Data processing method and related device

Abstract

A data processing method can be applied to the field of artificial intelligence and comprises the steps of obtaining first attribute information and at least one calibration feature of a first user and a first article; the calibration feature indicates at least one of a recommended scene, an item, or a provider of the item; obtaining a calibration coefficient corresponding to each calibration feature; obtaining a first recommendation score for recommending the first article to the first user through a recommendation model according to the first attribute information; and adjusting the first recommendation score according to the calibration coefficient. According to the method, the corresponding calibration coefficients are set for different calibration dimensions, and the calibration coefficient of the calibration dimension related to the current to-be-recommended data is utilized to adjust the output of the model, so that the recommendation effect and precision of the model can be improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a data processing method and related devices. Background Art

[0002] Artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is also the study of the design principles and implementation methods of various intelligent machines, so that machines have the functions of perception, reasoning and decision-making.

[0003] In recommendation models, it is crucial to accurately estimate user behavior. For example, for the click-through rate (CTR) estimation of advertisements, the model needs to accurately estimate the probability of a user clicking on an advertisement or news item, so as to guide the platform to perform personalized sorting based on the user's interests. In a typical online advertising recommendation system, the score value estimated by the model affects the sorting and bidding mechanism. Ideally, the score distribution estimated by the recommendation model needs to be consistent with the probability of real feedback. For example, if the average click-through rate estimated by the model is 5%, then it is expected that 5% of the ad exposures will be clicked. However, in actual situations, there is a difference between the score estimated by the model and the actual observed value of online data. This may be due to the inherent bias of the model on the one hand, or due to changes in data distribution on the other hand. Overestimation or underestimation of the model score may cause waste of exposure opportunities and advertisers' over-cost problems, and even lead to online operation accidents. Therefore, model calibration methods are usually used to post-process the estimated score of the model so that the score distribution estimated by the recommendation model fits the probability of real feedback.

[0004] In the existing technology, the prediction results of the recommendation model are post-processed and corrected based on the recent logs collected online, so that the estimated probability distribution of the model fits the real user feedback. However, the existing method calibrates the global output results of the model through a calibration coefficient, and the calibration effect is poor. Summary of the invention

[0005] In a first aspect, the present application provides a data processing method, comprising: obtaining first attribute information of a first user and a first item and at least one calibration feature; the calibration feature indicates at least one of a recommended scenario, an item, or a provider of an item; obtaining a calibration coefficient corresponding to each of the calibration features; based on the first attribute information, obtaining a first recommendation score for recommending the first item to the first user through a recommendation model; and adjusting the first recommendation score based on the calibration coefficient.

[0006] Among them, calibration features are features that will affect the recommendation score, for example, calibration features can be item ID, recommended layout, item provider (such as advertiser) ID, recommendation platform, recommendation scenario, etc. Recommendation scenario can be related to context (such as time, region, etc.). Different calibration features will lead to different data distribution, thus affecting the recommendation score and recommendation results.

[0007] In the embodiment of the present application, by setting corresponding calibration coefficients for different calibration dimensions and adjusting the output of the model using the calibration coefficients of the calibration dimensions related to the data currently to be recommended, the recommendation effect and accuracy of the model can be improved.

[0008] In a possible implementation, the method further includes:

[0009] A second recommendation score for recommending the second item to the second user and an actual interaction score of the second user for the second item are obtained; the second recommendation score is obtained through the recommendation model, and the second attribute information corresponds to a target calibration feature; the target standard feature is one of the at least one calibration feature; a first deviation degree between the actual interaction score and the second recommendation score is calculated by a binomial confidence interval calculation method; a calibration coefficient corresponding to the target calibration feature is generated according to the deviation degree.

[0010] The confidence level of the positive sample proportion (that is, the calibration coefficient) is calculated through the confidence interval of the binomial distribution, and the model is calibrated with different intensities according to the confidence level. Even if the confidence level of the accumulated data in a shorter time window is used, a good calibration effect can be obtained.

[0011] In a possible implementation, the method further includes: mapping the first deviation degree to obtain a second deviation degree by a target mapping method; the target mapping method is a monotonically increasing function; and calculating a third recommendation score obtained after the actual interaction score deviates by the second deviation degree by a binomial confidence interval calculation method;

[0012] A calibration coefficient corresponding to the target calibration feature is obtained according to a difference between the third recommended score and the second recommended score.

[0013] In a possible implementation, obtaining the calibration coefficient corresponding to the target calibration feature based on the difference between the third recommended score and the second recommended score includes: taking the ratio of the third recommended score to the second recommended score as the calibration coefficient corresponding to the target calibration feature.

[0014] In a possible implementation, the confidence interval is a Wilson interval.

[0015] In a possible implementation, the first attribute information corresponds to multiple calibration features, and the adjusting the first recommendation score according to the calibration coefficient includes: fusing multiple calibration coefficients corresponding to the multiple calibration features to obtain fused calibration coefficients; and adjusting the first recommendation score according to the fused calibration coefficients.

[0016] In a second aspect, the present application provides a data processing device, the device comprising:

[0017] An acquisition module, configured to acquire first attribute information of a first user and a first item and at least one calibration feature; the calibration feature indicates at least one of a recommended scene, an item, or a provider of the item;

[0018] Obtaining a calibration coefficient corresponding to each of the calibration features;

[0019] a processing module, configured to obtain, based on the first attribute information and through a recommendation model, a first recommendation score for recommending the first item to the first user;

[0020] The first recommendation score is adjusted according to the calibration factor.

