Method and apparatus for acquiring recommended material, device, medium and product

Abstract

One or more embodiments of this document provide a method, apparatus, device, medium, and product for obtaining recommended content. The method includes: responding to receiving a first operation triggered in a first document; obtaining first information associated with a first content block in the first document based on the document type of the first document, the first operation being used to trigger content recommendation for the first content block; obtaining first query information based on the first information associated with the first content block; performing a search based on the first query information to obtain multiple candidate recall information; and obtaining first recommended content based on the first information associated with the first content block and the multiple candidate recall information. For user-triggered content recommendation operations, the obtained recommended content is associated with both the document type and specific content blocks within the document, improving the matching degree between the recommended content and the document content to be improved within the content block, making the recommended content more targeted and accurate.

Description

Technical Field

[0001] One or more embodiments of this document relate to a method for acquiring recommendation materials, an apparatus for acquiring recommendation materials, an electronic device, a computer-readable storage medium, and a computer program product. Background Technology

[0002] A document processing system can be understood as a software system used to manage documents. Users can use the document processing system to perform document-related operations, such as creating documents, editing documents, and viewing documents.

[0003] When users are editing documents using a document processing system, accurate material recommendations are particularly important to help them improve editing efficiency. Summary of the Invention

[0004] This summary section is provided to briefly introduce the concepts, which will be described in detail in the detailed description section below. This summary section is not intended to identify key or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0005] At least one embodiment of this document provides a method for obtaining recommended materials, comprising: responding to receiving a first operation triggered in a first document; obtaining first information associated with a first content block in the first document according to the document type of the first document, wherein the first operation is used to trigger material recommendation for the first content block; obtaining first query information according to the first information associated with the first content block; performing a search based on the first query information to obtain multiple candidate recall information; and obtaining a first recommended material according to the first information associated with the first content block and the multiple candidate recall information, wherein the first recommended material is related to at least a portion of the document content to be improved in the first content block.

[0006] At least one embodiment of this document provides an apparatus for obtaining recommended materials, comprising: a first acquisition module configured to: in response to receiving a first operation triggered in a first document, acquire first information associated with a first content block in the first document according to the document type of the first document, wherein the first operation is used to trigger material recommendation for the first content block; a second acquisition module configured to: acquire first query information based on the first information associated with the first content block; a recall module configured to: perform a retrieval based on the first query information to obtain multiple candidate recall information; and a third acquisition module configured to: obtain a first recommended material based on the first information associated with the first content block and the multiple candidate recall information, wherein the first recommended material is related to at least a portion of the document content to be improved in the first content block.

[0007] At least one embodiment of this document provides an electronic device, including: at least one processor; and at least one memory, including one or more computer program instructions; wherein the one or more computer program instructions are executed by the processor to perform the method for obtaining recommended materials provided by at least one embodiment of this document.

[0008] At least one embodiment of this document provides a computer-readable storage medium for non-transitory storage of computer-readable instructions, wherein the computer-readable instructions, when executed by a processor, implement the method for obtaining recommended materials provided in at least one embodiment of this document.

[0009] At least one embodiment of this document provides a computer program product, including a computer program that, when executed by a processor, implements the method for obtaining recommended materials provided in at least one embodiment of this document.

[0010] In a method for obtaining recommended materials provided in at least one embodiment of this paper, in a document editing scenario, for a material recommendation operation triggered by a user, first information associated with the content block requiring material recommendation is obtained in conjunction with the document type. Then, query information is obtained from the first information, and retrieval is performed using the query information to obtain candidate recall information. The first information and candidate recall information are organized to obtain recommended materials, thereby realizing material recommendation to the user. In this way, the obtained recommended materials are associated with both the document type and the specific content block in the document, improving the matching degree between the recommended materials and the document content that needs to be improved in the content block (such as missing document content in the content block or document content that needs to be updated in the content block), making the recommended materials more targeted and accurate. Attached Figure Description

[0011] The above and other features, advantages, and aspects of the embodiments herein will become more apparent when taken in conjunction with the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0012] Figure 1 This illustration schematically depicts an application scenario of a method for obtaining recommended materials provided in at least one embodiment of this paper;

[0013] Figure 2 The illustration shows a flowchart of a method for obtaining recommendation materials provided in at least one embodiment of this document;

[0014] Figure 3 The illustration shows a flowchart of a method for obtaining recommendation materials provided in at least one embodiment of this document;

[0015] Figure 4 The schematic diagram illustrates a structural schematic of an apparatus for obtaining recommended materials according to at least one embodiment of this document; and

[0016] Figure 5 A schematic diagram of the structure of an electronic device suitable for implementing at least one embodiment of the present invention is shown. Detailed Implementation

[0017] One or more embodiments of this document will now be described in more detail with reference to the accompanying drawings. While some embodiments of this document are shown in the drawings, it should be understood that this document can be implemented in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided to provide a more thorough and complete understanding of this document. It should be understood that the accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of this document.

[0018] It should be understood that the steps described in the method embodiments herein may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this document is not limited in this respect.

[0019] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0020] It should be noted that the concepts of "first" and "second" mentioned in this article are only used to distinguish different devices, modules or units, and are not used to limit the order of the functions performed by these devices, modules or units or their interdependencies.

[0021] It should be noted that the terms "one" and "more" used in this document are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0022] The names of the messages or information exchanged between the various devices in the embodiments herein are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0023] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition, use, storage or deletion of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0024] It is understood that before using the technical solutions disclosed in the embodiments of this article, relevant users should be informed of the type, scope of use, and usage scenarios of the information involved in this article and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means. Relevant users may include any type of rights holder, such as individuals, enterprises, or groups.

[0025] For example, in response to receiving an active request from a user, a prompt message is sent to the relevant user to clearly inform the user that the requested operation will require obtaining and using the user's information, thereby enabling the relevant user to choose whether to provide information to the software or hardware such as the electronic device, application, server, or storage medium that performs the operation of any embodiment of the present document based on the prompt message.

[0026] As an optional but non-restrictive implementation, in response to a user's active request, a prompt message can be sent to the user, such as a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide information to the electronic device.

[0027] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation method described in this article. Other methods that comply with relevant laws and regulations may also be applied to the implementation method described in this article.

[0028] A document processing system can be understood as a software system used to manage documents. It is widely used in various scenarios such as office work, study, and creation. Users can use the document processing system to perform document-related operations, such as creating, editing, and viewing documents.

[0029] When users edit documents using a document processing system, they often need to use some materials to edit the document content. For example, users need to manually collect the necessary materials and then combine the collected materials to edit the document in the document processing system.

[0030] Taking the example of a user editing a technical introduction document for technology A in a document processing system, the user needs to search for materials related to technology A from different channels such as the internal document library and the Internet, organize the materials related to technology A, and then input the document content into the technical introduction document for technology A.

[0031] The above methods require users to manually collect and integrate materials, which consumes a lot of time in the material preparation process and reduces the user's document editing efficiency. Therefore, some document processing systems provide material recommendation capabilities, which proactively provide users with recommended materials related to the content of the currently edited document during the document editing process. Using recommended materials helps users improve document editing efficiency, and providing users with accurate recommended materials is particularly important.

[0032] To at least partially solve the above-mentioned technical problem, at least one embodiment of this paper provides a method for obtaining recommended materials. The method includes: in response to receiving a first operation triggered in a first document, obtaining first information associated with a first content block in the first document according to the document type of the first document, the first operation being used to trigger material recommendation for the first content block; obtaining first query information based on the first information associated with the first content block; performing a search based on the first query information to obtain multiple candidate recall information; and obtaining a first recommended material based on the first information associated with the first content block and the multiple candidate recall information, wherein the first recommended material is related to at least a portion of the document content to be improved in the first content block.

