System and method for collecting and sorting non-perpetual items based on digital conversion

Abstract

The invention discloses a system and a method for collecting and sorting non-abandoned items based on digital conversion, relates to the technical field of digital information processing, and aims to solve the technical problems that non-abandoned item data retrieval efficiency is low, specific data is difficult to quickly and accurately obtain, and non-abandoned cultural research, inheritance and popularization are hindered. Comprising an entity data storage module and a virtual data storage module, and is used for storing entity and virtual data of non-forgotten items; the non-abandoned project knowledge information extraction unit comprises a keyword extraction module and a picture extraction module and is used for extracting keywords and pictures from the data; and the non-abandoned knowledge digital conversion unit comprises an entity data digital processing module and a virtual data digital processing module and is used for carrying out digital conversion on the extracted data. According to the method, the corresponding storage data can be quickly searched according to the hyperlink, so that the retrieval efficiency is greatly improved, and the problem of low data retrieval efficiency is effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of digital information processing technology, and more specifically, to a system and method for collecting and organizing intangible cultural heritage items based on digital conversion. Background Art

[0002] In the field of digital management of intangible cultural heritage, existing technologies face the prominent problem of inefficient retrieval of intangible cultural heritage materials. As intangible cultural heritage materials continue to accumulate, their quantity and variety are increasing, encompassing both physical materials (such as ancient books, images of handicrafts, and live performance videos) and virtual materials (such as relevant web pages and digital audio).

[0003] However, the current lack of efficient data classification and associative retrieval mechanisms means researchers and enthusiasts often have to spend considerable time sifting through vast amounts of disorganized data when searching for specific intangible cultural heritage materials. For example, when searching for information on "traditional opera performance characteristics of a particular region," the lack of scientifically-based classification and associative analysis makes it difficult for users to quickly locate materials that include both textual descriptions of performance techniques and corresponding live performance videos or images. This severely hinders the research, inheritance, and promotion of intangible cultural heritage. To address this issue, we propose a system and method for collecting and organizing intangible cultural heritage items based on digital conversion. Summary of the Invention

[0004] The purpose of the present invention is to provide a system and method for collecting and organizing intangible cultural heritage items based on digital conversion, so as to solve the technical problems of low efficiency in retrieval of intangible cultural heritage items, difficulty in quickly and accurately obtaining specific information, and obstacles to the research, inheritance and promotion of intangible cultural heritage.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: a system for collecting and organizing intangible cultural heritage items based on digital conversion, comprising:

[0006] The intangible cultural heritage knowledge information storage unit includes a physical data storage module and a virtual data storage module, which are used to store the physical and virtual data of the intangible cultural heritage items;

[0007] The intangible cultural heritage project knowledge information extraction unit includes a keyword extraction module and an image extraction module, which are used to extract keywords and images from the data;

[0008] The intangible cultural heritage knowledge digital conversion unit includes a physical data digital processing module and a virtual data digital processing module, which is used to digitally convert the extracted data;

[0009] The digital intangible cultural heritage knowledge information storage unit is connected to the intangible cultural heritage knowledge digital conversion unit and includes an intangible cultural heritage knowledge classification storage module and a hyperlink establishment module for classifying and storing digital data and establishing keyword hyperlinks;

[0010] The intangible cultural heritage knowledge retrieval unit is connected to the digitized intangible cultural heritage knowledge information storage unit, and includes a keyword retrieval module and a picture retrieval module, which are used to retrieve digitized data through keywords or pictures.

[0011] Preferably, the keyword extraction module extracts keywords using the following algorithm:

[0012] Assume that the text content is T, and the word set W is obtained by word segmentation processing = {w1,w2,…,w n};

[0013] Calculate each word w i word frequency Where, TF(w i ) is used to measure w i The frequency of occurrence in a single document, w iT Indicates w i The number of times it appears in T, T W represents the total number of words in T;

[0014] Combined with inverse document frequency In the formula, IDF(w i ) is used to measure w i The universal importance of H, H represents the total number of documents, H wi Indicates that it contains w i The number of documents;

[0015] The comprehensive score of the final keyword extraction Score (w i )=TF(w i )×IDF(w i ), select several words with higher scores as keywords.