[0021] In a possible implementation, the acquisition module is further used to:

[0022] Obtaining a second recommendation score for recommending the second item to the second user and an actual interaction score of the second user on the second item; the second recommendation score is obtained by the recommendation model, the second attribute information corresponds to a target calibration feature; the target standard feature is one of the at least one calibration feature;

[0023] The processing module is also used

[0024] A first degree of deviation between the actual interaction score and the second recommendation score is calculated by a binomial confidence interval calculation method; and a calibration coefficient corresponding to the target calibration feature is generated according to the degree of deviation.

[0025] In a possible implementation, the processing module is further configured to:

[0026] By means of a target mapping device, the first deviation degree is mapped to obtain a second deviation degree; the target mapping device is a monotonically increasing function;

[0027] Calculating, by means of a binomial confidence interval calculation device, a third recommendation score obtained after the actual interaction score deviates by the second deviation degree;

[0028] A calibration coefficient corresponding to the target calibration feature is obtained according to a difference between the third recommended score and the second recommended score.

[0029] In a possible implementation, the processing module is specifically configured to:

[0030] A ratio of the third recommended score to the second recommended score is used as a calibration coefficient corresponding to the target calibration feature.

[0031] In a possible implementation, the confidence interval is a Wilson interval.

[0032] In a possible implementation, the processing module is specifically configured to:

[0033] fusing a plurality of calibration coefficients corresponding to the plurality of calibration features to obtain fused calibration coefficients;

[0034] The first recommendation score is adjusted according to the fused calibration coefficient.

[0035] 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.

[0036] 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.

[0037] In a fifth aspect, an embodiment of the present application provides a computer program product, including code, which, when executed, is used to implement the above-mentioned first aspect and any optional method.

[0038] In a sixth aspect, the present application provides a chip system, which includes a processor for supporting a data processing device to implement the functions involved in the above aspects, such as sending or processing the data involved in the above methods; or information. In one possible design, the chip system also includes a memory, which is used to store program instructions and data necessary for executing the device or training the device. The chip system can be composed of a chip, or it can include a chip and other discrete devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A structural diagram of the main framework of artificial intelligence;

[0040] Figure 2 A schematic diagram of a system architecture provided for an embodiment of the present application;

[0041] Figure 3 A schematic diagram of a system architecture provided for an embodiment of the present application;

[0042] Figure 4 A schematic diagram of a recommended scenario provided in an embodiment of the present application;

[0043] Figure 5 A flowchart of a data processing method provided in an embodiment of the present application;

[0044] Figure 6 It is a schematic diagram of an application architecture;

[0045] Figure 7 A schematic diagram of the structure of a data processing device provided in an embodiment of the present application;

[0046] Figure 8 A schematic diagram of an execution device provided in an embodiment of the present application;

[0047] Fig. 9 A schematic diagram of a training device provided in an embodiment of the present application;

[0048] Fig.10 A schematic diagram of a chip provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The embodiments of the present invention are described below in conjunction with the accompanying drawings in the embodiments of the present invention. The terms used in the implementation mode 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.

[0050] 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.

[0051] 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 need not be 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, which is only to describe the distinction mode adopted by the objects of the same attributes when describing 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.

[0052] First, the overall workflow of the artificial intelligence system is described. Figure 1 , Figure 1 The figure shows a structural diagram of the main framework of artificial intelligence. The following is an explanation of the above artificial intelligence theme framework from the two dimensions of "intelligent information chain" (horizontal axis) and "IT value chain" (vertical axis). Among them, the "intelligent information chain" reflects a series of processes from data acquisition to processing. For example, it can be a general process of intelligent information perception, intelligent information representation and formation, intelligent reasoning, intelligent decision-making, intelligent execution and output. In this process, the data has undergone a condensation process of "data-information-knowledge-wisdom". The "IT value chain" reflects the value that artificial intelligence brings to the information technology industry from the underlying infrastructure of human intelligence, information (providing and processing technology implementation) to the industrial ecology process of the system.

[0053] (1) Infrastructure

[0054] The infrastructure provides computing power support for the AI ​​system, enables communication with the outside world, and supports it through the basic platform. It communicates with the outside world through sensors; computing power is provided by smart chips (CPU, NPU, GPU, ASIC, FPGA and other hardware acceleration chips); the basic platform includes distributed computing frameworks and networks and other related platform guarantees and support, which can include cloud storage and computing, interconnected networks, etc. For example, sensors communicate with the outside world to obtain data, and these data are provided to the smart chips in the distributed computing system provided by the basic platform for calculation.

[0055] (2) Data

[0056] The data on the upper layer of the infrastructure is used to represent the data sources in the field of artificial intelligence. The data involves graphics, images, voice, text, and IoT data of traditional devices, including business data of existing systems and perception data such as force, displacement, liquid level, temperature, and humidity.

[0057] (3) Data processing

[0058] Data processing usually includes data training, machine learning, deep learning, search, reasoning, decision-making and other methods.

[0059] Among them, machine learning and deep learning can symbolize and formalize data for intelligent information modeling, extraction, preprocessing, and training.

[0060] 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.

[0061] Decision-making refers to the process of making decisions after intelligent information is reasoned, usually providing functions such as classification, sorting, and prediction.

[0062] (4) General ability

[0063] 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 an algorithm or a general system, for example, translation, text analysis, computer vision processing, speech recognition, image recognition, etc.

[0064] (5) Smart products and industry applications

[0065] 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 applications. Its application areas mainly include: smart terminals, smart transportation, smart medical care, autonomous driving, smart cities, etc.

[0066] The embodiments of the present application can be applied to the field of information recommendation, and the scenarios include but are 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 referred to as "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 operation behavior of users in the mobile application market for the recommended APP, and the recommendation model trained thereby is applicable to the above-mentioned mobile APP application market, or can be used in the APP application market of other types of terminals to recommend terminal APPs. The recommendation model will eventually calculate the recommendation probability or score of each recommended object. The recommendation system will select the recommendation results based on certain selection rules, such as sorting them by recommendation probability or score, and present them to users through corresponding applications or terminal devices. Users will operate on the objects in the recommendation results to generate user behavior logs, etc.

