uses the media (user attribute information can indicate preferences for the media and, therefore, can be considered as part of the media attribute information). This user attribute information may be information maintained by the DMP. Typically, this is represented by tags within a limited space. However, this representation method has poor generalization and does not provide rich semantic information.

[0015] In an embodiment of the present application, the attribute information of the user is enriched by constructing a prompt (third prompt). Specifically, a third prompt can be obtained, and the third prompt includes the historical behavior of the user, and the third prompt indicates that the portrait information of the user is enriched according to the historical behavior; wherein the historical behavior can be the historical operation information of the user on the media platform (such as advertisements that have been clicked, etc.). According to the third prompt, the portrait information of the user is obtained through the language model; the portrait information of the user obtained through the language model can be used as the attribute information of the media. The portrait information of the user obtained through the language model can include the regeneration result of the historical behavior of the user, and the result can include the behavior that the user has not done in the past but is very likely to do in the future. The result can also include other descriptions with richer semantics for the historical behavior of the user.

[0016] In one possible implementation, the method further includes: obtaining a fourth prompt, the fourth prompt including context information corresponding to when the user triggers an advertising access request on the media, the fourth prompt indicating determining the user's intention information based on the context information; and obtaining the user's intention information through a language model based on the fourth prompt.

[0017] In an embodiment of the present application, a user intention that is more in line with the current scenario and user can be generated for different user requests. On the one hand, this intention can provide the relevance between the recall advertisement and the user intention. On the other hand, this intention can more accurately portray the user portrait, thereby promoting the estimated click-through rate and conversion rate, and ultimately bringing about an improvement in advertising effectiveness.

[0018] In a second aspect, the present application provides a data processing device, the device comprising:

[0019] An acquisition module is configured to acquire a first prompt, wherein the first prompt includes description information of an advertisement and attribute information of a medium; the medium is a delivery platform for the advertisement, and the first prompt indicates generating a delivery requirement for the advertisement on the medium based on the description information and the attribute information of the medium;

[0020] a processing module, configured to obtain, based on the first prompt and using a language model, the delivery demand on the media;

[0021] Based on receiving the advertisement access request of the user on the media, it is determined whether to deliver the advertisement to the user according to the relationship between the delivery demand and the user's intention information.

[0022] In one possible implementation, the delivery requirement includes multiple slots and a description corresponding to each slot, wherein each slot corresponds to a requirement; the first prompt specifically instructs to generate multiple slots and a description corresponding to each slot based on the description information and the attribute information of the media, wherein each slot corresponds to a delivery requirement.

[0023] In a possible implementation, the processing module is specifically configured to:

[0024] According to the first prompt, a candidate delivery demand on the media is obtained through a language model;

[0025] Modification information for the candidate delivery requirement is received, and according to a second prompt, the delivery requirement on the media is obtained through a language model, and the second prompt instructs to modify the candidate delivery requirement according to the modification information.

[0026] In a possible implementation, the acquisition module is further configured to:

[0027] Obtaining a third prompt, the third prompt including the user's historical behavior, the third prompt instructing to enrich the user's portrait information based on the historical behavior;

[0028] The processing module is further configured to obtain attribute information of the user according to the third prompt through a language model; wherein the attribute information of the media includes portrait information of the user.

[0029] In a possible implementation, the acquisition module is further configured to:

[0030] Obtaining a fourth prompt, the fourth prompt including context information corresponding to when the user triggers an advertisement access request on the media, the fourth prompt indicating determining the user's intent information based on the context information;

[0031] The processing module is further configured to obtain the user's intention information based on the fourth prompt through a language model.

[0032] In a third aspect, an embodiment of the present application provides a data processing device, which may include a memory, a processor, and a bus system, wherein the memory is used to store programs, and the processor is used to execute the programs in the memory to perform any optional method as described in the first aspect above.

[0033] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer executes the above-mentioned first aspect and any optional method.

[0034] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising code, which, when executed, is used to implement the above-mentioned first aspect and any optional method.

[0035] In a sixth aspect, the present application provides a chip system comprising a processor for supporting an execution device or a training device in implementing the functions described in the aforementioned aspects, such as transmitting or processing data or information involved in the aforementioned methods. In one possible design, the chip system further comprises a memory for storing program instructions and data necessary for the execution device or the training device. The chip system may be comprised solely of a chip or may include a chip and other discrete components. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a schematic diagram of the structure of the artificial intelligence main framework;

[0037] FIG2 is a schematic diagram of a system architecture provided in an embodiment of the present application;

[0038] FIG3 is a schematic diagram of a system architecture provided in an embodiment of the present application;

[0039] FIG4 is a schematic diagram of a recommendation scenario provided in an embodiment of the present application;

[0040] FIG5 is a flow chart of a data processing method provided in an embodiment of the present application;

[0041] FIG6 is a flow chart of a data processing method provided in an embodiment of the present application;

[0042] FIG7 is a flow chart of a data processing method provided in an embodiment of the present application;

[0043] FIG8 is a schematic diagram of a bidding ranking and charging method provided in an embodiment of the present application;

[0044] FIG9 is a schematic structural diagram of a data processing device provided in an embodiment of the present application;

[0045] FIG10 is a schematic diagram of an execution device provided in an embodiment of the present application;

[0046] FIG11 is a schematic diagram of a training device provided in an embodiment of the present application;

[0047] FIG12 is a schematic diagram of a chip provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present invention in conjunction with the accompanying drawings. The terms used in the embodiments of the present invention are only used to explain the specific embodiments of the present invention, and are not intended to limit the present invention.

[0049] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0050] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, and this is merely a way of distinguishing the objects of the same attributes when describing them in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0051] First, let's describe the overall workflow of an AI system. See Figure 1, which shows a schematic diagram of the AI ​​framework. This framework will be explained from two perspectives: the "intelligent information chain" (horizontal axis) and the "IT value chain" (vertical axis). The "intelligent information chain" reflects the entire process from data acquisition to processing. For example, it could be the general process of intelligent information perception, intelligent information representation and formation, intelligent reasoning, intelligent decision-making, and intelligent execution and output. Throughout this process, data undergoes a condensed journey from "data-information-knowledge-wisdom." The "IT value chain," spanning the underlying infrastructure of human intelligence, information (provided and processed by technology), and the system's industrial ecosystem, reflects the value that AI brings to the information technology industry.

[0052] (1) Infrastructure

[0053] Infrastructure provides computing power for AI systems, enabling communication with the outside world and supporting this through a foundational platform. External communication occurs through sensors; computing power is provided by intelligent chips (CPUs, NPUs, GPUs, ASICs, FPGAs, and other hardware accelerators). The foundational platform includes a distributed computing framework and network-related platform guarantees and support, including cloud storage and computing, and interconnected networks. For example, sensors communicate with the outside world to acquire data, which is then fed into the intelligent chips within the distributed computing system provided by the foundational platform for computation.