[0016] Preferably, the hyperlink establishing module establishes a keyword hyperlink by the following steps:

[0017] A: Get the keywords K extracted from the digitized data = {k1, k2, ..., k s};

[0018] B: Establish keyword density adjustment factor k1 and calculate the first value Where N is the total number of characters in the digitized data, num(k i ) is the keyword k i Number of occurrences;

[0019] C: Establish keyword scarcity enhancement factor k2, Where a is the number of times the extracted keyword appears in the digitized data of the i-th item;

[0020] D: Establish the cross-document correlation factor k3, the calculation formula is Where b ij is the number of times the keyword extracted from the i-th digitized data appears in the j-th digitized data, Used to calculate the total number of times the keyword appears in all documents;

[0021] E: Arrange the digitized data according to the value of the cross-document correlation factor k3 and establish the association between keywords and stored data;

[0022] F: Randomly establish hyperlinks between keywords based on the value of the cross-document correlation factor k3.

[0023] Preferably, the hyperlink establishment module is further provided with a keyword association word extraction module, which randomly extracts words with synonyms or extended meanings of the keyword as keyword association words through a semantic analysis algorithm before establishing the hyperlink, and hyperlinks the keyword association words with the corresponding keyword.

[0024] Preferably, when establishing a hyperlink, the hyperlink establishment module calculates the percentage of each keyword in the extracted keywords in the digitized data corresponding to the keywords and keyword associations extracted by the hyperlink establishment module to measure the importance of the keywords in the data.

[0025] Preferably, the keyword retrieval module is provided with synonymous keywords and extended keywords, the synonymous keywords include first-level synonyms, second-level synonyms and third-level synonyms, the extended keywords include first-level extended words, second-level extended words and third-level extended words, and the synonymous keywords and extended keywords are all based on the digitized data, and according to the numerical value of the cross-document correlation factor in the digitized data where the synonymous keyword or extended keyword appears, multiple hyperlinks are established in order from first to last.

[0026] A method for collecting and arranging intangible cultural heritage items based on digital conversion comprises the following steps:

[0027] S1: Data scanning and information extraction, scanning physical and virtual data, extracting keywords and images;

[0028] S2: Data digitization processing: digitally converting the physical and virtual data after extracting keywords and images, which is achieved through the physical data digitization processing module and the virtual data digitization processing module;

[0029] S3: establishing hyperlinks, establishing hyperlinks between the entity and virtual data after the keywords and pictures are extracted and the extracted keywords respectively;

[0030] S4: Data storage, storing the digitized data according to classification rules;

[0031] S5: Data search, searching digitized data by keywords or pictures.

[0032] Preferably, in step S1, the specific steps of data scanning and information extraction are as follows:

[0033] S101: For physical documents, read the content of paper documents, extract keywords, and digitize the information after keyword extraction;

[0034] S102: Scanning paper documents using an image extraction module to extract images from the documents;

[0035] S103: For the virtual data, the keyword extraction module extracts information and keywords;

[0036] S104: Scanning the virtual data, extracting and digitizing images in the data;

[0037] S105: Use the video conversion module to convert the video into digital data and extract text and pictures from the video.

[0038] Preferably, in step S3, the specific steps of establishing a hyperlink are as follows:

[0039] S301: extracting information from each piece of digitized data and obtaining extracted keywords;

[0040] S302: Obtaining a first value of the number of occurrences of the extracted keyword and the total number of words when establishing a hyperlink;

[0041] S303: Obtain a second value and establish a keyword scarcity enhancement factor k2;

[0042] S304: Obtain the cross-document correlation factor k3 and its calculation formula;

[0043] S305: Arrange the digitized data according to the value of the cross-document correlation factor k3, and establish an association between the keyword and the stored data;

[0044] S306: randomly establishing hyperlinks between the extracted keywords according to the value of the cross-document relevance factor k3.

[0045] Preferably, in step S5, data is retrieved in the following manner:

[0046] The keyword search module searches for corresponding stored data through hyperlinks and supports synonymous keyword and extended keyword searches;

[0047] The image retrieval module searches for the digital data with the highest similarity by comparing feature vectors.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. The present invention provides an intangible cultural heritage knowledge classification storage module and a hyperlink establishment module to classify and store digitized intangible cultural heritage materials and establish keyword hyperlinks. Combined with the keyword retrieval module and the image retrieval module, the system can quickly locate the information required by the user. The system can quickly search for corresponding stored information based on the hyperlink, greatly improving retrieval efficiency and effectively solving the problem of low information retrieval efficiency.

[0050] 2. The present invention also utilizes a keyword association extraction module within the hyperlink establishment module to randomly extract words with similar or extended meanings to the keyword as keyword associations before establishing the hyperlink. This allows searches to not only find information that directly matches the keyword, but also access a wider range of related information. The system can then find additional information through associations, further enriching search results, deepening the depth and breadth of searches, and better meeting research needs.