[0067] Reference Figure 4 In the recommendation process, when a user interacts with the recommendation system, a recommendation request is triggered. The recommendation system inputs the request and its related feature information into the deployed recommendation model, and then predicts the click-through rate of all candidate objects. Subsequently, the candidate objects are sorted in descending order according to the predicted click-through rate, and the candidate objects are displayed in different positions in order as the recommendation results for the user. The user browses the displayed items and performs user behaviors, such as browsing, clicking, and downloading. These user behaviors will be stored in the log as training data, and the parameters of the recommendation model will be updated from time to time through the offline training module to improve the recommendation effect of the model.

[0068] For example, when a user opens the mobile app market, the recommendation module of the app market is triggered. The recommendation module of the app market predicts the possibility of the user downloading each given candidate application based on the user's historical download records, user click records, the application's own characteristics, time, location and other environmental characteristics. Based on the prediction results, the app market displays them in descending order of likelihood to increase the probability of application download. Specifically, applications that are more likely to be downloaded are ranked at the front, and applications that are less likely to be downloaded are ranked at the back. The user's behavior will also be stored in the log and the parameters of the prediction model will be trained and updated through the offline training module.

[0069] 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 build a user lifelong learning system framework. Lifelong companions can record past events of users based on system data and application data, understand the user's current intentions, predict the user's future actions or behaviors, and ultimately realize intelligent services. In the current first stage, based on music apps, video apps, browser apps, etc., the user's 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.

[0070] Next, the application architecture of the embodiment of the present application is introduced.

[0071] See attached Figure 2, an embodiment of the present invention provides a recommendation system architecture 200. The data collection 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). There can be multiple feature information, which can specifically include user feature information, object feature information and label features. User feature information is used to characterize user features, such as gender, age, occupation, hobbies, etc. Object feature information is used to characterize the features of the object pushed to the user. Different recommendation systems correspond to different objects, and the feature types 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 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, and 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 the sample is collected, it can be stored in the database 230, and 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 the model parameter matrix based on the sample training in the database 230 for generating the 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 for generating the recommendation model 201. The recommendation model 201 can be used to evaluate a large number of objects to obtain the score 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 result based on the evaluation result of the recommendation model 201 and recommends it to the client device through the I / O interface 212.

[0072] In the 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 the trained recommendation model; the implementation details of the calculation module 211 can refer to Figure 5 A detailed description of the method embodiment shown.

[0073] 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 the 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.

[0074] The execution device 210 is configured with an I / O interface 212 for data exchange with external devices. The execution device 210 can obtain user feature information, such as user identification, user identity, gender, occupation, hobbies, etc., from the client device 240 through the I / O interface 212. This part of information can also be obtained from the system database. The recommendation model 201 recommends target recommendation objects to the user based on the user feature information and the feature information of the recommended object. The execution device 210 can be set in a cloud server or in a user client.

[0075] The execution device 210 can call data, codes, 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, or can be set independently, or can be set in other network entities, and the number can be one or more.

[0076] 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 scores, wherein the objects with higher rankings will be recommended to the client device 240.

[0077] Finally, the I / O interface 212 returns the recommendation result to the client device 240 for presentation to the user.

[0078] 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.

[0079] It is worth noting that Figure 2 This 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. Figure 2In the embodiment, 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.

[0080] In an embodiment of the present application, the training device 220, the execution device 210, and the client device 240 may 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.

[0081] See attached Figure 3 , 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 on a physical site, or distributed on 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 scores, and recommends the object to be recommended to the user according to the sorting results. For example, the first 10 objects in the sorting results are recommended to the user.

[0082] Among them, the data storage system 250 is used to receive and store the parameters of the recommendation model sent by the training device, and to store the data of 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 or more devices deployed outside the execution device 210. At this time, 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, and 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, such as 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.

[0083] 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 smart phone, 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.

[0084] 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.

[0085] In another implementation, the execution device 210 may be implemented by a local device. For example, the local device 301 may implement the recommendation function of the execution device 210 based on the recommendation model to obtain user feature information and feedback the recommendation result to the user, or provide services to the user of the local device 302.

[0086] 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.

[0087] 1. Click-throughrate (CTR)

[0088] 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 in the recommendation system.

[0089] 2. Personalized recommendation system

[0090] A personalized recommendation system refers to a system that uses a machine learning algorithm to analyze the user's historical data (such as the operation information in the embodiment of the present application), and uses this to predict new requests and provide personalized recommendation results.

[0091] 3. Offline training

[0092] 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.

[0093] 4. Online prediction (onlineinference)

[0094] Online prediction refers to predicting the user's preference for the recommended item in the current context based on the offline trained model according to the characteristics of the user, item and context, and predicting the probability of the user selecting the recommended item.

[0095] For example, Figure 4 Schematic diagram of the recommendation system provided in the embodiment of the present application. Figure 4 As shown, when a user enters the system, a recommendation request will be triggered. The recommendation system will input the request and related information (such as the operation information in the embodiment of the present application) into the recommendation model, and then predict the user's selection rate for the items in the system. Further, 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 the recommendation results for the user. The user browses the 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.

[0096] For example, a user can trigger the recommendation system in the application market by opening the application market in a smart terminal (e.g., a mobile phone). The recommendation system in the application market will predict the probability of the user downloading each recommended candidate APP based on the user's historical behavior log, such as the user's historical download records, user selection records, and the application market's own characteristics, such as time, location and other environmental feature information. Based on the calculation results, the recommendation system in the application market can display the candidate APPs in descending order according to the predicted probability values, thereby increasing the download probability of the candidate APPs.

