Cloud disk file sharing method and device, equipment, medium and computer program product

Abstract

The invention discloses a cloud disk file sharing method and device, equipment, a medium and a computer program product, and the method comprises the steps: obtaining to-be-shared file content in response to a cloud disk file sharing request of a sharer; inputting the to-be-shared file content into a preset large language model for key information identification, and generating a preliminary sharing abstract; wherein the large language model is obtained by training according to the network hotspot event and user portraits and user preferences of sharers; obtaining a user portrait and historical behavior data of a shared person, and inputting the user portrait and historical behavior data of the shared person into the large language model to optimize the preliminary sharing abstract and generate a target sharing abstract; and pushing the target sharing abstract to the shared person. According to the method, personalized sharing abstracts can be generated for different users in combination with the user portraits, the user preferences, the network hotspot events and the user behavior data, the intelligent level of cloud disk file sharing is improved, and the user experience is improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a cloud drive file sharing method, apparatus, device, medium, and computer program product. Background Technology

[0002] Currently, cloud storage sharing commonly uses direct link sharing. Users upload files to the cloud storage platform, generate a sharing link, and send it to the recipient, who can then access and download the file by clicking the link. However, this method requires the recipient to download the entire file to understand its content. The shared content is presented in a simplistic way and fails to generate customized content summaries or recommendations based on the recipient's needs, thus negatively impacting the user experience. Summary of the Invention

[0003] The purpose of this invention is to provide a cloud drive file sharing method, apparatus, device, medium, and computer program product that can combine user profiles, user preferences, trending online events, and user behavior data to generate personalized sharing summaries for different users, thereby improving the intelligence level of cloud drive file sharing and enhancing user experience.

[0004] To achieve the above objectives, embodiments of the present invention provide a cloud drive file sharing method, including: In response to the sharer's cloud drive file sharing request, the content of the file to be shared is obtained according to the cloud drive file sharing request; The content of the file to be shared is input into a preset large language model for key information identification, generating a preliminary sharing summary; wherein, the large language model is trained based on trending online events and the sharer's user profile and preferences; Obtain the user profile and historical behavior data of the recipient, and input the user profile and historical behavior data of the recipient into the large language model to optimize the preliminary sharing summary and generate the target sharing summary; The target sharing summary is pushed to the recipient.

[0005] As an improvement to the above solution, the content of the file to be shared includes any one or more of text, images, and videos.

[0006] As an improvement to the above scheme, the training method of the large language model includes: The large language model is fine-tuned according to the preset sharing summary format; Acquire trending online events and user profiles and preferences of those who share them; The network hot topics, user profiles of the sharers, and user preferences are input into the finely tuned large language model for training at preset time intervals.

[0007] As an improvement to the above solution, the optimization of the preliminary sharing summary to generate the target sharing summary includes: The initial sharing summary is then personalized and optimized with content recommendations to generate a target sharing summary; wherein the target sharing summary includes personalized sharing summary and content recommendation data.

[0008] As an improvement to the above solution, after pushing the target sharing summary to the recipient, the method further includes: Obtain the interaction data of the recipient, and calculate the recommendation score based on the interaction data of the recipient, the user profile, and the target sharing summary; The large language model is optimized based on the interaction data, user profile, and target sharing summary of the share recipients whose recommended scores are greater than a preset score threshold.

[0009] As an improvement to the above solution, the step of calculating the recommendation score based on the interaction data of the recipient, user profile, and the target sharing summary includes: Based on the interaction data of the recipients, a weighted scoring matrix of the recipients for the target sharing summary is constructed. Matrix decomposition is performed on the weighted scoring matrix to obtain the latent feature matrix of the share recipient and the latent feature matrix of the target sharing summary; The recommendation score is calculated based on the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary.

[0010] As an improvement to the above scheme, the interactive data includes any one or more of the following: click rate, number of views, number of downloads, and number of saves.

[0011] As an improvement to the above scheme, the weighted scoring matrix is: ; In the formula, Represents the weighted scoring matrix; Indicates click-through rate. Indicates the number of views. Indicates the number of downloads. Indicates the number of transfers; These represent the weights of different types of interactive data.

[0012] As an improvement to the above scheme, the step of performing matrix decomposition on the weighted scoring matrix to obtain the latent feature matrix of the share recipient and the latent feature matrix of the target sharing summary includes: Singular value decomposition is performed on the weighted scoring matrix to obtain the latent feature matrix of the shared subject and the latent feature matrix of the target shared summary.

