An integrated sharing method and system for digital media content resources

Abstract

The present invention provides a method and system for integrated sharing of digital media content resources, which are applied to the technical field of data processing. This application preprocesses digital media content resources, cloud platform basic data, and user demand data to generate a standardized media content sequence, cloud platform features, user demand features, and resource classification features; processes resource association relationship data to generate resource association features; processes the policies in the sharing strategy update mechanism information based on a resource value evaluation system to generate updated sharing strategy data; processes the standardized media content sequence, cloud platform features, and resource association features based on a target integrated sharing model to generate target integrated sharing features; and processes the target integrated sharing features based on the target integrated sharing model and the updated sharing strategy data to generate the integrated sharing results of digital media content resources.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method and system for integrating and sharing digital media content resources. Background Art

[0002] The basic data of the cloud platform includes numerous performance indicators. However, some indicators are redundant or invalid. For example, the storage device temperature indicator in cloud storage data has a low correlation with storage capacity and usage in some cloud platforms and has little impact on resource sharing decisions; the occasionally occurring error sampling data in network data is also an invalid indicator. These redundant and invalid indicators interfere with the judgment of the true performance of the cloud platform, affect the accuracy of resource processing and sharing decisions, result in the inability to fully utilize the advantages of the cloud platform, and reduce the processing and sharing efficiency of resources in the cloud environment.

[0003] There are various problems in the integration and sharing of digital media content resources in the cloud environment. The resource formats are diverse and often contain invalid encodings, which bring difficulties to format unification and content parsing, and affect the efficiency of resource integration and sharing. The basic data of the cloud platform has redundant and invalid indicators, interfering with performance judgment and sharing decisions. The user demand data is in a complex form, making it difficult to accurately extract key demands and unable to meet personalized needs. The resource association relationships are complex, and it is difficult for existing technologies to deeply explore them, reducing the pertinence and efficiency of sharing. The sharing strategy update mechanism is complex, with insufficient strategy classification, value evaluation, and conflict coordination, and it cannot be optimized according to the real-time data of the cloud platform and the dynamic needs of users, resulting in poor sharing effects. These problems limit the development of the integration and sharing of digital media content resources in the cloud environment and urgently need to be solved to improve the quality and efficiency of resource sharing and better meet user needs.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and thus includes information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for the integrated sharing of digital media content resources, which at least to some extent overcomes the problems existing in the prior art. By unifying the formats and parsing the content of digital media content resources, a standardized media content sequence is generated; the basic data of the cloud platform is screened, and the cloud platform features are calculated; the user demand data is parsed, and the user demand features are summarized; the metadata of digital media content resources is extracted, and the resource classification features are generated through clustering analysis. The resource link and attribute association information are extracted, and the resource association tightness and direction change information are generated through model processing, and then classified and identified, summarized and analyzed to obtain the resource association features. Classify the policy data according to the strategic objectives, evaluate the value and screen, coordinate the conflicts to generate highly feasible policies, and combine with the real-time data of the cloud platform for weighted processing to update the shared policy data. Process the relevant data with the model, combine with the shared policy, optimize and adjust and check, and finally generate the integrated sharing result of digital media content resources to achieve the efficient integration and sharing of resources in the cloud environment.

[0006] Other features and advantages of this application will become apparent from the following detailed description, or be learned in part through the practice of the present invention.

[0007] According to one aspect of this application, a method for the integrated sharing of digital media content resources is provided, including: obtaining digital media content resources, cloud platform basic data, user demand data, resource association relationship data, and shared policy update mechanism information; preprocessing the digital media content resources, cloud platform basic data, and user demand data to generate a standardized media content sequence, cloud platform features, user demand features, and resource classification features; processing the resource association relationship data to generate resource association features; processing the policies in the shared policy update mechanism information based on the resource value evaluation system to generate updated shared policy data; processing the standardized media content sequence, cloud platform features, and resource association features based on the target integrated sharing model to generate target integrated sharing features; processing the target integrated sharing features based on the target integrated sharing model and the updated shared policy data to generate the integrated sharing result of digital media content resources.

[0008] Another aspect of the present application is an integrated sharing device for digital media content resources, which is characterized by including: an acquisition module for acquiring digital media content resources, cloud platform basic data, user demand data, resource association relationship data, and sharing policy update mechanism information; a processing module for preprocessing the digital media content resources, cloud platform basic data, and user demand data to generate a standardized media content sequence, cloud platform features, user demand features, and resource classification features; processing the resource association relationship data to generate resource association features; processing the policies in the sharing policy update mechanism information based on a resource value evaluation system to generate updated sharing policy data; processing the standardized media content sequence, cloud platform features, and resource association features based on a target integrated sharing model to generate target integrated sharing features; and processing the target integrated sharing features based on the target integrated sharing model and the updated sharing policy data to generate an integrated sharing result of digital media content resources.

[0009] According to another aspect of the present application, there is provided a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a second processor, the above-mentioned integrated sharing method of digital media content resources is implemented.

[0010] For the integrated sharing method and system of digital media content resources provided by the present application, the server collects digital media content, cloud platform basics, user demands, resource association relationships, and sharing policy update mechanism information from multiple channels to provide a data basis for subsequent processing. It unifies the format and parses the content of digital media content resources to generate a standardized media content sequence; filters the cloud platform basic data and calculates the cloud platform features; analyzes the user demand data and summarizes the user demand features; extracts the metadata of digital media content resources and generates resource classification features through clustering analysis. It extracts resource link and attribute association information, processes it through a model to generate resource association tightness and direction change information, and then classifies, identifies, summarizes, and analyzes to obtain resource association features. Classify the policy data according to the policy objectives, evaluate the value and filter, coordinate conflicts to generate highly feasible policies, and combine with real-time cloud platform data for weighted processing to update the sharing policy data. Obtain training and verification data samples and a basic model, process to generate prediction parameters, and optimize the training to obtain a target integrated sharing model. Use this model to process relevant data, combine with the sharing policy, optimize and adjust, and finally generate an integrated sharing result of digital media content resources to achieve efficient integration and sharing of resources in the cloud environment.

[0011] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. Description of the Drawings

[0012] Figure 1Flowchart showing a method for integrated sharing of digital media content resources provided by an embodiment of the present application;

[0013] Figure 2 Schematic structural diagram showing an apparatus for integrated sharing of digital media content resources provided by an embodiment of the present application. Detailed implementation manners

[0014] The following describes preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0015] The following combines Figure 1 to describe the method for integrated sharing of digital media content resources according to an exemplary embodiment of the present application. As Figure 1 shown, this method is applied to a server and includes:

[0016] S101, obtain digital media content resources, cloud platform basic data, user demand data, resource association relationship data, and shared policy update mechanism information.

[0017] In one implementation manner, digital media content is collected from multiple channels, such as video resources on online video platforms, audio files on music platforms, and image resources on picture material websites. For video resources, different types are covered, such as movies, TV series, and short videos; audio resources include music, audiobooks, radio programs, etc.; image resources include photographic works, illustrations, icons, etc. At the same time, the corresponding metadata is obtained when collecting the resources, such as the duration, resolution, and frame rate of the video, the format, bit rate, and singer of the audio, and the size, color mode, and creator of the image, etc. Various basic data of the cloud platform itself are collected, including cloud storage-related data, such as the total storage capacity, the used capacity, and the remaining available capacity; network-related data, such as the real-time rate, average rate, bandwidth peak and valley of the network bandwidth, and the average value, maximum value, and minimum value of the network latency; computing resource data, such as the usage rate, load condition, and number of cores of the CPU, and the usage amount, free amount, and memory frequency of the memory. These data reflect the operating state and performance level of the cloud platform and provide support for subsequent resource processing and sharing.