[0054] (2) Data

[0055] Data above the infrastructure layer represents data sources for AI. This data includes graphics, images, voice, and text, as well as IoT data from traditional devices. This includes business data from existing systems and sensor data such as force, displacement, liquid level, temperature, and humidity.

[0056] (3) Data processing

[0057] Data processing generally includes data training, machine learning, deep learning, search, reasoning, decision-making, etc.

[0058] Among them, machine learning and deep learning can symbolize and formalize data for intelligent information modeling, extraction, preprocessing, and training.

[0059] Reasoning refers to the process of simulating human intelligent reasoning in computers or intelligent systems, using formalized information to perform machine thinking and solve problems based on reasoning control strategies. Typical functions are search and matching.

[0060] Decision-making refers to the process of making decisions after intelligent information is reasoned, and usually provides functions such as classification, sorting, and prediction.

[0061] (4) General ability

[0062] After the data has undergone the data processing mentioned above, some general capabilities can be further formed based on the results of the data processing, such as algorithms or a general system, for example, translation, text analysis, computer vision processing, speech recognition, image recognition, etc.

[0063] (5) Smart products and industry applications

[0064] Smart products and industry applications refer to the products and applications of artificial intelligence systems in various fields. They are the encapsulation of the overall artificial intelligence solution, which productizes intelligent information decision-making and realizes practical application. Its application areas mainly include: smart terminals, smart transportation, smart medical care, autonomous driving, smart cities, etc.

[0065] The embodiments of the present application can be applied to the field of information recommendation, which includes but is not limited to scenarios involving e-commerce product recommendations, search engine result recommendations, application market recommendations, music recommendations, video recommendations, etc. The recommended items in various application scenarios can also be called "objects" to facilitate subsequent descriptions, that is, in different recommendation scenarios, the recommended object can be an APP, or a video, or music, or a certain product (such as the presentation interface of an online shopping platform, which will display different products for presentation according to different users, which can actually be presented through the recommendation results of a recommendation model). These recommendation scenarios usually involve user behavior log collection, log data preprocessing (for example, quantization, sampling, etc.), sample set training to obtain a recommendation model, and analysis and processing of the objects involved in the scenario corresponding to the training sample items (such as APP, music, etc.) according to the recommendation model. For example, the samples selected in the recommendation model training link come from the user's operation behavior on the recommended APP in the mobile application market. The recommendation model trained thereby is applicable to the above-mentioned mobile APP application market, or can be used for the APP application market of other types of terminals to recommend terminal APPs. The recommendation model will ultimately calculate the recommendation probability or score of each object to be recommended. The recommendation system selects the recommendation results based on certain selection rules, such as sorting them by recommendation probability or score, and presents them to users through corresponding applications or terminal devices. Users operate on the objects in the recommendation results to generate user behavior logs, etc.

[0066] Referring to Figure 4, during the recommendation process, when a user interacts with the recommendation system, a recommendation request is triggered. The recommendation system inputs this request and its associated feature information into the deployed recommendation model, which then predicts the user's click-through rate for all candidate items. The candidate items are then sorted in descending order based on the predicted click-through rates and displayed in sequential order at different locations as the recommended results for the user. Users browse the displayed items and perform user actions, such as browsing, clicking, and downloading. These user actions are logged as training data, and the offline training module periodically updates the recommendation model parameters to improve the model's recommendation effectiveness.

[0067] For example, when a user opens a mobile app market, the app market's recommendation module is triggered. This module predicts the user's likelihood of downloading each candidate app based on historical download and click history, app features, and contextual information such as time and location. Based on these predictions, the app market displays apps in descending order of likelihood, increasing the probability of download. Specifically, apps more likely to be downloaded are ranked higher, while apps less likely to be downloaded are ranked lower. User behavior is also logged, and the prediction model parameters are trained and updated through an offline training module.

[0068] For example, in applications related to lifelong companions, based on the user's historical data in video, music, news and other domains, various models and algorithms can be used to imitate the human brain mechanism to build a cognitive brain and establish a user lifelong learning system framework. Lifelong companions can record past events of users based on system data and application data, understand users' current intentions, predict users' future actions or behaviors, and ultimately realize intelligent services. In the current first phase, based on music apps, video apps, browser apps, etc., user behavior data (including end-side text messages, photos, email events, etc.) is obtained. On the one hand, a user portrait system is built, and on the other hand, learning and memory modules based on user information filtering, association analysis, cross-domain recommendations, causal reasoning, etc. are implemented to build a user's personal knowledge graph.

[0069] Next, the application architecture of the embodiment of the present application is introduced.

[0070] Referring to FIG2 , an embodiment of the present invention provides a recommendation system architecture 200. The data acquisition device 260 is used to collect samples. A training sample can be composed of multiple feature information (or described as attribute information, such as user attributes and item attributes). Feature information can be of multiple types, specifically including user feature information, object feature information, and label features. User feature information is used to characterize user characteristics, such as gender, age, occupation, hobbies, etc. Object feature information is used to characterize the characteristics of the object pushed to the user. Different recommendation systems correspond to different objects, and the types of features that need to be extracted from different objects are also different. For example, the object features extracted from the training samples of the APP market can be the name (identification), type, size, etc. of the APP. ; The object features mentioned in the training samples of e-commerce apps can be the name of the product, the category it belongs to, the price range, etc.; the label feature is used to indicate whether the sample is a positive example or a negative example. Usually, the label feature of the sample can be obtained through the user's operation information on the recommended object. The sample in which the user operates the recommended object is a positive example, and the sample in which the user does not operate the recommended object or only browses the recommended object is a negative example. For example, when the user clicks, downloads, or purchases the recommended object, the label feature is 1, indicating that the sample is a positive example. If the user does not perform any operation on the recommended object, the label feature is 0, indicating that the sample is a negative example. After collection, the sample can be stored in the database 230. Some or all of the feature information in the sample in the database 230 can also be directly obtained from the client device 240, such as user feature information, user operation information on the object (used to determine the type identification), object feature information (such as object identification), etc. The training device 220 obtains a model parameter matrix based on the sample training in the database 230 for generating a recommendation model 201 (such as the feature extraction network and neural network in the embodiment of the present application). The following will describe in more detail how the training device 220 trains to obtain the model parameter matrix used to generate the recommendation model 201. The recommendation model 201 can be used to evaluate a large number of objects to obtain the scores of each object to be recommended. Furthermore, a specified or preset number of objects can be recommended from the evaluation results of a large number of objects. The calculation module 211 obtains the recommendation results based on the evaluation results of the recommendation model 201 and recommends them to the client device through the I / O interface 212.

[0071] In an embodiment of the present application, the training device 220 can select positive and negative samples from the sample set in the database 230 and add them to the training set, and then use the recommendation model to train the samples in the training set to obtain a trained recommendation model; the implementation details of the calculation module 211 can refer to the detailed description of the method embodiment shown in Figure 5.