[0033] Based on the method for obtaining recommended materials provided in at least one embodiment of this document, at least one embodiment of this document also provides an apparatus, electronic device, computer-readable storage medium, and computer program product for obtaining recommended materials.

[0034] In a method for obtaining recommended materials provided in at least one embodiment of this paper, in a document editing scenario, for a material recommendation operation triggered by a user, first information associated with the content block requiring material recommendation is obtained in conjunction with the document type. Then, query information is obtained from the first information, and retrieval is performed using the query information to obtain candidate recall information. The first information and candidate recall information are organized to obtain recommended materials, thereby realizing material recommendation to the user. In this way, the obtained recommended materials are associated with both the document type and the specific content block in the document, improving the matching degree between the recommended materials and the document content that needs to be improved in the content block (such as missing document content in the content block or document content that needs to be updated in the content block), making the recommended materials more targeted and accurate.

[0035] The following detailed description, with reference to the accompanying drawings, describes one or more embodiments and some examples thereof.

[0036] Figure 1 The illustration shows an application scenario diagram of a method for obtaining recommended materials provided in at least one embodiment of this article.

[0037] like Figure 1As shown, the application scenario provided in this embodiment may include user 101, terminal device 102, and server 103. Terminal device 102 can be various electronic devices capable of providing interactive pages, such as smart wearable devices, smart appliances, smart cars, mobile phones, tablets, laptops, or desktop computers.

[0038] In one or more embodiments of this document, a client may be installed in the terminal device 102. The client may be a client of a document processing system. The server 103 may be a server that provides support for the operation of the client installed in the terminal device 102. That is, the server 103 may be a server of a document processing system. For example, the server 103 may be a server for a local area network or a wide area network, or a cloud server, etc. The one or more embodiments of this document do not limit this.

[0039] User 101 can be a user of a client installed on terminal device 102. For example, user 101 can be an editor who edits documents, a viewer who views documents, or a commenter who makes comments on documents in a document processing system.

[0040] In one or more embodiments described herein, the document processing system can be deployed in different ways. In some embodiments, the document processing system can be deployed locally, for example, it can be a document processing application (APP), document processing tool, document processing plugin, etc., installed on terminal device 102; or, for example, it can be deployed in the cloud, that is, it can provide document management services in the form of cloud services, for example, it can be an online document processing tool. Furthermore, the document processing system can be integrated into other systems, providing document processing functions as a functional module within those systems. For example, it can be integrated into an office collaboration system, providing document processing services (e.g., cloud document processing services) within that system.

[0041] The server 103 can communicate with the terminal device 102, for example, by providing the client installed on the terminal device 102 with relevant data (such as document content) required by the terminal device to run the client; or, for example, the server 103 can also receive relevant data returned by the terminal device 102 during the running of the client (such as operation data on the document triggered by user 101 on the client).

[0042] For example, the material recommendation method provided in one or more embodiments of this document can be implemented in software, hardware, firmware, or any combination thereof.

[0043] For example, the material recommendation method provided in one or more embodiments of this document is applicable to a terminal device, which can load and execute the material recommendation method. The embodiments of this document do not limit this. For example, the terminal device may include a central processing unit (CPU), graphics processing unit (GPU), digital signal processor (DSP), neural network processing unit (NPU), or other forms of processing units with data processing capabilities and / or instruction execution capabilities, storage units, etc. The server or terminal device may also have an operating system and various types of application programming interfaces (APIs) installed, implementing the material recommendation method provided in the embodiments of this document by running code or instructions.

[0044] The following will combine Figure 2 and Figure 3 A method for obtaining recommended materials, provided in at least one embodiment of this document, will be described in detail.

[0045] Figure 2 The illustration shows a flowchart of a method for obtaining recommended materials provided in at least one embodiment of this document.

[0046] like Figure 2 As shown, the method for obtaining recommendation materials in this embodiment includes steps S201 to S204. In some embodiments, the entity executing the method for obtaining recommendation materials can be an electronic device with a client deployed, an electronic device with a server deployed, or any electronic device that communicates between the client and the server; one or more embodiments herein do not limit this. The steps included in the method for obtaining recommendation materials are described below:

[0047] Step S201: In response to receiving a first operation triggered in the first document, obtain first information associated with the first content block in the first document according to the document type of the first document.

[0048] In one or more embodiments herein, the first document may be understood as any document, for example, the first document may be a document that is currently open and being viewed.

[0049] By dividing the first document in terms of structural logic, the first document is divided into at least one content block. That is, the first document may include one or more content blocks. A content block can be understood as a document section. A content block corresponds to at least a continuous portion of the document content in the first document. For example, the first document may include multiple subheadings. The document content under a subheading can form a content block. Or, for example, the first document may include a table. At least one adjacent area in the table can form a content block.

[0050] In one or more embodiments of this document, a user is supported in triggering a first operation in a first document. The first operation can be used to trigger material recommendations for a first content block in the first document. The first content block can be any content block among one or more content blocks included in the first document.

[0051] In other words, when a user is processing the first document (for example, editing the document), if there is a need for reference materials for the first content block, the first operation can be triggered to recommend materials for the first content block.

[0052] One or more embodiments of this document support triggering the first operation in different ways. For example, the document page of the first document can provide a material recommendation control, and the user can trigger the first operation by triggering the material recommendation control; or, for another example, the triggering logic of the first operation can be pre-configured, and the first operation is triggered when the operation triggered by the user in the first document meets the triggering logic of the first operation.

[0053] For example, the triggering logic for the first operation can be: when the user opens the first document, the first operation is triggered; or, the triggering logic for the first operation can also be: when the user opens the first document and moves to the first content block, the first operation is triggered; or, when the user edits the first content block, the first operation is triggered.

[0054] Material recommendation for the first content block in the first document can be understood as: recommending materials for the document content corresponding to the first content block in the first document; that is, the object of material recommendation is the first content block. In other words, in one or more embodiments of this paper, material recommendation is based on content blocks, providing targeted and fine-grained material recommendations for a specific content block in the first document.

[0055] Upon receiving the first operation, the first content block requiring material recommendation can be located within the first document based on the first operation. In some embodiments, the first content block can be obtained as follows: the first content block is determined based on the operation position of the first operation within the first document. Thus, based on the material recommendation needs indicated by the operation position of the operator in the first operation, the first content block satisfying the operator's material recommendation needs is located from the first document, ensuring that the material recommendation process targets the content block needed by the operator and guaranteeing the effectiveness of the material recommendation.

[0056] In other embodiments, the first document may include first document content, and the operator of the first operation corresponds to a first operator identifier, for example, the first operator identifier may be the account name of the operator of the first operation. In this case, the first content block can be obtained as follows: in response to the first document content including the first operator identifier, the first content block is determined based on the position information of the first operator identifier in the first document. Thus, by combining the operator identifier mentioned in the first document content, the content block targeted by the first operation is automatically located, and the first content block related to the operator of the first operation that needs to be recommended for materials is identified, without the operator of the first operation needing to specify the content block, making it convenient for the operator of the first operation to quickly start the material recommendation process.

[0057] Furthermore, considering that one or more embodiments of this paper recommend materials, recommending materials after the first content block has already included complete document content would lead to a waste of computing resources. Therefore, it is also possible to determine whether the first content block has at least some document content that needs to be improved. That is, to determine whether the document content of the first content block is complete. Incomplete document content of the first content block can be understood as the first content block being empty, the document content in the first content block not being fully edited, or the document content in the first content block needing to be updated.

[0058] For example, based on the content block title and content block body of the first content block, it is determined whether the first content block has at least some document content that needs to be improved. In response to the content block title being not empty and the content block body including incomplete content, the first content block has at least some document content that needs to be improved.

[0059] In one or more embodiments of this document, a first operation triggered in a first document is received, indicating that the user has a need for material recommendations for the first content block. In order to provide the user with accurate recommended materials that are strongly associated with the first content block, first information associated with the first content block in the first document is obtained according to the document type of the first document.