[0097] For example, an APP with a higher predicted user selection rate may be displayed in a front recommended position, and an APP with a lower predicted user selection rate may be displayed in a back recommended position.

[0098] The above-mentioned recommendation model may be a neural network model. The following introduces the relevant terms and concepts of the neural network that may be involved in the embodiments of the present application.

[0099] (1) Neural Network

[0100] A neural network may be composed of neural units, and a neural unit may refer to an operation unit that takes xs (i.e., input data) and intercept 1 as input, and the output of the operation unit may be:

[0101]

[0102] 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 in the neural unit into the output signal. The output signal of the activation function can be used as the input of the next convolution 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 characteristics of the local receptive field. The local receptive field can be an area composed of several neural units.

[0103] (2) Deep Neural Networks

[0104] 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. From the position of different layers of 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 DNN has many layers, the coefficient W and the offset vector The definition of these parameters in DNN is as follows: Taking 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 subscripts correspond to the output third layer index 2 and the input second layer index 4. In summary, the coefficients from the kth neuron in the L-1th layer to the jth neuron in the Lth layer are defined as It should be noted that the input layer does not have a W parameter. In a deep neural network, more hidden layers allow the network to better describe complex situations in the real world. Theoretically, the more parameters a model has, the higher its complexity and the greater its "capacity", which means it can complete more complex learning tasks. Training a deep neural network is the process of learning the weight matrix, and its ultimate goal is to obtain the weight matrix of all layers of the trained deep neural network (the weight matrix formed by many layers of vector W).

[0105] (3) Loss Function

[0106] In the process of training a deep neural network, because we hope that the output of the deep neural network is as close as possible to the value we really want to predict, we can compare the predicted value of the current network with the target value we really want, and then update the weight vector of each layer of the neural network according to the difference between the two (of course, there is usually an initialization process before the first update, that is, pre-configuring parameters for each layer in the deep neural network). For example, if the predicted value of the network is high, adjust the weight vector to make it predict a lower value, and keep adjusting until the deep neural network can predict the target value we really want or a value very close to the target value we really want. Therefore, it is necessary to pre-define "how to compare the difference between the predicted value and the target value", which is the loss function or objective function, which are important equations used to measure the difference between the predicted value and the target value. Among them, taking the loss function as an example, the higher the output value (loss) of the loss function, the greater the difference, so the training of the deep neural network becomes a process of minimizing this loss as much as possible.

[0107] (4) Back propagation algorithm

[0108] The error back propagation (BP) algorithm can be used to correct the size of the parameters in the initial model during the training process, so that the error loss of the model becomes smaller and smaller. Specifically, the forward transmission of the input signal to the output will generate error loss, and the parameters in the initial model are updated by back propagating the error loss information, so that the error loss converges. The back propagation algorithm is a back propagation movement dominated by error loss, aiming to obtain the optimal model parameters, such as the weight matrix.

[0109] (5) Machine Learning System

[0110] 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.

[0111] (6) Personalized recommendation system

[0112] A system that uses machine learning algorithms to analyze and build models based on historical user data, predict new user requests, and provide personalized recommendation results.

[0113] (7) Model calibration: By mapping the model output, the score distribution of the model output is made as close as possible to the distribution of the true label.

[0114] In recommendation models, it is crucial to accurately estimate user behavior. For example, for the click-through rate (CTR) estimation of advertisements, the model needs to accurately estimate the probability of a user clicking on an advertisement or news item, so as to guide the platform to perform personalized sorting based on the user's interests. In a typical online advertising recommendation system, the score value estimated by the model affects the sorting and bidding mechanism. Ideally, the score distribution estimated by the recommendation model needs to be consistent with the probability of real feedback. For example, if the average click-through rate estimated by the model is 5%, then it is expected that 5% of the ad exposures will be clicked. However, in actual situations, there is a difference between the score estimated by the model and the actual observed value of online data. This may be due to the inherent bias of the model on the one hand, or due to changes in data distribution on the other hand. Overestimation or underestimation of the model score may cause waste of exposure opportunities and advertisers' over-cost problems, and even lead to online operation accidents. Therefore, model calibration methods are usually used to post-process the estimated score of the model so that the score distribution estimated by the recommendation model fits the probability of real feedback.

[0115] In the existing technology, the prediction results of the recommendation model are post-processed and corrected based on the recent logs collected online, so that the estimated probability distribution of the model fits the real user feedback. However, the existing method calibrates the global output results of the model through a calibration coefficient, and the calibration effect is poor.

[0116] In order to solve the above problems, the present application provides a data processing method.

[0117] Reference Figure 5 , Figure 5 An example of a data processing method provided in the present application is shown in FIG. Figure 5 As shown, a data processing method provided by an embodiment of the present application includes:

[0118] 501. Acquire first attribute information of a first user and a first object and at least one calibration feature; the calibration feature indicates at least one of a recommended scene, an object, or a provider of the object.

[0119] In the embodiment of the present application, the execution subject of step 501 may be a terminal device, and the terminal device may be a portable mobile device, such as but not limited to a mobile or portable computing device (such as a smart phone), a personal computer, a server computer, a handheld device (such as a tablet) or a laptop device, a multi-processor system, a game console or controller, a microprocessor-based system, a set-top box, a programmable consumer electronic product, a mobile phone, a mobile computing and / or communication device with a wearable or accessory form factor (such as a watch, glasses, a headset or earplugs), a network PC, a minicomputer, a mainframe computer, a distributed computing environment including any of the above systems or devices, and the like.

[0120] In the embodiment of the present application, the execution entity of step 501 may be a server on the cloud side, and the server may receive the user's operation data sent from the terminal device, and then the server may obtain the user's operation data.

