Digital intelligence management system for ethnic sports culture

Abstract

The invention provides a digital intelligent management system for ethnic sports culture. The system comprises a data acquisition module which is used for acquiring ethnic sports culture resource data through an internet big database system; the data processing module is used for carrying out data processing operation on the collected national sports culture data; the data classification module is used for classifying the national sports culture data subjected to data processing operation; the data storage module is used for storing and managing the national sports culture data; the user interaction module is used for providing a user interaction interface and providing an intelligent question and answer function and personalized recommendation of national sports culture for a user; and the data visualization module is used for displaying the national sports culture data in the forms of charts and graphs by using a data visualization technology. Through collection, processing, classification, storage, interaction and data visualization of national sports culture related data, digital inheritance, protection and innovative development of the national sports culture are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of information and sports culture resource management, and particularly to a digital and intelligent management system for ethnic sports culture. Background Art

[0002] With the rapid development of information technology, the digitalization process in the cultural field has been continuously promoted. In the aspect of ethnic sports culture, there have been some related technical application attempts. However, in terms of ethnic sports culture data processing in the existing technology, only simple collection of data related to ethnic sports culture is carried out, and no further data processing is performed on the collected relevant data, which easily leads to errors in subsequent data analysis, dissemination, etc. In the aspect of ethnic sports culture dissemination, although there are Internet platforms for display, most of them are simple video playback or text introduction, with a relatively single dissemination form, lacking interactivity and personalized experience with users, unable to perform intelligent recommendations based on users' sports interest preferences, and unable to meet users' needs for in-depth understanding and participation in ethnic sports culture. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a digital and intelligent management system for ethnic sports culture to solve or at least partially solve the above problems existing in the prior art.

[0004] To achieve the above purpose, the present invention provides a digital and intelligent management system for ethnic sports culture, including:

[0005] A data acquisition module: used to collect ethnic sports culture resource data through an Internet big data system, and the ethnic sports culture resource data includes text, video and image data;

[0006] A data processing module: used to perform data processing operations on the text, video and image data in the collected ethnic sports culture data respectively;

[0007] A data classification module: used to classify the ethnic sports culture data after data processing operations;

[0008] A data storage module: used to store and manage the ethnic sports culture data after data classification;

[0009] A user interaction module: used to provide a user interaction interface and provide an intelligent question and answer function and personalized recommendation for users regarding ethnic sports culture;

[0010] A data visualization module: used to display ethnic sports culture data in the form of charts and graphs by using data visualization technology.

[0011] Furthermore, key features are extracted from the text data in the collected ethnic sports culture data, and features irrelevant to ethnic sports culture are deleted, as shown below:

[0012]

[0013] Among them, I(n) is the key feature of the text, m is the total number of categories of all texts, i is the i-th category, g(Ai) is the probability that the text of the Ai-th category appears in all texts, g(t) is the probability that the feature word t is included in all texts, is the probability that the feature word t is not included in all texts, and g(Ai|t) is the probability that the text containing the feature word t belongs to the text of the Ai-th category, is the probability that the text containing the feature word t does not belong to the text of the Ai-th category.

[0014] Furthermore, the processing operation on the video image data in the collected ethnic sports culture data specifically includes the following steps:

[0015] S11. Perform data cleaning operation on the video image data in the collected ethnic sports culture data, and perform missing data processing on it, as shown below:

[0016]

[0017] Among them, x m is the missing data point, and x a and x b are the adjacent data points before and after respectively;

[0018] S12. Based on step S11, perform feature extraction on the video image data after data cleaning and missing data processing, as shown below:

[0019] T(i) = ∑∣K(i) - K(i - 1)∣

[0020] Among them, T(i) is the difference value between the i-th frame and the previous frame of the video image data, and K(i) is the pixel matrix of the i-th frame.

