Resource pushing methods and apparatuses, electronic device, storage medium and program product

Abstract

The present disclosure provides resource pushing methods and apparatuses, an electronic device, a storage medium and a program product, and relates to the technical field of artificial intelligence, in particular to the field of resource pushing. A resource pushing method comprises: determining a first intention of a user for acquiring batch resources; on the basis of the first intention, performing feature analysis on a first trigger subject to obtain first feature information of the first trigger subject; and, on the basis of the first feature information, screening a plurality of candidate resources to obtain a target resource set and pushing the target resource set to the user.

Description

Resource delivery methods and devices, electronic devices, storage media and program products Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, particularly to the field of resource delivery, and more specifically, to a resource delivery method and apparatus, electronic device, storage medium, and program product. Background Technology

[0002] A file is a collection of digital information stored on a terminal device. For example, files can include various forms of data such as text, images, audio, video, and program code. Batch file filtering refers to the automated process of classifying, identifying, and selecting a large number of files in file management for terminal devices. Summary of the Invention

[0003] In view of this, the present disclosure provides a resource push method and apparatus, electronic device, storage medium and program product.

[0004] According to one aspect of this disclosure, a resource push method is provided, comprising: determining a user's first intention to obtain a batch of resources; performing feature analysis on a first triggering entity based on the first intention to obtain first feature information of the first triggering entity; and filtering multiple candidate resources according to the first feature information to obtain a target resource set and pushing the target resource set to the user.

[0005] According to one aspect of this disclosure, a resource push method is provided, comprising: determining a user's second intention to obtain resources; performing feature analysis on a second triggering entity based on the second intention to obtain second feature information of the second triggering entity; and filtering multiple candidate resources based on the second feature information to obtain a target resource and pushing the target resource to the user.

[0006] According to another aspect of this disclosure, a resource push device is provided, comprising: a first determining module, configured to determine a user's first intention to obtain a batch of resources; a first feature analysis module, configured to perform feature analysis on a first triggering entity based on the first intention to obtain first feature information of the first triggering entity; and a first filtering module, configured to filter multiple candidate resources according to the first feature information to obtain a target resource set and push the target resource set to the user.

[0007] According to another aspect of this disclosure, a resource push device is provided, comprising: a second determining module, configured to determine a user's second intention to obtain resources; a second feature analysis module, configured to perform feature analysis on a second triggering entity based on the second intention to obtain second feature information of the second triggering entity; and a second filtering module, configured to filter multiple candidate resources according to the second feature information to obtain a target resource and push the target resource to the user.

[0008] According to another aspect of this disclosure, an electronic device is provided, comprising: one or more processors; and a memory for storing one or more instructions, wherein, when the one or more instructions are executed by the one or more processors, the one or more processors cause the one or more processors to perform the method as described above.

[0009] According to another aspect of this disclosure, a computer-readable storage medium is provided having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the method described above.

[0010] According to another aspect of this disclosure, a computer program product is provided, which includes computer-executable instructions that, when executed, are used to implement the method described above. Attached Figure Description

[0011] The above and other objects, features and advantages of this disclosure will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0012] Figure 1 schematically illustrates a system architecture to which the resource push method can be applied according to an embodiment of the present disclosure;

[0013] Figure 2 schematically illustrates a flowchart of a resource push method according to an embodiment of the present disclosure;

[0014] Figure 3 schematically illustrates an example of a process for determining whether a user is in a preset scenario according to an embodiment of the present disclosure;

[0015] Figure 4 schematically illustrates an example of a process for determining a user's first intent to acquire bulk resources according to an embodiment of the present disclosure;

[0016] Figure 5 schematically illustrates an example of a process in which a first triggering subject is subjected to feature analysis based on a first intent to obtain first feature information of the first triggering subject according to an embodiment of the present disclosure;

[0017] Figure 6A schematically illustrates an example of a process of filtering multiple candidate resources based on first feature information to obtain a target resource set according to an embodiment of the present disclosure.

[0018] Figure 6B schematically illustrates an example of a process for filtering multiple candidate resources based on first feature information to obtain a target resource set according to another embodiment of the present disclosure;

[0019] Figure 6C schematically illustrates an example of a process of filtering multiple candidate resources based on first feature information to obtain a target resource set according to yet another embodiment of the present disclosure.

[0020] Figure 7 schematically illustrates an example of a process in which a first triggering subject is subjected to feature analysis based on a first intent to obtain first feature information of the first triggering subject according to an embodiment of the present disclosure;

[0021] Figure 8 schematically illustrates an example of a process for filtering multiple candidate resources based on first feature information to obtain a target resource set according to another embodiment of the present disclosure;

[0022] Figure 9 schematically illustrates a flowchart of a resource push method according to another embodiment of the present disclosure;

[0023] Figure 10 schematically illustrates an example of a process of filtering multiple candidate resources to obtain a target resource based on second feature information when the second triggering subject includes a second dialogue context or second voice data, according to an embodiment of the present disclosure.

[0024] Figure 11A schematically illustrates an example of a process of filtering multiple candidate resources to obtain a target resource based on second feature information when the second triggering subject includes user operation and a second dialogue context, according to an embodiment of the present disclosure.

[0025] Figure 11B schematically illustrates an example of a process of filtering multiple candidate resources to obtain a target resource based on second feature information when the second triggering subject includes a second user operation and a second dialogue context, according to an embodiment of the present disclosure.

[0026] Figure 12 schematically illustrates a block diagram of a resource push device according to an embodiment of the present disclosure;

[0027] Figure 13 schematically illustrates a block diagram of a resource push device according to another embodiment of the present disclosure; and

[0028] Figure 14 schematically illustrates a block diagram of an electronic device suitable for implementing a resource push method according to an embodiment of the present disclosure. Detailed Implementation

[0029] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0030] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0031] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0032] When using expressions such as "at least one of A, B, and C", they should generally be interpreted in accordance with the meaning that is commonly understood by a person skilled in the art (e.g., "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B, and C, etc.).

[0033] In the technical solution disclosed herein, the acquisition, storage, and application of user personal information comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and there is no violation of public order and good morals.

[0034] In the technical solution disclosed herein, the user's authorization or consent is obtained before acquiring or collecting the user's personal information.

[0035] In the process of batch file filtering, it can be difficult for users unfamiliar with command-line tools or scripts to perform batch filtering using command-line tools or by writing scripts. Furthermore, due to the large time span and sheer number of files, manual file filtering can be time-consuming and inefficient when filtering criteria are unclear or file attributes are complex. Some batch filtering tools typically only support specific file types or formats, limiting the scope and flexibility of the filtering process.

[0036] Therefore, how to accurately identify users' needs for batch file filtering and intelligently and quickly filter out the content that users need is an urgent problem to be solved.

[0037] Therefore, this disclosure provides a resource push method, apparatus, electronic device, storage medium, and program product, relating to the field of artificial intelligence technology, and particularly to the field of resource push. The resource push method includes: determining a user's first intention to obtain a batch of resources; based on the first intention, performing feature analysis on a first triggering entity to obtain first feature information of the first triggering entity; and, based on the first feature information, filtering multiple candidate resources to obtain a target resource set and pushing the target resource set to the user.

[0038] Figure 1 schematically illustrates a system architecture to which the resource push method can be applied according to an embodiment of the present disclosure. It should be noted that Figure 1 is merely an example of a system architecture to which the embodiments of the present disclosure can be applied, to help those skilled in the art understand the technical content of the present disclosure, but does not imply that the embodiments of the present disclosure cannot be used in other devices, systems, environments, or scenarios.

[0039] As shown in Figure 1, the system architecture 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between different devices.

[0040] It should be noted that the resource push method provided in this embodiment can generally be executed by server 105. Accordingly, the resource push device provided in this embodiment can generally be set in server 105.

[0041] Alternatively, the resource push method provided in this embodiment of the present disclosure can also be executed by the first terminal device 101, the second terminal device 102, or the third terminal device 103. Correspondingly, the resource push device provided in this embodiment of the present disclosure can also be disposed in the first terminal device 101, the second terminal device 102, or the third terminal device 103.

[0042] It should be understood that the number of terminal devices, networks, and servers shown in Figure 1 is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0043] It should be noted that the sequence numbers of the operations in the following methods are for descriptive purposes only and should not be considered as indicating the execution order of the operations. Unless explicitly stated otherwise, the method does not need to be executed in the exact order shown.

[0044] The system architecture for applying the resource push method provided in this disclosure has been described above. The resource push process of an embodiment of this disclosure will be further described below with reference to Figure 2.

[0045] Figure 2 schematically illustrates a flowchart of a resource push method according to an embodiment of the present disclosure.

[0046] As shown in Figure 2, the resource push method 200 includes operations S210 to S230.

[0047] In operation S210, determine the user's primary intent to acquire bulk resources.

[0048] In operation S220, based on the first intention, feature analysis is performed on the first triggering subject to obtain the first feature information of the first triggering subject.

[0049] In operation S230, multiple candidate resources are filtered based on the first feature information to obtain a target resource set and push the target resource set to the user.