[0121] In a possible implementation, when recommending items to a user, it is necessary to calculate the recommendation score of each item recommended to the user. The embodiment of the present application is described by taking the calculation of the recommendation score of recommending a first item to a first user as an example:

[0122] In a possible implementation, attribute information of the first user and the first object may be obtained.

[0123] Among them, the user's attribute information can be an attribute related to the user's preference characteristics, at least one of gender, age, occupation, income, hobbies and education level, among which gender can be male or female, age can be a number between 0-100, occupation can be teacher, programmer, chef, etc., hobbies can be basketball, tennis, running, etc., and education level can be elementary school, junior high school, high school, university, etc.; this application does not limit the specific type of user's attribute information.

[0124] Among them, the items can be physical items or virtual items, such as applications (APP), audio and video, web pages, advertisements, and news information. The attribute information of the items can be at least one of the item name, developer, installation package size, category, and popularity. Among them, taking the item as an application as an example, the category of the item can be chatting, parkour games, office, etc., and the popularity can be a score, comment, etc. for the item; the present application does not limit the specific type of the attribute information of the item.

[0125] In a possible implementation, at least one calibration feature may also be obtained. The calibration feature is a feature that affects the recommendation score. For example, the calibration feature may be the ID of the item, the recommended layout, the ID of the provider of the item (such as an advertiser), the recommendation platform, the recommendation scenario, etc. The recommendation scenario may be related to the context (such as time, region, etc.). Different calibration features may lead to different data distributions, thereby affecting the recommendation score and recommendation results.

[0126] Among them, the tasks implemented by the recommendation model in the embodiment of the present application can be but are not limited to a variety of the following tasks: purchase behavior prediction, shopping cart behavior prediction, sharing behavior prediction, browsing behavior prediction, completion rate prediction, like prediction, collection prediction, click prediction and click conversion prediction.

[0127] 502. Obtain a calibration coefficient corresponding to each of the calibration features.

[0128] In a possible implementation, a corresponding calibration coefficient may be set in advance for each calibration feature in a plurality of calibration features. When calculating a recommendation score (e.g., calculating the recommendation score of a first item for a first user), the calibration features related to the calculation may be obtained. For example, the ID of the first item, the ID of the provider of the first item, the current recommendation layout, and the current recommendation platform may be selected as calibration features. The calibration coefficient corresponding to each calibration feature may be obtained, and the calibration coefficient may represent the recommendation score and the real score (e.g., the actual interaction score of the user on the item, such as the actual click-through rate) obtained by the recommendation model when processing the recommendation data related to the calibration feature (e.g., including the attribute information of the user and the item, the context information, etc.). Furthermore, the calibration coefficient may be used to adjust the recommendation score output by the subsequent recommendation model, thereby obtaining an accurate recommendation score adapted to different calibration features.

[0129] Next, we will introduce how to configure the size of the calibration coefficient corresponding to each calibration feature.

[0130] In a possible implementation, when calculating the calibration coefficient corresponding to a calibration feature, a recommendation sample related to the calibration feature can be obtained, and the recommendation sample can include the recommendation score of the recommendation model for recommending items to the user when processing the recommendation data that meets the calibration feature, and the actual interaction score of the user on the item.

[0131] For example, taking the target calibration feature as an example, the second recommendation score for recommending the second item to the second user and the actual interaction score of the second user with the second item can be obtained; the second recommendation score is obtained through the recommendation model, and the second attribute information corresponds to the target calibration feature; the target standard feature is one of the at least one calibration feature.

[0132] The calibration coefficient can characterize the deviation between the model output and the actual interaction score. Therefore, in the embodiment of the present application, the first degree of deviation between the actual interaction score and the second recommendation score can be calculated by the calculation method of the binomial confidence interval; the calibration coefficient corresponding to the target calibration feature is generated according to the degree of deviation.

[0133] The confidence interval may be, but is not limited to, a Wilson interval.

[0134] Taking click-through rate prediction as an example, the actual interaction score can be the average click-through rate. For each calibration feature (or feature dimension), the Wilson interval formula is used to calculate the confidence. Based on the number of samples n, the average score of the recommendation model under this calibration feature is calculated. The observed actual average click rate p can be substituted into the following formula to calculate the deviation degree z from the model prediction result to the actual observation result (that is, the first deviation degree in the embodiment of the present application):

[0135]

[0136] The value of z can be solved by bisection method or Newton iteration method.

[0137] In a possible implementation, the first deviation degree can be mapped by a target mapping method to obtain a second deviation degree; the target mapping method is a monotonically increasing function; the third recommendation score obtained after the actual interaction score is deviated by the second deviation degree is calculated by a binomial confidence interval calculation method; and the calibration coefficient corresponding to the target calibration feature is obtained according to the difference between the third recommendation score and the second recommendation score. The target mapping method can also be used to reduce the value of the input data, that is, the second deviation degree is less than the first deviation degree.

[0138] For example, the calculated z can be transformed, and the following monotonic transformation function can be used:

[0139]

[0140] The new deviation degree z obtained by transformation ′ (that is, the second degree of deviation in the embodiment of the present application) is further brought into the first formula, combined with the sample size n and the observed true average click rate p, the calibrated average model score can be obtained: (i means the ith calibration feature). If The sign in the formula takes a positive value, otherwise it takes a negative value.

[0141] In a possible implementation, the ratio of the third recommendation score to the second recommendation score may be used as the calibration coefficient corresponding to the target calibration feature. That is, the calibration coefficient k under this calibration feature is calculated. i The corresponding calibration coefficient is the ratio of the average model scores before and after calibration, that is,

[0142] It should be understood that the binomial confidence interval calculation method of the embodiment of the present application can be extended to use other calculation methods in addition to the Wilson interval. Other monotonically increasing bounded functions can also be used for the transformation function of the deviation degree.