[0051] 3. This invention also measures the importance of keywords in the document by calculating the percentage of keywords in the hyperlink creation module when creating hyperlinks. Furthermore, the keyword retrieval module includes synonymous keywords and extended keywords, each of which creates hyperlinks based on the numerical value of the cross-document relevance factor. This enables the system to prioritize the presentation of documents that are more relevant and core to the search requirements. Documents that contain key performance techniques and are frequently mentioned are prioritized, optimizing the accuracy of search results and providing users with more valuable information. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 Schematic diagram of the system architecture of the present invention;

[0053] Figure 2 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0054] Example 1: Figure 1 As shown, the present invention relates to a system for collecting and arranging intangible cultural heritage items based on digital conversion, comprising an intangible cultural heritage knowledge information storage unit, an intangible cultural heritage item knowledge information extraction unit, an intangible cultural heritage knowledge digital conversion unit, a digital intangible cultural heritage knowledge information storage unit, and an intangible cultural heritage knowledge retrieval unit;

[0055] Intangible cultural heritage knowledge information storage unit, used to store physical and virtual data of intangible cultural heritage items;

[0056] In an embodiment of the present invention, the intangible cultural heritage knowledge information storage unit includes a physical data storage module and a virtual data storage module;

[0057] An entity data storage module for storing entity-form data such as books, calligraphy and paintings, pictures, audio, and video materials related to intangible cultural heritage projects;

[0058] A virtual data storage module for storing virtual-form data such as videos, audios, pictures, and web pages with hyperlinks related to intangible cultural heritage projects, providing a comprehensive storage foundation for intangible cultural heritage project data;

[0059] An intangible cultural heritage project knowledge information extraction unit for extracting keywords and pictures from the data;

[0060] In an embodiment of the present invention, the intangible cultural heritage project knowledge information extraction unit includes a keyword extraction module and a picture extraction module;

[0061] The keyword extraction module scans the entity data and virtual data, and extracts the keywords therein through natural language processing algorithms and text analysis techniques;

[0062] Among them, the extraction formula for keywords is: Let the text content be T, and perform word segmentation on it to obtain the word set W = {w1, w2,..., w n}, calculate the word frequency of each word w i In the formula, TF(w ) is used to measure the frequency of w i appearing in a single document. The higher the frequency, the stronger the representativeness of w i for the document, i represents the number of times w appears in T, T i represents the total number of words in T, T W =n; W =n;

[0063] Then combine with the inverse document frequency In the formula, IDF(w i ) is used to measure the general importance of w i . If w i appears frequently in multiple documents, its IDF value is lower, indicating that the discrimination of w i is lower (such as stop words like "of" and "is"), H represents the total number of documents, represents the number of documents containing w[[ID=五一]] i ; [[ID=5四十二]]

[0064] The final comprehensive score of keyword extraction Score(w i ) = TF(w i ) × IDF(w i ), select several words with higher scores as keywords. In the formula, Score(w i ) combines TF(w i ) and IDF(w i), the higher the score, the i The more it can represent the core content of the document, in the intangible cultural heritage materials, this formula is used to automatically identify representative professional terms;

[0065] The image extraction module is used to extract images from the physical and virtual materials of intangible cultural heritage projects and realize the preliminary extraction of key information of the materials;

[0066] Intangible cultural heritage knowledge digital conversion unit, used to digitally convert the extracted data;

[0067] In an embodiment of the present invention, the intangible cultural heritage knowledge digital conversion unit includes a physical data digital processing module and a virtual data digital processing module;

[0068] The physical data digitization processing module digitizes the physical data from which keyword information has been extracted, such as scanning paper books into electronic documents and digitally photographing physical images.

[0069] The virtual data digitization processing module is used to further digitize the virtual data from which image information has been extracted, such as extracting structured data from webpage content and converting the format of video and audio, making the intangible cultural heritage materials easier to store and manage;

[0070] A digital intangible cultural heritage knowledge information storage unit is connected to the intangible cultural heritage knowledge digital conversion unit and is used to classify and store digital data and establish keyword hyperlinks;

[0071] In an embodiment of the present invention, the digital intangible cultural heritage knowledge information storage unit includes an intangible cultural heritage knowledge classification storage module and a hyperlink establishment module;

[0072] The intangible cultural heritage knowledge classification storage module categorizes and stores digitized data according to the categories of intangible cultural heritage items, such as traditional skills, folk customs, and traditional music, to facilitate data search and management;