[0050] The primary intent to acquire bulk resources refers to a user's desire to acquire a set of resources rather than a single resource. For example, a user might want to acquire a set of documents or images related to a specific topic, indicating a primary intent to acquire bulk resources. In one example, when the primary trigger is a selected resource, path analysis of the selection action for that resource can determine whether the user has a primary intent to acquire bulk resources. In another example, when the primary trigger is a first dialogue context or first voice data, intent analysis of the dialogue context or voice data can determine whether the user has a primary intent to acquire bulk resources.

[0051] If it is determined that the user has the primary intention to acquire a batch of resources, feature analysis can be performed on the primary triggering entity based on this primary intention to obtain the primary feature information of the primary triggering entity. Feature analysis refers to the process of identifying and analyzing the key attributes or features of the primary triggering entity. The primary feature information refers to the data obtained through feature analysis, which describes the specific attributes of the triggering entity. In one example, when the primary triggering entity is a selected resource, feature analysis can be performed on different content based on the different types of resources selected by the user to obtain the primary feature information of the primary triggering entity. In another example, when the primary triggering entity is dialogue context or voice data, the primary feature information of the primary triggering entity can be obtained by performing semantic analysis on the dialogue context or voice data.

[0052] Resource types can include at least one of the following: images, files, emails, and chat logs. Images can refer to still images taken with a camera or a video formed by a series of still images played in sequence at high speed. Files refer to the physical storage format of data, such as presentations and spreadsheets. Emails refer to information sent and received through an email system. Chat logs refer to conversations generated during communication through instant messaging software or social media platforms.

[0053] For example, if the user selects an image as the resource type, feature analysis can be performed on the content characteristics of the selected image to obtain first feature information. Alternatively, if the user selects a file as the resource type, feature analysis can be performed on the content characteristics of the selected file to obtain first feature information. Alternatively, if the user selects an email as the resource type, feature analysis can be performed on the content characteristics of the selected email to obtain first feature information. Alternatively, if the user selects chat history as the resource type, feature analysis can be performed on the content characteristics of the selected chat history to obtain first feature information.

[0054] After obtaining the first feature information, multiple candidate resources can be filtered based on the first feature information to obtain a target resource set. During the filtering process based on the first feature information, multiple candidate resources belonging to the same resource type can be filtered according to the resource type of the selected resources, or multiple candidate resources belonging to different resource types can be filtered according to the first feature information obtained from the first dialogue context or first voice data. The target resource set may include at least one target resource that has the same first feature information as the first triggering entity. After obtaining the target resource set, it can be pushed to the user.

[0055] According to embodiments of this disclosure, automatically identifying a user's first intent to acquire a batch of resources helps improve the interactivity and responsiveness of resource push notifications. By performing feature analysis on the first triggering entity based on the identified first intent, the characteristics of the resources needed by the user can be intelligently analyzed. Based on this, by filtering multiple candidate resources according to the analyzed first feature information, the resources needed by the user can be automatically selected and proactively pushed to the user. Thus, users no longer need to use search functions or search for resources one by one, saving time spent on resource filtering and improving the automation and accuracy of resource push notifications. This allows for accurate identification of the user's intent to acquire a batch of resources and intelligent and rapid filtering of the target resource set required by the user.

[0056] In embodiments of this disclosure, the triggering entity refers to the source or entity of the triggering information. The triggering entity may include at least one of the following: a selected resource in response to a user's selection action, dialogue context, and voice data. In one example, a user's selection action refers to a decision and action made by a user when interacting with software, applications, websites, or other digital interfaces. The selection action may include at least one of the following: clicking a link or icon of a resource to be selected on the interface using a mouse or touchscreen; typing information about a resource to be selected in a text box or search bar; selecting one or more resources to be selected from a drop-down menu or tab; selecting a resource to be selected on a touchscreen device using a swipe gesture; or selecting a resource to be selected by dragging it to a specific location on the interface. In another example, dialogue context refers to the information and environmental background provided by the communication that has already occurred in the dialogue, which helps the server understand the current communication content and intent. Voice data refers to sound signals captured and processed by a terminal device.

[0057] According to embodiments of this disclosure, by comprehensively monitoring the user's selection operation, the user's dialogue context, or the user's voice data, the triggering subject can be determined in a diversified manner, achieving accurate capture and rapid response of the user's intent, which helps to improve the intelligence level of resource push.

[0058] In the embodiments of this disclosure, before determining the user's first intention, it can be determined whether the user is in a preset scenario. The process of determining whether the user is in a preset scenario is illustrated below with Figure 3 as an example.

[0059] Figure 3 schematically illustrates an example of a process for determining whether a user is in a preset scenario according to an embodiment of the present disclosure.

[0060] As shown in Figure 3, taking the trigger subject 301 as the selected resource in 300 as an example, the process of determining whether the user is in the preset scenario is explained.

[0061] The selected resource may have a resource type 303, for example, resource type 303 may include at least one of images, files, emails, and chat logs. The preset scenario may include a set of preset operations 304 for each of multiple resource types 303, and this set of preset operations 304 may be pre-configured for different resource types 303. For each resource type 303, the set of preset operations 304 may include at least one preset operation. The at least one preset operation may be a preset operation for different terminal devices or a preset operation for the same terminal device. The terminal device may include a mobile phone and a computer.

[0062] During the execution of the resource push method provided in this disclosure, the triggering entity 301 can be monitored. Upon detecting a selection operation 302, it can be determined whether the user is in a preset scenario based on the selection operation 302. For example, based on the resource type 303 of the triggering entity 301, a target operation set 305 can be determined from the preset operation sets 304 of each of the multiple resource types 303. The target operation set 305 may include at least one preset operation; for example, the target operation set 305 may include preset operation 305_1, preset operation 305_2, ..., preset operation 305_M, where M is a positive integer.

[0063] After determining the target operation set 305, operation S310 can be executed. In operation S310, it can be determined whether at least one preset operation in the target operation set 305 contains a target operation that matches the selected operation 302. If yes, it can be determined that the user is in a preset scenario 306. If no, it can be determined that the user is not in a preset scenario 307. The following examples, using resource type 303 as images, files, emails, and chat logs, further illustrate the process of determining whether the user is in a preset scenario based on the selected operation.

[0064] According to embodiments of this disclosure, by presetting multiple sets of preset operations corresponding to each resource type, determining a target set of operations based on the selected operation, and matching the selected operation with at least one preset operation in the target set, the system determines whether the user is in a preset scenario based on whether there is a target operation in at least one preset operation that matches the selected operation. This improves the accuracy and reliability of scenario recognition.

[0065] For example, when resource type 303 is images, preset operations for mobile devices could include long-pressing in the album to select multiple images, or opening the album directly in the chat window to support multiple selection. Preset operations for computers could include using the shortcut Ctrl to select multiple images from a folder, or opening the image folder in the chat window using the shortcut Ctrl to select multiple images. Selection operation 302 for mobile devices could include whether to open the mobile device's album or whether to open the album through the mobile chat window. Selection operation 302 for computers could include whether to open the computer's image folder or whether to open the image folder through the computer's chat window. In the above examples, if a target operation matching selection operation 302 exists among the preset operations for images, it can be determined that the user is in preset scenario 306.

[0066] Alternatively, when resource type 303 is "file," preset operations on the computer may include using the Ctrl key to select multiple files within a folder, or opening a folder directly from the chat window. Selection operations on the computer 302 may include whether to open a folder on the computer, or whether to open a folder through the chat window. In the above example, if any of the preset operations for files matches the target operation of selection operation 302, then the user is determined to be in preset scenario 306.

[0067] Alternatively, if resource type 303 is email, multiple email addresses can be selected directly. Selection action 302 can include whether to open the email account. In the above example, if a target action matching selection action 302 exists among the preset actions for email, then the user is determined to be in preset scenario 306.

[0068] Alternatively, when the resource type is chat history, the preset operation for mobile devices may include long-pressing in the chat window to select multiple items; the preset operation for computers may include right-clicking in the chat window to select multiple items. The selection operation 302 for mobile devices may include whether the user is in a mobile chat window; the selection operation 302 for computers may include whether the user is in a computer chat window. In the above example, if a target operation matching selection operation 302 exists among the preset operations for chat history, it can be determined that the user is in preset scenario 306.

[0069] In one example, if it is determined that the user is not in the preset scenario 307, the resource push method provided in this disclosure may not be executed further. If it is determined that the user is in the preset scenario 306, it can be further determined whether the user intends to obtain a batch of resources based on the selection operation 302.

[0070] According to embodiments of this disclosure, by determining whether a user is in a preset scenario, and if the user is in a preset scenario, further determining whether the user has an intent, this intelligent processing method not only improves response speed but also helps to improve the efficiency and accuracy of subsequent resource push services because it can proactively identify the user's scenario and user intent.

[0071] In the embodiments of this disclosure, after determining that the user is in a preset scenario, the user's first intention to obtain batch resources can be determined based on the user's selection operation. The process of determining the user's first intention to obtain batch resources based on the selection operation is illustrated below with Figure 4 as an example.

[0072] Figure 4 schematically illustrates an example of a process for determining a user’s first intent to acquire bulk resources according to an embodiment of the present disclosure.

[0073] As shown in Figure 4, taking the selected resource as the trigger subject in 400 as an example, the process of determining whether the user has the first intention to obtain a batch of resources is explained.