[0143] In an embodiment of the present application, the confidence level (that is, the calibration coefficient) of the positive sample ratio is calculated by the confidence interval of the binomial distribution, and the model is calibrated with different intensities according to the confidence level. Even if the confidence level of the accumulated data in a shorter time window is used, a good calibration effect can be obtained.

[0144] 503. Obtain a first recommendation score for recommending the first item to the first user through a recommendation model according to the first attribute information.

[0145] In a possible implementation, the first attribute information may be processed using a recommendation model to obtain a first recommendation score for recommending the first item to the first user.

[0146] 504. Adjust the first recommendation score according to the calibration coefficient.

[0147] In a possible implementation, after the calibration coefficient is obtained, the first recommendation score may be adjusted according to the calibration coefficient.

[0148] In a possible implementation, after obtaining the multiple calibration coefficients, the multiple calibration coefficients corresponding to the multiple calibration features may be fused to obtain fused calibration coefficients; and the first recommendation score may be adjusted according to the fused calibration coefficients.

[0149] It should be understood that the fusion method in the embodiment of the present application may be a geometric mean, or other mean calculation methods may be used, and the present application is not limited thereto.

[0150] Exemplarily, the calibration coefficients under different calibration features are aggregated according to the following formula:

[0151]

[0152] where w iRepresents the geometric mean weight of each calibration feature (the default is 1, which can also be set manually or determined by parameter search). For a specific sample, assuming that the model's score for the sample is y, then the calibrated score is

[0153] In the embodiment of the present application, by setting corresponding calibration coefficients for different calibration dimensions and adjusting the output of the model using the calibration coefficients of the calibration dimensions related to the data currently to be recommended, the recommendation effect and accuracy of the model can be improved.

[0154] In addition, the confidence level of the positive sample proportion (that is, the calibration coefficient) is calculated through the confidence interval of the binomial distribution, and the model is calibrated with different intensities according to the confidence level. Even if the confidence level of the accumulated data in a shorter time window is used, a good calibration effect can be obtained.

[0155] Next, combine Figure 6 An application framework schematic diagram of an embodiment of the present application is introduced, including:

[0156] 101 confidence-based model calibration module and 102 multi-dimensional calibration fusion module.

[0157] exist Figure 6 In the framework shown, the 101 confidence-based model calibration module considers the confidence of the positive sample proportion calculated by the collected samples through the Wilson confidence interval of the binomial distribution, and maps the degree of deviation between the true observation value and the model estimated distribution according to a mapping function, thereby calibrating the model with different intensities.

[0158] The 102 multi-dimensional calibration fusion module independently calculates the calibration scores of a sample in multiple dimensions and fuses the calibrated scores by weighted geometric averaging. It is a configurable module.

[0159] Next, the beneficial effects of the embodiments of the present application are introduced in combination with experiments:

[0160] The experimental evaluation indicators are the overall deviation between the estimated score and the true statistical value and the deviation in each feature dimension. Experiments are conducted on this dataset, and the results are shown in Table 1.

[0161] Table 1 Experimental results statistics

[0162]

[0163] As can be seen from Table 1, compared with the baseline, the embodiment of the present application can achieve better results and can consistently reduce the deviation of the model scoring in each dimension.

[0164] Next, a data processing device provided by an embodiment of the present application is introduced from the perspective of a device. Figure 7 , Figure 7 A schematic diagram of the structure of a data processing device provided in an embodiment of the present application is shown in FIG. Figure 7 As shown, a data processing device 700 provided in an embodiment of the present application includes:

[0165] The acquisition module 701 is used to acquire first attribute information of a first user and a first item and at least one calibration feature; the calibration feature indicates at least one of a recommended scene, an item, or a provider of the item; and acquire a calibration coefficient corresponding to each of the calibration features;

[0166] For a detailed introduction to the acquisition module 701 , reference may be made to the description of steps 501 and 502 in the above embodiment, which will not be repeated here.

[0167] The processing module 702 is used to obtain a first recommendation score for recommending the first item to the first user through a recommendation model according to the first attribute information; and adjust the first recommendation score according to the calibration coefficient.

[0168] For a detailed introduction to the processing module 702, reference may be made to the description of steps 503 and 504 in the above embodiment, which will not be repeated here.

[0169] In a possible implementation, the acquisition module 701 is further used to:

[0170] Obtaining a second recommendation score for recommending the second item to the second user and an actual interaction score of the second user on the second item; the second recommendation score is obtained by the recommendation model, the second attribute information corresponds to a target calibration feature; the target standard feature is one of the at least one calibration feature;

[0171] The processing module 702 is also used to

[0172] A first degree of deviation between the actual interaction score and the second recommendation score is calculated by a binomial confidence interval calculation method; and a calibration coefficient corresponding to the target calibration feature is generated according to the degree of deviation.

[0173] In a possible implementation, the processing module 702 is further configured to:

[0174] By means of a target mapping device, the first deviation degree is mapped to obtain a second deviation degree; the target mapping device is a monotonically increasing function;

[0175] Calculating, by means of a binomial confidence interval calculation device, a third recommendation score obtained after the actual interaction score deviates by the second deviation degree;

[0176] A calibration coefficient corresponding to the target calibration feature is obtained according to a difference between the third recommended score and the second recommended score.

[0177] In a possible implementation, the processing module 702 is specifically configured to:

[0178] A ratio of the third recommended score to the second recommended score is used as a calibration coefficient corresponding to the target calibration feature.

[0179] In a possible implementation, the confidence interval is a Wilson interval.

[0180] In a possible implementation, the processing module 702 is specifically configured to:

[0181] fusing a plurality of calibration coefficients corresponding to the plurality of calibration features to obtain fused calibration coefficients;

[0182] The first recommendation score is adjusted according to the fused calibration coefficient.