[0018] Collect user demand data through various methods. For example, set up a user feedback entry on the platform where users can submit their demands for digital media content, including the specific content types they want to watch or use, theme preferences, etc.; conduct questionnaires to ask users about their expectations for platform functions, such as the optimization requirements for the search function and improvement suggestions for the playback interface; analyze the historical behavior data of users on the platform, such as browsing records, favorite records, playback duration, search keywords, etc., so as to mine users' potential demands and behavior patterns. Sort out the association relationships between digital media content resources. From the perspective of resource links, collect the hyperlink reference relationships between different resources. For example, a graphic news article quotes the link of a specific video, or a video embeds the link of other audio resources. From the aspect of resource attribute associations, analyze the associations of resources in terms of attributes such as theme, subject matter, creator, and audience group. For example, a series of works produced by the same creator, or different types of resources for the same audience group (such as animated videos for children and children's book audio). These association relationships help to discover the internal connections between resources and improve the pertinence and efficiency of resource sharing.

[0019] Collect information related to the sharing strategies currently implemented on the cloud platform, including content dissemination strategies, such as the sharing permission settings of resources (public sharing, private sharing, sharing with specified users), the selection of dissemination channels (social media sharing, in-site sharing); user experience optimization strategies, such as the adaptive playback strategy for different network environments and the setting of personalized recommendation algorithms; resource management strategies, such as the storage allocation rules of resources and access permission control; copyright protection strategies, such as copyright statements and the application of digital rights management (DRM) technology, etc. At the same time, obtain the rules and triggering conditions for sharing strategy updates, such as the mechanism information for updating strategies based on factors such as changes in user behavior data, fluctuations in cloud platform performance, and adjustments in copyright regulations.

[0020] S102, preprocess the digital media content resources, cloud platform basic data, and user demand data to generate standardized media content sequences, cloud platform features, user demand features, and resource classification features.

[0021] In one implementation, the digital media content resources are processed to unify the formats, removing incompatible formats and invalid encodings to generate initial media content information. In a cloud environment, the sources of digital media content resources are extensive and the formats are diverse, with a large number of incompatible formats and invalid encodings. For example, images collected from different websites may include various formats such as WebP, PNG, JPEG, etc., and some images may have invalid encodings due to transmission errors or file corruption. During format unification processing, format conversion tools are used to convert various formats into a common format for the cloud platform. For example, video formats such as AVI and FLV are uniformly converted to the MP4 format, audio formats such as WMA and APE are uniformly converted to the MP3 format, and image formats such as ICO and BMP are uniformly converted to the JPEG format. At the same time, encoding detection algorithms are used to identify and remove invalid encodings, thereby obtaining complete and standardized initial media content information, laying a foundation for subsequent processing.

[0022] The initial media content information is parsed to generate a standardized media content sequence. Among them, the standardized media content sequence includes an image key element sequence, an audio spectrum feature sequence, and a video scene transition sequence. For images, algorithms such as edge detection and color feature extraction are used to identify key elements in the images, such as human outlines, landmark buildings, etc., to form an image key element sequence; for audio, spectrum analysis technology is used to obtain features such as the frequency and amplitude of the audio, and an audio spectrum feature sequence is constructed to describe the timbre and pitch changes of the audio; for video, scene segmentation algorithms are used to detect shot transitions and scene changes in the video to generate a video scene transition sequence. These standardized media content sequences can more accurately describe the core features of digital media content, facilitating subsequent resource integration and sharing.

[0023] The performance indicators of the cloud platform's basic data are screened and processed to remove redundant and invalid indicators, generating basic performance data. The cloud platform's basic data contains numerous performance indicators, but some indicators are redundant or invalid. For example, the storage device temperature indicator in cloud storage data has a low correlation with storage capacity and usage in some cloud platforms and has little impact on resource sharing decisions, belonging to a redundant indicator; the occasionally occurring error sampling data in network data belongs to an invalid indicator. By setting screening rules, such as through correlation analysis between indicators and importance assessment of business requirements, these redundant and invalid indicators are removed, and indicators such as total storage capacity, used capacity, real-time network bandwidth rate, and CPU usage rate, which play a key role in resource processing and sharing decisions, are retained to generate concise and effective basic performance data.

[0024] Calculate the features of the basic performance data to generate cloud platform features. Among them, the cloud platform features include the storage capacity availability rate feature, the network bandwidth stability feature, and the computing resource load balancing feature. Based on the filtered basic performance data, perform feature calculations to generate features that can reflect the operating status and performance of the cloud platform. Calculate the storage capacity availability rate, which is the ratio of the remaining available capacity to the total capacity, to measure the remaining space of the cloud storage resources; by monitoring the fluctuations of the network bandwidth over a period of time, calculate the network bandwidth stability feature, such as the standard deviation of the bandwidth fluctuations; analyze the load conditions of the CPU at different time periods, and calculate the computing resource load balancing feature, such as the difference in the average load rates of each core. These cloud platform features provide a quantitative basis for evaluating the support capabilities of the cloud platform for the processing and sharing of digital media content resources.

[0025] Perform semantic parsing and processing on the user demand data to extract the key demand semantics and generate a preliminary demand semantics set. The user demand data exists in various forms, such as the text feedback by users, the answers to the questionnaires, the user behavior data, etc. Use the semantic analysis algorithm in natural language processing (NLP) technology to parse this data. For example, for the user feedback "hoping to smoothly watch high-definition videos on the mobile phone and have more science fiction themed content", through semantic parsing, extract the key demand semantics such as "mobile phone viewing", "high-definition video", and "science fiction theme", and summarize the key demand semantics of numerous users to form a preliminary demand semantics set, providing materials for subsequent analysis of user demand features.

[0026] Perform classification and induction processing on the preliminary demand semantics set to generate user demand features. Among them, the user demand features include content preference features, function usage features, and experience optimization features. From the aspect of content preference, count the demand frequencies of users for various themes (such as science fiction, romance, history, etc.) and content forms (such as video, audio, image) to determine the content preference features; for function usage, analyze the demands and suggestions of users for functions such as platform search, play, and download to obtain function usage features, such as some users hoping that the search function can support fuzzy search and search by tags; from the perspective of experience optimization, based on the feedback of users on aspects such as play smoothness and interface design, summarize the experience optimization features, such as users expecting the play interface to be simple and ad-free and the loading speed to be faster.

[0027] Extract the metadata of digital media content resources to generate preliminary resource classification information. Metadata includes various descriptive information of digital media content resources. Extract metadata such as resource type (video, audio, image), theme (comedy, action, documentary, etc.), creator, creation time, audience group (children, teenagers, adults), dissemination channel (social media, professional platform, official website), etc. from digital media content resources. Based on these metadata, preliminarily classify the resources. For example, classify the resources into video, audio, and image categories, and then further subdivide. For example, under the video category, it can be divided into movies, TV dramas, short videos, etc., to form preliminary resource classification information, providing a basis for subsequent more accurate resource classification and management.