[0074] The selected resource can have a resource type 403, for example, resource type 403 can include at least one of images, files, emails, and chat logs. For each resource type 403, a candidate operation path set 404 can be pre-configured for each resource type 403, and the candidate operation path set 404 can include at least one candidate operation path. The candidate operation path set 404 refers to a pre-defined series of possible operation sequences for different resource types. The aforementioned at least one candidate operation path can be a candidate operation path for different terminal devices or a candidate operation path for the same terminal device. The terminal device can include mobile devices and computers.

[0075] When the triggering entity includes the selected resource 402, a target operation path set 405 can be determined from the preset candidate operation path sets 404 for each of the multiple resource types 403, based on the resource type 403 of the selected resource 402. The target operation path set 405 refers to a specific set of operation paths determined from the candidate operation path sets 404 for each of the multiple resource types 403, based on the resource type 403 of the selected resource 402. The target operation path set 405 may include at least one candidate operation path; for example, the target operation path set 405 may include candidate operation path 405_1, candidate operation path 405_2, ..., candidate operation path 405_N, where N is a positive integer.

[0076] After obtaining the target operation path set 405, the selected operation 401 can be validated based on the target operation path set 405 to determine whether the user has a primary intent. The validation process can refer to the process of determining whether there is an operation path that matches the selected operation 401 among at least one candidate operation path in the target operation path set 405.

[0077] According to embodiments of this disclosure, by pre-setting multiple candidate operation path sets for each resource type, when the triggering subject includes a selected resource, a target operation path set matching the selected resource type can be quickly determined from the pre-set multiple candidate operation path sets for each resource type. Based on this, by validating the user's selected operation according to the target operation path set, it is possible to effectively identify and confirm whether the user has intent, thereby improving the efficiency and accuracy of intent recognition.

[0078] In one example, after obtaining the target operation path set 405, operation S410 can be performed. In operation S410, it can be determined whether there is a target operation path that matches the selection operation 401 among at least one candidate operation path in the target operation path set 405. If yes, it can be determined that the user has a first intention 406. If no, it can be determined that the user does not have a first intention 407.

[0079] In one example, each candidate operation path may include at least two candidate sub-operations with an association relationship, and the selection operation 401 may include at least two trigger sub-operations with an association relationship. A candidate operation path refers to a pre-defined series of possible operation sequences, and a candidate sub-operation refers to a single operation step constituting a candidate operation path. The selection operation 401 refers to the operation sequence actually executed by the user, and the trigger sub-operation refers to a single operation step constituting the selection operation 401. An association relationship refers to the execution order between any two adjacent candidate sub-operations or any two adjacent trigger sub-operations. During the execution of operation S410, for each candidate operation path, each candidate sub-operation and trigger sub-operation can be compared according to the association relationship to determine whether the candidate operation path matches the selection operation.

[0080] For example, if the candidate operation path includes candidate sub-operation 1 -> candidate sub-operation 2 -> candidate sub-operation 3, and the selected operation 401 includes trigger sub-operation 1 -> trigger sub-operation 2 -> trigger sub-operation 3, then it can be determined one by one whether candidate sub-operation 1 matches trigger sub-operation 1, candidate sub-operation 2 matches trigger sub-operation 2, and candidate sub-operation 3 matches trigger sub-operation 3. If each trigger sub-operation in the selected operation 401 matches each candidate sub-operation in the candidate operation path, then it can be determined that the candidate operation path matches the selected operation 401. If there is a trigger sub-operation in the selected operation 401 that does not match a candidate sub-operation in the candidate operation path, then it can be determined that the candidate operation path does not match the selected operation 401.

[0081] The following examples, using resource type 403 as images, files, emails, and chat logs, further illustrate the process of determining whether there is a target operation path matching the selection operation 401 in the target operation path set 405 based on the selection operation 401.

[0082] For example, when resource type 403 is images, candidate operation paths for mobile devices could include opening the album -> long press -> multi-select -> clicking two photos consecutively, or opening the album through a chat window -> clicking two photos consecutively. Candidate operation paths for computers could include double-clicking to open an image folder -> right-clicking to select multiple images / using the Ctrl key to select multiple images, or opening an image folder through a chat window -> clicking two images consecutively. In the above examples, if each triggering sub-operation in selection operation 401 matches each candidate sub-operation in a certain candidate operation path, then it can be determined that the user has a first intent 406.

[0083] Alternatively, when resource type 403 is a file, candidate operation paths for the computer can include double-clicking to open a folder on the computer -> right-clicking for multiple selections / using the Ctrl key for multiple selections, or opening a folder through a chat window -> right-clicking for multiple selections / using the Ctrl key for multiple selections. In the above example, if each trigger sub-operation in operation 401 matches each candidate sub-operation in a candidate operation path, then it can be determined that the user has a first intent 406.

[0084] Alternatively, if resource type 403 is email, candidate operation paths could include opening the mailbox -> clicking to select two emails consecutively. In the example above, if each triggering sub-operation in selection operation 401 matches each candidate sub-operation in a candidate operation path, then it can be determined that the user has a first intent 406.

[0085] Alternatively, when resource type 403 is chat history, the candidate operation path for mobile devices could include long-press to select multiple options -> click on two message records consecutively; the candidate operation path for desktop devices could include right-click to select multiple options. In the above example, if each trigger sub-operation in selection operation 401 matches each candidate sub-operation in a certain candidate operation path, then it can be determined that the user has a first intent 406.

[0086] According to embodiments of this disclosure, since the candidate operation path includes at least two candidate sub-operations with specific relationships, and the selection operation also includes at least two trigger sub-operations with relationships, by comparing each candidate sub-operation with the trigger sub-operation one by one according to the relationships, it is possible to intelligently determine whether the user's actual operation matches the preset operation path, thereby improving the understanding and response capabilities of user behavior, achieving accurate identification of the selection operation, and helping to improve the accuracy of intent recognition.

[0087] In one example, if it is determined that the user does not have the first intent 407, the resource push method provided in this disclosure may not be executed further. If it is determined that the user has the first intent 406, feature analysis can be further performed on the first triggering subject based on the first intent to obtain the first feature information of the first triggering subject.

[0088] According to embodiments of this disclosure, by matching at least one candidate operation path with the selected operation respectively, it is determined whether there is a target operation path in the target operation path set that matches the selected operation. Then, based on whether there is a target operation path that matches the selected operation, it is determined whether the user has a first intention to obtain batch resources, thereby realizing intelligent recognition and confirmation of user intention.

[0089] In the embodiments of this disclosure, after determining the user's first intention to obtain a batch of resources, feature analysis can be performed on the first triggering subject based on the intention to obtain the first feature information of the first triggering subject. The following uses Figure 5 as an example to illustrate the process of obtaining the first feature information when the first triggering subject includes selected resources.

[0090] Figure 5 schematically illustrates an example of a process in which a first triggering subject is subjected to feature analysis based on a first intent, according to an embodiment of the present disclosure, to obtain first feature information of the first triggering subject.

[0091] As shown in Figure 5, in step 500, after determining that the user has a first intention 501, feature analysis can be performed on the first triggering subject based on the first intention to obtain the first feature information of the first triggering subject. The following example, taking the first triggering subject including selected resources, illustrates the process of determining the first feature information 505. After determining that the user has a first intention 501, operation S510 can be executed. In operation S510, it is determined whether there is only one selected resource.

[0092] If so, feature analysis can be directly performed on the selected resource 502 to obtain the first feature information 503. For example, when the resource type is an image, multiple selection via right-click / Ctrl key usually only requires selecting one selected resource 502. Alternatively, when the resource type is a file, usually only one selected resource 502 needs to be selected.

[0093] If not, the common features shared by at least two selected resources 504 can be analyzed to obtain first feature information 505. For example, when the resource type is an image, it is usually necessary to select at least two selected resources 504 when directly selecting from the album. Alternatively, when the resource type is an email, it is usually necessary to select at least two selected resources 504. The at least two selected resources 504 include selected resources 504_1, ..., selected resources 504_P, where P is a positive integer greater than 1. For example, taking P=3, in this step, the common features shared by selected resources 504_1, selected resources 504_2, and selected resources 504_3 can be analyzed to obtain first feature information 505.

[0094] According to embodiments of this disclosure, when a user selects a single selected resource, feature analysis is performed on the selected resource; when a user selects multiple selected resources, features common to these selected resources are analyzed. This enables the handling of feature analysis scenarios for different numbers of selected resources, thereby intelligently responding to the diverse needs of the user.

[0095] In the embodiments of this disclosure, after obtaining the first feature information, multiple candidate resources can be screened based on the first feature information to obtain a target resource set. The process of obtaining the target resource set is illustrated below using Figure 6A as an example.

[0096] Figure 6A schematically illustrates an example of a process of filtering multiple candidate resources based on first feature information to obtain a target resource set according to an embodiment of the present disclosure.

[0097] As shown in Figure 6A, in 600A, the feature information can include explicit feature 602 and implicit feature 603. Explicit feature 602 represents a feature directly determined by the triggering entity, i.e., a feature that the user can directly observe. Implicit feature 603 represents a feature determined by the background data of the triggering entity, i.e., a feature that the user cannot directly observe and needs to be determined by analyzing the background data. When there is more than one explicit feature 602, each explicit feature 602 can have a priority, which represents the degree of influence of the explicit feature 602 on the resource push effect. The order in which explicit feature 602 and implicit feature 603 are determined can be configured according to actual business needs and is not limited here. For example, explicit feature 602 can be determined first, followed by implicit feature 603. Alternatively, implicit feature 603 can be determined first, followed by explicit feature 602.

