Model training method, data recommendation method, device, electronic device and storage medium

Abstract

The present disclosure provides a model training method, a data recommendation method, a device, an electronic device and a storage medium, which relate to the field of artificial intelligence technology, and in particular to the fields of deep learning, natural language processing, data recommendation, etc. The specific implementation scheme is: obtaining target behavior data from a first feed information stream and a first channel in the first feed information stream; obtaining target resource data from the first feed information stream and a first channel in the first feed information stream; obtaining sample data based on the target behavior data and the target resource data; performing model training of cross-domain graph learning based on the sample data to obtain a trained first target model. The present disclosure improves the accuracy of model training.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to the fields of deep learning, natural language processing, and data recommendation. Background Art

[0002] In the field of data recommendation, the client can recommend various types of data of interest to users. For example, it can recommend an article or a playlist of interest to users. The client can push various types of data of interest to users in the form of a feed information flow.

[0003] With the development of technology, artificial intelligence can be used to accelerate data processing, and it is applicable to a variety of application scenarios. For example, artificial intelligence technology can be used in hardware design or software design for image processing, graphic processing, video processing, etc. in data recommendation to improve data processing speed and accuracy. For the field of data recommendation, a model with high accuracy in data recommendation is needed. Summary of the invention

[0004] The present disclosure provides a model training method, a data recommendation method, an apparatus, an electronic device, and a storage medium.

[0005] According to one aspect of the present disclosure, a model training method is provided, comprising:

[0006] Acquire target behavior data from a first feed information stream and a first channel in the first feed information stream;

[0007] Acquire target resource data from the first feed information stream and a first channel in the first feed information stream;

[0008] Obtaining sample data according to the target behavior data and the target resource data;

[0009] Model training for cross-domain graph learning is performed based on the sample data to obtain a trained first target model.

[0010] According to another aspect of the present disclosure, a model training method is provided, comprising:

[0011] Obtain target behavior data from a first channel in a first feed information flow;

[0012] Acquire target resource data from the first channel in the first feed information flow;

[0013] Obtaining sample data according to the target behavior data and the target resource data;

[0014] The graph neural network model is trained according to the sample data to obtain a trained second target model.

[0015] According to another aspect of the present disclosure, a data recommendation method is provided, comprising:

[0016] Obtaining touch operations on a first channel in a first feed information stream;

[0017] In response to the touch operation, resource data matching the touch operation is obtained according to the first target model obtained by the above-mentioned model training method and / or the second target model obtained by the above-mentioned model training method.

[0018] According to another aspect of the present disclosure, a model training device is provided, comprising:

[0019] A first processing unit, configured to obtain target behavior data from a first feed information stream and a first channel in the first feed information stream;

[0020] a second processing unit, configured to obtain target resource data from the first feed information stream and the first channel in the first feed information stream;

[0021] A third processing unit, configured to obtain sample data according to the target behavior data and the target resource data;

[0022] The first model training unit is used to perform model training of cross-domain graph learning according to the sample data to obtain a trained first target model.

[0023] According to another aspect of the present disclosure, a model training device is provided, comprising:

[0024] A data processing unit, configured to obtain target behavior data from a first channel in a first feed information flow;

[0025] A resource processing unit, configured to obtain target resource data from the first channel in the first feed information flow;

[0026] A sample determination unit, configured to obtain sample data according to the target behavior data and the target resource data;

[0027] The second model training unit is used to perform model training of the graph neural network according to the sample data to obtain a trained second target model.

[0028] According to another aspect of the present disclosure, a data recommendation device is provided, comprising:

[0029] An acquisition unit, configured to acquire a touch operation on a first channel in a first feed information stream;

[0030] A response unit is used to respond to the touch operation, and obtain resource data matching the touch operation according to the first target model obtained by the above-mentioned model training method, and / or the second target model obtained by the above-mentioned model training method.

[0031] According to another aspect of the present disclosure, there is provided an electronic device, comprising:

[0032] at least one processor; and

[0033] a memory communicatively connected to the at least one processor; wherein,

[0034] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method provided by any embodiment of the present disclosure.

[0035] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method provided by any embodiment of the present disclosure.

[0036] According to another aspect of the present disclosure, a computer program product is provided, including computer instructions, which implement the method provided by any embodiment of the present disclosure when executed by a processor.

[0037] By adopting the present disclosure, target behavior data can be obtained from the first feed information stream and the first channel in the first feed information stream, and target resource data can be obtained from the first feed information stream and the first channel in the first feed information stream. Sample data is obtained based on the target behavior data and the target resource data, and model training of cross-domain graph learning is performed based on the sample data to obtain a trained first target model, thereby performing data recommendation through the trained first target model, thereby improving the accuracy of data recommendation.

[0038] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0040] Figure 1 is a schematic diagram of a distributed cluster processing scenario according to an embodiment of the present disclosure;

[0041] Figure 2 is a flowchart of a model training method according to an embodiment of the present disclosure;

[0042] Figure 3 is a schematic diagram of a feed stream in an example according to an embodiment of the present disclosure;

[0043] Figure 4 is a schematic diagram of a feed stream in another example according to an embodiment of the present disclosure;

[0044] Figure 5 is a flowchart of another model training method according to an embodiment of the present disclosure;

[0045] Figure 6 is a flowchart of a data recommendation method according to an embodiment of the present disclosure;

[0046] Figure 7 is a schematic diagram of the structure of a model training device according to an embodiment of the present disclosure;

[0047] Figure 8 is a schematic diagram of the composition structure of another model training device according to an embodiment of the present disclosure;

[0048] Fig. 9 is a schematic diagram of the structure of a data recommendation device according to an embodiment of the present disclosure;

[0049] Fig.10 It is a block diagram of an electronic device used to implement the model training method or data recommendation method of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0050] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0051] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there may be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The term "at least one" in this article means any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C. The terms "first" and "second" in this article refer to multiple similar technical terms and distinguish them. They do not mean to limit the order or to limit them to only two. For example, the first feature and the second feature refer to two types / two features. The first feature can be one or more, and the second feature can also be one or more.