[0073] A hyperlink establishing module, for establishing a hyperlink with the keyword extracted by the keyword extracting module;

[0074] In an embodiment of the present invention, a hyperlink to a keyword extracted by a keyword extraction module is established, and the specific steps are as follows:

[0075] A: Obtain keywords: For each digitized data, the virtual data digitization processing module establishes an information extraction module, which scans the digitized data and obtains the keywords K = {k1, k2, ..., k s};

[0076] B: Get the first value: Establish the keyword density adjustment factor k1, k1 is a real number greater than 1, and the total number of words in the digitized data is N, and the keyword k i The number of occurrences is num(k i ), then the first value of the number of occurrences of the extracted keywords and the total number of words in the digitized data is obtained when a hyperlink is established. Where N is the total number of characters in the digitized data, num(k i ) is the keyword k i The number of occurrences is used to calculate the density (frequency) of the keyword in the document. The higher the density, the more important the keyword may be to the document, which helps to prioritize the establishment of hyperlinks to high-density keywords and improve retrieval efficiency. k1 is quantified through the calculation formula of R1. The two are essentially qualitative (factor) and quantitative (value) expressions of the same concept. R1 is the specific implementation of k1, which is used to measure keyword density, thereby guiding the priority of hyperlink establishment and ultimately optimizing the retrieval efficiency of intangible cultural heritage materials.

[0077] C: Get the second value: Establish the keyword scarcity enhancement factor k2, Where a is the number of times the extracted keyword appears in the i-th digitized data, i = 1,…,n. This coefficient is used to adjust the weight of the keyword. When the number of keyword appearances a is small, k2 is large, which enhances the weight of low-frequency but key terms (such as the names of specific techniques in intangible cultural heritage projects). Conversely, when a is large and k2 approaches 1, it prevents high-frequency words (such as "tradition" and "culture") from excessively affecting the establishment of hyperlinks.

[0078] D: At the same time, establish the cross-document correlation factor k3: the calculation formula is Where b ij is the number of times the keyword extracted from the i-th digitized data appears in the j-th digitized data, j = 1, ..., m, It is used to calculate the total number of times the keyword appears in all documents. k3 is used to measure the "universality" of the keyword. If a keyword appears frequently in multiple documents (k3 is large), the established hyperlink is more global; conversely, if it only appears in a few documents, the hyperlink is more specific, which helps to build a hierarchical association network. The intangible cultural heritage knowledge classification storage module has multiple digitized data. The values ​​of i and j in i≠j are different.

[0079] E: Establish a keyword database: Arrange the digitized data according to the value of the cross-document correlation factor k3, and associate the arranged keywords with the corresponding stored data;

[0080] F: Randomly establish hyperlinks: Based on the value of the cross-document correlation factor k3, randomly establish hyperlinks between m keywords extracted from the digitized data;

[0081] In an embodiment of the present invention, the hyperlink establishment module is further provided with a keyword association word extraction module, which, before establishing a hyperlink, randomly extracts words with similar or extended meanings to the keyword as keyword association words through a semantic analysis algorithm, and hyperlinks the keyword association words with the corresponding keyword to further enrich the association between the data;

[0082] In the embodiment of the present invention, when establishing a hyperlink, the hyperlink establishment module also calculates the percentage of the number of each keyword in the digitized data corresponding to the keyword and keyword association word extracted by the hyperlink establishment module. The calculation formula is: Used to measure the importance of keywords in the data, where: Represents k in a single digitized data i The number of K It represents the number of all keywords K in a single digitized data. The keyword set K = {k1, k2, ..., k n}, this formula is used to evaluate the relative importance of a single keyword in a document. If the percentage of keywords is high, it means that these words are the core content of the document and can be used to optimize the ranking of search results;

[0083] An intangible cultural heritage knowledge retrieval unit, connected to the digitized intangible cultural heritage knowledge information storage unit, for retrieving digitized data by keywords or pictures;

[0084] In an embodiment of the present invention, the intangible cultural heritage knowledge retrieval unit includes a keyword retrieval module and an image retrieval module;

[0085] Keyword search module, which searches for corresponding stored data through the hyperlink creation module;

[0086] In an embodiment of the present invention, the keyword search module is provided with synonymous keywords and extended keywords, the synonymous keywords include first-level synonyms, second-level synonyms, and third-level synonyms, and the extended keywords include first-level extended words, second-level extended words, and third-level extended words. The synonymous keywords and the extended keywords are each based on m pieces of digitized data, and according to the numerical value of the cross-document correlation factor k3 in the digitized data where the synonymous keyword or the extended keyword appears, multiple hyperlinks are established in order from first to last, so as to improve the accuracy and comprehensiveness of the search;

[0087] The image retrieval module searches for the digitized storage data with the highest similarity to the retrieved image through the intangible cultural heritage knowledge classification storage module, and uses image recognition algorithms, such as algorithms based on feature point matching and color histogram comparison, to perform similarity recognition.