[0072] After the training device 220 obtains the model parameter matrix based on sample training and uses it to construct the recommendation model 201, the recommendation model 201 is sent to the execution device 210, or the model parameter matrix is ​​directly sent to the execution device 210, and a recommendation model is constructed in the execution device 210 for making recommendations for the corresponding system. For example, the recommendation model obtained based on video-related sample training can be used to recommend videos to users on video websites or APPs, and the recommendation model obtained based on APP-related sample training can be used to recommend APPs to users in the application market.

[0073] Execution device 210 is equipped with an I / O interface 212 for data exchange with external devices. Execution device 210 can obtain user characteristic information, such as user ID, user identity, gender, occupation, and hobbies, from client device 240 via I / O interface 212. This information can also be obtained from a system database. Recommendation model 201 recommends target objects to the user based on the user characteristic information and the characteristic information of the recommended objects. Execution device 210 can be located in a cloud server or in a user client.

[0074] The execution device 210 can call data, code, etc. in the data storage system 250, and can also store output data in the data storage system 250. The data storage system 250 can be set in the execution device 210, can be set independently, or can be set in other network entities, and can be one or more.

[0075] The calculation module 211 uses the recommendation model 201 to process the user characteristic information and the characteristic information of the object to be recommended. For example, the calculation module 211 uses the recommendation model 201 to analyze and process the user characteristic information and the characteristic information of the object to be recommended, so as to obtain the score of the object to be recommended, and sort the objects to be recommended according to the score, among which the objects with higher rankings will be recommended to the client device 240.

[0076] Finally, the I / O interface 212 returns the recommendation result to the client device 240 and presents it to the user.

[0077] More deeply, the training device 220 can generate corresponding recommendation models 201 based on different sample feature information for different goals to provide users with better results.

[0078] It is worth noting that Figure 2 is only a schematic diagram of a system architecture provided by an embodiment of the present invention. The positional relationship between the devices, components, modules, etc. shown in the figure does not constitute any limitation. For example, in Figure 2, the data storage system 250 is an external memory relative to the execution device 210. In other cases, the data storage system 250 can also be placed in the execution device 210.

[0079] In an embodiment of the present application, the training device 220, the execution device 210, and the client device 240 can be three different physical devices respectively. It is also possible that the training device 220 and the execution device 210 are on the same physical device or a cluster. It is also possible that the execution device 210 and the client device 240 are on the same physical device or a cluster.

[0080] Refer to Figure 3, which is a system architecture 300 proposed in an embodiment of the present invention. In this architecture, the execution device 210 is implemented by one or more servers, and optionally cooperates with other computing devices, such as data storage, routers, load balancers and other devices; the execution device 210 can be arranged at one physical site, or distributed at multiple physical sites. The execution device 210 can use the data in the data storage system 250, or call the program code in the data storage system 250 to implement the object recommendation function. Specifically, the information of the object to be recommended is input into the recommendation model, and the recommendation model generates an estimated score for each object to be recommended, and then sorts them in order from high to low according to the estimated score, and recommends the object to be recommended to the user according to the sorting result. For example, the first 10 objects in the sorting result are recommended to the user.

[0081] Among them, the data storage system 250 is used to receive and store the parameters of the recommendation model sent by the training device, as well as the data for storing the recommendation results obtained by the recommendation model. Of course, it may also include the program code (or instructions) required for the normal operation of the storage system 250. The data storage system 250 can be a distributed storage cluster composed of one device or multiple devices deployed outside the execution device 210. In this case, when the execution device 210 needs to use the data on the storage system 250, the storage system 250 can send the data required by the execution device to the execution device 210. Accordingly, the execution device 210 receives and stores (or caches) the data. Of course, the data storage system 250 can also be deployed in the execution device 210. When deployed in the execution device 210, the distributed storage system can include one or more memories. Optionally, when there are multiple memories, different memories are used to store different types of data. For example, the model parameters of the recommendation model generated by the training device and the data of the recommendation results obtained by the recommendation model can be stored in two different memories respectively.

[0082] Users can operate their respective user devices (e.g., local device 301 and local device 302) to interact with execution device 210. Each local device can represent any computing device, such as a personal computer, a computer workstation, a smartphone, a tablet computer, a smart camera, a smart car or other type of cellular phone, a media consumption device, a wearable device, a set-top box, a game console, etc.

[0083] Each user's local device can interact with the execution device 210 through a communication network of any communication mechanism / communication standard. The communication network can be a wide area network, a local area network, a point-to-point connection, etc., or any combination thereof.

[0084] In another implementation, the execution device 210 can be implemented by a local device. For example, the local device 301 can implement the recommendation function of the execution device 210 based on the recommendation model to obtain user feature information and feedback the recommendation results to the user, or provide services to the user of the local device 302.

[0085] Since the embodiments of the present application involve the application of a large number of neural networks, in order to facilitate understanding, the relevant terms and related concepts such as neural networks involved in the embodiments of the present application are first introduced below.

[0086] 1. Click-throughrate (CTR)

[0087] Click probability, also known as click-through rate, refers to the ratio of the number of clicks and exposures of recommended information (for example, recommended items) on a website or application. Click-through rate is usually an important indicator for measuring recommendation systems in recommendation systems.

[0088] 2. Personalized recommendation system

[0089] A personalized recommendation system refers to a system that uses machine learning algorithms to analyze users' historical data (such as the operation information in the embodiments of the present application), predicts new requests based on this data, and provides personalized recommendation results.

[0090] 3. Offline training

[0091] Offline training refers to a module in a personalized recommendation system that iteratively updates the recommendation model parameters according to the machine learning algorithm based on the user's historical data (such as the operation information in the embodiment of the present application) until the set requirements are met.

[0092] 4. Online Inference

[0093] Online prediction refers to predicting the user's preference for recommended items in the current context based on the characteristics of the user, item, and context based on the offline trained model, and predicting the probability of the user selecting the recommended item.

[0094] For example, FIG3 is a schematic diagram of a recommendation system provided by an embodiment of the present application. As shown in FIG3, when a user enters the system, a recommendation request is triggered. The recommendation system inputs the request and related information (such as operation information in the embodiment of the present application) into the recommendation model, and then predicts the user's selection rate of items in the system. Furthermore, the items are sorted in descending order according to the predicted selection rate or a function based on the selection rate, that is, the recommendation system can display the items in different positions in sequence as recommendation results for the user. The user browses items in different positions and performs user behaviors, such as browsing, selecting, and downloading. At the same time, the user's actual behavior will be stored in the log as training data, and the parameters of the recommendation model will be continuously updated through the offline training module to improve the prediction effect of the model.