[0052] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0053] Feed information flow is a form of data that continuously provides content to users. It is a resource aggregator composed of multiple content sources. Users can actively subscribe to the message source and push data content to users.

[0054] The feed information stream can be the feed information stream in the main feed interface of the client, or the feed information stream under one or several specific channels in the main feed interface of the client. For one or several specific channels that are newly launched, such as women's columns, sports, etc., most of them use the feed information stream in the main feed interface as sample data for model training. For example, a large amount of behavioral data based on active users loaded directly in the main feed interface can be used as the sample data. This is different from the user distribution for one or several specific channels, and the resource distribution is also different, resulting in low accuracy of the model trained based on the sample data. Taking the field of data recommendation as an example, low model accuracy will lead to low accuracy of data recommendation based on the model. For another example, for one or several specific channels, an inverted index is established based on the labels and categories of all resources in the current scenario to recommend data through inverted retrieval based on user interests. This is useful in the initial stage when data collection is needed, but in the long run, the accuracy of data recommendation achieved through the inverted index is also low. In the scenario of matching polysemous words, the "apple" in fruit is different from the "apple" in mobile phone, which can easily introduce bad cases.

[0055] In summary, in order to improve the accuracy of model training, it is necessary to expand the current sample data or introduce more sample data for one or several specific channels.

[0056] According to an embodiment of the present disclosure, Figure 1 This is a schematic diagram of a distributed cluster processing scenario according to an embodiment of the present disclosure. The distributed cluster system is an example of a cluster system. It exemplarily describes how to use the distributed cluster system for model training. The present disclosure is not limited to model training on a single machine or multiple machines. Distributed processing can further improve the accuracy of model training. Figure 1As shown, the distributed cluster system includes multiple nodes (such as server cluster 101, server 102, server cluster 103, server 104, server 105, and server 105 can also be connected to electronic devices, such as mobile phone 1051 and desktop 1052), and one or more model training tasks can be performed together between multiple nodes, and between multiple nodes and connected electronic devices. Optionally, multiple nodes in the distributed cluster system can adopt a data-parallel model training method, and multiple nodes can perform model training tasks based on the same training method to better train the model; if multiple nodes in the distributed cluster system adopt a model-parallel model training method, multiple nodes can perform model training tasks based on different training methods to better train the model. Optionally, after each round of model training is completed, data exchange (such as data synchronization) can be performed between multiple nodes.

[0057] According to an embodiment of the present disclosure, a model training method is provided. Figure 2 It is a flowchart of the model training method according to the embodiment of the present disclosure. The method can be applied to a model training device. For example, the device can be deployed in a terminal or server or other processing device in a single machine, multiple machines or cluster system to implement model training and other processing. The terminal can be a user equipment (UE, User Equipment), a mobile device, a personal digital assistant (PDA, Personal Digital Assistant), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the method can also be implemented by a processor calling computer-readable instructions stored in a memory. For example Figure 2 As shown, this method is applied to Figure 1 Any node or electronic device (mobile phone or desktop computer, etc.) in the cluster system shown includes:

[0058] S201. Acquire target behavior data from a first feed information stream and a first channel in the first feed information stream.

[0059] S202: Acquire target resource data from the first feed information stream and a first channel in the first feed information stream.

[0060] S203: Obtain sample data according to the target behavior data and the target resource data.

[0061] S204: Perform cross-domain graph learning model training based on the sample data to obtain a trained first target model.

[0062] In an example of S201-S204, target behavior data can be obtained from the first feed information stream and the first channel in the first feed information stream (such as women's columns, sports, news, etc.) (for example, a user clicks on an article on clothing matching, a sports video, a news report, etc. that he is interested in, and then can obtain knowledge related to the clothing matching, play the sports video, or read the news, etc.), and target resource data can be obtained from the first feed information stream and the first channel in the first feed information stream (the form of target resource data can be diversified, such as graphic information, dynamic video information, short videos of 5-10 minutes, short videos within 1 minute, etc.). Sample data is obtained based on the target behavior data and the target resource data, and model training of cross-domain graph learning is performed based on the sample data to obtain the trained first target model.

[0063] By adopting the embodiment of the present disclosure, target behavior data can be obtained from the first feed information stream and the first channel in the first feed information stream, and target resource data can be obtained from the first feed information stream and the first channel in the first feed information stream. Sample data is obtained based on the target behavior data and the target resource data, and model training of cross-domain graph learning is performed based on the sample data to obtain the trained first target model, thereby improving the accuracy of model training.

[0064] In one embodiment, model training of cross-domain graph learning is performed according to sample data to obtain a trained first target model, including: obtaining a heterogeneous graph according to the sample data, and performing model training of the cross-domain graph learning according to the heterogeneous graph to obtain the trained first target model. With this embodiment, model training of the cross-domain graph learning can be performed according to the heterogeneous graph, thereby improving the accuracy of model training.