[0098] In one example, when the triggering subject includes the selected resource 601, both the explicit feature 602 and the implicit feature 603 can be associated with the resource type. The resource type can include at least one of the following: image 601_1, file 601_2, email 601_3, and chat log 601_4, etc.

[0099] For example, in the case of resource type 601_1, explicit feature 602 may include at least one of people, buildings, and objects in the image, and implicit feature 603 may include at least one of time information and location information. In this case, time information refers to the time when the image was taken, and location information refers to the location where the image was taken. It should be noted that the priority of people can be higher than the priority of buildings, and the priority of buildings can be higher than the priority of objects. That is, in the case of resource type 601_1, explicit feature 602 prioritizes identifying people, then buildings, and finally objects.

[0100] Alternatively, when the resource type is file 601_2, explicit feature 602 may include at least one of file name and file type, and implicit feature 603 may include at least one of time information and text content. In this case, time information refers to the file's creation or modification time, and text content can be determined by analyzing the text data within the file. It should be noted that the file name has a higher priority than the file type; that is, when the resource type is file 601_2, explicit feature 602 prioritizes identifying the file name, and then identifies the file type.

[0101] Alternatively, when the resource type is email 601_3, explicit feature 602 may include at least one of the subject, recipient, and sender, and implicit feature 603 may include at least one of time information and email content. In this case, time information refers to the sending or receiving time of the email, and the email content can be determined by analyzing the email body. It should be noted that the subject has a higher priority than the recipient and sender; that is, when the resource type is email 601_3, explicit feature 602 identifies the subject first, then the recipient and sender.

[0102] Alternatively, when the resource type is chat log 601_4, the explicit feature 602 may include at least one of the person and the keyword, and the implicit feature 603 may include time information. In this case, the time information can be determined by analyzing the timestamp of the chat log. It should be noted that the person has a higher priority than the keyword; that is, when the resource type is chat log 601_4, the explicit feature 602 prioritizes identifying the person, and then identifies the keyword.

[0103] According to embodiments of this disclosure, when the triggering subject includes selected resources, since both explicit and implicit features are associated with resource types, explicit and implicit features corresponding to different resource types can be adaptively extracted, thereby helping to improve the efficiency and accuracy of feature extraction.

[0104] According to embodiments of this disclosure, for different resource types, priority is set for the explicit features of each resource type. Since the priority represents the degree of influence of the explicit features on the resource push effect, resources can be pushed more accurately according to the user's needs, thereby improving the personalization of resource push and user satisfaction.

[0105] After obtaining the explicit feature 602 and implicit feature 603 of the triggering subject, during the screening of multiple candidate resources 604, the explicit feature 602 can serve as the basis for the first round of screening. That is, multiple candidate resources 604 can be screened based on at least one explicit feature 602 to obtain at least one candidate resource set 605. The candidate resource set 605 refers to a group of candidate resources screened based on the explicit feature 602. Based on this, the implicit feature 603 can serve as the basis for the second round of screening. That is, at least one candidate resource set 605 can be screened based on at least one implicit feature 603 to obtain the target resource set 606. The target resource set 606 refers to a group of candidate resources further screened based on the implicit feature 603, building upon the candidate resource set 605.

[0106] According to embodiments of this disclosure, by distinguishing between explicit and implicit features, when screening multiple candidate resources, a set of candidate resources is first selected based on explicit features, and then these candidate resources are further refined using implicit features. Since explicit features are features directly determined by the triggering subject, and implicit features are features determined based on the background data of the triggering subject, refined screening and management of resources are achieved, resulting in a target resource set that better meets the personalized needs of users, thereby improving the accuracy and personalization of resource recommendations.

[0107] The following example, using Figure 6B, illustrates the process of obtaining the target resource set when the first triggering subject includes the selected resource and has only one explicit feature.

[0108] Figure 6B schematically illustrates an example of a process for filtering multiple candidate resources based on first feature information to obtain a target resource set according to another embodiment of the present disclosure.

[0109] As shown in Figure 6B, taking a dominant feature 607 as an example, the process of filtering multiple candidate resources 608 based on the first feature information to obtain the target resource set 615 is illustrated in 600B. The above process may include a first round of filtering based on the dominant feature 607 and a second round of filtering based on the latent feature.

[0110] For the first round of screening based on explicit feature 607, multiple candidate resources 608 can be screened based on one explicit feature 607 to obtain a candidate resource set 609. This candidate resource set 609 may include at least one candidate resource. The display method of the candidate resource set 609 can be configured according to actual business needs and is not limited here. For example, the candidate resource set 609 can be displayed by popping up a dialog box asking "Do you want to include all content of explicit feature 607?" and showing a thumbnail list of at least one candidate resource.

[0111] For example, taking images as the resource type, user 612 selects image 1 and image 2. The common explicit feature 607 of images 1 and 2 is person A and person B. Therefore, based on this explicit feature 607, all candidate resources 608 containing person A and person B are determined as the candidate resource set 609. Alternatively, taking emails as the resource type, user 612 selects email 1 and email 2. The common explicit feature 607 of email 1 and email 2 is recipient C. Therefore, based on this explicit feature 607, all candidate resources 608 containing recipient C are determined as the candidate resource set 609.

[0112] According to embodiments of this disclosure, when there is only one explicit feature, multiple candidate resources can be screened directly based on that explicit feature, which can significantly reduce the interference of irrelevant information, directly screen a set of candidate resources that match the explicit feature from a large number of candidate resources, improve screening efficiency, and ensure that users can quickly obtain resources that match their needs.

[0113] For the second round of screening based on latent features, after user 612 clicks the feature selection control 613, the feature selection control 613 on the front-end interface 610 can display candidate feature identifiers 614 determined based on at least one latent feature. For example, if there are Q latent features, the candidate feature identifiers can include candidate feature identifier 1 determined based on latent feature 1, candidate feature identifier 2 determined based on latent feature 2, ..., candidate feature identifier Q determined based on latent feature Q, where Q is a positive integer. Furthermore, the display area of ​​the feature selection control 613 can be scrolled to display other candidate feature identifiers.

[0114] In response to detecting a selection operation by user 612 for a target feature identifier among at least one candidate feature identifier 614, the target feature identifier can be displayed in a dropdown list 611. For example, if user 612 selects "candidate feature identifier 2" as the target feature identifier, then the target feature identifier can be displayed in the dropdown list 611. Based on this, at least one candidate resource 609 can be filtered according to the latent features corresponding to the target feature identifier to obtain a target resource set 615.

[0115] According to embodiments of this disclosure, a feature selection control on the front-end interface displays candidate feature identifiers determined based on latent features, allowing users to intuitively select a target feature identifier from the latent features. Furthermore, in response to detecting a user's selection of a target feature identifier, at least one candidate resource is subjected to secondary filtering based on the latent features corresponding to the target feature identifier. This enables precise filtering of the candidate resource set obtained from the initial filtering, resulting in a target resource set that better meets the user's personalized needs, thus improving the accuracy and personalization of resource recommendations.

[0116] The following example, using Figure 6C, illustrates the process of obtaining the target resource set when the first triggering subject includes selected resources and has more than one explicit feature.

[0117] Figure 6C schematically illustrates an example of a process for filtering multiple candidate resources based on first feature information to obtain a target resource set according to yet another embodiment of the present disclosure.

[0118] As shown in Figure 6C, taking the case where there is more than one dominant feature 616 in 600C as an example, the process of filtering multiple candidate resources 617 based on feature information to obtain the target resource set 630 is illustrated. The above process may include a first round of filtering based on the dominant feature 616 and a second round of filtering based on the latent feature.

[0119] For the first round of screening based on dominant feature 616, multiple candidate resources 617 can be screened based on at least two dominant features 616 to obtain multiple candidate resource sets 618. In one example, the above screening process may include, for each dominant feature 616, screening multiple candidate resources 617 based on that dominant feature 616 to obtain a separate candidate resource set 618 for each dominant feature. In another example, the above screening process may further include, determining a feature set of at least two dominant features 616, and screening multiple candidate resources 617 based on that feature set to obtain a separate candidate resource set 618 for each feature set.

[0120] According to embodiments of this disclosure, when there is more than one explicit feature, the candidate resource set may include multiple candidate resources selected based on each individual explicit feature, or multiple candidate resources selected based on a feature set determined based on at least two explicit features, thereby improving the accuracy of resource retrieval and optimizing the organization and presentation of resources.

[0121] In one example, after obtaining the aforementioned candidate resource sets 618, a candidate set identifier for each candidate resource set 618 can be determined. For example, for a candidate resource set 618 obtained through filtering by explicit feature 616, the candidate set identifier can be determined based on explicit feature 616. Alternatively, for a candidate resource set 618 obtained through feature set, the candidate set identifier can be determined based on at least two explicit features 616.