[0095] For example, a user opening an app store on a smart device (e.g., a mobile phone) triggers the app store's recommendation system. The app store's recommendation system predicts the probability of the user downloading each recommended app based on the user's historical behavior logs, such as their download history and selection history, as well as the app store's own characteristics, such as time, location, and other environmental characteristics. Based on the calculated results, the app store's recommendation system can display the candidate apps in descending order of predicted probability, thereby increasing the download probability of the candidate apps.

[0096] For example, an APP with a higher predicted user selection rate may be displayed in a front recommendation position, and an APP with a lower predicted user selection rate may be displayed in a back recommendation position.

[0097] The above-mentioned recommendation model can be a neural network model. The following introduces the relevant terms and concepts of neural networks that may be involved in the embodiments of the present application.

[0098] (1) Neural Network

[0099] A neural network can be composed of neural units. A neural unit can refer to an operation unit that takes xs (i.e., input data) and intercept 1 as input. The output of the operation unit can be:

[0100] Where s = 1, 2, ... n, n is a natural number greater than 1, Ws is the weight of xs, and b is the bias of the neural unit. f is the activation function of the neural unit, which is used to introduce nonlinear characteristics into the neural network to convert the input signal of the neural unit into the output signal. The output signal of the activation function can be used as the input of the next convolutional layer, and the activation function can be a sigmoid function. A neural network is a network formed by connecting multiple single neural units mentioned above, that is, the output of one neural unit can be the input of another neural unit. The input of each neural unit can be connected to the local receptive field of the previous layer to extract the features of the local receptive field. The local receptive field can be an area composed of several neural units.

[0101] (2) Deep Neural Networks

[0102] Deep Neural Network (DNN), also known as multi-layer neural network, can be understood as a neural network with many hidden layers. There is no special metric for "many" here. Based on the position of different layers in DNN, the neural network inside DNN can be divided into three categories: input layer, hidden layer, and output layer. Generally speaking, the first layer is the input layer, the last layer is the output layer, and the layers in between are all hidden layers. The layers are fully connected, that is, any neuron in the i-th layer must be connected to any neuron in the i+1-th layer. Although DNN looks complicated, the work of each layer is actually not complicated. Simply put, it is the following linear relationship expression: in, is the input vector, is the output vector, is the offset vector, W is the weight matrix (also called coefficient), and α() is the activation function. Each layer is just an input vector After such a simple operation, the output vector Since there are many DNN layers, the coefficient W and the offset vector The definition of these parameters in DNN is as follows: Take the coefficient W as an example: Assume that in a three-layer DNN, the linear coefficient from the 4th neuron in the second layer to the 2nd neuron in the third layer is defined as The superscript 3 represents the layer number of the coefficient W, while the subscript corresponds to the output of the third layer index 2 and the input of the second layer index 4. In summary, the coefficient from the kth neuron in the L-1th layer to the jth neuron in the Lth layer is defined as It's important to note that the input layer has no W parameter. In deep neural networks, more hidden layers allow the network to better capture complex real-world situations. Theoretically, a model with more parameters has higher complexity and greater "capacity," meaning it can handle more complex learning tasks. Training a deep neural network is essentially the process of learning the weight matrix, with the ultimate goal of obtaining the weight matrices for all layers of a trained deep neural network (a weight matrix formed by the vectors W across many layers).

[0103] (3) Loss function

[0104] During the training of a deep neural network, because we want the output of the deep neural network to be as close as possible to the desired predicted value, we can compare the current network's predicted value with the desired target value and then update the weight vector of each layer of the neural network based on the difference between the two. (Of course, before the first update, there is usually an initialization process, which pre-configures the parameters for each layer in the deep neural network.) For example, if the network's predicted value is too high, the weight vector is adjusted to make it predict a lower value. This adjustment is continued until the deep neural network can predict the desired target value or a value very close to the desired target value. Therefore, it is necessary to predefine "how to compare the difference between the predicted value and the target value." This is the loss function (or objective function), which is an important equation used to measure the difference between the predicted value and the target value. For example, the loss function output value (loss) indicates a greater difference, so training a deep neural network becomes a process of minimizing this loss.

[0105] (4) Backpropagation algorithm

[0106] The back propagation (BP) algorithm can be used to correct the size of the initial model parameters during training, reducing the model's error loss. Specifically, forward propagation of the input signal to the output generates error loss. This error loss information is then backpropagated to update the parameters in the initial model, thereby converging the error loss. The BP algorithm is a backward propagation movement driven by error loss, aiming to obtain optimal model parameters, such as the weight matrix.

[0107] (5) Machine Learning Systems

[0108] Based on the input data and labels, the parameters of the machine learning model are trained through optimization methods such as gradient descent, and finally the trained model is used to complete the prediction of unknown data.

[0109] (6) Personalized recommendation system

[0110] A system that uses machine learning algorithms to analyze and model users' historical data, predict new user requests, and provide personalized recommendation results.

[0111] (7) Pre-trained language model: A pre-trained language model is a model that performs unsupervised learning on a large-scale corpus to obtain word or subword level representations, and then further trains specific natural language processing tasks on this basis. Pre-trained language models are usually modeled using neural networks, which can input a piece of text and output the probability distribution of the next word or subword in the text. Pre-trained language models have been widely used in the field of natural language processing, including machine translation, speech recognition, text classification, information retrieval and other tasks, and have achieved significant results.

[0112] (8) Prompt

[0113] A natural language term that can be categorized into hard templates and soft templates. Hard templates are generally natural language words or sentences with a specific meaning; soft templates are generally parameterized representation vectors without any meaning.

[0114] Machine learning systems, including personalized recommendation systems, train the parameters of machine learning models based on input data and labels using optimization methods such as gradient descent. Once the model parameters converge, the model can be used to predict unknown data. For example, in a personalized recommendation system, click-through rate prediction involves input data including user and product attributes. Predicting personalized recommendations based on user preferences plays a crucial role in improving the accuracy of the recommendation system.

[0115] With the release of large prediction models like ChatGPT, many companies are exploring the application of large language models for recommendation tasks. ChatGPT is a large language model that uses reinforcement learning with human feedback to align its generative capabilities with human intent. This approach enables ChatGPT to achieve comparable performance to human conversations.

[0116] In the ad delivery system, advertisers can set ad delivery tasks based on their needs within a demand-side platform (DSP). Tasks are anchored by user group tags and keywords, and advertisers select keywords that apply to their tasks. However, due to limited knowledge, advertisers may not select the most beneficial keywords that best reflect their needs, resulting in low ad delivery accuracy.

[0117] In order to solve the above problems, the present application provides a data processing method, which can be an inference process of a model.

[0118] 5 , which is a schematic diagram of an embodiment of a data processing method provided in an embodiment of the present application, as shown in FIG5 , a data processing method provided in an embodiment of the present application includes:

[0119] 501. Obtain a first prompt, which includes description information of the advertisement and attribute information of the media; the media is a delivery platform for the advertisement, and the first prompt indicates generating delivery requirements for the advertisement on the media based on the description information and the attribute information of the media.