[0065] In one embodiment, the first feed information stream and the first channel in the first feed information stream obtain target behavior data, including: obtaining the first behavior data matching the target user in the first feed information stream, determining all user behavior data matching the first channel in the first feed information stream as second behavior data, and obtaining the target behavior data based on the first behavior data and the second behavior data. With this embodiment, the target behavior data obtained by combining the first feed information stream and the multiple behavior data of the first channel in the first feed information stream can more accurately describe the user behavior.

[0066] In one embodiment, obtaining target resource data from a first feed information stream and a first channel in the first feed information stream includes: obtaining first resource data in the first feed information stream that matches the target resource data, determining resource data associated with all user behavior data in the first feed information stream that matches the first channel as second resource data, and obtaining the target resource data based on the first resource data and the second resource data. With this embodiment, the target resource data obtained by combining the first feed information stream and multiple resource data of the first channel in the first feed information stream can more accurately describe the resource type that matches the user behavior.

[0067] In one embodiment, obtaining the first behavior data matching the target user in the first feed information flow includes: obtaining the first behavior data matching the target user in the first feed information flow according to the user characteristics. The user characteristics include: at least one of the user ID, the user gender, the user age, the city where the user is located, the user's historical behavior, and the scene where the user is located. In this embodiment, the first feed information flow (main feed) is screened based on the matching of the user characteristics, thereby obtaining the first behavior data matching the target user in the first feed information flow, and then the first behavior data matching the target user is screened out from all user behavior data based on the user characteristics.

[0068] In one embodiment, all user behavior data matching the first channel in the first feed information stream is determined as the second behavior data, including: in response to a touch operation on the first channel, switching from the first feed stream interface to which the first feed information stream belongs to the second feed stream interface to which the first channel belongs, and determining all user behavior data matching the first channel collected based on the second feed stream interface as the second behavior data. In this embodiment, for one or several specific first channels (such as women's columns, sports, news, etc.), for all users of the first channel, all user behavior data matching the first channel collected after switching from the first feed information stream (main feed) interface to the feed stream interface where the first channel is located are the second behavior data.

[0069] Figure 3-4 is a schematic diagram of a feed stream in an example according to an embodiment of the present disclosure, such as Figure 3 As shown, it includes: a main feed stream interface 301, which includes multiple channels. Click the "Women's Column" channel to switch to the feed stream interface 303 where the channel is located. The feed stream interface 303 where the channel is located includes data of multiple resource types recommended to users, which can be graphic information, dynamic video information, short videos of 5-10 minutes, short videos within 1 minute, etc. Figure 4 As shown, the main feed stream interface 401 includes multiple channels and multiple feed information streams 403 in the main feed stream interface 401. Click on the "Women's Column" channel to switch to the feed stream interface 404 where the channel is located. The feed stream interface 404 where the channel is located includes data of various resource types recommended to users, which can be graphic information, dynamic video information, short videos of 5-10 minutes, short videos within 1 minute, etc.

[0070] In one embodiment, obtaining first resource data that matches target resource data in the first feed information stream includes: obtaining first resource data that matches target resource data in the first feed information stream according to resource characteristics. The resource characteristics include at least one of user ID, resource type, and type of content provider that produces the resource. In this embodiment, for the first feed information stream (main feed), screening is performed based on matching of resource characteristics (resource characteristics can be represented by an identifier distributed in the current first channel), thereby obtaining first resource data that matches the target user in the first feed information stream, and then filtering out first resource data that matches the target resource from resource data related to all user behaviors based on the resource characteristics. The content provider that produces the resource, for example, may be the author of the original article, for example, when the resource is an original article.

[0071] In one embodiment, resource data associated with all user behaviors matching the first channel in the first feed information stream is determined as second resource data, including: in response to a touch operation on the first channel, switching from the first feed stream interface to which the first feed information stream belongs to the second feed stream interface to which the first channel belongs, acquiring all user behavior data matching the first channel collected based on the second feed stream interface, and determining resource data associated with all user behavior data matching the first channel as the second resource data. With this embodiment, for one or several specific first channels (such as women's columns, sports, news, etc.), for all users of the first channel, after switching from the first feed information stream (main feed) interface to the feed stream interface where the channel is located, the resource data associated with all user behavior data matching the first channel collected are all second resource data.

[0072] In one embodiment, a heterogeneous graph is obtained based on sample data, including: obtaining a user node based on target behavior data, obtaining a resource node based on target resource data, determining that the target user has a touch behavior on the target resource, the target behavior data is associated with the target resource data, connecting the user node and the resource node to obtain an edge, and obtaining the heterogeneous graph based on the user node, the resource node and the edge. With this embodiment, the heterogeneous graph can be obtained based on the user node, the resource node and the edge, and the model training of the cross-domain graph learning is performed based on the heterogeneous graph, thereby improving the accuracy of the model training.

[0073] In one embodiment, model training for cross-domain graph learning is performed according to a heterogeneous graph to obtain a first target model after training, including: starting from a user node, establishing at least one first node link with each node in the heterogeneous graph, obtaining a first sequence according to at least one first node link, starting from a resource node, establishing at least one second node link with each node in the heterogeneous graph, and obtaining a second sequence according to at least one second node link. Model training for cross-domain graph learning is performed according to the first sequence and the second sequence to obtain the first target model after training. Using this embodiment, the first sequence and the second sequence can be obtained based on the heterogeneous graph, and the model training for the cross-domain graph learning (such as the Skip-Gram method) is performed according to the first sequence and the second sequence, thereby improving the accuracy of model training.