[0013] As an improvement to the above scheme, the formula for calculating the recommendation score based on the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary is as follows: ; In the formula, Indicates the recommended score; Indicates the recipient of the sharing The latent feature vector, i.e., the latent feature matrix of the recipient. The Row vectors; Indicates target sharing summary The latent feature vector, i.e., the latent feature matrix of the target sharing summary. The Row vectors.

[0014] This invention also provides a cloud drive file sharing device, comprising: The file content acquisition module is used to respond to the cloud drive file sharing request from the sharer and acquire the file content to be shared according to the cloud drive file sharing request. The sharing summary generation module is used to input the content of the file to be shared into a preset large language model for key information recognition and to generate a preliminary sharing summary; wherein, the large language model is trained based on online hot events and the user profile and user preferences of the sharer; The sharing summary optimization module is used to obtain the user profile and historical behavior data of the recipient, input the user profile and historical behavior data of the recipient into the large language model, and optimize the preliminary sharing summary to generate the target sharing summary. The sharing summary push module is used to push the target sharing summary to the recipient.

[0015] This invention also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the cloud disk file sharing method described above.

[0016] This invention also provides a computer-readable storage medium, which includes a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the cloud disk file sharing method described above.

[0017] This invention also provides a computer program product, which includes a computer program or computer instructions. When the computer program or computer instructions are executed by a processor, they implement the cloud disk file sharing method described above.

[0018] Compared to existing technologies, the beneficial effects of the cloud drive file sharing method, apparatus, device, medium, and computer program product provided by this invention are as follows: Responding to a sharer's cloud drive file sharing request, and obtaining the content of the file to be shared based on the request; inputting the content of the file to be shared into a preset large language model for key information recognition, generating a preliminary sharing summary; wherein, the large language model is trained based on online hot events and the sharer's user profile and user preferences; obtaining the recipient's user profile and historical behavior data, and inputting the recipient's user profile and historical behavior data into the large language model to optimize the preliminary sharing summary, generating a target sharing summary; and pushing the target sharing summary to the recipient. This invention can combine user profiles, user preferences, online hot events, and user behavior data to generate personalized sharing summaries for different users, improving the intelligence level of cloud drive file sharing and enhancing user experience. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a preferred embodiment of a cloud drive file sharing method provided by the present invention; Figure 2 This is a schematic diagram of a preferred embodiment of a cloud disk file sharing device provided by the present invention; Figure 3 This is a schematic diagram of a preferred embodiment of a terminal device provided by the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1 , Figure 1 This is a flowchart illustrating a preferred embodiment of a cloud drive file sharing method provided by the present invention. The cloud drive file sharing method includes: S1, responding to the sharer's cloud drive file sharing request, and obtaining the content of the file to be shared according to the cloud drive file sharing request; S2, the content of the file to be shared is input into a preset large language model for key information recognition, and a preliminary sharing summary is generated; wherein, the large language model is trained based on online hot events and the user profile and user preferences of the sharer; S3, obtain the user profile and historical behavior data of the recipient, input the user profile and historical behavior data of the recipient into the large language model to optimize the preliminary sharing summary and generate the target sharing summary; S4, push the target sharing summary to the recipient.

[0022] Specifically, this invention provides a cloud drive file sharing method. First, it responds to a sharer's cloud drive file sharing request and obtains the content of the file to be shared based on the request. The content of the file to be shared includes any one or more of text, images, and videos. Then, the multimodal content of the file to be shared (text, images, videos, etc.) is input into a preset large language model for key information recognition to understand the semantics and context of the content, generating a preliminary sharing summary so that the recipient can quickly understand the key points of the file. The large language model is trained based on trending online events and the sharer's user profile and preferences. Next, the user profile and historical behavior data of the recipients are obtained and input into the large language model. This allows the large language model to personalize and optimize the preliminary sharing summaries shared to different recipients, generating a target sharing summary that includes personalized sharing summaries and content recommendation data. The target sharing summary is then pushed to the recipients, achieving personalized sharing summaries and content recommendations.

[0023] The embodiments of the present invention can combine user profiles, user preferences, trending online events, and user behavior data to generate personalized sharing summaries for different users, thereby improving the personalization and relevance of cloud drive sharing content, enhancing the intelligence level of cloud drive file sharing, and improving user experience.

[0024] Furthermore, the training method for the large language model includes: The large language model is fine-tuned according to the preset sharing summary format; Acquire trending online events and user profiles and preferences of those who share them; The network hot topics, user profiles of the sharers, and user preferences are input into the finely tuned large language model for training at preset time intervals.