[0122] For each of the aforementioned candidate set identifiers, the candidate set identifiers 623 of multiple candidate resource sets 618 can be displayed through the set selection control 622 on the front-end interface 619. For example, if there are U candidate resource sets 618, the candidate set identifiers can include candidate set identifier 1 corresponding to candidate resource set 1, candidate set identifier 2 corresponding to candidate resource set 2, ..., and candidate set identifier U corresponding to candidate resource set U, where U is a positive integer greater than 1. Furthermore, the display area of ​​the set selection control 622 can be scrolled to display other candidate set identifiers.

[0123] In response to detecting user 621's selection of a target set identifier from multiple candidate set identifiers 623, the target set identifier can be displayed in dropdown list 620. For example, if user 622 selects "candidate set identifier 2" as the target set identifier, then that target set identifier can be displayed in dropdown list 620. Based on this, the candidate resource set 618 corresponding to the target set identifier can be determined as candidate resource set 624. This candidate resource set 624 refers to the candidate resource set to be selected for the second round of screening based on implicit features.

[0124] In another example, after obtaining the aforementioned candidate resource sets 618, explicit features 616 can be displayed on the front-end interface 619 via options, with each explicit feature 616 connected by "or / and". In this case, user 621 can select one or more options from the aforementioned candidate resource sets 618 to determine the candidate resource set 624 to be used for the second round of screening based on implicit features.

[0125] The following will use two explicit features 616 as examples to illustrate the first round of screening based on explicit features 616 for resource types of images or files.

[0126] Taking dominant feature 616 as an example, which includes dominant feature 1 and dominant feature 2, three candidate resource sets 618 can be obtained in this case: a candidate resource set 618 including dominant feature 1, a candidate resource set 618 including dominant feature 2, and a candidate resource set 618 including both dominant feature 1 and dominant feature 2.

[0127] Taking an image as the resource type as an example, explicit feature 1 can be "person A and person B", and explicit feature 2 can be "building C". Thus, three candidate resource sets 618 can be obtained, namely, the candidate resource set 618 including person A and person B, the candidate resource set 618 including building C, and the candidate resource set 618 including person A, person B and building C.

[0128] After obtaining the aforementioned candidate resource sets 618, in one example, the candidate set identifiers of the three candidate resource sets 618 can be displayed on the front-end interface 619. For example, the candidate set identifier of the candidate resource set 618 including character A and character B is "candidate set identifier 1", the candidate set identifier of the candidate resource set 618 including building C is "candidate set identifier 2", and the candidate set identifier of the candidate resource set 618 including character A, character B, and building C is "candidate set identifier 3". In another example, the aforementioned explicit features 616 can also be displayed on the front-end interface 619 as options, for example, "character A and character B or / and building C".

[0129] The following will use three explicit features 616 as an example to illustrate the first round of screening based on explicit features 616 for resource types of images or files.

[0130] Taking a dominant feature 618 that includes dominant feature 1, dominant feature 2, and dominant feature 3 as an example, in this case, seven candidate resource sets 618 can be obtained, namely, the candidate resource set 618 that includes dominant feature 1, the candidate resource set 618 that includes dominant feature 2, the candidate resource set 618 that includes dominant feature 3, the candidate resource set 618 that includes dominant features 1 and dominant feature 2, the candidate resource set 618 that includes dominant features 2 and dominant feature 3, the candidate resource set 618 that includes dominant features 1 and dominant feature 3, and the candidate resource set 618 that includes dominant features 1, dominant feature 2, and dominant feature 3.

[0131] Taking the image as an example again, the dominant feature 1 can be "person A and person B", the dominant feature 2 can be "building C", and the dominant feature 3 can be "object D". Thus, we can obtain seven candidate resource sets 618, namely, the candidate resource set 618 including person A and person B, the candidate resource set 618 including building C, the candidate resource set 618 including object D, the candidate resource set 618 including person A, person B and building C, the candidate resource set 618 including person A, person B and object D, the candidate resource set 618 including building C and object D, and the candidate resource set 618 including person A and person B, building C and object D.

[0132] After obtaining the aforementioned candidate resource sets 618, in one example, the candidate set identifiers of the seven candidate resource sets 618 can be displayed on the front-end interface 619. For example, the candidate set identifier of the candidate resource set 618 including character A and character B is "candidate set identifier 1", the candidate set identifier of the candidate resource set 618 including building C is "candidate set identifier 2", the candidate set identifier of the candidate resource set 618 including object D is "candidate set identifier 3", the candidate set identifier of the candidate resource set 618 including character A, character B and building C is "candidate set identifier 4", the candidate set identifier of the candidate resource set 618 including character A, character B and object D is "candidate set identifier 5", the candidate set identifier of the candidate resource set 618 including building C and object D is "candidate set identifier 6", and the candidate set identifier of the candidate resource set 618 including character A, character B, building C and object D is "candidate set identifier 7". In another example, the aforementioned explicit features 616 can also be displayed on the front-end interface 619 as options, such as "Person A and Person B or / and Building C or / and Object D".

[0133] According to embodiments of this disclosure, for a candidate resource set with explicit features, a candidate set identifier can be determined based on the explicit features. For a candidate resource set with a feature set, a candidate set identifier can be determined based on at least two explicit features. These multiple candidate set identifiers are displayed through a set selection control on the front-end interface, enabling intuitive display and management of multiple candidate resource sets. Furthermore, in response to detecting a user's selection operation on a target set identifier among multiple candidate set identifiers, the candidate resource set corresponding to the target set identifier can be determined as the candidate resource set, thereby providing the user with more accurate resource filtering results. This not only improves the efficiency and accuracy of resource retrieval but also enables resource filtering based on multiple features, enhancing the flexibility of resource management.

[0134] For the second round of screening based on latent features, after user 627 clicks the feature selection control 628, candidate feature identifiers 629 determined based on at least one latent feature can be displayed through the feature selection control 628 on the front-end interface 625. For example, if there are v latent features, the candidate feature identifiers may include candidate feature identifier 1 determined based on latent feature 1, candidate feature identifier 2 determined based on latent feature 2, ..., candidate feature identifier V determined based on latent feature V, where V is a positive integer. Furthermore, the display area of ​​the feature selection control 628 can be scrolled to display other candidate feature identifiers.

[0135] In response to detecting a selection operation by user 627 for a target feature identifier among at least one candidate feature identifier 629, the target feature identifier can be displayed in dropdown list 626. For example, if user 627 selects "candidate feature identifier 2" as the target feature identifier, then the target feature identifier can be displayed in dropdown list 626. Based on this, the candidate resource set 624 can be filtered according to the latent features corresponding to the target feature identifier to obtain the target resource set 630.

[0136] In the embodiments of this disclosure, after determining the user's first intention to obtain a batch of resources, feature analysis can be performed on the first triggering subject based on the first intention to obtain the first feature information of the first triggering subject. The following uses Figure 7 as an example to illustrate the process of obtaining the first feature information when the first triggering subject includes the first dialogue context or the first voice data.

[0137] Figure 7 schematically illustrates an example of a process in which a first triggering subject is subjected to feature analysis based on a first intent, according to an embodiment of the present disclosure, to obtain first feature information of the first triggering subject.

[0138] As shown in Figure 7, in step 700, when the first triggering subject includes a first dialogue context 701 or first voice data 702, a trained large model can be used to monitor the first dialogue context 701 or the first voice data 702. In response to detecting that the first dialogue context 701 or the first voice data 702 represents a first intent of the user to obtain bulk resources, feature analysis can be performed on the first dialogue context 701 or the first voice data 702 based on the first intent to obtain first feature information 704.

[0139] The trained large model can refer to a pre-trained large language model (LLM) that classifies intents based on retrieval-augmented generation (RAG) methods. During model training, the large model can embed multiple sample intents separately to obtain multiple sample intent features, which are then stored in a vector library. During intent prediction, the trained large model can embed either the first dialogue context 701 or the first speech data 702. Using the obtained matching features, it finds at least one similar sample intent feature in the vector library, ranks the at least one sample intent feature based on its similarity to the matching feature, and generates a prompt using the most similar sample intent feature to guide the trained large model in predicting the first intent of the first dialogue context 701 or the first speech data 702.

[0140] In one example, the first dialogue context 701 refers to the dialogue content when the user interacts with the server. Based on the first dialogue context 701, fields are extracted to obtain the first content field 703. Based on this, the first feature information 704 can be determined.

[0141] In another example, the first voice data 702 refers to the voice input when the user interacts with the server. Based on the first voice data 702, fields are extracted to obtain the first content field 703. Based on this, the first feature information 704 can be determined.

[0142] According to embodiments of this disclosure, accurate field extraction is achieved through intelligent analysis and processing of the first dialogue context or first voice data, which helps improve the accuracy of information processing. Based on this, the first feature information is determined according to the first content field obtained from the field extraction, improving the accuracy of the first feature information and thus helping to improve the effectiveness of subsequent resource delivery.

[0143] In the embodiments of this disclosure, after obtaining the first feature information, multiple candidate resources can be filtered according to the first feature information to obtain a target resource set. The following uses Figure 8 as an example to illustrate the process of obtaining the target resource set when the first triggering subject includes the first dialogue context or the first voice data.

[0144] Figure 8 schematically illustrates an example of a process for filtering multiple candidate resources based on first feature information to obtain a target resource set according to another embodiment of the present disclosure.