[0028] Perform clustering analysis on the preliminary resource classification information to generate resource classification features. Among them, resource classification features include classification features by theme, classification features by audience group, and classification features by dissemination channel. Based on theme clustering, resources with similar themes are grouped into one category. For example, all science fiction-themed movies, novels, comics, etc. are grouped into science fiction resources to form classification features by theme; starting from the audience group, resources targeted at the same audience group are clustered. For example, children's picture books, children's animations, and children's music are grouped into children's audience resources to obtain classification features by audience group; according to the dissemination channel, resources mainly disseminated through social media, resources published on professional platforms, etc. are clustered respectively to generate classification features by dissemination channel. These resource classification features contribute to the efficient organization and precise sharing of resources.

[0029] S103. Process the resource association relationship data to generate resource association features.

[0030] In one implementation, extract and screen the resource link information and resource attribute association information in the resource association relationship data to generate resource reference relationship factors, attribute association data, and basic resource association type information. In the cloud environment, there are complex association relationships between digital media content resources, and these relationships are mainly reflected in two aspects: resource links and resource attributes. Resource link information includes hyperlink references between different resources. For example, a video resource embeds a link to another audio resource, or a graphic and text news item quotes a link to a specific video. Resource attribute association information covers the associations of resources in attributes such as theme, theme, creator, audience group, etc. For example, a series of works produced by the same creator, or different types of resources targeted at the same audience group (such as children's animated videos and children's audiobooks).

[0031] When extracting resource link information, by parsing resource files or web page code, the link content therein is identified to determine the source resource and target resource of the link. For resource attribute association information, relevant attribute information is obtained from the metadata of the resource. For example, information such as theme and creator is extracted from the metadata of a video. Then, the extracted information is screened and processed to remove invalid or incorrect links, as well as attribute information that has no actual associated meaning. For example, some expired links or incorrect attribute tags that have nothing to do with the main body of the resource. After screening, a resource reference relationship factor is generated, which records the reference relationship between resources. For example, if video A references audio B, the corresponding reference relationship factor can be generated; the attribute association data contains the screened resource attribute association information, such as the attribute association between a specific video and other videos of the same theme; the basic information of the resource association type clarifies the type of resource association, such as link reference association, theme association, etc.

[0032] Based on the resource association analysis model library, the resource reference relationship factor, attribute association data, and basic information of the resource association type are processed to generate resource association tightness information and association direction change information. The resource association analysis model library contains various models for analyzing resource association relationships, such as the association analysis model based on graph theory and the clustering analysis model in machine learning. These models can deeply analyze the resource reference relationship factor, attribute association data, and basic information of the resource association type generated in the previous step.

[0033] Taking the model based on graph theory as an example, resources are regarded as nodes in a graph, and the resource reference relationship factor and attribute association data are used as the edges between nodes. By calculating indicators such as the shortest path and connection strength between nodes, the tightness of the association between resources is evaluated. If there are multiple direct or indirect links between two resources and there are also many overlaps through attribute association, then their association tightness is relatively high. For the association direction change information, it is determined by analyzing the change trend of the reference relationship and attribute association between resources. For example, if the reference frequency of one resource to another resource gradually increases, or the similarity between the two resources in terms of attributes continuously improves, it can be judged that the association direction changes towards a closer direction; conversely, if the reference frequency decreases or the attribute similarity decreases, the association direction changes towards a looser direction. Through the processing of these models, information that can accurately reflect the resource association tightness and association direction change is generated, providing strong support for subsequent resource management and sharing.

[0034] Classify and identify relevant data in the resource association relationship data based on the information of the closeness of resource association and the change information of the association direction to generate the screened result of the associated data. For the closeness of association, different thresholds are set to divide the association degree levels of resources, such as highly associated, moderately associated, and weakly associated. For highly associated resources, they are marked as key associated resources; moderately associated resources are marked as general associated resources; and weakly associated resources are marked as weakly associated resources. In terms of the change of the association direction, if the association direction changes towards a closer direction, it is marked as a positively changing resource; if it changes towards a looser direction, it is marked as a negatively changing resource. Through such classification and identification processing, clear attributes are assigned to each piece of data in the resource association relationship data. Then, according to these classifications and identifications, data with specific attributes are screened out, such as screening out all highly associated and positively changing resource data, or screening out weakly associated but potentially associable resources with the possibility of association improvement. These screened data form the screened result of the associated data, which can highlight the parts with important value or potential value in the resource association relationship, facilitating more targeted subsequent resource integration and sharing.

[0035] Summarize and analyze the screened result of the associated data to generate resource association characteristics, where the resource association characteristics are used to characterize the association degree and the association direction situation between digital media content resources. Summarize the screened result of the associated data generated in the previous step, integrate the screened data of different types and different sources together to form a comprehensive data set. Then, conduct in-depth analysis on this summarized data set and calculate various statistical indicators and characteristic values. For example, calculate the proportion of the number of highly associated resources in the total number of resources to measure the overall closeness of association between resources; analyze the distribution of positively changing resources in different resource types to understand the trend characteristics of the change of the resource association direction. Through these analyses, resource association characteristics are generated. These characteristics accurately describe the association degree between digital media content resources in a quantitative manner, such as the specific value of the closeness of association, the proportion of resources with different association degrees, etc.; at the same time, they also reflect the situation of the association direction, such as whether the association direction generally develops towards a closer direction or a looser direction, and the differences in different resource categories. The resource association characteristics provide key decision-making basis for the integrated sharing of digital media content resources in the cloud environment, helping the platform better organize resources, optimize the recommendation algorithm, and improve the efficiency and quality of resource sharing.

[0036] S104, process the strategies in the shared policy update mechanism information based on the resource value evaluation system to generate the updated shared policy data.

[0037] In one implementation, the policies in the shared policy update mechanism information are extracted and classified based on policy objectives and application scenarios to generate policy data for content dissemination, policy data for user experience optimization, policy data for resource management, and policy data for copyright protection. For the policies for content dissemination, their objective is to promote the dissemination of digital media content resources inside and outside the cloud platform. For example, policies such as the setting of sharing permissions for resources (public sharing, private sharing, sharing with specified users) and the selection of dissemination channels (social media sharing, in-site sharing) will be extracted to form policy data for content dissemination. For instance, the sharing policy of a certain video resource is set to public sharing and supports dissemination on multiple social media platforms, and this information constitutes the policy data for content dissemination of this resource. The policies for user experience optimization aim to enhance the user experience when using the cloud platform. Policies such as the adaptive playback policy for different network environments and the setting of personalized recommendation algorithms are classified into this category. For example, the cloud platform provides a policy to automatically reduce the video quality when the network is unstable to ensure smooth playback, and a policy to provide personalized content recommendations based on the user's historical behavior data, and these all belong to the policy data for user experience optimization.

[0038] The policies for resource management are mainly used to reasonably manage and allocate digital media content resources on the cloud platform. Policies such as the storage allocation rules for resources and access permission control belong to this category. For example, the cloud platform stipulates that certain high-definition video resources are stored in a specific high-performance storage area and can only be accessed by paid users, and these policies constitute the policy data for resource management. The policies for copyright protection are to protect the copyright of digital media content resources. Policies such as copyright notices and the application of digital rights management (DRM) technology are extracted to form policy data for copyright protection. For example, the copyright notice of a certain music resource clearly stipulates that it cannot be reproduced without authorization and uses DRM technology to prevent illegal copying and dissemination, and this information is the policy data for copyright protection.