[0074] In one example, starting from a user node, a certain probability is used to wander to one of its neighbor nodes (resource nodes), and then wander to the neighbor nodes (user nodes) of the resource node with a certain probability, and so on, a wandering sequence with the user as the starting point is obtained. Wandering several times from the same starting point, several wandering sequences A are obtained, and the wandering sequence A is recorded as the first sequence. Similarly, starting from a resource node, a certain probability is used to wander to one of its neighbor nodes (user nodes), and then wander to the neighbor nodes (resource nodes) of the user node with a certain probability, and so on, a wandering sequence with the resource as the starting point is obtained, and wandering several times from the same starting point, several wandering sequences B are obtained, and the wandering sequence B is recorded as the second sequence.

[0075] Furthermore, the user features (the user features are in the form of node vectors) and model parameters output by the above model can be saved in the cluster with timeliness to ensure that the trained model is applied to the application scenario of the current channel and matches the application scenario of the current channel, thereby improving the accuracy of data recommendation.

[0076] Furthermore, a resource vector index can be constructed to obtain four types of resource indexes. In the process of using the model, for the "user-resource recall" scenario, for the current user, the user characteristics of the current user are obtained through the above model, and the four types of resource indexes are matched according to the user characteristics of the current user, thereby obtaining the resource type matching the current user.

[0077] Furthermore, a user vector index can be constructed by obtaining the vectors of active users of the channel in the recent period of time and then constructing the user vector index. In the process of using the model, for the "user-user recall" scenario, for the current user, the user characteristics of the current user are obtained through the above model, and the user vector index is matched according to the user characteristics of the current user, so as to obtain similar users matching the current user.

[0078] Furthermore, a resource-resource cache can be constructed to obtain the full resource vector every hour as the key; use the vector of each key to find the resource vector index and obtain the top K resources that best match it (K can be 100, 200, 300, 400, etc.) as the value; cache the key and value in pairs as a resource-resource cache. In the process of using the model, for the "resource-resource recall" scenario, for the current user, the user characteristics of the current user are obtained through the above model. In the case where the user characteristics of the current user describe the user's click history, the resource list that the user has clicked is used to batch search the resource-resource cache to obtain similar resources that match the current user's resources.

[0079] According to an embodiment of the present disclosure, a model training method is provided. Figure 5 It is a flowchart of the model training method according to the embodiment of the present disclosure. The method can be applied to a model training device. For example, the device can be deployed in a terminal or server or other processing device in a single machine, multiple machines or cluster system to implement model training and other processing. The terminal can be a user equipment (UE, User Equipment), a mobile device, a personal digital assistant (PDA, Personal Digital Assistant), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the method can also be implemented by a processor calling computer-readable instructions stored in a memory. For example Figure 5 As shown, this method is applied to Figure 1 Any node or electronic device (mobile phone or desktop computer, etc.) in the cluster system shown includes:

[0080] S501: Acquire target behavior data from a first channel in a first feed information flow.

[0081] S502: Acquire target resource data from the first channel in the first feed information flow.

[0082] S503: Obtain sample data according to the target behavior data and the target resource data.

[0083] S504: Perform model training of the graph neural network based on the sample data to obtain a trained second target model.

[0084] In an example of S501-S504, target behavior data can be obtained from the first channel in the first feed information flow (such as women's columns, sports, news, etc.) (for example, a user clicks on an article on clothing matching, a sports video, a news report, etc. that he is interested in, and then can obtain knowledge related to the clothing matching, play the sports video, or read the news, etc.), and target resource data can be obtained from the first channel in the first feed information flow (the form of target resource data can be diversified, such as graphic information, dynamic video information, short videos of 5-10 minutes, short videos within 1 minute, etc.). Sample data is obtained based on the target behavior data and the target resource data, and model training of cross-domain graph learning is performed based on the sample data to obtain a trained second target model.

[0085] By adopting the embodiment of the present disclosure, target behavior data can be obtained from the first channel in the first feed information stream, and target resource data can be obtained from the first channel in the first feed information stream. Sample data is obtained based on the target behavior data and the target resource data, and model training of cross-domain graph learning is performed based on the sample data to obtain a trained second target model, thereby improving the accuracy of model training.

[0086] In one embodiment, a graph neural network model is trained based on sample data to obtain a trained second target model, including: obtaining a heterogeneous graph based on the sample data, and performing model training of the graph neural network based on the heterogeneous graph to obtain a trained second target model. With this embodiment, the model training of the cross-domain graph learning can be performed based on the heterogeneous graph, thereby improving the accuracy of model training.

[0087] In one embodiment, obtaining target behavior data from a first channel in a first feed information stream includes: determining all user behavior data in the first feed information stream that matches the first channel as the target behavior data. In this embodiment, all user behavior data in the first feed information stream that matches the first channel is determined as the target behavior data, which can more accurately describe user behavior and accelerate the iteration speed of model training.

[0088] In one embodiment, all user behavior data matching the first channel in the first feed information stream is determined as target behavior data, including: in response to a touch operation on the first channel, switching from the first feed stream interface to which the first feed information stream belongs to the second feed stream interface to which the first channel belongs, and determining all user behavior data matching the first channel collected based on the second feed stream interface as the target behavior data. With this embodiment, for one or several specific first channels (such as women's columns, sports, news, etc.), for all users of the first channel, all user behavior data matching the first channel collected after switching from the first feed information stream (main feed) interface to the feed stream interface where the first channel is located are target behavior data.