[0120] In existing advertising systems, advertising delivery tasks are formulated with group labels and keywords as anchors, and advertisers select keywords that are suitable for them to formulate tasks. This keyword-based approach has disadvantages: (1) it is difficult to dig out deep semantic information; (2) it cannot solve the problem of user behavior differences under different advertising media ends. In addition, advertisers select keywords from their own perspective to establish tasks, and due to the limitations of knowledge, they may not be able to select the keywords that are most beneficial to themselves and best reflect their needs. Keyword extraction based on traditional technology is limited by the model's knowledge reasoning ability and discriminant model ability, and cannot determine the advertiser's true intention. Moreover, the advertisements suitable for the same user in different scenarios and contexts are different, and how to combine the current context to generate advertisements suitable for recall with high relevance is also very critical.

[0121] As a traffic portal, media outlets will develop ad spaces for advertisers to generate advertising revenue. Media outlets can develop ad spaces through ADN (advertising network or ad alliance), ADX (advertising exchange), or by directly connecting with advertisers. ADN is a platform where advertising alliances sell media ad spaces to advertisers. Advertising alliances have pricing power, which is disadvantageous to advertisers and media holders. ADX allows advertisers to obtain ad spaces through bidding based on the value of the ad space to themselves, maximizing the interests of both parties. Advertisers can set the targeting criteria, budget, bid, creative, etc. of their target audiences on the DSP platform based on their own needs. DSPs automatically optimize the delivery results and provide data reports through technology and algorithms. The DMP platform is used for data management, unifying scattered data and labeling users as groups to provide to DSPs to help advertisers assign tasks.

[0122] The logic of the advertising delivery system is as follows: the user enters the media, triggering a media advertising space request. The request enters the ADX and other systems. The system completes the ad recall based on the scene context information and user information carried in the request, and then performs precise ranking calculations based on the detailed information and features of the recalled ads, user portraits, and contextual information. The winning ads are then determined through a bidding mechanism, and then displayed to the user. After the user browses, clicks, downloads, and other operations on the ads, the advertising system chooses whether to deduct fees from the advertiser's budget according to the requirements of the advertising task.

[0123] Before bidding on ads or placing ads on media platforms, advertisers must first specify their advertising tasks, such as the targeting criteria, budget, bid, and creative, as described above. These tasks are, to a certain extent, dependent on the media data and user behavior data managed by the DMP data management platform.

[0124] In order to better select and generate media context information, help advertisers to build tasks more efficiently and accurately, and more accurately understand the user's true intentions in the current context, the embodiment of the present application proposes an advertising delivery mechanism based on prompt generalization.

[0125] In one possible implementation, information about the advertising task set by the advertiser (e.g., including ad description information) can be obtained. This ad description information is based on the advertiser's subjective description and often cannot accurately express the advertiser's true intent. In this embodiment of the present application, a first prompt can be constructed to guide the language model to determine the advertiser's intent (i.e., ad placement requirements in this embodiment of the present application) when placing an ad on the media based on the ad description information and attribute information of the media to be placed.

[0126] Among them, the media can be advertising platforms such as APP (such as short video platforms) and websites.

[0127] For example, the description information of the advertisement may include a user group label selected from candidate labels, or an advertiser inputs a paragraph of text to describe their own needs, and the DSP platform may perform operations such as word segmentation on the description information to obtain the user group label.

[0128] In an embodiment of the present application, the first prompt is used to guide the language model to combine the original description information and the attribute information of the media to regenerate the description information of the advertisement. The generated advertisement delivery requirements can have richer semantics than the original advertisement description information, can express the advertiser's intention more accurately, and are more suitable for the scenario of each media.

[0129] For example, the first prompt can be used to guide the language model to regenerate the advertiser's input description or selected group tags. Because the first prompt carries the media attribute information, the language model can combine different scene information to select appropriate group tags or keyword tags for the advertiser in different media.

[0130] It should be understood that although the embodiments of the present application describe the generated before and after information as two different dimensions of "description information of the advertisement" and "demand for advertisement delivery", in fact the two can be different information of the same dimension, for example, "demand for advertisement delivery" is a new result of regenerating "description information of the advertisement".

[0131] In this embodiment of the present application, more generalized delivery requirements can be generated based on the advertiser's specified ad description information (such as keywords or provided natural language descriptions), thereby formulating more comprehensive ad delivery tasks. Furthermore, user-provided keywords or natural language descriptions are context-independent, but different media scenarios require specific tasks that better suit the scenario and user needs. This embodiment of the present application, based on prompts and language models, provides a scenario-adaptive task construction capability.

[0132] In one possible implementation, the delivery demand generated by the language model includes multiple slots and a description corresponding to each slot, wherein each slot corresponds to a demand; specifically, the multiple slots can be specified in the prompt, and the first prompt can be used to guide the language model to generate a description of the delivery demand corresponding to each slot, that is, the first prompt can specifically indicate to generate multiple slots and a description corresponding to each slot based on the description information and the attribute information of the media, and each slot corresponds to a delivery demand.

[0133] For example, referring to FIG7 , FIG7 is a schematic diagram of setting a delivery task. An advertiser may input the description information of the advertisement in the manner shown in FIG7 . In addition, optionally, the advertiser may be informed that the system will enhance the delivery demand of the advertisement.

[0134] Furthermore, media attribute information typically includes descriptive information about the media and may also include attribute information about the user who uses the media (user attribute information can indicate preferences for the media and, therefore, can be considered as part of the media attribute information). This user attribute information may be information maintained by the DMP. Typically, this is represented by tags within a limited space. However, this representation method has poor generalization and does not provide rich semantic information.

[0135] In an embodiment of the present application, the attribute information of the user is enriched by constructing a prompt (third prompt). Specifically, a third prompt can be obtained, and the third prompt includes the historical behavior of the user, and the third prompt indicates that the portrait information of the user is enriched according to the historical behavior; wherein the historical behavior can be the historical operation information of the user on the media platform (such as advertisements that have been clicked, etc.). According to the third prompt, the portrait information of the user is obtained through the language model; the portrait information of the user obtained through the language model can be used as the attribute information of the media. The portrait information of the user obtained through the language model can include the regeneration result of the historical behavior of the user, and the result can include the behavior that the user has not done in the past but is very likely to do in the future. The result can also include other descriptions with richer semantics for the historical behavior of the user.

[0136] The DMP data management platform stores data such as user portraits, features, and scenario information. On the one hand, these data will be used by the DSP demand platform to formulate advertising plans and tasks. On the other hand, when a user request arrives and the ADX conducts a transaction, the user-related data and features in the DSP will be used to predict the potential value of the advertisement.

[0137] As shown in Figure 6, based on LLM combined with Prompt's media data and user portrait generation module, the original media data and user portrait can be used as input, and by constructing a reasonable prompt strategy and utilizing the large language model capabilities to generate more generalized media descriptions and user portrait data with richer semantic information.