[0060] Document types can be used to represent the characteristics of document content. For example, based on whether a document is updated periodically, document types can include periodically updated documents and non-periodic updated documents. Periodically updated documents can be understood as documents that are updated and maintained periodically, such as weekly reports and quarterly meeting documents. Non-periodic updated documents can be understood as documents whose update and maintenance cycles are irregular, such as technical documents and creative documents.

[0061] The first information associated with the first content block can be understood as information in the first document that is related to the first content block and can help generate the first recommended material. In other words, the first information associated with the first content block serves as the basis for generating the first recommended material, and the first recommended material is related to the first information associated with the first content block.

[0062] In one or more embodiments of this document, the first information associated with the first content is related to the document type of the first document. That is, for documents of different document types, first recommended materials are generated based on different first information associated with the first content. By configuring different recommended material generation logics for different document types, a more reasonable and accurate material recommendation acquisition process is achieved.

[0063] In one or more embodiments of this document, the document type of the first document can be obtained by combining at least one of the document title of the first document and the content of the first document in the first document. The document title of the first document can be understood as the name of the first document, and the content of the first document in the first document can be understood as the existing document content in the first document. That is, the document type of the first document is determined by using the document title and the content of the first document together.

[0064] Figure 3 The illustration shows a flowchart of a method for obtaining recommended materials provided in at least one embodiment of this document.

[0065] like Figure 3 As shown, by obtaining the document type of the first document, it is determined whether the document type of the first document is a periodically updated document. Periodically updated documents and non-periodically updated documents have different first information associated with the first content block.

[0066] In some embodiments, obtaining first information associated with a first content block in a first document according to the document type of the first document includes: in response to the document type of the first document being a non-periodic update document, obtaining context information of the first content block in the first document and the content block title corresponding to the first content block; or in response to the document type of the first document being a periodically update document, obtaining context information of the first content block in the first document and at least one of the following: the content block title corresponding to the first content block and the content of a historical content block corresponding to the first content block, wherein the content of the historical content block includes: the content of the first content block in the previous update cycle.

[0067] In other words, when the first document is a non-periodicly updated document, the first information associated with the first content block includes: the context information of the first content block in the first document and the content block title corresponding to the first content block. When the first document is a periodically updated document, the first information associated with the first content block includes: the context information of the first content block in the first document and at least one of the following: the content block title corresponding to the first content block and the content of the historical content block.

[0068] The context information of the first content block can be understood as the document content of other content blocks in the first document that are associated with the first content block. For example, the context information of the first content block may include the document content of the content block in the first document that precedes the first content block (e.g., including the content block title and content block body) and the document content of the content block in the first document that follows the first content block (e.g., including the content block title and content block body).

[0069] Since the context information of the first content block is related to the document content of the first content block in terms of position, the context information of the first content block is of reference value for editing the document content of the first content block. Therefore, regardless of whether the document is updated periodically or non-periodically, the first recommended material can be generated by combining the context information of the first content block.

[0070] The content block title corresponding to the first content block can reflect the theme of the first content block. For example, multi-level headings related to the first content block can be identified. These multi-level headings can include the first heading in the first content block and at least one parent heading associated with the first heading. By concatenating these multi-level headings, the content block title is obtained. In other words, considering the accuracy of the content block title, in one or more embodiments of this paper, not only the first heading in the first content block is identified, but also higher-level parent headings associated with the first content block are additionally identified. By concatenating these parent headings, the content block title includes multi-level headings from broad to narrow scope, ensuring the semantic coherence and accuracy of the content block title. Thus, by concatenating multi-level headings, an accurate content block title for the first content block is obtained. Based on the accurate content block title of the first content block, the accurate and complete content of the content block text can be obtained, thereby improving the accuracy of the content block text integrity determination.

[0071] Since the title of the first content block can indicate the document content that the first content block should include to a certain extent, the title of the first content block is of reference value for editing the document content of the first content block. Therefore, whether the document is updated periodically or non-periodically, the first recommended material can be generated by combining the title of the first content block.

[0072] Furthermore, considering that periodically updated documents contain historical content blocks of the same content block (i.e., the first content block) from previous update cycles, the content of these historical content blocks (such as the main text and comments) can serve as a reference for writing the first content block in the current update cycle. Therefore, for the first document of a periodically updated document, the content of the historical content blocks can be combined to generate the first recommended material.

[0073] In some possible implementations, obtaining the context information of the first content block in the first document and at least one of the following: the content block title corresponding to the first content block and the content of the historical content block corresponding to the first content block includes: obtaining the historical content block corresponding to the first content block; in response to not obtaining the historical content block or the content of the historical content block being empty, obtaining the context information of the first content block in the first document and the content block title corresponding to the first content block; or in response to the content of the historical content block not being empty, obtaining the context information of the first content block in the first document, the content block title corresponding to the first content block, and the content of the historical content block.

[0074] The historical content block corresponding to the first content block can be understood as the first content block in the previous update cycle of the current update cycle. For example, if the first document is a weekly report document with an update cycle of one week, the first content block is section A in the weekly report document, and the historical content block corresponding to the first content block can be section A in the weekly report document of last week.

[0075] In other words, firstly, the historical content block corresponding to the first content block is obtained. If the historical content block cannot be obtained, or the content of the obtained historical content block is empty, it indicates that there is no historical content that can be referenced or learned from. In this case, the content of the meaningful historical content block cannot be obtained. Therefore, the first information associated with the first content block does not include the content of the historical content block. If the content of the historical content block can be obtained, it indicates that there is historical content that can be referenced or learned from. In this case, the first information associated with the first content block includes the content of the historical content block.

[0076] In this way, by obtaining the historical content blocks corresponding to the first content block, it is determined whether meaningful historical content blocks can be obtained. Then, in different situations of periodically updating the document, the specific information included in the first information associated with the first content block is determined, ensuring that the processing logic for obtaining the first information associated with the first content block is smooth.

[0077] One or more embodiments herein do not limit the method of obtaining historical content blocks. In some embodiments, obtaining the historical content block corresponding to the first content block includes: in response to the first document including multiple associated content blocks corresponding to different update cycles, obtaining the historical content block corresponding to the first content block in the first document; in other embodiments, obtaining the historical content block corresponding to the first content block in the previous update cycle includes: obtaining the historical document corresponding to the first document in the previous update cycle based on the document title of the first document, and obtaining the historical content block corresponding to the first content block in the historical document.

[0078] Periodically updated documents can be updated and maintained in different ways. In the case of a document containing content from multiple update cycles, the historical content block can be directly obtained from the first document. For example, if the first document is a weekly report document and the first content block is section A in the weekly report document, the document content is edited in the first document in each update cycle. The same content block forms related content blocks in multiple cycles. Therefore, the document content of the previous update cycle can be located from the first document, and then the historical content block (i.e., section A) can be located from the document content of the previous update cycle.

[0079] For cases where document content from different update cycles is scattered across different documents, i.e., a new document is created for each update cycle, and the content of that update cycle is updated within the new document, considering that document titles from different update cycles are usually quite similar, we first use the document title to retrieve the historical document, and then retrieve the historical content block from the historical document. For example, if the first document is a weekly report document, and the first content block is section A in the weekly report document, we first use the document title of the first document (e.g., the 10th weekly report) to retrieve the historical document (e.g., the document title of the historical document could be the 9th weekly report), and then locate the historical content block (i.e., section A) in the historical document.

[0080] Therefore, for periodically updated documents, different periodic update methods are fully considered, and different methods are used to obtain historical content blocks under different periodic update methods to ensure that the processing logic for obtaining historical content blocks is smooth and that historical content blocks can be accurately obtained under different periodic update methods.