[0025] Specifically, in this embodiment of the invention, after selecting a large language model, it is first fine-tuned according to a preset sharing summary format to obtain a fine-tuned large language model capable of generating sharing summary content. Then, it acquires user profiles and preferences of trending online events and sharers. User profiles include age, gender, and occupation; user preferences are derived from analysis of user activity records on cloud storage platforms, such as frequently viewed or downloaded file types; trending online events include popular trends in related fields. The trending online events, sharers' user profiles, and user preferences are input into the fine-tuned large language model at preset time intervals for training, enabling the trained large language model to generate preliminary sharing summaries. These preliminary summaries provide a quick overview of the content and attract user attention.

[0026] This invention can process multimodal content, including text, images, and videos. Leveraging the deep learning capabilities of a large language model, it provides richer and more accurate content understanding. Simultaneously, it can combine current online trends to generate concise and engaging summaries, enhancing the appeal of shared content. By analyzing user usage records and constructing detailed user profiles, it achieves a deep understanding of user preferences and needs, supporting personalized recommendations. Furthermore, the large language model in this invention possesses adaptive learning capabilities, continuously optimizing summary content based on user feedback and behavioral data to achieve more accurate personalized recommendations.

[0027] In a preferred embodiment, after pushing the target sharing summary to the recipient, the method further includes: Obtain the interaction data of the recipient, and calculate the recommendation score based on the interaction data of the recipient, the user profile, and the target sharing summary; The large language model is optimized based on the interaction data, user profile, and target sharing summary of the share recipients whose recommended scores are greater than a preset score threshold.

[0028] Specifically, in this embodiment of the invention, after pushing the target sharing summary to the recipient, the interaction data of the recipient is obtained, and a recommendation score is calculated based on the recipient's interaction data, user profile, and target sharing summary. The recommendation score is associated with the target sharing summary. Based on the interaction data, user profile, and target sharing summary of the recipient corresponding to a recommendation score greater than a preset score threshold, the large language model is optimized to form a positive recommendation loop, enabling the large language model to adapt to changes in user preferences.

[0029] This invention pushes sharing summaries and recommended content to the recipients in real time, collects user interaction data in a timely manner, combines user interaction data with user profiles and content features, calculates recommendation scores using matrix factorization algorithms, and continuously optimizes the large language model based on actual user interaction feedback.

[0030] Furthermore, the step of calculating the recommendation score based on the interaction data of the recipient, the user profile, and the target sharing summary includes: Based on the interaction data of the recipients, a weighted scoring matrix of the recipients for the target sharing summary is constructed. Matrix decomposition is performed on the weighted scoring matrix to obtain the latent feature matrix of the share recipient and the latent feature matrix of the target sharing summary; The recommendation score is calculated based on the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary.

[0031] Specifically, in this embodiment of the invention, the interaction data of the recipients is converted into ratings, and different weights are assigned to different types of interaction data to construct a weighted rating matrix of the recipients' opinions on the target sharing summary. The interaction data includes any one or more of click-through rate, number of views, number of downloads, and number of saves. Based on the recipients' click-through rate, number of views, number of downloads, and number of saves, the weighted rating matrix of the recipients' opinions on the target sharing summary is constructed as follows: ; In the formula, Represents the weighted scoring matrix; Indicates click-through rate. Indicates the number of views. Indicates the number of downloads. Indicates the number of transfers; These represent the weights of different types of interactive data.

[0032] For the weighted scoring matrix Perform matrix factorization to obtain the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary.

[0033] For example, embodiments of the present invention apply weighted scoring matrices. Perform singular value decomposition (SVD) to obtain the latent feature matrix of the shared entity. Share the latent feature matrix of the summary with the target .

[0034] Based on the latent feature matrix of the recipient Share the latent feature matrix of the summary with the target The recommended score is calculated as follows: ; In the formula, Indicates the recommended score; Indicates the recipient of the sharing The latent feature vector, i.e., the latent feature matrix of the recipient. The Row vectors; Indicates target sharing summary The latent feature vector, i.e., the latent feature matrix of the target sharing summary. The Row vectors.

[0035] This invention uses user interaction data from a test set to evaluate the model's recommendation score. Metrics such as root mean square error (RMSE) can be used to measure the difference between the recommended score and the actual recommended score. The weighted scoring matrix is ​​periodically updated using newly collected user interaction data. And refactor the matrix to adapt to changes in user preferences.