[0183] Next, a terminal device provided by an embodiment of the present application is introduced. Figure 8 , Figure 8 This is a schematic diagram of a terminal device provided in an embodiment of the present application. The terminal device 800 can be specifically a mobile phone, a tablet, a laptop computer, a smart wearable device, etc., which is not limited here. Among them, the terminal device 800 can be used as a training device to implement Figure 5 The function of the data processing method in the corresponding embodiment, or as an execution device to execute based on Figure 5 The trained model obtained by the data processing method in the corresponding embodiment. Specifically, the terminal device 800 includes: a receiver 801, a transmitter 802, a processor 803 and a memory 804 (wherein the number of processors 803 in the terminal device 800 can be one or more), wherein the processor 803 may include an application processor 8031 ​​and a communication processor 8032. In some embodiments of the present application, the receiver 801, the transmitter 802, the processor 803 and the memory 804 may be connected via a bus or other means.

[0184] The memory 804 may include a read-only memory and a random access memory, and provides instructions and data to the processor 803. A portion of the memory 804 may also include a non-volatile random access memory (NVRAM). The memory 804 stores processor and operation instructions, executable modules or data structures, or subsets thereof, or extended sets thereof, wherein the operation instructions may include various operation instructions for implementing various operations.

[0185] The processor 803 controls the operation of the execution device. In a specific application, the various components of the execution device are coupled together through a bus system, wherein the bus system includes not only a data bus but also a power bus, a control bus, and a status signal bus, etc. However, for the sake of clarity, various buses are referred to as bus systems in the figure.

[0186] The method disclosed in the above embodiment of the present application can be applied to the processor 803, or implemented by the processor 803. The processor 803 can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware or software instructions in the processor 803. The above processor 803 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 (tensorprocessing 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 803 can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present application. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as being executed by a hardware decoding processor, or being executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 804, and the processor 803 reads the information in the memory 804, and completes the steps 501 to 504 in the above embodiment in combination with its hardware.

[0187] The receiver 801 can be used to receive input digital or character information and generate signal input related to the relevant settings and function control of the execution device. The transmitter 802 can be used to output digital or character information through the first interface; the transmitter 802 can also be used to send instructions to the disk group through the first interface to modify the data in the disk group; the transmitter 802 can also include a display device such as a display screen.

[0188] The present application also provides a server. Fig. 9 , Fig. 9 : This is a structural diagram of a server provided in an embodiment of the present application. Specifically, the server 900 is implemented by one or more servers. The server 900 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 99 (for example, one or more processors) and a memory 932, and one or more storage media 930 (for example, one or more mass storage devices) storing application programs 942 or data 944. Among them, the memory 932 and the storage medium 930 can be short-term storage or permanent storage. The program stored in the storage medium 930 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the training device. Furthermore, the central processing unit 99 can be configured to communicate with the storage medium 930 and execute a series of instruction operations in the storage medium 930 on the server 900.

[0189] The server 900 may also include one or more power supplies 926, one or more wired or wireless network interfaces 950, one or more input and output interfaces 958; or, one or more operating systems 941, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0190] Specifically, the server may serve as a training device to execute steps 501 to 504 in the above embodiment.

[0191] In a possible implementation, the terminal device 800 or the server 900 can be used as a training device to execute steps 501 to 504 in the above embodiment to obtain a trained model, and deploy the trained model to an execution device, which can also be in the form of a terminal device 800 or a server 900. When the execution device executes the trained model, it can refer to Figure 5 The model feed-forward process in the corresponding embodiment.

[0192] Also provided in an embodiment of the present application is a computer program product which, when executed 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.

[0193] 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.

[0194] The execution device, training device or terminal device provided in the embodiments of the present application may specifically be a chip, and the chip includes: a processing unit and a communication unit, wherein the processing unit may be, for example, a processor, and the communication unit may be, for example, an input / output interface, a pin or a circuit, etc. The processing unit may 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 may 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.

[0195] For details, please refer to Fig.10 , Fig.10 A schematic diagram of the structure of a chip provided in an embodiment of the present application, the chip can be expressed as a neural network processor NPU1000, NPU 1000 is mounted on the host CPU (Host CPU) as a coprocessor, and the host CPU assigns tasks. The core part of the NPU is the operation circuit 1003, which is controlled by the controller 1004 to extract matrix data from the memory and perform multiplication operations.

[0196] NPU 1000 can achieve Figure 5 The data processing method provided in the described embodiments.

[0197] More specifically, in some implementations, the operation circuit 1003 in the NPU 1000 includes a plurality of processing units (Process Engine, PE) inside. In some implementations, the operation circuit 1003 is a two-dimensional systolic array. The operation circuit 1003 can also be a one-dimensional systolic array or other electronic circuits capable of performing mathematical operations such as multiplication and addition. In some implementations, the operation circuit 1003 is a general-purpose matrix processor.

[0198] For example, assume there is an input matrix A, a weight matrix B, and an output matrix C. The operation circuit takes the corresponding data of matrix B from the weight memory 1002 and caches it on each PE in the operation circuit. The operation circuit takes the matrix A data from the input memory 1001 and performs matrix operation with matrix B, and the partial result or final result of the matrix is ​​stored in the accumulator 1008.

[0199] The unified memory 1006 is used to store input data and output data. The weight data is directly transferred to the weight memory 1002 through the direct memory access controller (DMAC) 1005. The input data is also transferred to the unified memory 1006 through the DMAC.

[0200] BIU stands for Bus Interface Unit, i.e., bus interface unit 1010 , which is used for interaction between AXI bus, DMAC and instruction fetch buffer (IFB) 1009 .