[0081] Step S202: Obtain the first query information based on the first information associated with the first content block.

[0082] The first query information can be understood as a query statement used to retrieve candidate materials that can generate the first recommended material. The first query information is generated based on the first information associated with the first content block. That is, the first query information has a strong correlation with the document type and the first content block of the first document, ensuring that the retrieval process targets the first content block in the first document.

[0083] As mentioned earlier, compared to non-periodicly updated documents, the first information associated with the first content block corresponding to periodically updated documents can additionally include the content of historical content blocks. The granularity of the first query information corresponding to periodically updated documents is finer. Therefore, the first query information corresponding to non-periodicly updated documents can be understood as a coarse query, while the first query information corresponding to periodically updated documents can be understood as a detailed query.

[0084] The first query information can be used to retrieve candidate materials related to at least part of the document content to be improved in the first content block. For example, the first query information can be used to retrieve candidate materials related to the content block title of the first content block.

[0085] In some embodiments, considering that periodically updated documents are updated and maintained in each update cycle, the first query information can be time-limited to reduce the retrieval of candidate materials unrelated to the current update cycle. For example, obtaining the first query information based on the first information associated with the first content block includes: in response to the document type of the first document being a periodically updated document, obtaining the first query information, which is used to query information updated within the time cycle corresponding to the first document, and the time cycle corresponding to the first document is determined by the document title of the first document or the update cycle of the first document.

[0086] The time period corresponding to the first document can be understood as the time period corresponding to this update. The time period corresponding to the first document is the same as the update cycle of the first document. The starting point of the time period corresponding to the first document can be the end of the previous update cycle. For example, the first document is a weekly report document, the update cycle of the first document is one week, the previous update cycle is from January 1 to January 7, the first content block is section A in the weekly report document, and the first query information can be "What new results were there in section A between January 8 and January 14?".

[0087] The time period corresponding to the first document can be obtained from the document title. For example, if the document title of the first document is "Project Progress Update from January 1st to January 7th", then the time period corresponding to the first document can be from January 1st to January 7th. The time period corresponding to the first document can also be obtained from the update cycle of the first document. For example, if the update cycle of the first document is one week and the end time of the previous update cycle is December 31st, then the time period corresponding to the first document can be from January 1st to January 7th.

[0088] Thus, for periodically updated documents, since past candidate materials have low reference value for document editing in the current update cycle (i.e., the time period corresponding to the first document), content for limiting the update time is added to the first query information. This ensures that the first query information can only be used to retrieve candidate materials within the time period corresponding to the first document, guaranteeing that the candidate materials retrieved using the first query information (i.e., candidate recall information) are all candidate materials updated within the current update cycle. This reduces the adverse impact of candidate materials updated too early on document editing in the current update cycle. By adjusting the first query information, the accuracy of the first recommended material is improved.

[0089] One or more embodiments of this document do not limit the method of obtaining the first query information. In some embodiments, a query template for the first query information is pre-configured, and the first query information is obtained by extracting some key information from the first information associated with the first content block and filling the key information into the query template.

[0090] In other embodiments, the first query information is obtained automatically using a model. For example, the first query information is obtained based on the first information associated with the first content block, including: sending the first information associated with the first content block to a first model, and obtaining the first query information based on the output of the first model.

[0091] The first model can be understood as any artificial intelligence model with the ability to recognize element information. For example, the first model can include any one or a combination of multiple large-scale models such as language models, speech models, vision models, and multimodal models. For example, the first model can be a model built on a transformer architecture, a model built on a recurrent neural network, a model built on an attention mechanism, etc. Alternatively, the first model can also be a model obtained by improving on the transformer architecture, such as a mixture of experts (MoE) model.

[0092] One or more embodiments in this paper do not limit the method of obtaining the first model. For example, the first model can be an existing, open-source artificial intelligence model. Alternatively, the first model can be an artificial intelligence model obtained by fine-tuning the pre-trained artificial intelligence model based on the pre-trained artificial intelligence model and using training data related to query information generation (such as the first information associated with other content blocks, query information generation requirements, generated query information, etc.).

[0093] The first model can generate initial query information, prompt words, based on prompt learning. In generative tasks (such as text generation, question answering, and dialogue tasks), these prompt words can be used to guide the AI ​​model to make specific outputs. By configuring prompt words, the AI ​​model can understand the background and requirements of the task, enabling it to handle different types of processing tasks without retraining, thus increasing the scalability and flexibility of the AI ​​model.

[0094] For example, a first prompt word is generated and sent to a first model. The first prompt word may include: first information associated with the first content block, query information generation requirements, and prompt information for indicating that query information that meets the query information generation requirements is generated based on the first information associated with the first content block. The output of the first model is received (e.g., output results that include the query information). Based on the output of the first model (e.g., information extraction from the output of the first model), the first query information is obtained.

[0095] By sending the first prompt word to the first model, the first model can analyze the query information generation requirements by leveraging the prompt word's prompting capabilities. It can then filter the necessary information from the first information associated with the first content block, and combine the filtered information to output the results including the first query information. This achieves automatic generation of the first query information and improves the automation and efficiency of first query information generation.

[0096] Step S203: Based on the first query information, perform a retrieval to obtain multiple candidate recall information.

[0097] Step S204: Obtain the first recommended material based on the first information associated with the first content block and multiple candidate recall information.

[0098] In one or more embodiments of this document, recommended materials can be understood as material information used to assist in editing document content. The first recommended material can be related to at least part of the document content to be improved in the first content block. The at least part of the document content to be improved can be understood as lacking at least part of the document content or having at least part of the document content that needs to be updated. That is, the first recommended material can be used to assist the operator of the first operation in editing the at least part of the document content that is missing in the first content block, or the first recommended material can be used to assist the operator of the first operation in editing the at least part of the document content that needs to be updated in the first content block.

[0099] For example, the first content block should include five document contents from (1) to (5). The first content block is missing two document contents from (3) and (4). The document contents from (5) in the first content block need to be updated. The first recommended material is related to the three document contents from (3) to (5) and can be used to assist the operator of the first operation in editing the three document contents from (3) to (5) in the first content block.

[0100] The first recommended material can include material information in different forms, such as text material information, video material information, image material information, audio material information, file material information, etc.

[0101] For example, the first recommended material may include one or more of the following: information related to the first document and the first content block, detailed information about the recommended material, and source information about the recommended material. The information related to the first document and the first content block helps the operator of the first operation to quickly understand the document and content block corresponding to the first recommended material. The detailed information about the recommended material helps the operator of the first operation to quickly determine whether the recommended material is available. The source information about the recommended material helps the operator of the first operation to view the original text of the recommended material and verify its authenticity.

[0102] In some possible implementations, considering that the first recommended material is used to assist the operator of the first operation in editing the first content block, the first recommended material provided to the operator of the first operation should all be material information that the operator of the first operation has access to. For example, based on the first query information, a search is conducted in a data source that the operator of the first operation has access to obtain multiple candidate recall information.

[0103] The access permissions of the operator of the first operation can be used to represent the range of information that the operator of the first operation can access. By ensuring that the data source used for retrieval complies with the access permissions of the operator of the first operation, it is ensured that the material information included in the first recommended material complies with the access permissions of the operator of the first operation, thereby reducing the possibility of recommending first recommended material outside the access permissions of the operator of the first operation.

[0104] In some possible implementations, the data source can also correspond to the document type. For example, based on the first query information, a search is performed in at least one data source corresponding to the first document type to obtain multiple candidate recall information.

[0105] In other words, different document types are configured with different data sources. That is, when recommending materials for documents of different document types, the recommended materials can be generated from different data sources as material acquisition channels.