[0145] As shown in Figure 8, taking the first triggering subject including the first dialogue context 801 or the first voice data 802 as an example, the process of filtering multiple candidate resources 805 based on the first feature information to obtain the target resource set 812 is illustrated. The above process may include a first round of filtering based on explicit features 804 and a second round of filtering based on implicit features.

[0146] For the first round of screening based on explicit feature 804, fields can be extracted from the first dialogue context 801 or the first voice data 802 to obtain the first content field 803. Based on this, explicit feature 8054 can be determined according to the first content field 803.

[0147] After obtaining the explicit feature 804, multiple candidate resources 805 can be filtered based on the explicit feature 804 to obtain a candidate resource set 806. This candidate resource set 806 may include at least one candidate resource. The display method of the candidate resource set 806 can be configured according to actual business needs and is not limited here. For example, the candidate resource set 806 can be displayed by popping up a dialog box asking "Do you want to include all content of explicit feature 804?" and showing a thumbnail list of at least one candidate resource.

[0148] For the second round of screening based on latent features, after user 809 clicks the feature selection control 810, candidate feature identifiers 811 determined based on at least one latent feature can be displayed through the feature selection control 810 on the front-end interface 807. For example, if there are W latent features, the candidate feature identifiers can include candidate feature identifier 1 determined based on latent feature 1, candidate feature identifier 2 determined based on latent feature 2, ..., candidate feature identifier W determined based on latent feature W, where W is a positive integer. Furthermore, the display area of ​​the feature selection control 811 can be scrolled to display other candidate feature identifiers.

[0149] In response to detecting a selection operation by user 809 for a target feature identifier among at least one candidate feature identifier 811, the target feature identifier can be displayed in a dropdown list 808. For example, if user 809 selects "candidate feature identifier 2" as the target feature identifier, then the target feature identifier can be displayed in the dropdown list 808. Based on this, the candidate resource set 806 can be filtered according to the latent features corresponding to the target feature identifier to obtain the target resource set 812.

[0150] The above are merely exemplary embodiments, but are not limited thereto. Other resource push methods known in the art may also be included, as long as they can improve the automation and accuracy of resource push.

[0151] The system architecture for applying the resource push method provided in this disclosure has been described above. The resource push process of another embodiment of this disclosure will be further described below with reference to Figure 9.

[0152] Figure 9 schematically illustrates a flowchart of a resource push method according to another embodiment of the present disclosure.

[0153] As shown in Figure 9, the resource push method 900 includes operations S910 to S930.

[0154] In operation S910, determine the user's secondary intent to obtain resources.

[0155] During operation S920, based on the second intent, feature analysis is performed on the second triggering subject to obtain the second feature information of the second triggering subject.

[0156] During operation of S930, multiple candidate resources are filtered based on the second feature information to obtain the target resource and push the target resource to the user.

[0157] The secondary intent to acquire resources refers to a user's desire to acquire a single resource. For example, a user might want to acquire a document or image related to a specific topic, indicating a secondary intent to acquire a batch of resources. In one example, where the secondary trigger is a second dialogue context or second voice data, the secondary intent can be determined based on the secondary trigger. For instance, intent analysis of the second dialogue context or second voice data can be used to determine whether the user has a secondary intent to acquire resources.

[0158] If it is determined that the user has a secondary intention to obtain resources, feature analysis can be performed on the secondary triggering entity based on this secondary intention to obtain secondary feature information of the secondary triggering entity. Feature analysis refers to the process of identifying and analyzing the key attributes or characteristics of the secondary triggering entity. Secondary feature information refers to the data obtained through feature analysis, which describes the specific attributes of the triggering entity. After obtaining the secondary feature information, multiple candidate resources can be filtered based on the secondary feature information to obtain the target resource. After obtaining the target resource, the target resource can be pushed to the user.

[0159] According to embodiments of this disclosure, automatically identifying a user's second intent to obtain resources helps improve the interactivity and responsiveness of resource push notifications. By performing feature analysis on the second triggering entity based on the identified second intent, the characteristics of the resources needed by the user can be intelligently analyzed. Based on this, by filtering multiple candidate resources according to the analyzed second feature information, the resources needed by the user can be automatically selected and proactively pushed to the user. Thus, users do not need to use search functions or search for resources one by one, saving time spent on resource filtering and improving the automation and accuracy of resource push notifications. This allows for accurate identification of the user's resource acquisition intent and intelligent and rapid filtering of the target resources needed by the user.

[0160] In the embodiments of this disclosure, after determining the user's second intention to obtain resources, feature analysis can be performed on the second triggering subject based on the second intention to obtain the second feature information of the second triggering subject. Based on this second feature information, multiple candidate resources are then filtered to obtain the target resource. The following uses Figure 10 as an example to illustrate the process of obtaining the second feature information and the target resource when the second triggering subject includes a second dialogue context or second voice data.

[0161] Figure 10 schematically illustrates an example of a process of filtering multiple candidate resources to obtain a target resource based on second feature information when the second triggering subject includes a second dialogue context or second voice data, according to an embodiment of the present disclosure.

[0162] As shown in Figure 10, in 1000, taking the second triggering subject including the second dialogue context 1001 or the second voice data 1002 as an example, the process of filtering multiple candidate resources 1005 according to the second feature information 1004 to obtain the target resource 1012 is explained.

[0163] In one example, fields can be extracted from the second dialogue context 1001 or the second voice data 1002 to obtain the second content field 1003. Based on this, the second feature information 1004 can be determined according to the second content field 1003. After obtaining the second feature information 1004, multiple candidate resources 1005 can be filtered according to the second feature information 1004 to obtain at least one candidate resource 1006.

[0164] After obtaining at least one candidate resource 1006, the candidate resource identifier 1011 of each candidate resource can be displayed through the resource selection control 1010 on the front-end interface 1007. The display method of the resource selection control 1010 can be configured according to actual business needs and is not limited here. For example, the resource selection control 1010 can be displayed through a floating box 1009 in a dialog box. Alternatively, the resource selection control 1010 can also be displayed through a pop-up window.

[0165] The display method of at least one candidate resource can be configured according to actual business needs and is not limited here. For example, as shown in 1000, a list of candidate resources can be displayed in the resource selection control 1010, which includes the candidate resource identifier of each of the at least one candidate resource for the user to select. Alternatively, the resource path of each of the at least one candidate resource can also be displayed in the resource selection control 1010 for the user to select. Alternatively, a resource thumbnail of each of the at least one candidate resource can also be displayed in the resource selection control 1010 for the user to select.

[0166] During the process of displaying at least one candidate resource via the resource selection control 1010, the display position of the resource selection control 1010 can be configured to be at a preset position interval from the second content field 1003 of the display area of ​​the front-end interface 1007. The preset position interval can be configured according to actual business needs and is not limited here. In one example, the preset position interval can be determined based on the coordinates 1009_1 of the top-left corner of the floating box 1009 used to display the resource selection control 1010 and the coordinates 1003_1 of the top-left corner of the second content field 1003. For example, the preset position interval can be set to be 1 cm apart horizontally and 2 cm apart vertically between the top-left corner coordinates 1009_1 and 1003_1.

[0167] It should be noted that the display position of the resource selection control 1010 can adapt to the display position of the second content field 1003 as the second dialogue context 1001 or the second voice data 1002 progresses, so as to realize real-time changes in the front-end interface 1007. For example, as the second dialogue context 1001 progresses, the display position of the second content field 1003 in the front-end interface 1007 moves 1 cm vertically while remaining unchanged horizontally. Then, the corresponding resource selection control 1010 can also move 1 cm vertically while remaining unchanged horizontally in the front-end interface 1007.

[0168] After the user clicks the resource selection control 1010, the resource selection control 1010 on the front-end interface 1007 can display candidate resource identifiers 1011 based on at least one candidate resource. For example, if there are X candidate resource identifiers, then the candidate resource identifiers 1011 can include candidate resource identifier 1, candidate resource identifier 2, ..., candidate resource identifier X, where X is a positive integer. Furthermore, the display area of ​​the resource selection control 1010 can be scrolled to display other candidate resource identifiers.

[0169] In response to detecting a user's selection of a target resource identifier from at least one candidate resource identifier 1011, the target resource identifier can be displayed in a dropdown list 1008. For example, if the user selects "candidate resource identifier 2", then the target resource identifier can be displayed in the dropdown list 1008. Based on this, the candidate resource corresponding to the target resource identifier can be determined as the target resource 1012.

[0170] The process of obtaining the second feature information and the target resource when the second triggering subject includes the second dialogue context or the second voice data has been described above. The process of obtaining the second feature information and the target resource when the second triggering subject includes user operation and the second dialogue context will be described below using Figures 11A and 11B as examples.

[0171] Figure 11A schematically illustrates an example of a process of filtering multiple candidate resources to obtain a target resource based on second feature information when the second triggering subject includes user operation and a second dialogue context, according to an embodiment of the present disclosure.

[0172] As shown in Figure 11A, taking the second triggering subject including user operation 1101 and second dialogue context 1102 as an example, the process of filtering multiple candidate resources 1105 according to the second feature information 1104 to obtain the target resource 1112 is explained.