[0138] For example, advertisers need to set information such as "promoted product details," "targeting," and "keywords." For example, for keywords, advertisers need to match keywords based on content from search and user browsing behaviors to increase exposure. In this embodiment of the application, the content to be selected by the user is not limited to the original samples from original user behaviors such as search and browsing, but also includes keywords in content information generated based on LLM. Although such words may not appear in the current behavior, they may appear in the future and have a strong correlation with the current behavior.

[0139] It should be understood that the specific form and format of the prompt is not limited in the embodiments of the present application. For example, it can be natural language or a representation obtained through embedding processing.

[0140] 502. According to the first prompt, obtain the delivery demand on the media through a language model.

[0141] Determining delivery tasks is often not a one-time process. After the language model determines delivery requirements based on the first prompt, advertisers can adjust the language model's results, and the language model can regenerate delivery requirements based on these adjustments. For example, tasks assigned by the language model can be adjusted and regenerated based on other interactive methods such as conversations.

[0142] In one possible implementation, candidate delivery requirements for the media can be obtained using a language model based on the first prompt. Upon receiving modification information for the candidate delivery requirements, the delivery requirements for the media can be obtained using the language model based on a second prompt, with the second prompt instructing the candidate delivery requirements to be modified based on the modification information. In other words, the language model can be guided by the prompt to regenerate the delivery requirements based on the user-specified modification information.

[0143] 503. Based on the received advertisement access request from the user on the media, determine whether to deliver the advertisement to the user according to the relationship between the delivery demand and the user's intention information.

[0144] For example, the advertising delivery decision may be determined based on the matching relationship between the delivery demand and the user's intention information (as well as information such as advertising bidding and charging).

[0145] In one possible implementation, the result of advertising delivery can be determined by AND or ADX. Generally, the result of advertising delivery needs to be determined based on advertising bidding and charging, and advertising bidding and charging are related to the estimated click-through rate or conversion rate (for example, as shown in Figure 8). The accuracy of the estimated conversion rate is related to the recalled advertisements on the one hand, and the user on the other hand (for example, related to the user's intent information). Therefore, improving the accuracy of the estimation of the user's intent information can correspondingly improve the click-through rate and conversion rate of the advertisement.

[0146] In an embodiment of the present application, a user intention that is more in line with the current scenario and user can be generated for different user requests. On the one hand, this intention can provide the relevance between the recall advertisement and the user intention. On the other hand, this intention can more accurately portray the user portrait, thereby promoting the estimated click-through rate and conversion rate, and ultimately bringing about an improvement in advertising effectiveness.

[0147] Specifically, a prompt can be used to guide the language model to generate user intent that is more consistent with the current scenario and the user. In one implementation, a fourth prompt can be obtained, the fourth prompt including the context information corresponding to when the user triggers the advertising access request on the media (for example, time context, location context, specifically, the current market environment, public opinion environment, and hot spots, etc.), the fourth prompt indicating that the user's intent information is determined based on the context information; based on the fourth prompt, the user's intent information is obtained through the language model.

[0148] Since users may have different needs on different media ends under different contextual conditions, the advertising platform needs to generate more accurate and reasonable user tags (that is, the user intention in the embodiment of this application) for user requests based on the current market environment, public opinion environment and hot spots. More accurate and reasonable user tags are more conducive to recalling advertisements with high user click-through rates.

[0149] When the media end receives the user's traffic, it will trigger the corresponding advertising request. Different users in different scenarios and contexts have different intentions. The embodiment of the present application can more accurately generate real-time scene-aware intentions for users.

[0150] The existing technology uses traditional data mining technology based on raw data to obtain user tags and other information. The embodiment of the present application generalizes and deeply semantically mines the raw data based on LLM, has stronger migration capabilities, and can alleviate the problem of data sparsity; the task construction of the existing technology is generally that advertisers independently select keywords and other settings to construct tasks. Advertisers cannot understand the differences between all scenarios and it is difficult to accurately describe their own needs with DSP keywords. LLM can solve this pain point, and LLM can also achieve interactive task construction; current requests generally enter DSP with user ID and scenario information for a series of operations such as advertising recall and bidding. The user ID can manage related behaviors for modeling, but does not have a deeper understanding of the current user's intentions. LLM can solve this pain point.

[0151] On the one hand, the third prompt guides the language model to regenerate a more generalized and semantically rich user portrait, reducing data sparsity.

[0152] On the other hand, the fourth prompt guides the language model to generate scenario-adaptive user intent for different user requests, accurately describing the current request intent, improving the recall efficiency and relevance of ads, and ultimately enhancing advertising effectiveness.

[0153] On the other hand, by guiding the language model through the first prompt, advertisers can efficiently and accurately build scene-granular advertising tasks without any media background knowledge, and can adjust advertising tasks in an interactive manner.

[0154] Next, a data processing device provided by an embodiment of the present application will be described from the perspective of a device. Referring to FIG. 9 , FIG. 9 is a schematic diagram of the structure of a data processing device provided by an embodiment of the present application. As shown in FIG. 9 , a data processing device 900 provided by an embodiment of the present application includes:

[0155] Acquisition module 901 is configured to acquire a first prompt, the first prompt including description information of an advertisement and attribute information of a medium; the medium being a delivery platform for the advertisement, the first prompt indicating generation of a delivery requirement for the advertisement on the medium based on the description information and attribute information of the medium;

[0156] For a detailed introduction to the acquisition module 901 , reference may be made to the description of step 501 in the above embodiment, which will not be repeated here.

[0157] The processing module 902 is configured to obtain the media placement demand based on the first prompt using a language model;

[0158] Based on receiving the advertisement access request of the user on the media, it is determined whether to deliver the advertisement to the user according to the relationship between the delivery demand and the user's intention information.

[0159] The detailed introduction of the processing module 902 can refer to the description of step 502 to step 503 in the above embodiment, which will not be repeated here.

[0160] In one possible implementation, the delivery requirement includes multiple slots and a description corresponding to each slot, wherein each slot corresponds to a requirement; the first prompt specifically instructs to generate multiple slots and a description corresponding to each slot based on the description information and the attribute information of the media, wherein each slot corresponds to a delivery requirement.

[0161] In a possible implementation, the processing module 902 is specifically configured to:

[0162] According to the first prompt, obtaining candidate delivery requirements on the media through a language model;

[0163] Modification information for the candidate delivery requirement is received, and according to a second prompt, the delivery requirement on the media is obtained through a language model, and the second prompt instructs to modify the candidate delivery requirement according to the modification information.

[0164] In a possible implementation, the obtaining module 901 is further configured to:

[0165] Obtaining a third prompt, the third prompt including the user's historical behavior, the third prompt instructing to enrich the user's portrait information based on the historical behavior;

[0166] The processing module 902 is further configured to obtain attribute information of the user based on the third prompt through a language model; wherein the attribute information of the media includes portrait information of the user.