[0106] For example, for the first document, which is a periodically updated document, since periodically updated documents need to be updated and maintained periodically, and the material information for updating and maintaining periodically updated documents is usually scattered in different functional modules of the office collaboration system, at least one data source corresponding to the periodically updated document can include different functional modules of the office collaboration system, such as instant messaging module, meeting module, document module, etc. Based on the information in different functional modules of the office collaboration system, multiple candidate recall information is obtained.

[0107] For the first document, which is a technical document, since technical documents are related to technology and the material information for editing technical documents usually exists in Internet materials or other materials, at least one data source corresponding to the technical document can include the Internet and internal knowledge bases, such as online search engines. Based on the information in the Internet and internal knowledge bases, multiple candidate recall information is obtained.

[0108] In this way, by combining the characteristics of different document types, multiple candidate recall information can be obtained from the data source corresponding to the document type, so that the source of the recommended material matches the first document type corresponding to the first document, making the generation process of recommended material more targeted and efficient.

[0109] Candidate recall information can be understood as candidate materials used to generate the first recommended material. For example, candidate recall information can be data from at least one data source that corresponds to the document type and has access rights to the operator of the first operation.

[0110] In other words, the process of generating the first recommended material can be divided into two stages: candidate material retrieval and candidate material integration. In the candidate material retrieval stage, a retrieval is performed based on the first query information to obtain multiple candidate recall information and select candidate materials. In the candidate material integration stage, the multiple candidate recall information is analyzed and organized to generate the first recommended material that meets the material recommendation requirements and can be provided to the operator of the first operation.

[0111] In this way, through the two stages of candidate material retrieval and candidate material integration, candidate materials (i.e., multiple candidate recall information) related to at least part of the document content to be improved in the first content block are obtained, ensuring the richness of candidate materials. Then, the candidate materials are automatically integrated to form the first recommended material that can be provided to the operator of the first operation. This eliminates the need for the operator of the first operation to manually organize candidate materials, and the first recommended material can provide intuitive assistance in the document editing scenario.

[0112] In some embodiments, considering the accuracy of the first recommended material and the efficiency of generating the first recommended material, multiple candidate recall information can be filtered to remove some of them. For example, after retrieving multiple candidate recall information based on the first query information, the multiple candidate recall information can be filtered according to at least one of the sources and update times of the multiple candidate recall information to retain at least a portion of the multiple candidate recall information.

[0113] On the one hand, by filtering and screening multiple candidate recall information, candidate recall information that has reference and learning value for at least part of the document content to be improved in the first content block is retained, thereby reducing the number of candidate recall information and improving the generation efficiency of the first recommended material. On the other hand, multiple candidate recall information are screened from two aspects: the source and the update time. That is, the reference value of candidate recall information to at least part of the document content to be improved in the first content block is measured from the two aspects of the source and the update time, so that the source and update time of the candidate recall information retained after screening meet the generation requirements of the first recommended material, thereby improving the generation quality of the first recommended material.

[0114] For example, the unretained candidate recall information may include at least one of the following: candidate recall information from the first document, candidate recall information from multiple historical documents, and candidate recall information that does not match the time period corresponding to the first document. The multiple historical documents include documents corresponding to the first document in multiple historical update periods. The time period corresponding to the first document is obtained through the document title of the first document or the update period of the first document.

[0115] Candidate recall information from the first document can be understood as document content belonging to the first document, that is, document content retrieved from the first document based on the first query information. For the first document currently being edited, the document content within the first document has low reference value for the document editing process; therefore, candidate recall information from the first document is filtered out.

[0116] Candidate recall information from multiple historical documents can be understood as the document content of the first document within multiple historical update cycles; that is, the document content retrieved from multiple historical documents based on the first query information. Since the document content of historical documents belongs to different update cycles than the first document, the document content of historical documents has low reference value for editing the first document. Therefore, candidate recall information from multiple historical documents is filtered out.

[0117] The time period corresponding to the first document can be obtained from the document title. For example, if the document title of the first document is "Project Progress Update from January 1st to January 8th", then the time period corresponding to the first document can be from January 1st to January 8th. The time period corresponding to the first document can also be obtained from the update cycle of the first document. For example, if the update cycle of the first document is one week and the end time of the previous update cycle is December 31st, then the time period corresponding to the first document can be from January 1st to January 8th.

[0118] Candidate recall information that does not match the time period corresponding to the first document can be understood as candidate recall information whose update time is outside the time period corresponding to the first document. Since the update time of the above candidate recall information does not match the time period corresponding to the first document, the above candidate recall information has low reference value for document editing of the first document. Therefore, candidate recall information that does not match the time period corresponding to the first document is filtered out.

[0119] In some embodiments, the model automatically integrates candidate recall information to generate a first recommended content. For example, obtaining the first recommended content based on first information associated with the first content block and multiple candidate recall information includes: sending the first information associated with the first content block and multiple candidate recall information to a second model, and obtaining the first recommended content based on the output of the second model.

[0120] The second model can be understood as any artificial intelligence model with the ability to recognize element information. For example, the second model can include any one or a combination of multiple large-scale models such as language models, speech models, vision models, and multimodal models. For example, the second model can be a model built on a transformer architecture, a model built on a recurrent neural network, a model built on an attention mechanism, etc. Alternatively, the second model can also be a model obtained by improving on the transformer architecture, such as a mixture of experts (MoE) model.

[0121] It should be noted that in some embodiments, the first model and the second model can be the same artificial intelligence model, while in other embodiments, the first model and the second model can be different artificial intelligence models. One or more embodiments herein do not limit this.

[0122] One or more embodiments in this paper do not limit the method of obtaining the second model. For example, the second model can be an existing, open-source artificial intelligence model. Alternatively, the second model can be an artificial intelligence model obtained by fine-tuning the pre-trained artificial intelligence model based on the pre-trained artificial intelligence model using training data related to the generation of recommendation materials (such as other candidate recall information, recommendation materials corresponding to other candidate recall information, etc.).

[0123] The second model can generate the first recommended material based on cue learning. For example, it can generate a second cue word and send the second cue word to the second model. The second cue word may include: first information associated with the first content block, candidate recall information, and cue information used to indicate the integration of candidate recall information to form recommended material. The model can receive the output of the second model (e.g., output results that indicate recommended material) and determine the first recommended material based on the output of the second model (e.g., information extraction from the output of the second model).

[0124] By sending the second prompt word to the second model, the second model can analyze the content of the candidate recall information and organize the candidate recall information into recommended materials, thereby outputting the output results including the first recommended material. This achieves automatic generation of the first recommended material without requiring the first operator to manually organize the candidate materials.

[0125] Furthermore, to better assist users in editing the first content block of the document without requiring significant manual adjustments and modifications to the first recommended material, personalized material recommendations can be provided. For example, the first operation can be triggered by the first operator. In this case, the first recommended material is obtained based on the first information associated with the first content block and multiple candidate recall information, including: obtaining the first recommended material based on the first information associated with the first content block, multiple candidate recall information, and the writing characteristics corresponding to the first operator, and matching the first recommended material with the writing characteristics corresponding to the first operator.

[0126] Writing characteristics can be understood as writing habits. These characteristics can include those related to the content being written. For example, when editing the same content block in a document, some users prefer to edit only the parts that have changed since the previous update, while others prefer to edit the entire document. Writing characteristics can also include those related to the writing structure. For example, when editing the same content, some users prefer to use a general-to-specific structure, while others prefer to use objects as a structure. Writing characteristics can also include those related to the writing style. For example, when editing the same material, some users prefer to select most of the material to edit, while others prefer to select only a small portion. Some users prefer to edit in one writing order, while others prefer to choose a different writing order.

[0127] Considering that different users have different writing characteristics, when generating the first recommended material, specific writing characteristics can also be combined to provide the first user with targeted first recommended material. This makes the first recommended material similar to the first user's writing characteristics. When recommending materials to the first user, the first user does not need to perform too much additional processing on the first recommended material and can use the first recommended material in the document editing process.