[0173] In one example, in response to the detection of user action 1101, a second dialogue context 1102 can be determined based on user action 1101. The user action 1101 to be detected can be configured according to actual business needs and is not limited here. For example, user action 1101 could be clicking the "+" button in a chat scenario. Alternatively, user action 1101 could be clicking the "send file" button in a chat scenario. Another alternative is initiating a group chat in a chat scenario.

[0174] The method for determining the second dialogue context 1102 can be configured according to actual business needs and is not limited here. For example, the dialogue content within a preset time interval from the time when user operation 1101 is detected can be determined as the second dialogue context 1102. Alternatively, the dialogue content within a preset number of lines from the position corresponding to the time when user operation 1101 is detected can be determined as the second dialogue context 1102.

[0175] After obtaining the second dialogue context 1102, fields can be extracted from the second dialogue context 1102 to obtain the second content field 1103. Based on this, the second feature information 1104 can be determined according to the second content field 1103. After obtaining the second feature information 1104, multiple candidate resources 1105 can be filtered according to the second feature information 1104 to obtain at least one candidate resource 1106.

[0176] After obtaining at least one candidate resource 1106, the candidate resource identifier 1111 of each candidate resource can be displayed through the resource selection control 1110 on the front-end interface 1107. The display method of the resource selection control 1110 can be configured according to actual business needs and is not limited here. For example, the resource selection control 1110 can be displayed through a floating box 1009 in a dialog box. Alternatively, the resource selection control 1110 can also be displayed through a pop-up window.

[0177] The display method of at least one candidate resource can be configured according to actual business needs and is not limited here. For example, as shown in 1100A, a list of candidate resources can be displayed in the resource selection control 1110. This list includes the candidate resource identifier of each of the at least one candidate resource for the user to select. Alternatively, the resource path of each of the at least one candidate resource can also be displayed in the resource selection control 1110 for the user to select. Alternatively, a resource thumbnail of each of the at least one candidate resource can also be displayed in the resource selection control 1110 for the user to select.

[0178] During the process of displaying at least one candidate resource via the resource selection control 1110, the display position of the resource selection control 1110 can be determined based on whether the second content field 1103 is located in the display area 1107_1 of the front-end interface 1107. For example, in 1100A, if the second content field 1103 is located in the display area 1107_1 of the front-end interface 1107, the display position of the resource selection control 1110 is configured to be at a preset position interval from the second content field 1103 in the display area 1107_1 of the front-end interface 1107. The preset position interval can be configured according to actual business needs and is not limited here. In one example, the preset position interval can be determined based on the coordinates 1109_1 of the upper left corner of the floating box 1109 used to display the resource selection control 1110 and the coordinates 1103_1 of the upper left corner of the second content field 1103. For example, the preset position interval can be set so that the coordinates of the top left corner point 1109_1 and the coordinates of the top left corner point 1103_1 are 1 cm apart in the horizontal direction and 2 cm apart in the vertical direction.

[0179] It should be noted that the display position of the resource selection control 1110 can adapt to the display position of the second content field 1103 as the second dialogue context 1101 progresses, so as to realize real-time changes in the front-end interface 1107. For example, as the second dialogue context 1101 progresses, the display position of the second content field 1103 moves 1 cm vertically in the display area 1107_1 of the front-end interface 1107, while remaining unchanged horizontally. Then, the corresponding resource selection control 1110 can also move 1 cm vertically in the display area 1107_1 of the front-end interface 1107, while remaining unchanged horizontally.

[0180] After the user clicks the resource selection control 1110, the resource selection control 1110 on the front-end interface 1107 can display candidate resource identifiers 1111 based on at least one candidate resource. For example, if there are Y candidate resource identifiers, then the candidate resource identifiers 1111 can include candidate resource identifier 1, candidate resource identifier 2, ..., candidate resource identifier Y, where Y is a positive integer. Furthermore, the display area of ​​the resource selection control 1110 can be scrolled to display other candidate resource identifiers.

[0181] In response to detecting a user's selection of a target resource identifier from at least one candidate resource identifier 1111, the target resource identifier can be displayed in a dropdown list 1108. For example, if the user selects "candidate resource identifier 2" as the target resource identifier, then the target resource identifier can be displayed in the dropdown list 1108. Based on this, the candidate resource corresponding to the target resource identifier can be determined as the target resource 1112.

[0182] Figure 11B schematically illustrates an example of a process of filtering multiple candidate resources to obtain a target resource based on second feature information when the second triggering subject includes a second user operation and a second dialogue context, according to an embodiment of the present disclosure.

[0183] As shown in Figure 11B, compared to 1100A, in 1100B, when the second content field 1103 is not located in the display area 1107_1 of the front-end interface 1107 (i.e., when the aforementioned preset time period is long or the preset number of rows is large), the position of the second content field 1103 can be indicated by icon 1109_2. The specific form of icon 1109_2 can be configured according to actual business needs and is not limited here. For example, icon 1109_2 can be in the form of an arrow.

[0184] It should be noted that in the embodiments of the present invention, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of the present invention. However, they do not mean that the present invention has used or necessarily used such solutions.

[0185] Based on the above-described resource push method, the present invention also provides a resource push device. The device will be described in detail below with reference to FIG12.

[0186] Figure 12 schematically illustrates a block diagram of a resource push device according to an embodiment of the present disclosure.

[0187] As shown in Figure 12, the resource push device 1200 may include a first determination module 1210, a first feature analysis module 1220, and a first filtering module 1230.

[0188] The first determining module 1210 is used to determine the user's intent to obtain a batch of resources.

[0189] The first feature analysis module 1220 is used to perform feature analysis on the triggering subject based on intent, and obtain the feature information of the triggering subject.

[0190] The first filtering module 1230 is used to filter multiple candidate resources based on feature information, obtain a target resource set, and push the target resource set to the user.

[0191] Based on the above-described resource push method, the present invention also provides a resource push device. The device will be described in detail below with reference to FIG13.

[0192] Figure 13 schematically illustrates a block diagram of a resource push device according to another embodiment of the present disclosure.

[0193] As shown in Figure 13, the resource push device 1300 may include a second determination module 1310, a second feature analysis module 1320, and a second filtering module 1330.

[0194] The second determining module 1310 is used to determine the user's intent to obtain resources.

[0195] The second feature analysis module 1320 is used to perform feature analysis on the triggering subject based on intent, and obtain the feature information of the triggering subject.

[0196] The second filtering module 1330 is used to filter multiple candidate resources based on feature information, obtain target resources, and push target resources to the user.

[0197] Any one or more of the modules according to embodiments of this disclosure, or at least a portion thereof, may be implemented in one module. Any one or more of the modules according to embodiments of this disclosure may be implemented by dividing them into multiple modules. Any one or more of the modules according to embodiments of this disclosure may be at least partially implemented as hardware circuitry, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a System-on-Chip, a System-on-Substrate, a System-on-Package, an Application-Specific Integrated Circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, one or more of the modules according to embodiments of this disclosure may be at least partially implemented as computer program modules, which, when run, can perform corresponding functions.

[0198] It should be noted that the resource push device part in the embodiments of this disclosure corresponds to the resource push method part in the embodiments of this disclosure. For a detailed description of the resource push device part, please refer to the resource push method part, which will not be repeated here.

[0199] Figure 14 schematically illustrates a block diagram of an electronic device suitable for implementing a resource push method according to an embodiment of the present disclosure. The electronic device shown in Figure 14 is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present disclosure.

[0200] As shown in FIG. 14, a computer electronic device 1400 according to an embodiment of the present disclosure includes a processor 1401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1402 or a program loaded from a storage portion 1409 into a random access memory (RAM) 1403. The processor 1401 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1401 may also include onboard memory for caching purposes. The processor 1401 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0201] RAM 1403 stores various programs and data required for the operation of electronic device 1400. Processor 1401, ROM 1402 and RAM 1403 are interconnected via bus 1404.

[0202] According to embodiments of this disclosure, the electronic device 1400 may further include an input / output (I / O) interface 1405, which is also connected to a bus 1404. The electronic device 1400 may also include one or more of the following components connected to the input / output (I / O) interface 1405: an input section 1406 including a keyboard, mouse, etc.; an output section 1407 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1408 including a hard disk, etc.; and a communication section 1409 including a network interface card such as a LAN card, modem, etc. The communication section 1409 performs communication processing via a network such as the Internet. A drive 1410 is also connected to the input / output (I / O) interface 1405 as needed. A removable medium 1411, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1410 as needed so that computer programs read from it can be installed into the storage section 1408 as needed.

[0203] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The aforementioned computer-readable storage medium carries one or more programs, which, when executed, implement the resource push method according to the embodiments of this disclosure.

[0204] In this disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0205] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods provided in the embodiments of this disclosure. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the resource push method provided in the embodiments of this disclosure.

[0206] When the computer program is executed by the processor 1401, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0207] According to embodiments of this disclosure, program code for executing computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages.

[0208] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. It should also be noted that in some alternative implementations, the functions indicated in the boxes may occur in a different order than those shown in the drawings.

[0209] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A resource push method, comprising: Determine the user's primary intent to obtain bulk resources; Based on the first intent, feature analysis is performed on the first triggering entity to obtain the first feature information of the first triggering entity; as well as Based on the first feature information, multiple candidate resources are filtered to obtain a target resource set, which is then pushed to the user.