[0021] Furthermore, the data classification module specifically includes the following steps:

[0022] S21. Based on the processed ethnic sports culture data, take out all the sample data of the ethnic sports culture of the first category, mark them as I, mark all the sample data of the remaining ethnic sports culture data as II, and use all the samples marked as I and II as input samples to train a learning model and mark their serial numbers;

[0023] S22. Based on the processed ethnic sports culture data, extract all the sample data of the second category of ethnic sports culture, mark it as I, mark all the sample data of the remaining ethnic sports culture data as II. The following steps are the same as those in step S21. Finally, mark the trained learning model with a serial number to distinguish it from the learning model in step S21;

[0024] S23. Repeat steps S21 - S22 until all categories in the ethnic sports culture data are traversed, and finally obtain multiple trained learning models;

[0025] S24. Based on steps S21 - S23, when inputting a sample data of an ethnic sports culture to be classified, use multiple trained learning models to conduct inductive classification on it.

[0026] Furthermore, the data storage module includes using a distributed file storage system, integrating multiple storage analysis frameworks, and providing functions for storage data extraction, storage data management, and storage data expansion.

[0027] Furthermore, the user interaction module specifically includes the following steps:

[0028] S10. Based on the data classification module, provide an intelligent Q&A function for the user about ethnic sports culture, and infer the sports types liked by the user according to the intelligent Q&A function;

[0029] S11. Use random data augmentation to generate multiple different sequence views of the sports types liked by the user, and input them into the attention mechanism and neural network module respectively to generate multiple user sports preferences;

[0030] S12. Conduct contrastive loss learning on the user sports preferences to optimize the user sports preferences;

[0031] S13. Dynamically fuse the optimized user sports preferences with the ethnic sports culture data classified by data, output the final user sports preferences, and provide personalized recommendations for the user about ethnic sports culture based on it, as follows:

[0032] s x =(1 - λ x )a x +λ x b x

[0033] where s x is the final sports preference of user x, λ x is the dynamic fusion, and a x and b x are the set sports preferences of user x respectively.

[0034] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0035] The present invention provides a digital and intelligent management system for ethnic sports culture. Through the data processing module, data processing operations are respectively performed on the text and video image data in the ethnic sports culture data to improve the accuracy of the data. Through the data classification module, the ethnic sports culture data is classified by category. Through the data storage module, the storage efficiency of the ethnic sports culture data is improved, and the management cost is reduced. Through the user interaction module, an interaction function is provided for users, and the user participation and satisfaction are improved. The present invention uses digital and intelligent technologies to digitally manage and intelligently analyze the ethnic sports culture data, realizing the collection, processing, classification, storage, interaction, and data visualization of the data related to ethnic sports culture, and achieving the digital inheritance, protection, and innovative development of ethnic sports culture. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only the preferred embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0037] Figure 1 It is a schematic structural diagram of a digital and intelligent management system for ethnic sports culture provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] The principles and features of the present invention will be described below with reference to the accompanying drawings. The listed embodiments are only used to explain the present invention and are not intended to limit the scope of the present invention.

[0039] Referring to Figure 1 , this embodiment provides a digital and intelligent management system for ethnic sports culture, including:

[0040] Data acquisition module: used to acquire ethnic sports culture resource data through the Internet big data system, and the ethnic sports culture resource data includes text and video image data.

[0041] Data processing module: used to perform data processing operations on the text and video image data in the acquired ethnic sports culture data respectively, specifically including:

[0042] Performing data processing operations on the text in the acquired ethnic sports culture data, extracting the key features of ethnic sports culture from the text data, and deleting the features irrelevant to ethnic sports culture, which is expressed as follows:

[0043]

[0044] Among them, I(n) is the key feature of the text, m is the total number of categories of all texts, i is the i-th category, g(Ai) is the probability that the text of the Ai-th category appears in all texts, g(t) is the probability that the feature word t is included in all texts, is the probability that the feature word t is not included in all texts, and g(Ai|t) is the probability that the text containing the feature word t belongs to the text of the Ai-th category. is the probability that the text containing the feature word t does not belong to the text of the Ai-th category; the key features of the national sports culture may include, but are not limited to, national sports culture features, the origin region and time of national sports culture, etc.