[0128] In some embodiments, a model is used to generate first recommended material that matches the writing features of the first operator. For example, obtaining the first recommended material based on first information associated with the first content block, multiple candidate recall information, and the writing features corresponding to the first operator includes: sending the first information associated with the first content block, multiple candidate recall information, and the historical writing information of the first operator to a third model, and obtaining the first recommended material based on the output of the third model.

[0129] The third model can be understood as any artificial intelligence model with the ability to recognize element information. For example, the third model can include any one or a combination of multiple large-scale models such as language models, speech models, vision models, and multimodal models. For example, the third model can be a model built on a transformer architecture, a model built on a recurrent neural network, a model built on an attention mechanism, etc. Alternatively, the third model can also be a model obtained by improving on the transformer architecture, such as a mixture of experts (MoE) model.

[0130] It should be noted that in some embodiments, the third model can be the same artificial intelligence model as the first model and the second model, while in other embodiments, the third model can be a different artificial intelligence model from the first model and the second model. One or more embodiments herein do not impose any restrictions on this.

[0131] One or more embodiments in this paper do not limit the method of obtaining the third model. For example, the third model can be an existing, open-source artificial intelligence model. Alternatively, the third model can be an artificial intelligence model obtained by fine-tuning the pre-trained artificial intelligence model based on the pre-trained artificial intelligence model using training data related to the generation of recommendation materials (such as historical writing information of other operators, other candidate recall information, and recommendation materials corresponding to other candidate recall information).

[0132] The third model can generate the first recommended material based on prompt learning. For example, it can generate a third prompt word and send the third prompt word to the third model. The third prompt word may include: first information associated with the first content block, candidate recall information, historical writing information of the first operator, and prompt information for indicating that the candidate recall information is integrated to form recommended material that matches the historical writing information of the first operator. The model can receive the output of the third model (e.g., the output result that includes the recommendation material) and determine the first recommended material based on the output of the third model (e.g., information extraction from the output of the third model).

[0133] By sending a third prompt word to a third model, the third model can learn the writing characteristics of the first operator based on the first operator's historical writing information, and at the same time analyze the content of the candidate recall information, organize the candidate recall information into recommended materials, and then output the output results including the first recommended materials that match the writing characteristics of the first operator, thereby realizing the generation of targeted recommended materials for the first operator.

[0134] By generating a first recommended material, the first recommended material is provided as response data for the first operation to the operator of the first operation. For example, the first recommended material is sent to the client installed on the terminal device used by the operator of the first operation, so that the first recommended material can be rendered and displayed on the terminal device used by the operator of the first operation.

[0135] In one or more embodiments of this document, the first recommended material can provide multifaceted assistance to the operator of the first operation in the document editing scenario of the first content block. For example, the first recommended material can be sent to the operator of the first operation so that the first recommended material is presented to the operator of the first operation at a first position associated with the first content block, or the first recommended material can be sent to the operator of the first operation so that the second document content generated based on the first recommended material is presented to the operator of the first operation in the first content block.

[0136] The first position associated with the first content block can be understood as a position that is related to the location of the first content block. For example, the first position associated with the first content block could be above the first content block.

[0137] On the one hand, the first recommended material can be directly pushed to the operator of the first operation. By presenting the first recommended material to the operator, the operator can view the first recommended material and use it to edit at least some of the document content that needs to be improved in the first content block. For example, the operator can write at least some of the missing document content in the first content block, or modify at least some of the document content that needs to be updated in the first content block. On the other hand, the first recommended material can also be used to generate second document content. The second document content can be at least some of the missing document content in the first content block, or at least some of the updated document content in the first content block. By presenting the second document content to the operator of the first operation, the operator does not need to manually edit the document, thus realizing the automatic continuation or automatic rewriting of the first content block.

[0138] Based on the method for obtaining recommendation materials provided in at least one embodiment of this document, at least one embodiment of this document also provides an apparatus for obtaining recommendation materials. The following will be combined with... Figure 4 A detailed description is provided of the apparatus used to acquire recommended materials.

[0139] Figure 4 The illustration shows a schematic diagram of a device for obtaining recommended materials, provided in at least one embodiment of this document.

[0140] like Figure 4As shown, the apparatus 400 for acquiring recommended materials in this embodiment includes a first acquisition module 401, a second acquisition module 402, a recall module 403, and a third acquisition module 404. For example, the first acquisition module 401, the second acquisition module 402, the recall module 403, and the third acquisition module 404 can be implemented by hardware (e.g., circuit) modules or software modules. The following embodiments are similar and will not be repeated. For example, the first acquisition module 401, the second acquisition module 402, the recall module 403, and the third acquisition module 404 can be implemented by a central processing unit (CPU), a general-purpose graphics processor (GPGPU), a graphics processing unit (GPU), a tensor processor (TPU), a field-programmable gate array (FPGA), or other processing units with data processing capabilities and / or instruction execution capabilities, along with corresponding computer instructions.

[0141] The first acquisition module 401 is configured to: in response to receiving a first operation triggered in a first document, acquire first information associated with a first content block in the first document according to the document type of the first document, wherein the first operation is used to trigger material recommendation for the first content block. For example, the first acquisition module 401 can be configured to execute step S201 described above; its specific implementation principle can be found in the relevant description of step S201, and will not be repeated here.

[0142] The second acquisition module 402 is configured to: acquire first query information based on the first information associated with the first content block. For example, the second acquisition module 402 can be configured to execute step S202 as described above; its specific implementation principle can be found in the relevant description of step S202, and will not be repeated here.

[0143] The recall module 403 is configured to perform a retrieval based on the first query information to obtain multiple candidate recall information. For example, the recall module 403 can be configured to execute step S203 described above. The specific implementation principle can be referred to the relevant description of step S203, which will not be repeated here.

[0144] The third acquisition module 404 is configured to: obtain a first recommended material based on the first information associated with the first content block and the plurality of candidate recall information, wherein the first recommended material is related to at least a portion of the document content to be improved in the first content block. For example, the third acquisition module 404 can be configured to execute step S204 described above; its specific implementation principle can be found in the relevant description of step S204, and will not be repeated here.

[0145] In at least one embodiment of this document, the first acquisition module 401 is further configured to: in response to the document type of the first document being a non-periodic update document, acquire the context information of the first content block in the first document and the content block title corresponding to the first content block; or in response to the document type of the first document being a periodically update document, acquire the context information of the first content block in the first document and at least one of the following: the content block title corresponding to the first content block and the content of the historical content block corresponding to the first content block, wherein the content of the historical content block includes: the content of the first content block in the previous update cycle.

[0146] In at least one embodiment of this document, the first acquisition module 401 is further configured to: acquire a historical content block corresponding to the first content block; in response to not acquiring the historical content block or the content of the historical content block being empty, acquire the context information of the first content block in the first document and the content block title corresponding to the first content block; or in response to the content of the historical content block not being empty, acquire the context information of the first content block in the first document, the content block title corresponding to the first content block, and the content of the historical content block.

[0147] In at least one embodiment of this document, the first acquisition module 401 is further configured to: in response to the first document including a plurality of associated content blocks corresponding to different update cycles, acquire a historical content block corresponding to the first content block in the first document; or acquire a historical document corresponding to the first document in the previous update cycle based on the document title of the first document, and acquire a historical content block corresponding to the first content block in the historical document.

[0148] In at least one embodiment of this document, the second acquisition module 402 is further configured to: in response to the document type of the first document being a periodically updated document, acquire the first query information, wherein the first query information is used to query information updated within the time period corresponding to the first document, and the time period corresponding to the first document is determined by the document title of the first document or the update cycle of the first document.

[0149] In at least one embodiment of this document, the second acquisition module 402 is further configured to: send the first information associated with the first content block to the first model, and obtain the first query information based on the output of the first model.