[0089] In one embodiment, obtaining target resource data from a first channel in a first feed information stream includes: determining resource data associated with all user behavior data matching the first channel in the first feed information stream as the target resource data. In this embodiment, determining resource data associated with all user behavior data matching the first channel in the first feed information stream as target behavior data can more accurately describe resource types and accelerate the iteration speed of model training.

[0090] In one embodiment, resource data associated with all user behavior data matching the first channel in the first feed information stream is determined as target resource data, including: in response to a touch operation on the first channel, switching from the first feed stream interface to which the first feed information stream belongs to the second feed stream interface to which the first channel belongs. Obtain all user behavior data matching the first channel collected based on the second feed stream interface, and determine the resource data associated with all user behavior data matching the first channel as the target resource data. With this embodiment, for one or several specific first channels (such as women's columns, sports, news, etc.), for all users of the first channel, after switching from the first feed information stream (main feed) interface to the feed stream interface where the channel is located, the resource data associated with all user behavior data matching the first channel collected are all target resource data.

[0091] In one embodiment, a heterogeneous graph is obtained based on sample data, including: obtaining a user node based on target behavior data, obtaining a resource node based on target resource data, determining that the target user has a touch behavior on the target resource, the target behavior data is associated with the target resource data, connecting the user node and the resource node to obtain an edge, and obtaining the heterogeneous graph based on the user node, the resource node and the edge. With this embodiment, the heterogeneous graph can be obtained based on the user node, the resource node and the edge, and the model training of the cross-domain graph learning is performed based on the heterogeneous graph, thereby improving the accuracy of the model training.

[0092] In one embodiment, a graph neural network model is trained according to a heterogeneous graph to obtain a trained second target model, including: starting from the user node, establishing at least one first node link with each node in the heterogeneous graph, obtaining a first sequence according to the at least one first node link, starting from the resource node, establishing at least one second node link with each node in the heterogeneous graph, obtaining a second sequence according to the at least one second node link, and training the graph neural network model according to the first sequence and the second sequence to obtain the trained second target model. By adopting this embodiment, the first sequence and the second sequence can be obtained based on the heterogeneous graph, and the graph neural network model training can be performed according to the first sequence and the second sequence, thereby improving the accuracy of model training.

[0093] In one example, starting from a user node, a certain probability is used to wander to one of its neighbor nodes (resource nodes), and then wander to the neighbor nodes (user nodes) of the resource node with a certain probability, and so on, a wandering sequence with the user as the starting point is obtained. Wandering several times from the same starting point, several wandering sequences A are obtained, and the wandering sequence A is recorded as the first sequence. Similarly, starting from a resource node, a certain probability is used to wander to one of its neighbor nodes (user nodes), and then wander to the neighbor nodes (resource nodes) of the user node with a certain probability, and so on, a wandering sequence with the resource as the starting point is obtained, and wandering several times from the same starting point, several wandering sequences B are obtained, and the wandering sequence B is recorded as the second sequence.

[0094] Furthermore, the user features (the user features are in the form of node vectors) and model parameters output by the above model can be saved in the cluster with timeliness to ensure that the trained model is applied to the application scenario of the current channel and matches the application scenario of the current channel, thereby improving the accuracy of data recommendation. Since the amount of sample data for the second target model during the model training process is relatively small compared to the first target model training, multiple rounds of training can be performed using daily data to further improve the model accuracy.

[0095] Furthermore, a resource vector index can be constructed to obtain four types of resource indexes. In the process of using the model, for the "user-resource recall" scenario, for the current user, the user characteristics of the current user are obtained through the above model, and the four types of resource indexes are matched according to the user characteristics of the current user, thereby obtaining the resource type matching the current user.

[0096] Furthermore, a user vector index can be constructed by obtaining the vectors of active users of the channel in the recent period of time and then constructing the user vector index. In the process of using the model, for the "user-user recall" scenario, for the current user, the user characteristics of the current user are obtained through the above model, and the user vector index is matched according to the user characteristics of the current user, so as to obtain similar users matching the current user.

[0097] Furthermore, a resource-resource cache can be constructed to obtain the full resource vector every hour as the key; use the vector of each key to find the resource vector index and obtain the top K resources that best match it (K can be 100, 200, 300, 400, etc.) as the value; cache the key and value in pairs as a resource-resource cache. In the process of using the model, for the "resource-resource recall" scenario, for the current user, the user characteristics of the current user are obtained through the above model. In the case where the user characteristics of the current user describe the user's click history, the resource list that the user has clicked is used to batch search the resource-resource cache to obtain similar resources that match the current user's resources.

[0098] According to an embodiment of the present disclosure, a data recommendation method is provided. Figure 6 This is a flowchart of a data recommendation method according to an embodiment of the present disclosure. The method can be applied to a data recommendation device. For example, the device can be deployed in a terminal or server or other processing device in a single machine, multiple machines or cluster system to implement model training and other processing. The terminal can be a user equipment (UE, User Equipment), a mobile device, a personal digital assistant (PDA, Personal Digital Assistant), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the method can also be implemented by a processor calling computer-readable instructions stored in a memory. Figure 6 As shown, it is applied to terminal equipment, including:

[0099] S601: Acquire a touch operation on a first channel in a first feed information flow.