[0088] Example 2: Figure 1 、 Figure 2 As shown, a method for collecting and organizing intangible cultural heritage items based on digital conversion includes the following steps:

[0089] S1: Data scanning and information extraction;

[0090] Using the intangible cultural heritage knowledge information extraction unit to scan the data in the database including the physical data storage module and the virtual data storage module, and extract keywords and pictures;

[0091] In an embodiment of the present invention, the specific steps of data scanning and information extraction in step S1 are as follows:

[0092] S101: For physical materials, the intangible cultural heritage project knowledge information extraction unit reads the content of the paper materials, extracts keywords using the above-mentioned keyword extraction algorithm, and digitizes the information after the keyword extraction;

[0093] S102: Scanning paper documents using an image extraction module to extract images from the documents;

[0094] S103: For the virtual data, the keyword extraction module extracts information and extracts keywords using the same keyword extraction algorithm;

[0095] S104: Scanning the virtual data, extracting and digitizing images in the data;

[0096] S105: using a video conversion module to convert the video into digital data, extracting text and images from the video and converting them into digital text information and images, and converting paper data and virtual data into electronic data or network data after digitization;

[0097] S2: Data digitization processing;

[0098] The physical data and virtual data after the keywords and pictures are extracted are digitally converted respectively, which is achieved through the physical data digital processing module and the virtual data digital processing module;

[0099] S3: Create a hyperlink;

[0100] After extracting keywords and pictures, hyperlinks are established between the physical data and virtual data and the extracted keywords;

[0101] In an embodiment of the present invention, in step S3, the specific steps of establishing a hyperlink are as follows:

[0102] S301: extracting information from each piece of digitized data and obtaining keywords extracted from the digitized data;

[0103] S302: Obtaining a first value of the number of occurrences of a keyword extracted from the digitized data and the total number of words when establishing a hyperlink;

[0104] S303: Obtain a second value and establish a keyword scarcity enhancement factor k2;

[0105] S304: Obtain the cross-document correlation factor k3 and its calculation formula;

[0106] S305: Arrange the digitized data according to the values ​​of the cross-document correlation factor k3, and associate the arranged keywords with the corresponding stored data;

[0107] S306: randomly establishing hyperlinks between m keywords extracted from the digitized data according to the value of the cross-document correlation factor k3;

[0108] S4: Data storage;

[0109] The digitized data is stored according to the classification rules of the intangible cultural heritage knowledge classification storage module;

[0110] S5: Data search;

[0111] Searching for digitized materials through the intangible cultural heritage knowledge retrieval unit is achieved using the keyword retrieval module and the image retrieval module;

[0112] Among them, the keyword search module searches for corresponding stored data through hyperlinks and supports synonymous keyword and extended keyword searches;

[0113] The image retrieval module searches for the digital data with the highest similarity by comparing feature vectors.

[0114] Example 3: Example based on the intangible cultural heritage project of Ordos City:

[0115] 1. Target Groups

[0116] Two representative intangible cultural heritage items from Ordos City were selected:

[0117] 1. Ordos short-tune folk songs (traditional music)

[0118] 2. Genghis Khan Festival (Folk Customs)

[0119] 2. System Module Implementation

[0120] (1) Intangible Cultural Heritage Knowledge Information Storage Unit

[0121]

[0122]

[0123] (2) Intangible Cultural Heritage Knowledge Information Extraction Unit

[0124] 1. Keyword extraction module (taking "Ordos short-tune folk songs" as the object)

[0125] (1) Text processing: word segmentation yields word sets such as “Ordos”, “short-tune folk song”, “Mongolian”, “melody”, “lyrics”, and “improvisation”.

[0126] (2) TF-IDF calculation

[0127] Term frequency TF: "Short-tune folk song" appears 5 times in the document. The total number of words in the document is 100, and TF = 5 / 100 = 0.05.

[0128] Inverse document frequency (IDF): The number of documents containing “short-tune folk songs” is 2, the total number of documents is 10, and IDF = log (10 / 2) = 1.609.