[0167] In a possible implementation, the obtaining module 901 is further configured to:

[0168] Obtaining a fourth prompt, the fourth prompt including context information corresponding to when the user triggers an advertisement access request on the media, the fourth prompt indicating determining the user's intent information based on the context information;

[0169] The processing module 902 is further configured to obtain the user's intention information based on the fourth prompt through a language model.

[0170] Next, a terminal device provided in an embodiment of the present application is introduced. Please refer to Figure 10. Figure 10 is a structural schematic diagram of a terminal device provided in an embodiment of the present application. The terminal device 1000 can be specifically manifested as a mobile phone, a tablet, a laptop computer, a smart wearable device, etc., which is not limited here. Among them, the terminal device 1000 implements the function of the data processing method in the embodiment corresponding to Figure 5. Specifically, the terminal device 1000 includes: a receiver 1001, a transmitter 1002, a processor 1003 and a memory 1004 (wherein the number of processors 1003 in the terminal device 1000 can be one or more), wherein the processor 1003 may include an application processor 10031 and a communication processor 10032. In some embodiments of the present application, the receiver 1001, the transmitter 1002, the processor 1003 and the memory 1004 may be connected via a bus or other means.

[0171] The memory 1004 may include a read-only memory and a random access memory, and provides instructions and data to the processor 1003. A portion of the memory 1004 may also include non-volatile random access memory (NVRAM). The memory 1004 stores processor and operation instructions, executable modules, or data structures, or subsets or extended sets thereof. The operation instructions may include various operation instructions for implementing various operations.

[0172] Processor 1003 controls the operation of the terminal device. In specific applications, the various components of the terminal device are coupled together via a bus system. In addition to a data bus, the bus system may also include a power bus, a control bus, and a status signal bus. However, for clarity, all bus systems are referred to as a bus system in the figure.

[0173] The method disclosed in the above embodiment of the present application can be applied to the processor 1003, or implemented by the processor 1003. The processor 1003 can be an integrated circuit chip with signal processing capabilities. During the implementation process, the steps of the above method can be completed by the integrated logic circuit of the hardware in the processor 1003 or the instructions in the form of software. The above-mentioned processor 1003 can be a general-purpose processor, a digital signal processor (digital signal processing, DSP), a microprocessor or a microcontroller, and a vision processor (vision processing unit, VPU), a tensor processor (tensor processing unit, TPU) and other processors suitable for AI computing, and can further include an application specific integrated circuit (application specific integrated circuit, ASIC), a field programmable gate array (field-programmable gate array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The processor 1003 can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of this application can be directly implemented as being executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in memory 1004, and processor 1003 reads the information in memory 1004 and, in conjunction with its hardware, completes steps 501 to 503 in the above embodiment.

[0174] Receiver 1001 can be used to receive input digital or character information and generate signal input related to the relevant settings and function control of the terminal device. Transmitter 1002 can be used to output digital or character information through the first interface. Transmitter 1002 can also be used to send instructions to the disk pack through the first interface to modify the data in the disk pack. Transmitter 1002 can also include a display device such as a display screen.

[0175] The embodiment of the present application also provides a server. Please refer to Figure 11. Figure 11 is a schematic diagram of the structure of a server provided by the embodiment of the present application. Specifically, the server 1100 is implemented by one or more servers. The server 1100 may have relatively large differences due to different configurations or performance. It may include one or more central processing units (CPUs) 1111 (for example, one or more processors) and memory 1132, and one or more storage media 1130 (for example, one or more mass storage devices) for storing application programs 1142 or data 1144. Among them, the memory 1132 and the storage medium 1130 can be temporary storage or permanent storage. The program stored in the storage medium 1130 may include one or more modules (not shown in the figure), each module may include a series of instruction operations in the server. Furthermore, the central processing unit 1111 can be configured to communicate with the storage medium 1130 to execute a series of instruction operations in the storage medium 1130 on the server 1100.

[0176] The server 1100 may also include one or more power supplies 1126, one or more wired or wireless network interfaces 1150, one or more input and output interfaces 1158; or one or more operating systems 1141, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0177] Specifically, the server may perform steps 501 to 503 in the above embodiment.

[0178] An embodiment of the present application also provides a computer program product, which, when running on a computer, enables the computer to execute the steps executed by the aforementioned execution device, or enables the computer to execute the steps executed by the aforementioned training device.

[0179] A computer-readable storage medium is also provided in an embodiment of the present application, which stores a program for signal processing. When the computer-readable storage medium is run on a computer, it enables the computer to execute the steps executed by the aforementioned execution device, or enables the computer to execute the steps executed by the aforementioned training device.

[0180] The execution device, training device or terminal device provided in the embodiments of the present application can specifically be a chip, and the chip includes: a processing unit and a communication unit, the processing unit can be, for example, a processor, and the communication unit can be, for example, an input / output interface, a pin or a circuit, etc. The processing unit can execute the computer execution instructions stored in the storage unit, so that the chip in the execution device executes the data processing method described in the above embodiment, or so that the chip in the training device executes the data processing method described in the above embodiment. Optionally, the storage unit is a storage unit in the chip, such as a register, a cache, etc. The storage unit can also be a storage unit located outside the chip in the wireless access device, such as a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM), etc.

[0181] Specifically, see Figure 12, which is a schematic diagram of the structure of a chip provided in an embodiment of the present application. The chip can be represented as a neural network processor NPU 1200. NPU 1200 is mounted on the host CPU (host CPU) as a coprocessor and is assigned tasks by the host CPU. The core of the NPU is arithmetic circuit 1203, which is controlled by controller 1204 to extract matrix data from memory and perform multiplication operations.

[0182] The NPU 1200 can implement the data processing method provided in the embodiment described in FIG. 5 through the mutual cooperation between various internal components.

[0183] More specifically, in some implementations, the arithmetic circuit 1203 in the NPU 1200 includes multiple processing units (PEs). In some implementations, the arithmetic circuit 1203 is a two-dimensional systolic array. The arithmetic circuit 1203 can also be a one-dimensional systolic array or other electronic circuitry capable of performing mathematical operations such as multiplication and addition. In some implementations, the arithmetic circuit 1203 is a general-purpose matrix processor.

[0184] For example, assume there are input matrix A, weight matrix B, and output matrix C. The arithmetic circuit retrieves the corresponding data of matrix B from weight memory 1202 and caches it on each PE in the arithmetic circuit. The arithmetic circuit retrieves the data of matrix A from input memory 1201 and performs a matrix operation on matrix B. The partial or final matrix result is stored in accumulator 1208.

[0185] Unified memory 1206 is used to store input and output data. Weight data is directly transferred to weight memory 1202 through the Direct Memory Access Controller (DMAC) 1205. Input data is also transferred to unified memory 1206 through the DMAC.