[0100] S602: In response to the touch operation, obtain resource data matching the touch operation according to the first target model obtained in any one of the above embodiments and / or the second target model obtained in any one of the above embodiments.

[0101] In an example of S601-S602, a touch operation on a first channel in a first feed information stream can be obtained, and in response to the touch operation, resource data matching the touch operation (such as graphic and text information, dynamic video information, short videos of 5-10 minutes, short videos within 1 minute, etc.) can be obtained according to the first target model obtained in any of the above embodiments (such as the first target model obtained by model training of the cross-domain graph learning) and / or the second target model obtained in any of the above embodiments (such as the second target model obtained by model training of the neural network).

[0102] By adopting the embodiments of the present disclosure, the first target model and / or the second target model obtained by the above-mentioned model training can be used to recommend corresponding resources to the user on the client based on the output results of the first target model and / or the second target model (the outputs of these two target models can both be user features), thereby realizing user-resource recall, user-user recall, and resource-resource recall. Therefore, data recommendation is performed through the trained first target model and / or the second target model, thereby improving the accuracy of data recommendation.

[0103] According to an embodiment of the present disclosure, a model training device is provided. Figure 7 is a schematic diagram of the structure of the model training device according to an embodiment of the present disclosure, such as Figure 7 As shown, the model training device 700 includes: a first processing unit 701, used to obtain target behavior data from a first feed information stream and a first channel in the first feed information stream; a second processing unit 702, used to obtain target resource data from the first feed information stream and the first channel in the first feed information stream; a third processing unit 703, used to obtain sample data based on the target behavior data and the target resource data; a first model training unit 704, used to perform model training for cross-domain graph learning based on the sample data to obtain a trained first target model.

[0104] In one embodiment, the first model training unit is used to obtain a heterogeneous graph based on the sample data; perform model training for the cross-domain graph learning based on the heterogeneous graph to obtain a trained first target model.

[0105] In one embodiment, the first processing unit is used to obtain first behavior data matching a target user in a first feed information stream; determine all user behavior data matching a first channel in the first feed information stream as second behavior data; and obtain the target behavior data based on the first behavior data and the second behavior data.

[0106] In one embodiment, the second processing unit is used to obtain first resource data matching the target resource data in the first feed information stream; determine the resource data associated with all user behavior data matching the first channel in the first feed information stream as second resource data; and obtain the target resource data based on the first resource data and the second resource data.

[0107] In one embodiment, the first processing unit is used to obtain first behavior data matching the target user in the first feed information flow based on user characteristics; wherein the user characteristics include: at least one of: user ID, user gender, user age, user city, user historical behavior, and user scenario.

[0108] In one embodiment, the first processing unit is used to respond to a touch operation on the first channel, switch from a first feed stream interface to which the first feed information stream belongs to a second feed stream interface to which the first channel belongs; and determine all user behavior data matching the first channel collected based on the second feed stream interface as the second behavior data.

[0109] In one embodiment, the second processing unit is used to obtain first resource data matching the target resource data in the first feed information flow according to resource characteristics; wherein the resource characteristics include at least one of: user ID, resource type, and type of content provider producing the resource.

[0110] In one embodiment, the second processing unit is used to respond to a touch operation on the first channel, switch from a first feed stream interface to which the first feed information stream belongs to a second feed stream interface to which the first channel belongs; obtain all user behavior data collected based on the second feed stream interface and matching the first channel; and determine resource data associated with all user behavior data matching the first channel as the second resource data.

[0111] In one embodiment, the first model training unit is used to obtain a user node based on the target behavior data; obtain a resource node based on the target resource data; when it is determined that the target user has a touch behavior on the target resource, the target behavior data is associated with the target resource data, and the user node and the resource node are connected to obtain an edge; and the heterogeneous graph is obtained based on the user node, the resource node and the edge.

[0112] In one embodiment, the first model training unit is used to establish at least one first node link with each node in the heterogeneous graph with the user node as the starting point, and obtain a first sequence based on the at least one first node link; to establish at least one second node link with each node in the heterogeneous graph with the resource node as the starting point, and obtain a second sequence based on the at least one second node link; and to perform model training for the cross-domain graph learning based on the first sequence and the second sequence to obtain a trained first target model.

[0113] According to an embodiment of the present disclosure, a model training device is provided. Figure 8 is a schematic diagram of the structure of the model training device according to an embodiment of the present disclosure, such as Figure 8 As shown, the model training device 800 includes: a data processing unit 801, used to obtain target behavior data from the first channel in the first feed information stream; a resource processing unit 802, used to obtain target resource data from the first channel in the first feed information stream; a sample determination unit 803, used to obtain sample data based on the target behavior data and the target resource data; a second model training unit 804, used to perform model training of the graph neural network based on the sample data to obtain a trained second target model.

[0114] In one embodiment, the second model training unit is used to obtain a heterogeneous graph based on the sample data; perform model training of the graph neural network based on the heterogeneous graph to obtain a trained second target model.

[0115] In one implementation, the data processing unit is used to determine all user behavior data in the first feed information flow that matches the first channel as the target behavior data.

[0116] In one implementation, the resource processing unit is used to determine resource data associated with all user behavior data matching the first channel in the first feed information flow as the target resource data.

[0117] In one embodiment, the second model training unit is used to obtain a user node based on the target behavior data; obtain a resource node based on the target resource data; when it is determined that the target user has a touch behavior on the target resource, the target behavior data is associated with the target resource data, and the user node and the resource node are connected to obtain an edge; and the heterogeneous graph is obtained based on the user node, the resource node and the edge.