[0036] For example, in the embodiments of the present invention, the root mean square error ; In the formula, This represents the total number of scores in the test set; Indicates the recipient of the sharing Summary of target sharing The actual recommended score.

[0037] Accordingly, the present invention also provides a cloud disk file sharing device, which can implement all the processes of the cloud disk file sharing method in the above embodiments.

[0038] Please see Figure 2 , Figure 2 This is a schematic diagram of a preferred embodiment of a cloud drive file sharing device provided by the present invention. The cloud drive file sharing device includes: The file content acquisition module 201 is used to respond to the cloud disk file sharing request from the sharer and acquire the file content to be shared according to the cloud disk file sharing request. The sharing summary generation module 202 is used to input the content of the file to be shared into a preset large language model for key information recognition and to generate a preliminary sharing summary; wherein, the large language model is trained based on online hot events and the user profile and user preferences of the sharer; The sharing summary optimization module 203 is used to obtain the user profile and historical behavior data of the recipient, input the user profile and historical behavior data of the recipient into the large language model, so as to optimize the preliminary sharing summary and generate the target sharing summary. The sharing summary push module 204 is used to push the target sharing summary to the recipient.

[0039] Preferably, the content of the file to be shared includes any one or more of text, images, and videos.

[0040] Preferably, the training method for the large language model includes: The large language model is fine-tuned according to the preset sharing summary format; Acquire trending online events and user profiles and preferences of those who share them; The network hot topics, user profiles of the sharers, and user preferences are input into the finely tuned large language model for training at preset time intervals.

[0041] Preferably, optimizing the preliminary sharing summary to generate the target sharing summary includes: The initial sharing summary is then personalized and optimized with content recommendations to generate a target sharing summary; wherein the target sharing summary includes personalized sharing summary and content recommendation data.

[0042] Preferably, the device further includes a large language model optimization module, used for: After the target sharing summary is pushed to the recipient, the recipient's interaction data is obtained, and a recommendation score is calculated based on the recipient's interaction data, user profile, and the target sharing summary. The large language model is optimized based on the interaction data, user profile, and target sharing summary of the share recipients whose recommended scores are greater than a preset score threshold.

[0043] Preferably, the step of calculating the recommendation score based on the interaction data of the recipient, the user profile, and the target sharing summary includes: Based on the interaction data of the recipients, a weighted scoring matrix of the recipients for the target sharing summary is constructed. Matrix decomposition is performed on the weighted scoring matrix to obtain the latent feature matrix of the share recipient and the latent feature matrix of the target sharing summary; The recommendation score is calculated based on the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary.

[0044] Preferably, the interactive data includes any one or more of the following: click-through rate, number of views, number of downloads, and number of saves.

[0045] Preferably, the weighted scoring matrix is: ; In the formula, Represents the weighted scoring matrix; Indicates click-through rate. Indicates the number of views. Indicates the number of downloads. Indicates the number of transfers; These represent the weights of different types of interactive data.

[0046] Preferably, the step of performing matrix decomposition on the weighted scoring matrix to obtain the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary includes: Singular value decomposition is performed on the weighted scoring matrix to obtain the latent feature matrix of the shared subject and the latent feature matrix of the target shared summary.

[0047] Preferably, the formula for calculating the recommendation score based on the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary is as follows: ; In the formula, Indicates the recommended score; Indicates the recipient of the sharing The latent feature vector, i.e., the latent feature matrix of the recipient. The Row vectors; Indicates target sharing summary The latent feature vector, i.e., the latent feature matrix of the target sharing summary. The Row vectors.

[0048] In specific implementation, the working principle, control process and technical effects of the cloud disk file sharing device provided in the embodiments of the present invention are the same as those of the cloud disk file sharing method in the above embodiments, and will not be repeated here.

[0049] Please see Figure 3 , Figure 3 This is a schematic diagram of a preferred embodiment of a terminal device provided by the present invention. The terminal device includes a processor 301, a memory 302, and a computer program stored in the memory 302 and configured to be executed by the processor 301. When the processor 301 executes the computer program, it implements the cloud disk file sharing method described in any of the above embodiments.

[0050] Preferably, the computer program can be divided into one or more modules / units (such as computer program 1, computer program 2, ...), and the one or more modules / units are stored in the memory 302 and executed by the processor 301 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0051] The processor 301 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor 301 can be any conventional processor. The processor 301 is the control center of the terminal device, connecting various parts of the terminal device through various interfaces and lines.