[0150] In at least one embodiment of this document, the recall module 403 is further configured to: based on the first query information, search in a data source that the operator of the first operation has access to obtain the plurality of candidate recall information.

[0151] In at least one embodiment of this document, the third acquisition module 404 is further configured to: filter the plurality of candidate recall information based on at least one of the source of the plurality of candidate recall information and the update time of the plurality of candidate recall information, and retain at least a portion of the plurality of candidate recall information.

[0152] In at least one embodiment of this document, the third acquisition module 404 is further configured to: send the first information associated with the first content block and the plurality of candidate recall information to the second model, and obtain the first recommended material based on the output of the second model.

[0153] In at least one embodiment of this document, the first operation is triggered by a first operator; the third acquisition module 404 is further configured to: acquire the first recommended material based on the first information associated with the first content block, the plurality of candidate recall information and the writing features corresponding to the first operator, wherein the first recommended material matches the writing features corresponding to the first operator.

[0154] In at least one embodiment of this document, the third acquisition module 404 is further configured to: send the first information associated with the first content block, the plurality of candidate recall information and the historical writing information of the first operator to the third model, and obtain the first recommended material based on the output of the third model.

[0155] It should be noted that, for clarity and brevity, at least one embodiment herein does not show all the constituent units of the apparatus 400 for acquiring recommended materials. To achieve the necessary functions of the apparatus 400 for acquiring recommended materials, those skilled in the art can provide and configure other constituent units (not shown) according to specific needs, and one or more embodiments herein do not limit this.

[0156] At least one embodiment of this document also provides an electronic device, including a processing device and a storage device, the storage device including one or more computer program modules; wherein the one or more computer program modules are stored in the storage device and configured to be executed by the processing device, the one or more computer program modules being used to implement the method for obtaining recommended materials provided in any embodiment of this document.

[0157] For example, the processing device may be a processor, such as a central processing unit (CPU), digital signal processor (DSP), image processor (GPU), general-purpose graphics processor (GPGPU), or other form of processing unit with data processing capabilities and / or instruction execution capabilities. It may be a general-purpose processor or a dedicated processor and may control other components in the electronic device to perform the desired functions.

[0158] For example, the storage device may be a memory, and may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and a processing device may execute the program instructions to implement the functions (implemented by the processing device) in at least one embodiment herein and / or other desired functions. Various application programs and various data may also be stored in the computer-readable storage medium, which is not limited by one or more embodiments herein.

[0159] The following is for reference. Figure 5 This document illustrates a structural schematic of an electronic device (e.g., a terminal device or a server) 500 suitable for implementing at least one embodiment of this document. The terminal device in at least one embodiment of this document may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of at least one embodiment herein.

[0160] like Figure 5 As shown, electronic device 500 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 502 or a program loaded from storage device 508 into random access memory (RAM) 503. RAM 503 also stores various programs and data required for the operation of electronic device 500. Processing unit 501, ROM 502, and RAM 503 are interconnected via bus 504. Input / output (I / O) interface 505 is also connected to bus 504.

[0161] Typically, the following devices can be connected to I / O interface 505: input devices 506 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 507 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 508 including, for example, magnetic tapes, hard disks, etc.; and communication devices 509. Communication device 509 allows electronic device 500 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 5An electronic device 500 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0162] In particular, according to one or more embodiments herein, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, one or more embodiments herein include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 509, or installed from storage device 508, or installed from ROM 502. When the computer program is executed by processing device 501, it performs the functions defined in the methods of at least one embodiment herein.

[0163] It should be noted that the computer-readable medium described above can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this document, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0164] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0165] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0166] The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the aforementioned method for obtaining recommendation materials.

[0167] Computer program code for performing the operations described herein may be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0168] One or more embodiments of this document also provide a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in any embodiment of this document are generated.

[0169] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0170] When the computer program product is executed by a computer, the computer executes any of the aforementioned methods for obtaining recommendation materials. The computer program product can be a software installation package; when any of the aforementioned methods for obtaining recommendation materials needs to be used, the computer program product can be downloaded and executed on the computer.

[0171] 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 herein. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0172] The units or modules described in at least one embodiment herein can be implemented in software or hardware. The names of the units or modules do not, in some cases, constitute a limitation on the unit or module itself.

[0173] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0174] In the context of this document, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0175] According to one or more embodiments of this document, Example 1 provides a method for obtaining recommended content, including:

[0176] In response to receiving a first operation triggered in a first document, first information associated with a first content block in the first document is obtained according to the document type of the first document, wherein the first operation is used to trigger material recommendation for the first content block;

[0177] Based on the first information associated with the first content block, obtain the first query information;

[0178] Based on the first query information, a retrieval is performed to obtain multiple candidate recall information;

[0179] Based on the first information associated with the first content block and the multiple candidate recall information, a first recommended material is obtained, wherein the first recommended material is related to at least a portion of the document content to be improved in the first content block.

[0180] According to one or more embodiments of this document, Example 2 provides the method of obtaining first information associated with a first content block in the first document based on the document type of the first document, as in Example 1, including:

[0181] In response to the fact that the document type of the first document is a non-periodic update document, the context information of the first content block in the first document and the title of the content block corresponding to the first content block are obtained; or

[0182] In response to the fact that the document type of the first document is a periodically updated document, the context information of the first content block in the first document and at least one of the following are obtained: the content block title corresponding to the first content block and the content of the historical content block corresponding to the first content block, wherein the content of the historical content block includes: the content of the first content block in the previous update cycle.

[0183] According to one or more embodiments of this document, Example 3 provides the method of obtaining the context information of the first content block in the first document as described in Example 2, and at least one of the following: the content block title corresponding to the first content block and the content of the historical content block corresponding to the first content block, including:

[0184] Obtain the historical content block corresponding to the first content block;

[0185] In response to the failure to obtain the historical content block or the content of the historical content block being empty, the context information of the first content block in the first document and the content block title corresponding to the first content block are obtained; or, in response to the fact that the content of the historical content block is not empty, the context information of the first content block in the first document, the content block title corresponding to the first content block and the content of the historical content block are obtained.

[0186] According to one or more embodiments of this document, Example 4 provides the method for obtaining the historical content block corresponding to the first content block in Example 3, including:

[0187] In response to the first document including multiple associated content blocks with different update periods, the historical content block corresponding to the first content block is retrieved from the first document; or

[0188] Based on the document title of the first document, obtain the historical document corresponding to the first document in the previous update cycle, and obtain the historical content block corresponding to the first content block in the historical document.

[0189] According to one or more embodiments of this document, Example 5 provides the method of obtaining first query information based on the first information associated with the first content block, as described in Example 1, including:

[0190] In response to the fact that the document type of the first document is a periodically updated document, the first query information is obtained, wherein the first query information is used to query information updated within the time period corresponding to the first document, and the time period corresponding to the first document is determined by the document title of the first document or the update cycle of the first document.

[0191] According to one or more embodiments of this document, Example Six provides the method of obtaining first query information based on the first information associated with the first content block, as in Example One, including:

[0192] The first information associated with the first content block is sent to the first model, and the first query information is obtained based on the output of the first model.

[0193] According to one or more embodiments of this document, Example 7 provides a retrieval based on the first query information in any of Examples 1 to 6 to obtain multiple candidate recall information, including:

[0194] Based on the first query information, a search is conducted in the data sources that the operator of the first operation has access to obtain the multiple candidate recall information.

[0195] According to one or more embodiments of this document, Example 8 provides a method from any of Examples 1 through 6, and further includes:

[0196] Based on at least one of the sources of the multiple candidate recall information and the update time of the multiple candidate recall information, the multiple candidate recall information is filtered, and at least a portion of the multiple candidate recall information is retained.