[0118] In one embodiment, the second model training unit is used to establish at least one first node link with each node in the heterogeneous graph with the user node as the starting point, and obtain a first sequence based on the at least one first node link; to establish at least one second node link with each node in the heterogeneous graph with the resource node as the starting point, and obtain a second sequence based on the at least one second node link; and to train the model of the graph neural network based on the first sequence and the second sequence to obtain a trained second target model.

[0119] According to an embodiment of the present disclosure, a data recommendation device is provided. Fig. 9 is a schematic diagram of the structure of the model training device according to an embodiment of the present disclosure, such as Fig. 9 As shown, the model training device 900 includes: an acquisition unit 901, used to acquire a touch operation on a first channel in a first feed information stream; a response unit 902, used to respond to the touch operation, and obtain resource data matching the touch operation according to the first target model obtained according to any one of the above embodiments, and / or the second target model obtained according to any one of the above embodiments.

[0120] In one implementation, the resource data includes at least one of graphic information, dynamic video information, short video information, and mini video information.

[0121] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0122] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0123] Fig.10A schematic block diagram of an example electronic device 1000 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0124] like Fig.10 As shown, the electronic device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. In the RAM 1003, various programs and data required for the operation of the electronic device 1000 can also be stored. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0125] Multiple components in the electronic device 1000 are connected to the I / O interface 1005, including: an input unit 1006, such as a keyboard, a mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a disk, an optical disk, etc.; and a communication unit 1009, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1009 allows the electronic device 1000 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0126] The computing unit 1001 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1001 performs the various methods and processes described above, such as a model training method or a data recommendation method. For example, in some embodiments, the model training method or the data recommendation method may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the model training method or the data recommendation method described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to execute the model training method or the data recommendation method in any other suitable manner (e.g., by means of firmware).

[0127] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0128] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0129] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0130] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0131] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0132] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0133] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of this disclosure can be achieved, and this document is not limited here.

[0134] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A model training method, comprising: Acquire target behavior data from a first feed information stream and a first channel in the first feed information stream; Acquire target resource data from the first feed information stream and a first channel in the first feed information stream; Obtaining sample data according to the target behavior data and the target resource data; Performing cross-domain graph learning model training according to the sample data to obtain a trained first target model; The step of performing cross-domain graph learning model training according to the sample data to obtain a trained first target model includes: Obtaining a heterogeneous graph according to the sample data; Performing model training for the cross-domain graph learning according to the heterogeneous graph to obtain a trained first target model; Wherein, obtaining a heterogeneous graph according to the sample data includes: Obtaining a user node according to the target behavior data; Obtaining a resource node according to the target resource data; When it is determined that the target user has a touch behavior on the target resource, the target behavior data is associated with the target resource data, and the user node and the resource node are connected to obtain an edge; Obtaining the heterogeneous graph according to the user nodes, the resource nodes and the edges; The step of performing the model training of the cross-domain graph learning according to the heterogeneous graph to obtain the trained first target model includes: Taking the user node as a starting point, establishing at least one first node link with each node in the heterogeneous graph, and obtaining a first sequence according to the at least one first node link; Taking the resource node as a starting point, establishing at least one second node link with each node in the heterogeneous graph, and obtaining a second sequence according to the at least one second node link; Model training of the cross-domain graph learning is performed according to the first sequence and the second sequence to obtain a trained first target model.

2. The method according to claim 1, wherein: The acquiring target behavior data from the first feed information stream and the first channel in the first feed information stream includes: Obtaining the first behavior data matching the target user in the first feed information flow; Determine all user behavior data matching the first channel in the first feed information flow as second behavior data; The target behavior data is obtained according to the first behavior data and the second behavior data.

3. The method according to claim 1, wherein: The acquiring target resource data from the first feed information stream and the first channel in the first feed information stream includes: Acquire first resource data matching the target resource data in the first feed information flow; Determine resource data associated with all user behavior data matching the first channel in the first feed information flow as second resource data; The target resource data is obtained according to the first resource data and the second resource data.

4. The method according to claim 2, wherein: The obtaining of first behavior data matching the target user in the first feed information flow includes: According to the user characteristics, first behavior data matching the target user in the first feed information flow is obtained; The user characteristics include: at least one of user ID, user gender, user age, user city, user historical behavior, and user scenario.

5. The method according to claim 2, wherein: The step of determining all user behavior data matching the first channel in the first feed information flow as second behavior data includes: In response to a touch operation on the first channel, switching from a first feed stream interface to which the first feed information stream belongs to a second feed stream interface to which the first channel belongs; All user behavior data collected based on the second feed stream interface and matching the first channel are determined as the second behavior data.

6. The method according to claim 3, wherein: The obtaining of first resource data matching the target resource data in the first feed information flow includes: According to the resource characteristics, first resource data matching the target resource data in the first feed information flow is obtained; The resource characteristics include: at least one of: user ID, resource type, and type of content provider producing the resource.

7. The method according to claim 3, wherein: The step of determining the resource data associated with all user behaviors matching the first channel in the first feed information flow as the second resource data includes: In response to a touch operation on the first channel, switching from a first feed stream interface to which the first feed information stream belongs to a second feed stream interface to which the first channel belongs; Acquire all user behavior data matching the first channel collected based on the second feed stream interface; Resource data associated with all user behavior data matching the first channel is determined as the second resource data.