[0052] The memory 302 mainly includes a program storage area and a data storage area. The program storage area can store the operating system, applications required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory 302 can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, and a flash card, etc., or the memory 302 can also be other volatile solid-state storage devices.

[0053] It should be noted that the aforementioned terminal devices may include, but are not limited to, processors and memory, as will be understood by those skilled in the art. Figure 3 The structural diagram is merely an example of the terminal device described above and does not constitute a limitation on the terminal device described above. It may include more or fewer components than shown in the diagram, or combine certain components, or use different components.

[0054] This invention also provides a computer-readable storage medium, which includes a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the cloud disk file sharing method described in any of the above embodiments.

[0055] This invention also provides a computer program product, which includes a computer program or computer instructions. When the computer program or computer instructions are executed by a processor, they implement the cloud disk file sharing method described in any of the above embodiments.

[0056] This invention provides a cloud drive file sharing method, apparatus, device, medium, and computer program product. It responds to a sharer's cloud drive file sharing request and obtains the content of the file to be shared based on the request. The content is then input into a preset large language model for key information recognition, generating a preliminary sharing summary. This large language model is trained based on trending online events and the sharer's user profile and preferences. The user profile and historical behavior data of the recipient are obtained and input into the large language model to optimize the preliminary sharing summary, generating a target sharing summary. Finally, the target sharing summary is pushed to the recipient. This invention combines user profiles, user preferences, trending online events, and user behavior data to generate personalized sharing summaries for different users, improving the intelligence level of cloud drive file sharing and enhancing user experience.

[0057] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0058] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for sharing files on a cloud drive, characterized in that, include: In response to the sharer's cloud drive file sharing request, the content of the file to be shared is obtained according to the cloud drive file sharing request; The content of the file to be shared is input into a preset large language model for key information identification, generating a preliminary sharing summary; wherein, the large language model is trained based on trending online events and the sharer's user profile and preferences; Obtain the user profile and historical behavior data of the recipient, and input the user profile and historical behavior data of the recipient into the large language model to optimize the preliminary sharing summary and generate the target sharing summary; The target sharing summary is pushed to the recipient.

2. The cloud drive file sharing method as described in claim 1, characterized in that, The content of the file to be shared includes any one or more of the following: text, images, and videos.

3. The cloud drive file sharing method as described in claim 1, characterized in that, The training methods for the large language model include: The large language model is fine-tuned according to the preset sharing summary format; Acquire trending online events and user profiles and preferences of those who share them; The network hot topics, user profiles of the sharers, and user preferences are input into the finely tuned large language model for training at preset time intervals.

4. The cloud drive file sharing method as described in claim 1, characterized in that, The step of optimizing the preliminary sharing summary to generate the target sharing summary includes: The initial sharing summary is then personalized and optimized with content recommendations to generate a target sharing summary; wherein the target sharing summary includes personalized sharing summary and content recommendation data.

5. The cloud drive file sharing method as described in claim 1, characterized in that, After pushing the target sharing summary to the recipient, the method further includes: Obtain the interaction data of the recipient, and calculate the recommendation score based on the interaction data of the recipient, the user profile, and the target sharing summary; The large language model is optimized based on the interaction data, user profile, and target sharing summary of the share recipients whose recommended scores are greater than a preset score threshold.

6. The cloud drive file sharing method as described in claim 5, characterized in that, The step of calculating a recommendation score based on the interaction data of the recipient, user profile, and the target sharing summary includes: Based on the interaction data of the recipients, a weighted scoring matrix of the recipients for the target sharing summary is constructed. Matrix decomposition is performed on the weighted scoring matrix to obtain the latent feature matrix of the share recipient and the latent feature matrix of the target sharing summary; The recommendation score is calculated based on the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary.

7. The cloud drive file sharing method as described in claim 6, characterized in that, The interactive data includes any one or more of the following: click-through rate, number of views, number of downloads, and number of saves.

8. The cloud drive file sharing method as described in claim 7, characterized in that, The weighted scoring matrix is ​​as follows: ； In the formula, Represents the weighted scoring matrix; Indicates click-through rate. Indicates the number of views. Indicates the number of downloads. Indicates the number of transfers; These represent the weights of different types of interactive data.

9. The cloud drive file sharing method as described in claim 6, characterized in that, The step of performing matrix decomposition on the weighted scoring matrix to obtain the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary includes: Singular value decomposition is performed on the weighted scoring matrix to obtain the latent feature matrix of the shared subject and the latent feature matrix of the target shared summary.