[0039] Based on the resource value evaluation system, process the policy data for content dissemination, user experience optimization, resource management, and copyright protection to generate the content dissemination value evaluation value, user experience optimization value evaluation value, resource management value evaluation value, and copyright protection value evaluation value. For the policy data for content dissemination, evaluate its content dissemination value by analyzing the impact of the policy on aspects such as the scope, speed, and effect of content dissemination. For example, if a dissemination policy can significantly increase the playback volume of video resources in a short period, then its content dissemination value evaluation value is relatively high. For the policy data for user experience optimization, evaluate from aspects such as user satisfaction, usage convenience, and function satisfaction to obtain the user experience optimization value evaluation value. For example, a personalized recommendation policy that can accurately recommend content of interest to users, improving user satisfaction and platform stickiness, will have a relatively high user experience optimization value evaluation value.

[0040] The policy data for resource management evaluates the resource management value based on factors such as resource utilization rate, management cost, and security. For example, a reasonable storage allocation rule can improve the utilization rate of storage resources and reduce management costs, and its resource management value evaluation value will be relatively high. The policy data for copyright protection is evaluated based on aspects such as the effect of copyright protection and the degree of reduction in infringement risk to obtain the copyright protection value evaluation value. If a copyright protection policy can effectively prevent copyright infringement and reduce the risk of copyright disputes, its copyright protection value evaluation value will be relatively high.

[0041] Based on the content dissemination value evaluation value, user experience optimization value evaluation value, resource management value evaluation value, and copyright protection value evaluation value, mark and screen various policy data to generate high-value policy screening results, medium-value policy screening results, and low-value policy screening results. After obtaining the value evaluation values of various policies, mark and screen these policy data according to the preset thresholds. Set a certain numerical range as the standard for high-value policies. For example, policy data with a content dissemination value evaluation value higher than 80 points (out of 100), a user experience optimization value evaluation value higher than 75 points, a resource management value evaluation value higher than 85 points, and a copyright protection value evaluation value higher than 90 points are marked as high-value policies. Similarly, set the threshold ranges for medium-value and low-value policies. For example, policy data with a content dissemination value evaluation value between 60 - 80 points, a user experience optimization value evaluation value between 50 - 75 points, a resource management value evaluation value between 60 - 85 points, and a copyright protection value evaluation value between 70 - 90 points are marked as medium-value policies; those below the threshold range of medium-value policies are marked as low-value policies.

[0042] Then, based on these tags, filter out the policy data of different value levels to form the high-value policy screening results, medium-value policy screening results, and low-value policy screening results respectively. For example, after screening, it is found that the public sharing and multi-channel dissemination policies of a certain video resource score 90 points in the content dissemination value assessment and belong to high-value policies, so they are included in the high-value policy screening results; while the storage allocation policy of a certain resource scores 65 points in the resource management value assessment and belongs to medium-value policies, so it is included in the medium-value policy screening results.

[0043] Based on the policy conflict coordination rules, perform policy conflict detection and coordination processing on the high, medium, and low-value policy screening results respectively to generate high-feasibility policy information. There may be conflicts between different policies. For example, a policy oriented to content dissemination may conflict with a policy oriented to copyright protection, and excessive content dissemination may increase the risk of copyright infringement. Therefore, it is necessary to perform conflict detection and coordination processing on the policy screening results of different value levels based on the policy conflict coordination rules. The policy conflict coordination rules are a set of pre-established rules used to judge and resolve conflicts between policies. For example, it is stipulated that under the principle of copyright protection priority, when the content dissemination policy may lead to an increase in the risk of copyright infringement, the content dissemination policy is adjusted.

[0044] When performing conflict detection on the high-value policy screening results, check whether there are conflicts between these policies and with other policies. If conflicts are found, adjust and coordinate according to the coordination rules. For example, if a high-value content dissemination policy may lead to the risk of copyright infringement, then this policy needs to be modified, such as restricting the dissemination scope or adding copyright statements and other measures to eliminate the conflict. Similarly, similar conflict detection and coordination processing are also performed on the medium-value and low-value policy screening results. After conflict detection and coordination processing, high-feasibility policy information is generated. These information are an optimized and adjusted set of policies, ensuring that each policy coordinates with each other and can be effectively implemented in practical applications.

[0045] Obtain the real-time resource status data of the cloud platform and the dynamic information of user behavior. The real-time resource status data of the cloud platform and the dynamic information of user behavior are crucial for formulating effective sharing policies. The real-time resource status data includes cloud storage-related data (such as total storage capacity, used capacity, remaining available capacity, etc.), network-related data (such as real-time rate, average rate, bandwidth peak and valley of network bandwidth, average value, maximum value, minimum value of network latency), and computing resource data (such as CPU usage rate, load situation, number of cores, memory usage, free amount, memory frequency, etc.). These data reflect the current operating state and resource availability of the cloud platform.

[0046] The dynamic information of user behavior is obtained by analyzing the historical behavior data (such as browsing records, collection records, play duration, search keywords, etc.) and real-time behavior data (such as the content currently being browsed, operation behaviors, etc.) of users on the platform. For example, by analyzing the browsing records and search keywords of users, the interest preferences and needs of users can be understood; by monitoring the current operation behaviors of users, such as pausing and fast-forwarding during video playback, the usage habits and experience feelings of users can be understood. After obtaining these data, the sharing strategy can be further optimized and adjusted according to the actual resource status of the cloud platform and the behavior needs of users.

[0047] Based on the real-time resource status data of the cloud platform and the dynamic information of user behavior, the highly feasible strategy information is weighted to generate updated sharing strategy data. The real-time resource status data of the cloud platform and the dynamic information of user behavior provide a basis for the adjustment of the sharing strategy. According to these data, the highly feasible strategy information is weighted. For example, if the network bandwidth of the cloud platform is low, then in the content dissemination strategy, the weight of the dissemination strategy for high-definition videos needs to be reduced, and the video format with a lower bit rate is preferably selected for dissemination to ensure smooth playback; if the dynamic information of user behavior shows that the demand of users for a certain type of content increases, then in the personalized recommendation strategy, the recommendation weight of this type of content needs to be increased.

[0048] By weighted adjustment of each strategy in the highly feasible strategy information according to the real-time resource status data and the dynamic information of user behavior, updated sharing strategy data is generated. These updated strategy data can better adapt to the actual situation of the cloud platform and the needs of users, and improve the integrated sharing effect of digital media content resources. For example, after weighted processing, the generated updated sharing strategy data may stipulate that when the network bandwidth is low, video resources with lower resolution but higher smoothness are preferably recommended to users, and at the same time, according to the real-time interest preferences of users, the priority of the recommended content is adjusted, so as to improve the user experience and the effective utilization of resources.

[0049] S105, process the standardized media content sequence, cloud platform features, and resource association features based on the target integrated sharing model to generate target integrated sharing features.

[0050] In one implementation, target data extraction and classification processing are performed on the standardized media content sequence, cloud platform features, and resource association features to generate media content key element data, cloud platform performance key indicator data, and resource association core relationship data. The standardized media content sequence, cloud platform features, and resource association features are important data sets obtained through preliminary processing. For the standardized media content sequence, in the image key element sequence, image recognition technology is used to locate key elements such as people and objects. For example, in a landscape image, elements such as mountains and rivers are extracted; in the audio spectrum feature sequence, information such as the frequency and amplitude of the audio is analyzed to obtain the key frequency bands and energy distribution of the audio; in the video scene transition sequence, the shot transitions and scene changes in the video are detected, and the key frames and transition points of each scene are marked. Key elements are extracted from these sequences to form media content key element data.