[0129] Comprehensive score: 0.05×1.609=0.08. Select the top 5 words as keywords.

[0130] (3) Extraction results: Ordos short-tune folk songs, Mongolian traditional music, improvisation, long-tune comparison, and grassland culture.

[0131] 2. Image extraction module

[0132] (1) Scan the photos of folk song performances in the physical data and extract the pictures of the singer holding the horse-head fiddle; extract the panoramic pictures of thousands of people offering hadas during the festival.

[0133] (2) Capture clothing details and 3D modeling images of sacred objects on the web page from virtual data.

[0134] (3) Intangible Cultural Heritage Knowledge Digital Conversion Unit

[0135] 1. Physical data digital processing module

[0136] (1) Paper music scores: Converted to PDF using a high-definition scanner, and OCR recognizes the music score text and generates searchable text.

[0137] (2) Videotape / DVD: Use the video conversion module to convert VHS format to MP4 and extract key frame images and audio clips from the video.

[0138] 2. Virtual data digital processing module

[0139] (1) Web page content: parse the HTML structure, extract plain text and embedded images, and store them as structured data (JSON).

[0140] (2) Audio files: Convert to FLAC format and add metadata tags (such as track name, festival link, and recording time).

[0141] (4) Digital intangible cultural heritage knowledge information storage unit

[0142] 1. Intangible Cultural Heritage Knowledge Classification Storage Module

[0143] (1) Classification rules: Stored in the secondary categories of “traditional music” and “folk customs”, and each category is subdivided into “text materials”, “picture materials” and “audio and video materials”.

[0144] (2) Storage example:

[0145] Traditional Music → Ordos Short-Tune Folk Songs → Written Materials: Music PDFs, Research Papers

[0146] Folk Customs → Genghis Khan Festival → Audio and Video Materials: Full video recording of the ceremony and audio of the eulogy.

[0147] 2. Hyperlink creation module

[0148] (1) Keyword association

[0149] Main keywords: "Ordos short-tune folk songs" related associative words "Mongolian short-tune" and "grassland folk songs".

[0150] Cross-document association: "Genghis Khan Festival" and "Ordos Culture" appear in multiple documents. After calculating the cross-document association factor, the link is established first.

[0151] (2) Scarcity enhancement: The word “Suled” only appears in two documents in the festival. The keyword scarcity enhancement factor k2=1+10 / 2=6, which improves its hyperlink priority.

[0152] (5) Intangible Cultural Heritage Knowledge Retrieval Unit

[0153] 1. Keyword search module

[0154] (1) Synonymous / Extended Keywords

[0155] First-level synonyms: "short-tune folk song" → "short song" "short tune".

[0156] Second-level extended word: "Genghis Khan Festival" → "Mongolian Festival".

[0157] (2) Search example: Enter "short tune", the system returns all the materials containing "short tune folk songs" and "short songs" through hyperlinks, and sorts them by cross-document correlation, giving priority to displaying high-frequency documents involving both "Ordos" and "short tune".

[0158] 2. Image retrieval module

[0159] Upload a picture of an Ordos folk singer wearing silver jewelry. The system will extract feature vectors (such as clothing color and decorative patterns) and match them with pictures and related text descriptions of similar clothing used in the festival.

[0160] 3. Implementation of Methods and Steps

[0161] S1: Data scanning and information extraction;

[0162] S101-S102: physical data processing;

[0163] Scan the "Ordos Folk Song Collection" and extract keywords such as "difference between long and short tunes" and "rhyme patterns of lyrics"; extract illustrations in the book (such as pictures of folk song manuscripts from different historical periods).

[0164] Take photos of actual objects used in sacrificial ceremonies (such as copper kettles) and extract keywords such as "material of sacrificial vessels" and "meaning of patterns".

[0165] S103-S105: virtual data processing;

[0166] Crawl the content of the Ordos Intangible Cultural Heritage official website and extract keywords such as "time of applying for intangible cultural heritage for short-tune folk songs" and "representative inheritors"; download and digitize the short videos of the festival displayed on the official website.

[0167] S2: Data digitization processing;

[0168] Convert folk song tapes from the 1950s into WAV format, generate transcripts using speech recognition technology, and associate them with the corresponding audio files;

[0169] Vectorize ceremony-related CAD design drawings (such as altar layouts) and save them in scalable SVG format.

[0170] S3: Create a hyperlink;

[0171] 1. Calculation of the keyword density adjustment factor k1: In the document "Analysis of Short-Tune Folk Song Melodies," "improvisation" appears three times, and the document has 500 words. Therefore, the first value R1 = 3 / 500 = 0.006, which determines the importance of this keyword in the document.