[0186] BIU stands for Bus Interface Unit, i.e., bus interface unit 1210 , which is used for interaction between the AXI bus, DMAC, and instruction fetch buffer (IFB) 1209 .

[0187] The bus interface unit 1210 (BIU) is used for the instruction fetch memory 1209 to obtain instructions from the external memory, and is also used for the storage unit access controller 1205 to obtain the original data of the input matrix A or the weight matrix B from the external memory.

[0188] DMAC is mainly used to transfer input data in the external memory DDR to the unified memory 1206 or transfer weight data to the weight memory 1202 or transfer input data to the input memory 1201.

[0189] The vector calculation unit 1207 includes multiple operation processing units. When necessary, it further processes the output of the operation circuit 1203, such as vector multiplication, vector addition, exponential operation, logarithmic operation, size comparison, etc. It is mainly used for non-convolutional / fully connected layer network calculations in neural networks, such as batch normalization, pixel-level summation, and upsampling of feature planes.

[0190] In some implementations, the vector calculation unit 1207 can store the processed output vector in the unified memory 1206. For example, the vector calculation unit 1207 can apply a linear function or a nonlinear function to the output of the operation circuit 1203, such as linear interpolation of the feature plane extracted by the convolution layer, or accumulate a vector of values ​​to generate an activation value. In some implementations, the vector calculation unit 1207 generates a normalized value, a pixel-level summed value, or both. In some implementations, the processed output vector can be used as an activation input to the operation circuit 1203, for example, for use in subsequent layers in a neural network.

[0191] An instruction fetch buffer 1209 connected to the controller 1204 is used to store instructions used by the controller 1204;

[0192] Unified memory 1206, input memory 1201, weight memory 1202, and instruction fetch memory 1209 are all on-chip memories. External memories are private to the NPU hardware architecture.

[0193] The processor mentioned in any of the above places can be a general-purpose central processing unit, a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of the above program.

[0194] It should also be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided in this application, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines.

[0195] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course can also be implemented by special hardware including application-specific integrated circuits, special CPUs, special memories, special components, etc. In general, all functions performed by computer programs can be easily implemented with corresponding hardware, and the specific hardware structures used to implement the same function can also be diverse, such as analog circuits, digital circuits or special circuits, etc. However, for the present application, software program implementation is a better implementation method in most cases. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer's floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk or optical disk, etc., and includes a number of instructions to enable a computer device (which can be a personal computer, training equipment, or network equipment, etc.) to execute the methods described in each embodiment of the present application.

[0196] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0197] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, a computer, a training device or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website, a computer, a training device or a data center. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

Claims

1. A data processing method, characterized in that: The method comprises: Obtaining a first prompt, wherein the first prompt includes description information of the advertisement and attribute information of the media; the media is a delivery platform for the advertisement, and the first prompt indicates generating a delivery requirement for the advertisement on the media according to the description information and the attribute information of the media; According to the first prompt, obtaining the delivery demand on the media through a language model; Based on receiving the advertisement access request of the user on the media, it is determined whether to deliver the advertisement to the user according to the relationship between the delivery demand and the user's intention information.

2. The method according to claim 1, characterized in that The delivery demand includes multiple slots and a description corresponding to each slot, wherein each slot corresponds to a demand; the first prompt specifically instructs to generate multiple slots and a description corresponding to each slot according to the description information and the attribute information of the media, wherein each slot corresponds to a delivery demand.

3. The method according to claim 1 or 2, characterized in that: The obtaining the media delivery demand through a language model according to the first prompt includes: According to the first prompt, a candidate delivery demand on the media is obtained by using a language model; Modification information for the candidate delivery demand is received, and the delivery demand on the media is obtained through a language model according to a second prompt, and the second prompt instructs to modify the candidate delivery demand according to the modification information.

4. The method according to any one of claims 1 to 3, characterized in that: The method further comprises: Obtaining a third prompt, wherein the third prompt includes the historical behavior of the user, and the third prompt indicates enriching the user's portrait information according to the historical behavior; According to the third prompt, attribute information of the user is obtained through a language model; wherein the attribute information of the media includes portrait information of the user.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Obtaining a fourth prompt, the fourth prompt including context information corresponding to when the user triggers an advertisement access request on the medium, the fourth prompt indicating determining the user's intention information according to the context information; According to the fourth prompt, the user's intention information is obtained through a language model.

6. A data processing device, characterized in that: The device comprises: An acquisition module is used to acquire a first prompt, wherein the first prompt includes description information of the advertisement and attribute information of the media; the media is a delivery platform for the advertisement, and the first prompt indicates generating a delivery requirement for the advertisement on the media according to the description information and the attribute information of the media; A processing module, configured to obtain the delivery demand on the media through a language model according to the first prompt; Based on receiving the advertisement access request of the user on the media, it is determined whether to deliver the advertisement to the user according to the relationship between the delivery demand and the user's intention information.

7. The device according to claim 6, characterized in that The delivery demand includes multiple slots and a description corresponding to each slot, wherein each slot corresponds to a demand; the first prompt specifically instructs to generate multiple slots and a description corresponding to each slot according to the description information and the attribute information of the media, wherein each slot corresponds to a delivery demand.

8. The device according to claim 6 or 7, characterized in that The processing module is specifically used for: According to the first prompt, a candidate delivery demand on the media is obtained by using a language model; After receiving the modification information for the candidate delivery demand, according to the second prompt, the language model is used to obtain the media Delivery demand, the second prompt indicates to modify the candidate delivery demand according to the modification information.

9. The device according to any one of claims 6 to 8, characterized in that: The acquisition module is further used for: Obtaining a third prompt, wherein the third prompt includes the historical behavior of the user, and the third prompt indicates enriching the user's portrait information according to the historical behavior; The processing module is further used to obtain attribute information of the user according to the third prompt through a language model; wherein the attribute information of the media includes portrait information of the user.

10. The device according to any one of claims 6 to 9, characterized in that: The acquisition module is further used for: Obtaining a fourth prompt, the fourth prompt including context information corresponding to when the user triggers an advertisement access request on the medium, the fourth prompt indicating determining the user's intention information according to the context information; The processing module is further used to obtain the user's intention information through a language model according to the fourth prompt.

11. A computing device, characterized in that: The computing device includes a memory and a processor; the memory stores codes, and the processor is configured to obtain the codes and execute the method according to any one of claims 1 to 5.

12. A computer storage medium, characterized in that: The computer storage medium stores one or more instructions, which, when executed by one or more computers, enable the one or more computers to implement the method of any one of claims 1 to 5.

13. A computer program product comprising code, characterized in that When the code is executed, it is used to implement the method according to any one of claims 1 to 5.

14. A chip, comprising a processor, characterized in that: The processor is used to support a data processing device to implement the method according to any one of claims 1 to 5.

Images