[0051] For cloud platform features, features such as storage capacity availability rate, network bandwidth stability, and computing resource load balancing have quantified the performance of the cloud platform. Key indicators that have a crucial impact on resource sharing and processing are extracted from these features, such as the storage capacity availability rate, average rate and peak rate of network bandwidth, and load rate of CPU cores, to generate cloud platform performance key indicator data. Resource association features contain association information between resources. The core association relationships between resources are extracted from the resource association features. For example, in the resource reference relationship factor, it is determined which resources have strong reference relationships; in the attribute association data, close association relationships based on attributes such as theme and creator are found to form resource association core relationship data.

[0052] Based on the target integration and sharing model, analysis and processing are performed on the media content key element data, cloud platform performance key indicator data, and resource association core relationship data to generate content sharing value importance assessment information and integration and sharing difficulty assessment information. The target integration and sharing model is the core tool for realizing the integrated sharing of digital media content resources. The media content key element data, cloud platform performance key indicator data, and resource association core relationship data are input into the target integration and sharing model. Through the analysis of the media content key element data, the model evaluates the value of different media content during the sharing process. For example, media resources with popular themes and high-quality content have higher content sharing value; through the analysis of the cloud platform performance key indicator data, the model judges the support ability of the cloud platform for resource sharing. For example, a cloud platform with sufficient storage capacity and stable network bandwidth is more conducive to resource sharing; through the analysis of the resource association core relationship data, the model understands the degree of closeness and direction of the association between resources. Resources with close and positive associations have higher sharing value. Based on these analysis results, content sharing value importance assessment information is generated.

[0053] The model evaluates the possible difficulties in the integration and sharing process based on factors such as the performance status of the cloud platform, the complexity of the associations between resources, and the characteristics of the media content. For example, if the computing resource load on the cloud platform is unbalanced, it may lead to a slower resource processing speed and increase the difficulty of integration and sharing; complex resource association relationships, such as a large number of cross-references and complex attribute associations, will also increase the difficulty of integration and sharing. Through these analyses, integration and sharing difficulty assessment information is generated.

[0054] Based on the content sharing value importance assessment information and the integration and sharing difficulty assessment information, the target data in the standardized media content sequence, cloud platform characteristics, and resource association characteristics are screened and associated to generate the screening result of the integration and sharing target data. According to the content sharing value importance assessment information, a value threshold is set to screen out the key element data of media content with high sharing value, the key performance index data of the cloud platform, and the core relationship data of resource associations. For example, for media content, only the key element data with a content sharing value importance assessment score higher than a certain score is retained; for cloud platform performance indicators, the indicator data that has a positive impact on resource sharing and meets a certain standard is selected; for resource association relationships, the core relationship data with close associations and important significance for sharing is retained. According to the integration and sharing difficulty assessment information, further screen the data that is more easily shared under the current cloud platform conditions and resource association situations. For data with greater difficulty, if it cannot be effectively shared under the existing conditions, it is temporarily excluded. At the same time, the screened data is associated to establish logical connections between them. For example, match the media content with high sharing value with the cloud platform performance indicators that support its sharing and the relevant resource association relationships to generate the screening result of the integration and sharing target data.

[0055] Integrate and calculate the screening result of the integration and sharing target data, and process it in combination with the weight information in the target integration and sharing model and the updated sharing policy data to generate the target integration and sharing characteristics. Among them, the target integration and sharing characteristics are used to characterize the value and implementation difficulty of the digital media content resource integration and sharing. Integrate and calculate the screening result of the integration and sharing target data, and comprehensively calculate different types of data (key element data of media content, key performance index data of the cloud platform, core relationship data of resource associations) according to a certain algorithm to form a unified data set. This data set contains the key information related to the digital media content resource integration and sharing.

[0056] Combining the weight information in the target integrated sharing model and the updated shared policy data, the target integrated sharing model assigns different importance weights to different data elements, and the updated shared policy data also determines corresponding weights according to resource value and application scenarios. Applying these weights to the integrated dataset, the data is weighted to highlight the influence of important data and balance the relationships between different data. After the above processing, target integrated sharing features are generated. These features characterize the value and implementation difficulty of digital media content resource integration and sharing in a quantitative and intuitive manner. For example, the value of integrated sharing is represented by a comprehensive score, where a higher score indicates greater value; the implementation difficulty is described by some indicators and parameters, such as resource association complexity, cloud platform performance bottlenecks, etc., thus providing a key basis for subsequent resource integration and sharing decisions.

[0057] S106, process the target integrated sharing features based on the target integrated sharing model and the updated shared policy data to generate the integrated sharing results of digital media content resources.

[0058] In one implementation, obtain digital media content resource samples for training, cloud platform basic data samples, user demand data samples, resource association relationship data samples, shared policy data samples, corresponding data samples for verification, and a preset integrated sharing basic model. Collect digital media content resource samples from multiple sources. For example, obtain different types of videos from online video platforms, including movies, TV series, short videos, etc.; obtain various audio files from music platforms, such as pop music, classical music, audiobooks, etc.; obtain various images from picture material websites, such as photographic works, illustrations, icons, etc. At the same time, record the metadata of these resources, such as the duration, resolution, frame rate of videos, the format, bit rate, singer of audio, the size, color mode, creator of images, etc.

[0059] Collect cloud platform basic data samples, covering cloud storage-related data, such as total storage capacity, used capacity, remaining available capacity; network-related data, such as real-time rate, average rate, bandwidth peak and valley of network bandwidth, average value, maximum value, minimum value of network latency; computing resource data, such as CPU usage rate, load condition, number of cores, memory usage, free amount, memory frequency, etc. These data reflect the operating status and performance level of the cloud platform. Collect user demand data samples. By setting user feedback entrances on the platform, conducting questionnaire surveys, and analyzing the historical behavior data of users on the platform (such as browsing records, favorite records, play duration, search keywords, etc.), obtain the user's demand for digital media content, including information on content type, theme preference, functional requirements, etc.

[0060] Organize resource association data samples. From the perspective of resource links, collect hyperlink reference relationships between different resources, such as a link to a specific video cited in a graphic information, or a link to other audio resources embedded in a video; from the perspective of resource attribute association, analyze the association of resources in terms of themes, subjects, creators, audiences, etc., such as a series of works produced by the same creator, or different types of resources for the same audience (such as children's animation videos and children's book audio). Obtain sharing strategy data samples, including the content dissemination strategy currently implemented by the cloud platform (such as resource sharing permission settings, dissemination channel selection), user experience optimization strategy (such as adaptive playback strategy, personalized recommendation algorithm setting), resource management strategy (such as storage allocation rules, access permission control), and copyright protection strategy (such as copyright statement, digital rights management technology application). Prepare corresponding data samples for verification. These samples are consistent with the training samples in data type and structure, but from different sources, and are used for subsequent model verification. At the same time, obtain the preset integrated shared basic model, which is a preliminary model framework that provides a basis for subsequent model training.

[0061] The training digital media content resource samples, cloud platform basic data samples, user demand data samples, resource association data samples, and sharing strategy data samples are processed to generate preprocessed training feature data. The training digital media content resource samples are formatted in a unified manner. Due to the wide range of resource sources and diverse formats, there are a large number of incompatible formats and invalid codes. Using the format conversion tool, the video formats such as AVI and FLV are uniformly converted to MP4 format, the audio formats such as WMA and APE are uniformly converted to MP3 format, and the image formats such as ICO and BMP are uniformly converted to JPEG format. At the same time, the coding detection algorithm is used to identify and remove invalid codes to obtain the initial media content information. The initial media content information is parsed. For images, edge detection, color feature extraction and other algorithms are used to identify key elements in the image, such as character outlines, landmark buildings, etc., to form an image key element sequence; for audio, spectrum analysis technology is used to obtain audio frequency, amplitude and other features to construct an audio spectrum feature sequence; for video, with the help of scene segmentation algorithm, the camera switching and scene transition in the video are detected to generate a video scene transition sequence, thereby obtaining a standardized media content sequence.