[0045] The processing operation on the video image data in the collected national sports culture data specifically includes the following steps:

[0046] S11. Perform data cleaning operations on the video image data in the collected national sports culture data and perform missing data processing on it, which is expressed as follows:

[0047]

[0048] where x m is the missing data point, and x a and x b are the adjacent data points before and after respectively;

[0049] S12. Based on step S11, perform feature extraction on the video image data after data cleaning and missing data processing, which is expressed as follows:

[0050] T(i) = ∑∣K(i) - K(i - 1)∣

[0051] where T(i) is the difference value between the i-th frame and the previous frame of the video image data, and K(i) is the pixel matrix of the i-th frame. When T(i) exceeds the preset threshold, this frame is marked as a key frame.

[0052] Data classification module: used to classify the national sports culture data after data processing operations, specifically including the following steps:

[0053] S21. Based on the national sports culture data after data processing, take out all the sample data of the national sports culture of the first category and mark them as I, mark all the sample data of the remaining national sports culture data as II, and use all the samples marked as I and II as input samples to train a learning model and mark its serial number as ①, indicating that serial number ① is used to distinguish the national sports culture of the first category from the rest of the categories.

[0054] S22. Based on the processed ethnic sports culture data, extract all the sample data of the second category of ethnic sports culture, mark it as I, mark all the sample data of the remaining ethnic sports culture data as II. The next steps are the same as those in step S21. Finally, mark the trained learning model with a serial number to distinguish it from the learning model in step S21;

[0055] S23. Repeat steps S21 - S22 until all categories in the ethnic sports culture data are traversed, and finally obtain multiple trained learning models;

[0056] S24. Based on steps S21 - S23, when inputting a sample data of an ethnic sports culture to be classified, use multiple trained learning models to classify it inductively. If the learning model with the i-th serial number classifies the sample data into category I and the remaining learning models classify it into category II, then determine that the sample data belongs to the ethnic sports culture of the first category in this learning model; if multiple learning models classify the sample into category I at the same time while the remaining learning models classify it into category II, then it is determined as misclassified; if all learning models classify the sample into category II, it is also determined as misclassified.

[0057] Data storage module: used to store and manage the ethnic sports culture data after data classification, specifically including:

[0058] The data storage module includes a distributed file storage system and integrates multiple storage analysis frameworks, providing the ability of storage data extraction, storage data management, and support for multiple computing engines for the data storage module. Due to the high scalability of the distributed file storage system, the storage architecture can also provide the function of expanding the storage of ethnic sports culture data to cope with the growth of data volume.

[0059] User interaction module: used to provide a user interaction interface and provide intelligent question - answering functions and personalized recommendations for users regarding ethnic sports culture, specifically including the following steps:

[0060] S31. Based on the data classification module, provide an intelligent question - answering function for users regarding ethnic sports culture, and infer the sports types liked by users according to the intelligent question - answering function;

[0061] S32. Use random data augmentation to generate multiple different sequence views of the sports types liked by users, and input them into the attention mechanism and neural network module respectively to generate multiple user sports preferences;

[0062] S33. Perform contrastive loss learning on the user sports preferences to optimize the user sports preferences;

[0063] S34. Dynamically fuse the optimized user sports preferences with the ethnosports culture data classified by data, output the final user sports preferences, and provide personalized recommendations on ethnosports culture to the user based on them, as follows:

[0064] λ x = sigmoid([a x ,b x ,z t Q x +c x )

[0065] s x =(1 - λ x )a x +λ x b x

[0066] where s x is the final sports preference of user x, λ x is the dynamic fusion, Q x and c x are the trainable weights and biases respectively, a x and b x are the set sports preferences of user x respectively, and z t is the ethnosports culture data that interacts with the user.

[0067] Data visualization module: used to display the ethnosports culture data in the form of charts and graphs by using data visualization technology, specifically including:

[0068] The data visualization module effectively correlates and integrates the ethnosports culture data processed by the data classification module according to multiple dimensions such as origin region and time by using data visualization tools, and displays them in the form of charts, graphs, etc., to help users understand the ethnosports culture more intuitively and comprehensively.