10. The cloud drive file sharing method as described in claim 9, characterized in that, The formula for calculating the recommendation score based on the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary is as follows: ； In the formula, Indicates the recommended score; Indicates the recipient of the sharing The latent feature vector, i.e., the latent feature matrix of the recipient. The Row vectors; Indicates target sharing summary The latent feature vector, i.e., the latent feature matrix of the target sharing summary. The Row vectors.

11. A cloud drive file sharing device, characterized in that, include: The file content acquisition module is used to respond to the cloud drive file sharing request from the sharer and acquire the file content to be shared according to the cloud drive file sharing request. The sharing summary generation module is used to input the content of the file to be shared into a preset large language model for key information recognition and to generate a preliminary sharing summary; wherein, the large language model is trained based on online hot events and the user profile and user preferences of the sharer; The sharing summary optimization module is used to obtain the user profile and historical behavior data of the recipient, input the user profile and historical behavior data of the recipient into the large language model, and optimize the preliminary sharing summary to generate the target sharing summary. The sharing summary push module is used to push the target sharing summary to the recipient.

12. The cloud disk file sharing device as described in claim 11, characterized in that, The content of the file to be shared includes any one or more of the following: text, images, and videos.

13. The cloud disk file sharing device as described in claim 11, characterized in that, The training methods for the large language model include: The large language model is fine-tuned according to the preset sharing summary format; Acquire trending online events and user profiles and preferences of those who share them; The network hot topics, user profiles of the sharers, and user preferences are input into the finely tuned large language model for training at preset time intervals.

14. The cloud disk file sharing device as described in claim 11, characterized in that, The step of optimizing the preliminary sharing summary to generate the target sharing summary includes: The initial sharing summary is then personalized and optimized with content recommendations to generate a target sharing summary; wherein the target sharing summary includes personalized sharing summary and content recommendation data.

15. The cloud drive file sharing device as described in claim 11, characterized in that, The device also includes a large language model optimization module, used for: After the target sharing summary is pushed to the recipient, the recipient's interaction data is obtained, and a recommendation score is calculated based on the recipient's interaction data, user profile, and the target sharing summary. The large language model is optimized based on the interaction data, user profile, and target sharing summary of the share recipients whose recommended scores are greater than a preset score threshold.

16. The cloud drive file sharing device as described in claim 15, characterized in that, The step of calculating a recommendation score based on the interaction data of the recipient, user profile, and the target sharing summary includes: Based on the interaction data of the recipients, a weighted scoring matrix of the recipients for the target sharing summary is constructed. Matrix decomposition is performed on the weighted scoring matrix to obtain the latent feature matrix of the share recipient and the latent feature matrix of the target sharing summary; The recommendation score is calculated based on the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary.

17. The cloud disk file sharing device as described in claim 16, characterized in that, The interactive data includes any one or more of the following: click-through rate, number of views, number of downloads, and number of saves.

18. The cloud drive file sharing device as described in claim 17, characterized in that, The weighted scoring matrix is ​​as follows: ； In the formula, Represents the weighted scoring matrix; Indicates click-through rate. Indicates the number of views. Indicates the number of downloads. Indicates the number of transfers; These represent the weights of different types of interactive data.

19. The cloud disk file sharing device as described in claim 16, characterized in that, The step of performing matrix decomposition on the weighted scoring matrix to obtain the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary includes: Singular value decomposition is performed on the weighted scoring matrix to obtain the latent feature matrix of the shared subject and the latent feature matrix of the target shared summary.

20. The cloud drive file sharing device as described in claim 19, characterized in that, The formula for calculating the recommendation score based on the latent feature matrix of the recipient and the latent feature matrix of the target sharing summary is as follows: ； In the formula, Indicates the recommended score; Indicates the recipient of the sharing The latent feature vector, i.e., the latent feature matrix of the recipient. The Row vectors; Indicates target sharing summary The latent feature vector, i.e., the latent feature matrix of the target sharing summary. The Row vectors.

21. A terminal device, characterized in that, The device includes a processor and a memory, wherein the memory stores a computer program and the computer program is configured to be executed by the processor, wherein the processor, when executing the computer program, implements the cloud disk file sharing method as described in any one of claims 1 to 10.

22. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the device containing the computer-readable storage medium executes the computer program, it implements the cloud disk file sharing method as described in any one of claims 1 to 10.

23. A computer program product, characterized in that, The computer program product includes a computer program or computer instructions, which, when executed by a processor, implement the cloud disk file sharing method as described in any one of claims 1 to 10.

Images