[0062] Perform performance index screening and processing on the basic data samples of the cloud platform to remove redundant and invalid indexes. For example, remove the storage device temperature index in cloud storage data (which has a low correlation with storage capacity and usage and little impact on resource sharing decisions in some cloud platforms) and occasional error sampling data in network data. Retain indexes such as total storage capacity, used capacity, real-time network bandwidth rate, and CPU usage rate that play a key role in resource processing and sharing decisions to generate basic performance data. Calculate features from the basic performance data to generate cloud platform features. For example, calculate the available rate of storage capacity (the ratio of remaining available capacity to total capacity) to measure the remaining space of cloud storage resources; calculate the network bandwidth stability feature (such as the standard deviation of bandwidth fluctuations) by monitoring the fluctuations of network bandwidth over a period of time; analyze the CPU load in different time periods and calculate the computing resource load balancing feature (such as the difference in average load rates of each core). Perform semantic parsing and processing on the user demand data samples, and use semantic analysis algorithms in natural language processing (NLP) technology to extract key demand semantics. For example, for the user feedback "hoping to smoothly watch high-definition videos on the mobile phone and have more science fiction content", extract key demand semantics such as "mobile phone viewing", "high-definition video", and "science fiction theme", and summarize them to form a preliminary demand semantics set.

[0063] Perform classification and induction processing on the preliminary demand semantics set. From the aspect of content preference, count the demand frequencies of users for various themes (such as science fiction, romance, history, etc.) and content forms (such as video, audio, image) to determine content preference features; for function usage, analyze the demands and suggestions of users for functions such as platform search, play, and download to obtain function usage features; from the perspective of experience optimization, summarize experience optimization features based on user feedback on aspects such as play smoothness and interface design. Extract the metadata of digital media content resources to generate preliminary resource classification information. For example, extract metadata such as resource type (video, audio, image), theme (comedy, action, documentary, etc.), creator, creation time, audience group (children, teenagers, adults), and dissemination channel (social media, professional platform, official website), and classify the resources preliminarily according to these metadata. Perform clustering analysis processing on the preliminary resource classification information. Based on theme clustering, classify resources with similar themes into one category, such as classifying all science fiction movies, novels, comics, etc. as science fiction resources; starting from the audience group, cluster resources targeting the same audience group, such as classifying children's picture books, children's animations, and children's music as children's audience resources; according to the dissemination channel, cluster resources mainly disseminated through social media, resources published on professional platforms, etc. respectively to generate resource classification features. Integrate the processed data of various types above to generate preprocessed training feature data.

[0064] Process the preprocessed training feature data to generate integrated shared prediction parameters. Among them, the integrated shared prediction parameter vector is used to represent the prediction information on the integrated sharing results of digital media content resources, the key factors affecting the integrated sharing effect, and the optimization strategies. By establishing regression models, neural network models, etc., explore the potential relationships between data. Analyze the connections among the standardized media content sequences, cloud platform features, user demand features, resource classification features, as well as resource association relationship data and sharing strategy data. Extract from the analysis results the factors that have a key impact on the integrated sharing results of digital media content resources, such as the available rate of the storage capacity of the cloud platform and the impact of network bandwidth stability on resource transmission, and the impact of users' content preference features on resource recommendation. At the same time, predict the possible integrated sharing results under different circumstances, such as the sharing effects under different resource combinations and different sharing strategies. According to the analysis and prediction results, generate the integrated shared prediction parameter vector. This vector contains the prediction information on the integrated sharing results of digital media content resources, the key factors affecting the integrated sharing effect, and the optimization strategies. For example, predict which resource sharing method can obtain the best sharing effect under a certain network environment and user demand, and propose corresponding optimization strategies for the possible problems.

[0065] Based on the integrated shared prediction parameter vector, perform optimization training processing on the preset integrated shared basic model to generate the trained integrated shared model. Input the generated integrated shared prediction parameter vector into the preset integrated shared basic model. The model adjusts and optimizes its own structure and parameters according to these parameters. For example, if the prediction parameters indicate that the sharing effect of a certain type of resource under specific cloud platform conditions is quite different from the expectation, the model will adjust the resource allocation algorithm or recommendation strategy. Through multiple iterative trainings, let the model continuously learn and adapt to the training data. During the training process, the model will perform weighted processing on different data features according to the prediction parameters, highlighting the influence of important factors and weakening the influence of secondary factors. For example, assign higher weights to the cloud platform performance indicators and user demand features that have a greater impact on the integrated sharing results. After a certain number of trainings, the model gradually converges to form the trained integrated shared model. This model has better performance and adaptability in dealing with the integrated sharing problems of digital media content resources compared with the preset basic model, and can more accurately predict and process the tasks related to integrated sharing.

[0066] Based on the digital media content resource samples for verification, cloud platform basic data samples, user demand data samples, resource association relationship data samples, and sharing policy data samples, perform simulated digital media content resource integration and sharing processing on the trained integrated sharing model to generate verification results. Input the digital media content resource samples for verification, cloud platform basic data samples, user demand data samples, resource association relationship data samples, and sharing policy data samples into the trained integrated sharing model. The model simulates the integration and sharing process of digital media content resources and generates a series of output results according to the input data and its own algorithms. For example, the model will predict the results in aspects such as the dissemination effect of resources, user satisfaction, and resource utilization rate under given cloud platform conditions, user demands, and sharing policies. Organize these prediction results into verification results, which contain the predicted values of the model for various integration and sharing related metrics and are used to evaluate the accuracy and effectiveness of the model.

[0067] Based on the verification results, evaluate and adjust the trained integrated sharing model to generate a target integrated sharing model. Analyze the verification results to evaluate the performance of the trained integrated sharing model. For example, compare the difference between the predicted resource dissemination effect of the model and the actual situation and calculate the prediction accuracy rate; analyze whether the resources recommended by the model meet the actual needs of users and evaluate the accuracy of the user satisfaction prediction. According to the evaluation results, determine whether the model needs to be adjusted. If the model performs poorly in some aspects, such as having a low prediction accuracy rate or not being able to well adapt to specific sharing scenarios, the model needs to be adjusted. The adjustment methods include but are not limited to adjusting the parameters of the model, optimizing the algorithm structure, increasing or decreasing the weights of certain data features, etc. After multiple evaluations and adjustments, make the model meet the expected performance indicators and generate a target integrated sharing model. This model can more accurately handle the integration and sharing problems of digital media content resources and provide reliable support for subsequent resource integration and sharing.

[0068] In another implementation, based on the target integrated sharing model, comprehensive arithmetic processing is performed on the key element data of media content, the key performance indicator data of the cloud platform, and the core relationship data of resource association in the target integrated sharing features to generate initial integrated sharing data. The key element data of media content includes key person and object elements in images, key frequency band information of audio, key scene transition points of videos, etc.; the key performance indicator data of the cloud platform covers the available rate of storage capacity, the average rate and peak rate of network bandwidth, the core load rate of the CPU, etc.; the core relationship data of resource association reflects the close reference relationship between resources and the association relationship based on attributes such as theme and creator. By analyzing the key element data of media content, the value and attractiveness of different media content during the sharing process are evaluated; combined with the key performance indicator data of the cloud platform, the support ability and potential bottlenecks of the cloud platform for resource sharing are judged; based on the core relationship data of resource association, the collaborative sharing method between resources is determined. Through these operations, various types of data are integrated and transformed to generate initial integrated sharing data. These initial integrated sharing data are the preliminary results of the integrated sharing of digital media content resources, containing the basic integration information of the resources, but still need further optimization and processing.