[0172] 2. Calculation of cross-document correlation k3: "Genghis Khan" appears in 10 documents, 8 of which appear in festival-related documents and 2 in historical research documents. k3 = (8 + 2) / 10 = 1. Establish cross-document links with "Mongolian History" and "Ordos Tribe".

[0173] S4: Data storage;

[0174] The data is stored in a three-level directory of "Intangible Cultural Heritage Category → Project Name → Data Type", for example: Folklore → Genghis Khan Festival → Photo Data → Photos of the 2023 Festival;

[0175] Create a metadata index, including project numbers (such as MD-001 short-tune folk songs, MS-002 Genghis Khan Festival), keyword tags, and storage paths.

[0176] S5: Data search;

[0177] Keyword search verification: Enter "Ordos Festival", the system returns:

[0178] Main document: "Ceremony Rules for Genghis Khan Festival" (with hyperlinks to festival videos).

[0179] Related documents: "Comparative Study of Mongolian Festival Culture" (linked through the associative word "Mongolian Festival").

[0180] Image search verification: Upload a blurry image of a festival mask. The system will accurately locate the graphic materials and craftsman interview videos of "2015 Festival Mask Making Process" through feature comparison.

[0181] IV. Summary of Implementation Effects

[0182] Improved search efficiency: Traditional searches require manually flipping through 30+ paper documents. Now, keyword hyperlinks can locate target information within 10 seconds, improving efficiency by over 90%.

[0183] Deepening of related search: Through associative words and cross-document links, users can expand their search from "Ordos short-tune folk songs" to related materials such as "History of the Development of Mongolian Music" and "Grassland Cultural Ecology", expanding the coverage by 5 times.

[0184] Multimedia fusion retrieval: supports cross-modal retrieval of "image → text → audio and video", such as associating clothing images with the audio and video of corresponding folk song performances to achieve a three-dimensional presentation of intangible cultural heritage materials.

[0185] V. Summary

[0186] This embodiment verifies the practicality of the intangible cultural heritage project collection and organization system and method based on digital conversion in real intangible cultural heritage projects, especially its significant advantages in multimodal data management, intelligent association retrieval and efficient classification storage, providing a practical digital solution for the research, inheritance and promotion of Ordos City's intangible cultural heritage.

[0187] The embodiments disclosed in the present invention are preferred embodiments, but are not limited to them. Ordinary technicians in this field can easily understand the spirit of the present invention based on the above embodiments and make different extensions and changes. As long as they do not deviate from the spirit of the present invention, they are all within the scope of protection of the present invention.

Claims

1. A system for collecting and arranging intangible cultural heritage items based on digital conversion, characterized in that: include: The intangible cultural heritage knowledge information storage unit includes a physical data storage module and a virtual data storage module, which are used to store the physical and virtual data of the intangible cultural heritage items; The intangible cultural heritage project knowledge information extraction unit includes a keyword extraction module and an image extraction module, which are used to extract keywords and images from the data; The intangible cultural heritage knowledge digital conversion unit includes a physical data digital processing module and a virtual data digital processing module, which is used to digitally convert the extracted data; The digital intangible cultural heritage knowledge information storage unit is connected to the intangible cultural heritage knowledge digital conversion unit and includes an intangible cultural heritage knowledge classification storage module and a hyperlink establishment module for classifying and storing digital data and establishing keyword hyperlinks; The intangible cultural heritage knowledge retrieval unit is connected to the digitized intangible cultural heritage knowledge information storage unit, and includes a keyword retrieval module and a picture retrieval module, which are used to retrieve digitized data through keywords or pictures.

2. The system for collecting and arranging intangible cultural heritage items based on digital conversion according to claim 1 is characterized in that: The keyword extraction module extracts keywords through the following algorithm: Assume that the text content is T, and the word set W is obtained by word segmentation processing = {w1,w2,…,w n }; Calculate each word w i word frequency Where, TF(w i ) is used to measure w i The frequency of occurrence in a single document, represents w i The number of times it appears in T, T W represents the total number of words in T; Combined with inverse document frequency In the formula, IDF(w i ) is used to measure w i The universal importance of H represents the total number of documents. Indicates that it contains w i The number of documents; The comprehensive score of the final keyword extraction Score (w i )=TF(w i )×IDF(w i ), select several words with higher scores as keywords.