[0069] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A digital intelligent management system for ethnic sports culture, characterized in that, Including: Data acquisition module: used to collect ethnic sports culture resource data through the Internet big data system, and the ethnic sports culture resource data includes text, video and image data; Data processing module: used to perform data processing operations on the text and video and image data in the collected ethnic sports culture data respectively; Data classification module: used to classify the ethnic sports culture data after data processing operations; Data storage module: used to store and manage the ethnic sports culture data after data classification; User interaction module: used to provide a user interaction interface, and provide an intelligent question and answer function and personalized recommendation for the user on ethnic sports culture; Data visualization module: used to use data visualization technology to display ethnic sports culture data in the form of charts and graphs.

2. The digital intelligent management system for ethnic sports culture according to claim 1, characterized in that, The processing operation on the text data in the collected ethnic sports culture data includes: Extracting key features from the text data in the collected ethnic sports culture data, and deleting features irrelevant to ethnic sports culture, as follows: Among them, I(n) is the key feature of the text, m is the total number of categories of all texts, i is the i-th category, g(Ai) is the probability that the text of the Ai-th category appears in all texts, g(t) is the probability that the feature word t is included in all texts, is the probability that the feature word t is not included in all texts, and g(Ai|t) is the probability that the text containing the feature word t belongs to the text of the Ai-th category, is the probability that the text containing the feature word t does not belong to the text of the Ai-th category.

3. The digital intelligent management system for ethnic sports culture according to claim 1, wherein, The processing operation on the video and image data in the collected ethnic sports culture data specifically includes the following steps: S11. Perform data cleaning operation on the video and image data in the collected ethnic sports culture data, and perform missing data processing on it, as follows: where x m is a missing data point, and x a and x b are adjacent data points before and after, respectively; S12. Based on step S11, perform feature extraction on the video and image data after data cleaning and missing data processing, as follows: T(i) = ∑∣K(i) - K(i - 1)∣ Where, T(i) is the difference value between the i-th frame and the previous frame of the video and image data, and K(i) is the pixel matrix of the i-th frame.

4. The digital intelligent management system for ethnic sports culture according to claim 1, wherein The data classification module specifically includes the following steps: S21. Based on the ethnic sports culture data after data processing, take out all the sample data of the first category of ethnic sports culture, and mark them as I, mark all the sample data of the remaining ethnic sports culture data as II, and use all the samples marked as I and II as input samples to train a learning model, and perform serial number marking on it; S22. Based on the ethnic sports culture data after data processing, take out all the sample data of the second category of ethnic sports culture, and mark them as I, mark all the sample data of the remaining ethnic sports culture data as II, and the next steps are the same as step S21. Finally, perform serial number marking on the trained learning model to distinguish it from the learning model in step S21; S23. Repeat steps S21 - S22 until all categories in the ethnic sports culture data are traversed, and finally obtain multiple trained learning models; S24. Based on steps S21 - S23, when inputting a sample data of an ethnic sports culture to be classified, use multiple trained learning models to perform inductive classification on it.

5. The digital intelligent management system for ethnic sports culture according to claim 1, characterized in that The data storage module includes using a distributed file storage system, integrating multiple storage analysis frameworks, and providing storage data extraction, storage data management, and storage data expansion functions.

6. The digital intelligent management system for ethnic sports culture according to claim 1, wherein The user interaction module specifically includes the following steps: S10. Provide an intelligent Q&A function for ethnic sports culture based on the data classification module, and infer the sports types liked by the user according to the intelligent Q&A function; S11. Use random data augmentation to generate multiple different sequence views of the sports types liked by the user, and input them into the attention mechanism and neural network module respectively to generate multiple user sports preferences; S12. Perform contrastive loss learning on the user sports preferences to optimize the user sports preferences; S13. Dynamically fuse the optimized user sports preferences with the ethnic sports culture data classified by data, output the final user sports preferences, and provide personalized recommendations for ethnic sports culture to the user based on them, as follows: s x = (1 - λ x )a x + λ x b x Among them, s x is the final sports preference of user x, λ x is the dynamic fusion, a x and b x are respectively the set sports preferences of user x.

Images