8. A model training method, comprising: Obtain target behavior data from a first channel in a first feed information flow; Acquire target resource data from the first channel in the first feed information flow; Obtaining sample data according to the target behavior data and the target resource data; Performing model training of a graph neural network according to the sample data to obtain a trained second target model; The step of performing cross-domain graph learning model training according to the sample data to obtain a trained second target model includes: Obtaining a heterogeneous graph according to the sample data; Performing model training for the cross-domain graph learning according to the heterogeneous graph to obtain a trained second target model; Wherein, obtaining a heterogeneous graph according to the sample data includes: Obtaining a user node according to the target behavior data; Obtaining a resource node according to the target resource data; When it is determined that the target user has a touch behavior on the target resource, the target behavior data is associated with the target resource data, and the user node and the resource node are connected to obtain an edge; Obtaining the heterogeneous graph according to the user nodes, the resource nodes and the edges; The step of performing model training of the graph neural network according to the heterogeneous graph to obtain a trained second target model includes: Taking the user node as a starting point, establishing at least one first node link with each node in the heterogeneous graph, and obtaining a first sequence according to the at least one first node link; Taking the resource node as a starting point, establishing at least one second node link with each node in the heterogeneous graph, and obtaining a second sequence according to the at least one second node link; According to the model training of the graph neural network of the first sequence and the second sequence, a trained second target model is obtained.

9. The method according to claim 8, wherein: The step of acquiring target behavior data from a first channel in a first feed information flow includes: All user behavior data matching the first channel in the first feed information flow are determined as the target behavior data.

10. The method according to claim 8, wherein: The acquiring target resource data from the first channel in the first feed information flow includes: The resource data associated with all user behavior data matching the first channel in the first feed information flow is determined as the target resource data.

11. A data recommendation method, comprising: Obtaining touch operations on a first channel in a first feed information stream; In response to the touch operation, resource data matching the touch operation is obtained according to the first target model obtained according to any one of claims 1 to 7 above, and / or the second target model obtained according to any one of claims 8 to 10 above.

12. The method according to claim 11, wherein: The resource data includes: at least one of graphic information, dynamic video information, short video information, and mini video information.

13. A model training device, comprising: A first processing unit, configured to obtain target behavior data from a first feed information stream and a first channel in the first feed information stream; a second processing unit, configured to obtain target resource data from the first feed information stream and the first channel in the first feed information stream; A third processing unit, configured to obtain sample data according to the target behavior data and the target resource data; A first model training unit, used for performing model training of cross-domain graph learning according to the sample data to obtain a trained first target model; The step of performing cross-domain graph learning model training according to the sample data to obtain a trained first target model includes: Obtaining a heterogeneous graph according to the sample data; Performing model training for the cross-domain graph learning according to the heterogeneous graph to obtain a trained first target model; Wherein, obtaining a heterogeneous graph according to the sample data includes: Obtaining a user node according to the target behavior data; Obtaining a resource node according to the target resource data; When it is determined that the target user has a touch behavior on the target resource, the target behavior data is associated with the target resource data, and the user node and the resource node are connected to obtain an edge; Obtaining the heterogeneous graph according to the user nodes, the resource nodes and the edges; The step of performing the model training of the cross-domain graph learning according to the heterogeneous graph to obtain the trained first target model includes: Taking the user node as a starting point, establishing at least one first node link with each node in the heterogeneous graph, and obtaining a first sequence according to the at least one first node link; Taking the resource node as a starting point, establishing at least one second node link with each node in the heterogeneous graph, and obtaining a second sequence according to the at least one second node link; Model training of the cross-domain graph learning is performed according to the first sequence and the second sequence to obtain a trained first target model.

14. A model training device, comprising: A data processing unit, configured to obtain target behavior data from a first channel in a first feed information flow; A resource processing unit, configured to obtain target resource data from the first channel in the first feed information flow; A sample determination unit, configured to obtain sample data according to the target behavior data and the target resource data; A second model training unit, used to perform model training of the graph neural network according to the sample data to obtain a trained second target model; The step of performing cross-domain graph learning model training according to the sample data to obtain a trained second target model includes: Obtaining a heterogeneous graph according to the sample data; Performing model training for the cross-domain graph learning according to the heterogeneous graph to obtain a trained second target model; Wherein, obtaining a heterogeneous graph according to the sample data includes: Obtaining a user node according to the target behavior data; Obtaining a resource node according to the target resource data; When it is determined that the target user has a touch behavior on the target resource, the target behavior data is associated with the target resource data, and the user node and the resource node are connected to obtain an edge; Obtaining the heterogeneous graph according to the user nodes, the resource nodes and the edges; The step of performing model training of the graph neural network according to the heterogeneous graph to obtain a trained second target model includes: Taking the user node as a starting point, establishing at least one first node link with each node in the heterogeneous graph, and obtaining a first sequence according to the at least one first node link; Taking the resource node as a starting point, establishing at least one second node link with each node in the heterogeneous graph, and obtaining a second sequence according to the at least one second node link; According to the model training of the graph neural network of the first sequence and the second sequence, a trained second target model is obtained.

15. A data recommendation device, comprising: An acquisition unit, configured to acquire a touch operation on a first channel in a first feed information stream; A response unit is used to respond to the touch operation, and obtain resource data matching the touch operation according to the first target model obtained according to any one of claims 1 to 7 and / or the second target model obtained according to any one of claims 8 to 10.

16. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 12.

17. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-12.

18. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 12.

Images