[0197] According to one or more embodiments of this document, Example 9 provides a method for obtaining the first recommended material based on the first information associated with the first content block and the plurality of candidate recall information, as described in any of Examples 1 to 6, including:

[0198] The first information associated with the first content block and the multiple candidate recall information are sent to the second model, and the first recommended material is obtained based on the output of the second model.

[0199] According to one or more embodiments of this document, Example 10 provides that a first operation in any of Examples 1 to 6 is triggered by a first operator; the step of obtaining the first recommended material based on the first information associated with the first content block and the plurality of candidate recall information includes:

[0200] Based on the first information associated with the first content block, the multiple candidate recall information, and the writing features corresponding to the first operator, the first recommended material is obtained, wherein the first recommended material matches the writing features corresponding to the first operator.

[0201] According to one or more embodiments of this document, Example 11 provides the method of obtaining the first recommended material based on the first information associated with the first content block, the plurality of candidate recall information, and the writing features corresponding to the first operator, as described in Example 10, including:

[0202] The first information associated with the first content block, the multiple candidate recall information, and the historical writing information of the first operator are sent to the third model, and the first recommended material is obtained based on the output of the third model.

[0203] According to one or more embodiments of this document, Example Twelve provides an apparatus for obtaining recommended materials, comprising:

[0204] The first acquisition module is configured to: in response to receiving a first operation triggered in a first document, acquire first information associated with a first content block in the first document according to the document type of the first document, wherein the first operation is used to trigger material recommendation for the first content block;

[0205] The second acquisition module is configured to: acquire first query information based on the first information associated with the first content block;

[0206] The recall module is configured to: perform a retrieval based on the first query information to obtain multiple candidate recall information;

[0207] The third acquisition module is configured to: obtain a first recommended material based on the first information associated with the first content block and the multiple candidate recall information, wherein the first recommended material is related to at least a portion of the document content to be improved in the first content block.

[0208] According to one or more embodiments herein, Example Thirteen provides an electronic device comprising:

[0209] At least one processor; and

[0210] At least one memory, including one or more computer program instructions;

[0211] The one or more computer program instructions are executed by a processor at runtime, performing the method for obtaining recommended materials provided in at least one embodiment of this document.

[0212] According to one or more embodiments of this document, Example Fourteen provides a computer-readable storage medium that non-transitory stores computer-readable instructions, wherein the computer-readable instructions, when executed by a processor, implement the method for obtaining recommended materials provided in at least one embodiment of this document.

[0213] According to one or more embodiments of this document, Example Fifteen provides a computer program product including a computer program that, when executed by a processor, implements the method for obtaining recommended materials provided in at least one embodiment of this document.

[0214] The above description is merely a preferred embodiment and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure herein is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed herein that have similar functions.

[0215] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in sequential order. Multitasking and parallel processing may be advantageous in certain environments. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this document. Certain features described in the context of a single embodiment may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0216] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for obtaining recommendation materials, comprising: In response to receiving a first operation triggered in a first document, first information associated with a first content block in the first document is obtained according to the document type of the first document, wherein the first operation is used to trigger material recommendation for the first content block; Based on the first information associated with the first content block, obtain the first query information; Based on the first query information, a retrieval is performed to obtain multiple candidate recall information; Based on the first information associated with the first content block and the multiple candidate recall information, a first recommended material is obtained, wherein the first recommended material is related to at least a portion of the document content to be improved in the first content block.

2. The method according to claim 1, wherein, The step of obtaining the first information associated with the first content block in the first document according to the document type of the first document includes: In response to the fact that the document type of the first document is a non-periodic update document, the context information of the first content block in the first document and the title of the content block corresponding to the first content block are obtained; or In response to the fact that the document type of the first document is a periodically updated document, the context information of the first content block in the first document and at least one of the following are obtained: the content block title corresponding to the first content block and the content of the historical content block corresponding to the first content block, wherein the content of the historical content block includes: the content of the first content block in the previous update cycle.

3. The method according to claim 2, wherein, The step of obtaining the context information of the first content block in the first document and at least one of the following: the content block title corresponding to the first content block and the content of the historical content block corresponding to the first content block, includes: Obtain the historical content block corresponding to the first content block; In response to the failure to obtain the historical content block or the content of the historical content block being empty, the context information of the first content block in the first document and the content block title corresponding to the first content block are obtained; or, in response to the fact that the content of the historical content block is not empty, the context information of the first content block in the first document, the content block title corresponding to the first content block and the content of the historical content block are obtained.

4. The method according to claim 3, wherein, The step of obtaining the historical content block corresponding to the first content block includes: In response to the first document including multiple associated content blocks with different update periods, the historical content block corresponding to the first content block is retrieved from the first document; or Based on the document title of the first document, obtain the historical document corresponding to the first document in the previous update cycle, and obtain the historical content block corresponding to the first content block in the historical document.

5. The method according to claim 1, wherein, The step of obtaining the first query information based on the first information associated with the first content block includes: In response to the fact that the document type of the first document is a periodically updated document, the first query information is obtained, wherein the first query information is used to query information updated within the time period corresponding to the first document, and the time period corresponding to the first document is determined by the document title of the first document or the update cycle of the first document.

6. The method according to claim 1, wherein, The step of obtaining the first query information based on the first information associated with the first content block includes: The first information associated with the first content block is sent to the first model, and the first query information is obtained based on the output of the first model.

7. The method according to any one of claims 1 to 6, wherein, The retrieval based on the first query information yields multiple candidate recall information, including: Based on the first query information, a search is conducted in the data sources that the operator of the first operation has access to obtain the multiple candidate recall information.

8. The method according to any one of claims 1 to 6, wherein, The method further includes: Based on at least one of the sources of the multiple candidate recall information and the update time of the multiple candidate recall information, the multiple candidate recall information is filtered, and at least a portion of the multiple candidate recall information is retained.

9. The method according to any one of claims 1 to 6, wherein, The step of obtaining the first recommended material based on the first information associated with the first content block and the multiple candidate recall information includes: The first information associated with the first content block and the multiple candidate recall information are sent to the second model, and the first recommended material is obtained based on the output of the second model.

10. The method according to any one of claims 1 to 6, wherein, The first operation is triggered by the first operator; The step of obtaining the first recommended material based on the first information associated with the first content block and the multiple candidate recall information includes: The first recommended material is obtained based on the first information associated with the first content block, the multiple candidate recall information, and the writing features corresponding to the first operator, wherein the first recommended material matches the writing features corresponding to the first operator.

11. The method according to claim 10, wherein, The step of obtaining the first recommended material based on the first information associated with the first content block, the multiple candidate recall information, and the writing features corresponding to the first operator includes: The first information associated with the first content block, the multiple candidate recall information, and the historical writing information of the first operator are sent to the third model, and the first recommended material is obtained based on the output of the third model.

12. An apparatus for acquiring recommended materials, comprising: The first acquisition module is configured to: in response to receiving a first operation triggered in a first document, acquire first information associated with a first content block in the first document according to the document type of the first document, wherein the first operation is used to trigger material recommendation for the first content block; The second acquisition module is configured to: acquire first query information based on the first information associated with the first content block; The recall module is configured to: perform a retrieval based on the first query information to obtain multiple candidate recall information; The third acquisition module is configured to: obtain a first recommended material based on the first information associated with the first content block and the multiple candidate recall information, wherein the first recommended material is related to at least a portion of the document content to be improved in the first content block.

13. An electronic device, comprising: At least one processor; as well as At least one memory, including one or more computer program instructions; The one or more computer program instructions are executed by the processor to perform the method according to any one of claims 1 to 11.

14. A computer-readable storage medium for non-transitory storage of computer-readable instructions, wherein, The method of any one of claims 1 to 11 is implemented when the computer-readable instructions are executed by a processor.

15. A computer program product comprising a computer program that, when executed by a processor, implements the method of any one of claims 1 to 11.

Images