[0069] The initial integrated sharing data is matched and fused with the content dissemination strategy, user experience optimization strategy, resource management strategy, and copyright protection strategy in the updated sharing strategy data to generate fusion result information. The updated sharing strategy data contains information in multiple aspects such as content dissemination strategy, user experience optimization strategy, resource management strategy, and copyright protection strategy. The content dissemination strategy stipulates the sharing permissions of resources (such as public sharing, private sharing, sharing with specified users) and dissemination channels (such as social media sharing, in-site sharing); the user experience optimization strategy covers adaptive playback strategies for different network environments, personalized recommendation algorithms, etc.; the resource management strategy involves storage allocation rules of resources, access permission control, etc.; the copyright protection strategy includes copyright statements, the application of digital rights management (DRM) technology, etc.

[0070] Match and integrate the initial integrated shared data with these sharing strategies. For example, according to the content dissemination strategy, determine appropriate sharing methods and dissemination channels for the resources in the initial integrated shared data to ensure that the resources can be disseminated as expected; according to the user experience optimization strategy, adjust the presentation form and recommendation method of the resources to improve the user experience. For example, optimize the recommended list of resources based on the user's historical behavior data and real-time needs; according to the resource management strategy, set the storage location and access permissions of the resources to ensure the reasonable storage and secure access of the resources; in combination with the copyright protection strategy, add corresponding copyright marks and protection measures to the resources to prevent copyright infringement. Through these matching and integration operations, generate integrated result information, closely combine the initial integrated shared data with the sharing strategies, and provide a more targeted data basis for subsequent processing.

[0071] Based on the integrated result information, combined with the real-time resource status data and user behavior dynamic information of the cloud platform, perform real-time optimization and dynamic adjustment on the processed data to generate an intermediate result of the integrated sharing of digital media content resources with timeliness and pertinence. The real-time resource status data of the cloud platform reflects the current operating conditions of the cloud platform, including cloud storage-related data (such as total storage capacity, used capacity, remaining available capacity), network-related data (such as real-time rate, average rate, bandwidth peak and valley of network bandwidth, average value, maximum value, minimum value of network latency), and computing resource data (such as CPU usage rate, load condition, number of cores, memory usage, free amount, memory frequency, etc.). The user behavior dynamic information is obtained by analyzing the user's historical behavior data (such as browsing records, favorite records, play duration, search keywords, etc.) and real-time behavior data (such as currently browsing content, operation behavior, etc.) on the platform, and is used to understand the user's interest preferences, demand changes, and usage habits.

[0072] Combine the integrated result information with the above real-time data to perform real-time optimization and dynamic adjustment on the processed data. For example, if the network bandwidth of the cloud platform is low, according to the real-time resource status data, adjust the transmission format of the resources in the content dissemination strategy, and preferentially select low-bitrate video or audio formats for dissemination to ensure smooth playback; according to the sudden increase in the user's demand for a certain type of content in the user behavior dynamic information, timely adjust the resource recommendation strategy and increase the recommendation intensity of relevant content. Through these real-time optimization and dynamic adjustments, generate an intermediate result of the integrated sharing of digital media content resources with timeliness and pertinence, making it more in line with the actual operating conditions of the cloud platform and the real-time needs of users.

[0073] Conduct quality assessment and compliance inspection on the intermediate results of digital media content resource integration and sharing, eliminate data that does not meet quality standards and regulatory requirements, and generate the results of digital media content resource integration and sharing. The quality assessment criteria can include aspects such as the integrity of resources, clarity, and quality indicators of audio and video. For example, whether the resolution of the video reaches a certain standard, whether there is noise in the audio, etc.; the compliance inspection rules are based on relevant laws and regulations, such as copyright law, privacy protection law, etc., to check whether there are problems such as copyright infringement and user privacy leakage in the resources. Conduct a comprehensive inspection of the intermediate results of integration and sharing, mark or eliminate resources that do not meet quality standards. For example, delete media content with too low resolution or poor audio quality; for resources with copyright problems or violations of other regulations, also take corresponding measures to ensure that the final integration and sharing results are legal and compliant. After quality assessment and compliance inspection, the remaining data is the result of digital media content resource integration and sharing. These results can be shared efficiently, securely, and legally on the cloud platform to meet users' needs for digital media content resources and enhance the utilization value and sharing effect of resources.

[0074] The server collects information on digital media content, cloud platform foundation, user needs, resource association relationships, and sharing strategy update mechanisms from multiple channels to provide a data basis for subsequent processing. Unify the formats of digital media content resources and parse the content to generate a standardized media content sequence; screen the cloud platform basic data and calculate the cloud platform characteristics; parse the user needs data and summarize the user need characteristics; extract the metadata of digital media content resources and generate resource classification characteristics through clustering analysis.

[0075] Extract resource link and attribute association information, generate resource association tightness and direction change information through model processing, and then classify, identify, summarize and analyze to obtain resource association characteristics. Classify the policy data according to the policy objectives, evaluate the value and screen, coordinate conflicts to generate highly feasible policies, and combine with the real-time data of the cloud platform for weighted processing to update the sharing policy data. Obtain training and verification data samples and basic models, process to generate prediction parameters, and optimize the training to obtain the target integration and sharing model. Use this model to process relevant data, combine with the sharing policy, optimize and adjust and check, and finally generate the results of digital media content resource integration and sharing, realizing the efficient integration and sharing of resources in the cloud environment.

[0076] In one implementation, as Figure 2 shown, the present application also provides an integrated sharing device for digital media content resources, including:

[0077] An acquisition module 201, configured to acquire digital media content resources, cloud platform basic data, user needs data, resource association relationship data, and sharing strategy update mechanism information;

[0078] A processing module 202 is configured to preprocess digital media content resources, cloud platform basic data, and user demand data to generate a standardized media content sequence, cloud platform features, user demand features, and resource classification features; process resource association relationship data to generate resource association features; process the policies in the sharing policy update mechanism information based on a resource value evaluation system to generate updated sharing policy data; process the standardized media content sequence, cloud platform features, and resource association features based on a target integrated sharing model to generate target integrated sharing features; and process the target integrated sharing features based on the target integrated sharing model and the updated sharing policy data to generate an integrated sharing result of digital media content resources.

[0079] Each embodiment in this application is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the integrated sharing method, electronic device, electronic equipment, and readable storage medium for evaluating digital media content resources, since they are basically similar to the embodiment of the integrated sharing method of digital media content resources described above, the description is relatively simple. For the relevant parts, reference can be made to the partial description of the embodiment of the integrated sharing method of digital media content resources described above.