2. The method according to claim 1, wherein, The first triggering entity includes the selected resource; The step of performing feature analysis on the first triggering entity based on the first intent to obtain the first feature information of the first triggering entity includes: When there is only one selected resource, feature analysis is performed on the selected resource to obtain the first feature information; and When there is more than one selected resource, the characteristics shared by the at least two selected resources are analyzed to obtain the first characteristic information.

3. The method according to claim 1, wherein, The first feature information includes explicit features and implicit features. The explicit features represent features directly determined by the first triggering subject, and the implicit features represent features determined by the background data of the first triggering subject. The step of filtering multiple candidate resources based on the first feature information to obtain a target resource set includes: Based on at least one of the explicit features, the plurality of candidate resources are filtered to obtain at least one set of candidate resources; as well as The target resource set is obtained by filtering the at least one candidate resource set based on at least one of the implicit features.

4. The method according to claim 3, wherein, When there is only one dominant trait, The step of filtering the plurality of candidate resources based on the at least one explicit feature to obtain at least one candidate resource set includes: Based on the explicit features, the multiple candidate resources are filtered to obtain the candidate resource set.

5. The method according to claim 3, wherein, When there is more than one dominant trait, The step of filtering the plurality of candidate resources based on the at least one explicit feature to obtain at least one candidate resource set includes: Based on the dominant features, the multiple candidate resources are filtered to obtain a candidate resource set for each dominant feature; and Based on the feature set determined by the at least two explicit features, the plurality of candidate resources are filtered to obtain a candidate resource set for each feature set.

6. The method according to claim 5, further comprising, before filtering the at least one candidate resource set according to the at least one latent feature to obtain the target resource set: The candidate set identifiers of the multiple candidate resource sets are displayed through a set selection control on the front-end interface. For the candidate resource set with the dominant feature, the candidate set identifier is determined based on the dominant feature; for the candidate resource set with the feature set, the candidate set identifier is determined based on the at least two dominant features. as well as In response to detecting the user's selection operation for a target set identifier among a plurality of candidate set identifiers, the candidate resource set corresponding to the target set identifier is determined as the candidate resource set.

7. The method according to claim 4 or 5, wherein, The candidate resource set includes at least one candidate resource; The step of filtering the at least one candidate resource set based on the at least one latent feature to obtain the target resource set includes: The feature selection control on the front-end interface displays candidate feature identifiers determined based on at least one latent feature; and In response to detecting the user's selection operation for a target feature identifier among the at least one candidate feature identifiers, the at least one candidate resource is filtered according to the latent features corresponding to the target feature identifier to obtain the target resource set.

8. The method according to claim 1, wherein, The first triggering entity includes the first dialogue context or the first voice data; The step of performing feature analysis on the first triggering entity based on the first intent to obtain the first feature information of the first triggering entity includes: Field extraction is performed on the first dialogue context or the first voice data to obtain the first content field; and The first feature information is determined based on the first content field.

9. The method according to claim 1, wherein, The first intention to determine the acquisition of bulk resources includes: The selected resource is determined based on the user's first selection action; and Based on the first selection operation, the first intention to acquire batch resources is determined.

10. The method according to claim 9, wherein, The step of determining the first intent to acquire batch resources based on the first selection operation includes: Based on the resource type of the selected resources, a target operation path set is determined from the preset candidate operation path sets for each of the multiple resource types; and The selected operation is validated based on the set of target operation paths to determine whether the user has the first intent.

11. The method according to claim 10, wherein, The target operation path set includes at least one candidate operation path; The step of validating the selected operation based on the target operation path set to determine whether the user has the first intent includes: If a target operation path matching the selected operation exists among the at least one candidate operation path, it is determined that the user has the first intent. as well as If no target operation path matching the selected operation exists among the at least one candidate operation path, it is determined that the user does not have the first intent.

12. The method according to claim 11, wherein, The candidate operation path includes at least two candidate sub-operations that are related, and the selection operation includes at least two triggering sub-operations that are related. The method further includes: For each candidate operation path, each candidate sub-operation and the triggering sub-operation are compared according to the association relationship to determine whether the candidate operation path matches the selected operation.

13. The method according to claim 1, wherein, Determining a user's primary intent to acquire bulk resources includes: Determine whether the user is in a preset scenario; and In response to the user being in the preset scenario, determine whether the user has the first intention.

14. The method according to claim 13, wherein, The preset scenario includes a set of preset operations for each of multiple resource types, and each preset operation set includes at least one preset operation. Determining whether the user is in a preset scenario includes: If a target operation matching the selected operation exists in at least one preset operation of the target operation set, the user is determined to be in the preset scenario, wherein the target operation set is determined based on the resource type of the triggering subject, within the preset operation sets of each of the plurality of resource types; and If the target operation is not present in at least one of the preset operations in the target operation set, it is determined that the user is not in the preset scenario.

15. The method according to claim 2, wherein, Both the explicit and implicit features are associated with resource types; When the resource type is an image, the explicit features include at least one of people, buildings, and objects, and the implicit features include at least one of time information and location information. When the resource type is a file, the explicit features include at least one of the file name and the file type, and the implicit features include at least one of the time information and the text content; When the resource type is email, the explicit features include at least one of the subject, recipient, and sender, and the implicit features include at least one of the time information and email content; and When the resource type is chat history, the explicit features include at least one of the person and the keyword, and the implicit features include the time information.

16. The method according to claim 15, wherein, When the resource type is an image, the priority of the person is higher than the priority of the building, and the priority of the building is higher than the priority of the object, wherein the priority characterizes the degree of influence of the explicit feature on the resource push effect; When the resource type is a file, the file name has a higher priority than the file type. When the resource type is email, the subject has a higher priority than the recipient and the sender; and When the resource type is chat history, the priority of the person is higher than the priority of the keyword.

17. A resource push method, comprising: Determine the user's secondary intent in acquiring resources; Based on the second intent, feature analysis is performed on the second triggering subject to obtain the second feature information of the second triggering subject; as well as Based on the second feature information, multiple candidate resources are filtered to obtain the target resource and push the target resource to the user.

18. The method according to claim 17, wherein, The second triggering entity includes a second dialogue context or second voice data; The step of performing feature analysis on the second triggering entity based on the second intent to obtain the second feature information of the second triggering entity includes: Field extraction is performed on the second dialogue context or the second voice data to obtain the second content field; and The second feature information is determined based on the second content field.

19. The method according to claim 18, wherein, The step of filtering multiple candidate resources based on the second feature information to obtain the target resources includes: The resource selection control on the front-end interface displays the candidate resource identifiers of at least one candidate resource, wherein the at least one candidate resource is obtained by filtering multiple candidate resources based on the second feature information, and the resource selection control is configured to be spaced at a preset distance from the second content field; and In response to detecting a user's selection operation for a target resource identifier among the at least one candidate resource identifiers, the candidate resource corresponding to the target resource identifier is determined as the target resource.

20. The method of claim 17, wherein, The second triggering entity includes user actions and a second dialogue context; The step of performing feature analysis on the second triggering entity based on the second intent to obtain the second feature information of the second triggering entity includes: In response to the detection of the user operation, the second dialogue context determined based on the user operation is subjected to field extraction to obtain the second content field; as well as The second feature information is determined based on the second content field.

21. The method according to claim 20, wherein, The step of filtering multiple candidate resources based on the second feature information to obtain the target resources includes: The resource selection control on the front-end interface displays the candidate resource identifiers of at least one candidate resource, wherein the at least one candidate resource is obtained by filtering multiple candidate resources based on the second feature information; when the second content field is located in the display area of ​​the front-end interface, the resource selection control is configured to be spaced at a preset position interval from the second content field; when the second content field is not located in the display area, an icon indicates the position of the second content field; and In response to detecting a user's selection operation for a target resource identifier among the at least one candidate resource identifiers, the candidate resource corresponding to the target resource identifier is determined as the target resource.

22. The method according to claim 18 or 20, wherein, The second intention to determine the user's access to resources includes: The second intent is determined based on the second triggering entity.

23. A resource delivery device, comprising: The first determination module is used to determine the user's primary intent to obtain bulk resources; The first feature analysis module is used to perform feature analysis on the first triggering subject based on the first intent, and obtain the first feature information of the first triggering subject; as well as The first filtering module is used to filter multiple candidate resources based on the first feature information, obtain a target resource set, and push the target resource set to the user.

24. A resource delivery device, comprising: The second determination module is used to determine the user's second intention to obtain resources; The second feature analysis module is used to perform feature analysis on the second triggering subject based on the second intent, and obtain the second feature information of the second triggering subject; as well as The second filtering module is used to filter multiple candidate resources based on the second feature information, obtain the target resource, and push the target resource to the user.

25. An electronic device, comprising: One or more processors; Memory, used to store one or more instructions. When the one or more instructions are executed by the one or more processors, the one or more processors cause the one or more processors to implement the method of any one of claims 1 to 22.

26. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the method of any one of claims 1 to 22.

27. A computer program product comprising computer-executable instructions, which, when executed, are used to implement the method of any one of claims 1 to 22.

Images