[0201] The bus interface unit 1010 (Bus Interface Unit, BIU for short) is used for the instruction fetch memory 1009 to obtain instructions from the external memory, and is also used for the storage unit access controller 1005 to obtain the original data of the input matrix A or the weight matrix B from the external memory.

[0202] DMAC is mainly used to transfer input data in the external memory DDR to the unified memory 1006 or to transfer weight data to the weight memory 1002 or to transfer input data to the input memory 1001.

[0203] The vector calculation unit 1007 includes multiple operation processing units, and further processes the output of the operation circuit 1003 when necessary, 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, upsampling of feature planes, etc.

[0204] In some implementations, the vector calculation unit 1007 can store the processed output vector to the unified memory 1006. For example, the vector calculation unit 1007 can apply a linear function; or a nonlinear function to the output of the operation circuit 1003, such as linear interpolation of the feature plane extracted by the convolution layer, and then, for example, a vector of accumulated values ​​to generate an activation value. In some implementations, the vector calculation unit 1007 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 1003, for example, for use in a subsequent layer in a neural network.

[0205] An instruction fetch buffer 1009 connected to the controller 1004 is used to store instructions used by the controller 1004;

[0206] Unified memory 1006, input memory 1001, weight memory 1002 and instruction fetch memory 1009 are all on-chip memories. External memories are private to the NPU hardware architecture.

[0207] The processor mentioned in any of the above places may 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.

[0208] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, 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 over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0209] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an 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 floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0210] 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.

[0211] 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 may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may 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 may be transmitted from a website site, 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 site, computer, training device, or data center. The computer-readable storage medium may 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 may 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)), etc.

Claims

1. A data processing method, characterized in that: The method comprises: Acquire first attribute information of a first user and a first object and at least one calibration feature; the calibration feature indicates at least one of a recommended scene, an object, or a provider of the object; Obtaining a calibration coefficient corresponding to each of the calibration features; Obtaining, based on the first attribute information, a first recommendation score for recommending the first item to the first user through a recommendation model; The first recommendation score is adjusted according to the calibration factor.

2. The method according to claim 1, characterized in that The method further comprises: Obtaining a second recommendation score for recommending the second item to the second user and an actual interaction score of the second user on the second item; the second recommendation score is obtained by the recommendation model, the second attribute information corresponds to a target calibration feature; the target standard feature is one of the at least one calibration feature; A first degree of deviation between the actual interaction score and the second recommendation score is calculated by a binomial confidence interval calculation method; and a calibration coefficient corresponding to the target calibration feature is generated according to the degree of deviation.

3. The method according to claim 2, characterized in that The method further comprises: Mapping the first deviation degree by a target mapping method to obtain a second deviation degree; the target mapping method is a monotonically increasing function; Calculate, by using a binomial confidence interval calculation method, a third recommendation score obtained after the actual interaction score deviates by the second deviation degree; A calibration coefficient corresponding to the target calibration feature is obtained according to a difference between the third recommended score and the second recommended score.

4. The method according to claim 2 or 3, characterized in that: The obtaining, according to the difference between the third recommended score and the second recommended score, a calibration coefficient corresponding to the target calibration feature, comprises: A ratio of the third recommended score to the second recommended score is used as a calibration coefficient corresponding to the target calibration feature.

5. The method according to any one of claims 2 to 4, characterized in that: The confidence interval is the Wilson interval.

6. The method according to any one of claims 1 to 5, characterized in that: The first attribute information corresponds to a plurality of calibration features, and the adjusting the first recommendation score according to the calibration coefficient includes: fusing a plurality of calibration coefficients corresponding to the plurality of calibration features to obtain fused calibration coefficients; The first recommendation score is adjusted according to the fused calibration coefficient.

7. A data processing device, characterized in that: The device comprises: An acquisition module, configured to acquire first attribute information of a first user and a first item and at least one calibration feature, wherein the calibration feature indicates at least one of a recommended scene, an item, or a provider of the item; and to acquire a calibration coefficient corresponding to each of the calibration features; A processing module is used to obtain a first recommendation score for recommending the first item to the first user through a recommendation model according to the first attribute information; and adjust the first recommendation score according to the calibration coefficient.

8. The device according to claim 7, characterized in that The acquisition module is further used for: Obtaining a second recommendation score for recommending the second item to the second user and an actual interaction score of the second user on the second item; the second recommendation score is obtained by the recommendation model, the second attribute information corresponds to a target calibration feature; the target standard feature is one of the at least one calibration feature; The processing module is further used for: Calculating a first degree of deviation between the actual interaction score and the second recommendation score by a binomial confidence interval calculation device; The calibration coefficient corresponding to the target calibration feature is generated according to the degree of deviation.

9. The device according to claim 8, characterized in that The processing module is further used for: By means of a target mapping device, the first deviation degree is mapped to obtain a second deviation degree; the target mapping device is a monotonically increasing function; Calculate, by using a binomial confidence interval calculation method, a third recommendation score obtained after the actual interaction score deviates by the second deviation degree; A calibration coefficient corresponding to the target calibration feature is obtained according to a difference between the third recommended score and the second recommended score.

10. The device according to claim 8 or 9, characterized in that The processing module is specifically used for: A ratio of the third recommended score to the second recommended score is used as a calibration coefficient corresponding to the target calibration feature.

11. The device according to any one of claims 8 to 10, characterized in that: The confidence interval is the Wilson interval.

12. The device according to any one of claims 7 to 11, characterized in that The processing module is specifically used for: fusing a plurality of calibration coefficients corresponding to the plurality of calibration features to obtain fused calibration coefficients; The first recommendation score is adjusted according to the fused calibration coefficient.

13. 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 6.

14. 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 6.

15. 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 6.

16. 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 6.

Images