Claims

1. An integrated sharing method for digital media content resources, characterized in that, Including: Obtain digital media content resources, cloud platform basic data, user demand data, resource association relationship data, and sharing policy update mechanism information; Preprocess digital media content resources, cloud platform basic data, and user demand data to generate a standardized media content sequence, cloud platform features, user demand features, and resource classification features; Process the resource association relationship data to generate resource association features; Process the policies in the sharing policy update mechanism information based on the resource value evaluation system to generate updated sharing policy data; Obtain a target integrated sharing model, including: obtain digital media content resource samples, cloud platform basic data samples, user demand data samples, resource association relationship data samples, and sharing policy data samples for training, corresponding data samples for verification, and a preset integrated sharing basic model; process the digital media content resource samples, cloud platform basic data samples, user demand data samples, resource association relationship data samples, and sharing policy data samples for training to generate preprocessed training feature data; process the preprocessed training feature data to generate integrated sharing prediction parameters; perform optimization training on the preset integrated sharing basic model based on the integrated sharing prediction parameter vector to generate a trained integrated sharing model; perform simulated digital media content resource integration sharing processing on the trained integrated sharing model based on the digital media content resource samples, cloud platform basic data samples, user demand data samples, resource association relationship data samples, and sharing policy data samples for verification to generate a verification result; evaluate and adjust the trained integrated sharing model based on the verification result to generate a target integrated sharing model; Process the standardized media content sequence, cloud platform features, and resource association features based on the target integrated sharing model to generate target integrated sharing features; Process the target integrated sharing features based on the target integrated sharing model and the updated sharing policy data to generate digital media content resource integration sharing results.

2. The method according to claim 1, characterized in that, Preprocess digital media content resources, cloud platform basic data, and user demand data to generate a standardized media content sequence, cloud platform features, user demand features, and resource classification features, including: Perform format unification processing on digital media content resources, remove incompatible formats and invalid encodings, and generate initial media content information; Perform content parsing on the initial media content information to generate a standardized media content sequence, where the standardized media content sequence includes an image key element sequence, an audio spectrum feature sequence, and a video scene transition sequence; Perform performance index screening processing on the cloud platform basic data, remove redundant and invalid indicators, and generate basic performance data; Perform feature calculation on the basic performance data to generate cloud platform features, where the cloud platform features include storage capacity availability rate features, network bandwidth stability features, and computing resource load balancing features; Perform semantic parsing processing on user demand data, extract key demand semantics, and generate a preliminary demand semantic set; Classify and summarize the preliminary demand semantic set to generate user demand characteristics, where the user demand characteristics include content preference characteristics, function usage characteristics, and experience optimization characteristics; Extract the metadata of digital media content resources to generate preliminary resource classification information; Perform clustering analysis on the preliminary resource classification information to generate resource classification characteristics, where the resource classification characteristics include classification characteristics by theme, classification characteristics by audience group, and classification characteristics by dissemination channel.

3. The method according to claim 1, wherein Process the resource association relationship data to generate resource association characteristics, including: Extract and screen the resource link information and resource attribute association information in the resource association relationship data to generate resource reference relationship factors, attribute association data, and basic resource association type information; Based on the resource association analysis model library, process the resource reference relationship factors, attribute association data, and basic resource association type information to generate resource association tightness information and association direction change information; Based on the resource association tightness information and association direction change information, classify and label the relevant data in the resource association relationship data to generate the association data screening result; Summarize and analyze the association data screening result to generate resource association characteristics, where the resource association characteristics are used to characterize the association degree and association direction situation between digital media content resources.

4. The method according to claim 1, wherein Based on the resource value evaluation system, process the strategies in the shared strategy update mechanism information to generate updated shared strategy data, including: Extract and classify the strategies in the shared strategy update mechanism information based on the strategy objectives and application scenarios to generate strategy data for content dissemination, strategy data for user experience optimization, strategy data for resource management, and strategy data for copyright protection; Based on the resource value evaluation system, process the strategy data for content dissemination, strategy data for user experience optimization, strategy data for resource management, and strategy data for copyright protection to generate content dissemination value evaluation values, user experience optimization value evaluation values, resource management value evaluation values, and copyright protection value evaluation values; Based on the content dissemination value evaluation values, user experience optimization value evaluation values, resource management value evaluation values, and copyright protection value evaluation values, mark and screen the various types of strategy data to generate high-value strategy screening results, medium-value strategy screening results, and low-value strategy screening results; Based on the strategy conflict coordination rules, perform strategy conflict detection and coordination processing on the high, medium, and low-value strategy screening results respectively to generate high-feasibility strategy information; Obtain the real-time resource status data of the cloud platform and the dynamic information of user behavior; Based on the real-time resource status data of the cloud platform and the dynamic information of user behavior, perform weighted processing on the high-feasibility strategy information to generate updated shared strategy data.

5. The method according to claim 1, characterized in that, Based on the target integrated sharing model, process the standardized media content sequence, cloud platform characteristics, and resource association characteristics to generate target integrated sharing characteristics, including: Perform target data extraction and classification processing on standardized media content sequences, cloud platform features, and resource association features to generate media content key element data, cloud platform performance key indicator data, and resource association core relationship data; Based on the target integrated sharing model, the key element data of media content, the key indicator data of cloud platform performance, and the core relationship data of resource association are analyzed and processed to generate the content sharing value importance assessment information and the integrated sharing difficulty assessment information; Based on the content sharing value importance assessment information and the integrated sharing difficulty assessment information, the target data in the standardized media content sequence, cloud platform characteristics, and resource association characteristics are screened and associated to generate the integrated sharing target data screening results; The screening results of the integrated sharing target data are integrated and processed in combination with the target integrated sharing model and the weight information in the updated sharing strategy data to generate target integrated sharing features, where the target integrated sharing features are used to characterize the value and implementation difficulty of the integrated sharing of digital media content resources.

6. The method according to claim 1, wherein The target integrated sharing features are processed based on the target integrated sharing model and the updated sharing strategy data to generate digital media content resource integrated sharing results, including: Based on the target integrated sharing model, the key element data of media content, key indicator data of cloud platform performance, and core relationship data of resource association in the target integrated sharing features are comprehensively processed to generate initial integrated sharing data; Match and fuse the content dissemination strategy, user experience optimization strategy, resource management strategy, and copyright protection strategy in the initial integrated shared data with the updated shared policy data to generate fusion result information; Based on the fusion result information combined with the real-time resource status data and user behavior dynamic information of the cloud platform, the processed data is optimized and dynamically adjusted in real time to generate timely and targeted digital media content resource integration and sharing intermediate results; Conduct quality assessment and compliance checks on the intermediate results of digital media content resource integration and sharing, eliminate data that does not meet quality standards and regulatory requirements, and generate digital media content resource integration and sharing results.

7. An integrated sharing device for digital media content resources, characterized in that, For implementing the method of claim 1, the apparatus comprises: The acquisition module is used to obtain digital media content resources, cloud platform basic data, user demand data, resource association relationship data, and sharing strategy update mechanism information; The processing module is used to pre-process digital media content resources, cloud platform basic data and user demand data to generate standardized media content sequences, cloud platform characteristics, user demand characteristics and resource classification characteristics; process resource association relationship data to generate resource association characteristics; process the policies in the sharing policy update mechanism information based on the resource value evaluation system to generate updated sharing policy data; process the standardized media content sequences, cloud platform characteristics and resource association characteristics based on the target integrated sharing model to generate target integrated sharing characteristics; process the target integrated sharing characteristics based on the target integrated sharing model and the updated sharing policy data to generate digital media content resource integrated sharing results.

8. An electronic device, characterized in that, include: a first processor; and a memory for storing executable instructions of the first processor; wherein, the first processor is configured to execute the integrated sharing method of the digital media content resource according to any one of claims 1 to 6 by executing the executable instructions.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a second processor, it implements the integrated sharing method of the digital media content resource according to any one of claims 1 to 6.

Images