3. The intangible cultural heritage item collection and arrangement system based on digital conversion according to claim 1 is characterized in that: The hyperlink establishment module establishes a keyword hyperlink by the following steps: A: Get the keywords K extracted from the digitized data = {k1, k2, ..., k s }; B: Establish keyword density adjustment factor k1 and calculate the first value Where N is the total number of characters in the digitized data, num(k i ) is the keyword k i Number of occurrences; C: Establish keyword scarcity enhancement factor k2, Where a is the number of times the extracted keyword appears in the digitized data of the i-th item; D: Establish the cross-document correlation factor k3, the calculation formula is Where b ij is the number of times the keyword extracted from the i-th digitized data appears in the j-th digitized data, Used to calculate the total number of times the keyword appears in all documents; E: Arrange the digitized data according to the value of the cross-document correlation factor k3 and establish the association between keywords and stored data; F: Randomly establish hyperlinks between keywords based on the value of the cross-document correlation factor k3.

4. The system for collecting and arranging intangible cultural heritage items based on digital conversion according to claim 1 is characterized in that: The hyperlink establishment module is further provided with a keyword association word extraction module, which randomly extracts words with similar or extended meanings to the keyword as keyword association words through a semantic analysis algorithm before establishing the hyperlink, and hyperlinks the keyword association words with the corresponding keyword.

5. The system for collecting and arranging intangible cultural heritage items based on digital conversion according to claim 4 is characterized in that: When establishing a hyperlink, the hyperlink establishment module calculates the percentage of each keyword in the extracted keywords in the digitized data corresponding to the keywords and keyword associations extracted by the hyperlink establishment module to measure the importance of the keywords in the data.

6. The system for collecting and arranging intangible cultural heritage items based on digital conversion according to claim 1 is characterized in that: The keyword search module is provided with synonymous keywords and extended keywords, wherein the synonymous keywords include first-level synonyms, second-level synonyms and third-level synonyms, and the extended keywords include first-level extended words, second-level extended words and third-level extended words. Synonymous keywords and extended keywords are all based on the digitized data, and according to the numerical value of the cross-document correlation factor in the digitized data where the synonymous keyword or extended keyword appears, multiple hyperlinks are established in order from first to last.

7. A method for collecting and arranging intangible cultural heritage items based on digital conversion, which is applicable to a system for collecting and arranging intangible cultural heritage items based on digital conversion according to any one of claims 1 to 6, characterized in that: The following steps are involved: S1: Data scanning and information extraction, scanning physical and virtual data, extracting keywords and images; S2: Data digitization processing: digitally converting the physical and virtual data after extracting keywords and images, which is achieved through the physical data digitization processing module and the virtual data digitization processing module; S3: establishing hyperlinks, establishing hyperlinks between the entity and virtual data after the keywords and pictures are extracted and the extracted keywords respectively; S4: Data storage, storing the digitized data according to classification rules; S5: Data search, searching digitized data by keywords or pictures.

8. The method for collecting and arranging intangible cultural heritage items based on digital conversion according to claim 7 is characterized in that: In step S1, the specific steps of data scanning and information extraction are as follows: S101: For physical documents, read the content of paper documents, extract keywords, and digitize the information after keyword extraction; S102: Scanning paper documents using an image extraction module to extract images from the documents; S103: For the virtual data, the keyword extraction module extracts information and keywords; S104: Scanning the virtual data, extracting and digitizing images in the data; S105: Use the video conversion module to convert the video into digital data and extract text and pictures from the video.

9. The method for collecting and arranging intangible cultural heritage items based on digital conversion according to claim 8 is characterized in that: In step S3, the specific steps of establishing a hyperlink are as follows: S301: extracting information from each piece of digitized data and obtaining extracted keywords; S302: Obtaining a first value of the number of occurrences of the extracted keyword and the total number of words when establishing a hyperlink; S303: Obtain a second value and establish a keyword scarcity enhancement factor k2; S304: Obtain the cross-document correlation factor k3 and its calculation formula; S305: Arrange the digitized data according to the value of the cross-document correlation factor k3, and establish an association between the keyword and the stored data; S306: randomly establishing hyperlinks between the extracted keywords according to the value of the cross-document relevance factor k3.

10. The method for collecting and arranging intangible cultural heritage items based on digital conversion according to claim 9 is characterized in that: In step S5, data is retrieved in the following manner: The keyword search module searches for corresponding stored data through hyperlinks and supports synonymous keyword and extended keyword searches; The image retrieval module searches for the digital data with the highest similarity by comparing feature vectors.

Images