Scholar-oriented digital identity node analysis method, device and system

Abstract

The invention discloses a scholar-oriented digital identity node analysis method, device and system, and belongs to the technical field of digital identity recognition. The method comprises the following steps: initiating a node addressing request to a root node based on an identifier analysis request initiated by a user side, so that the root node identifies a personnel identifier type corresponding to the node addressing request; selecting an analysis node Vj according to the personnel identification type, so that the analysis node Vj generates an analysis result of the identification analysis request; wherein the analysis node Vj is used for analyzing a personnel identification type q; and returning the analysis result to the user side. According to the method, interconnection and intercommunication of cross-platform data can be realized, the analysis efficiency and accuracy are improved, the user experience is optimized, and convenient and efficient academic achievement management and retrieval services are provided for scholars.

Description

Technical Field

[0001] The present application belongs to the field of digital identity recognition technology, and specifically relates to a method, device and system for analyzing digital identity nodes for scholars. Background Art

[0002] In the field of scientific research, the identification of scholars and the association of academic achievements are important links to ensure academic integrity and promote knowledge sharing and communication. With the diversified development of academic information platforms, different databases and platforms provide their own scholar identity identification systems, such as ORCID, Scopus Author ID, ResearcherID, BRID, etc. Although these systems effectively solve the problem of the uniqueness of scholar identities and the association of academic achievements, they also bring about data compatibility and complexity of analysis. Therefore, it is particularly important to develop a scholar digital identity node analysis method that can be compatible with and analyze multiple personnel identifications.

[0003] ORCID (Open Researcher and Contributor ID): An open researcher and contributor identification number provided by an international non-profit organization jointly funded by multiple research institutions, funding agencies and publishers. It is currently the most widely used author identifier. ORCID avoids name confusion of duplicate authors (or authors with very similar names) by providing a long-term digital ID card, and realizes the automatic association of the author's own research results.

[0004] Scopus Author ID: Scholar identification code automatically generated by Scopus database. Scholars do not need to apply or register. Once their achievements are included in Scopus database, they will automatically obtain this identification code. Scopus Author ID automatically matches and merges scholars' achievements through algorithms, making it convenient for scholars to retrieve and manage their academic achievements in Scopus database.

[0005] ResearcherID (Publons): A scholar identification code provided by the Web of Science platform. After a scholar registers a personal homepage, the platform will assign a Researcher ID. This ID can not only help scholars manage their personal works and citation information, but also synchronize with systems such as ORCID.

[0006] BRID (Basic Researcher ID): BRID is established in accordance with my country's national standard GB / T 32843-2016 "Science and Technology Resource Identification" (example: 03592.00.68682). The National Natural Science Foundation of China adopts the principle of assigning codes to researchers, and has now completed the assignment of codes to researchers with accounts in the Science Foundation Network Information System (hereinafter referred to as the Information System). Each BRID corresponds to only one researcher, and the code is fixed after assignment and does not reflect the researcher's personal information.

[0007] Although these systems have solved the problem of the uniqueness of scholar identities and the association of academic achievements to a certain extent, there are still many deficiencies in the data compatibility and parsing methods between the systems:

[0008] 1. Data island phenomenon: Data between different systems cannot be directly interoperable, resulting in scholars having to maintain their identity information and academic achievements on multiple platforms.

[0009] 2. Parsing complexity: Due to the different data formats and parsing methods of each system, scholars need to face complex parsing processes when searching and managing academic results across platforms, which reduces work efficiency.

[0010] 3. Diversified platform access: In each journal publishing system, authors need to associate their own personal ID when publishing papers, which means that each publishing system needs to be connected one by one, increasing the integration workload.

[0011] In view of the above problems, it is urgent to develop a scholar digital identity node parsing method that can be compatible with parsing multiple personnel identifiers. This method should have the following characteristics:

[0012] Compatibility: It can parse multiple personnel identifiers such as ORCID, Scopus Author ID, ResearcherID, BRID, etc., and realize the interconnection and interoperability of cross-platform data.

[0013] Efficiency: It has efficient data analysis and processing capabilities and can quickly and accurately complete the analysis and synchronization of large-scale data.

[0014] Ease of use: Provides a simple and clear operating interface and friendly user experience to lower the threshold for scholars to use.

[0015] Security: Ensure the security and privacy of data during parsing and transmission.

[0016] Low coupling: Reduce data requirements and interact with data through API interfaces. Summary of the invention

[0017] The present invention aims to solve the data compatibility and parsing complexity of various heterogeneous personnel identification systems in the academic field. By designing a digital identity node parsing method, device and system for scholars, the interconnection of cross-platform data can be achieved, the parsing efficiency and accuracy can be improved, and the user experience can be optimized, so as to provide scholars with convenient and efficient academic achievement management and retrieval services.

[0018] To achieve the above objectives, the technical solution of the present invention includes the following contents.

[0019] A method for resolving digital identity nodes for scholars, applied to a resolution node V i , the method comprising:

[0020] Based on the identity resolution request initiated by the client, a node addressing request is initiated to the root node, so that the root node identifies the personnel identity type corresponding to the node addressing request;

[0021] Select a parsing node V according to the personnel identification type j , so that the parsing node V j Generate a resolution result of the identity resolution request; wherein the resolution node V i and the parsing node V j They are used to parse personnel identification type p and personnel identification type q respectively;

[0022] The parsing result is returned to the client.

[0023] Further, the initiating a node addressing request to the root node includes:

[0024] Parsing Node V i Send the node addressing request to the API gateway

[0025] The API gateway resolves the node V based on the information in the node addressing request. i Perform identity verification;

[0026] In the parsing node V i If the identity authentication is passed, the API gateway forwards the node addressing request to the root node.

[0027] Further, the root node identifies the type of personnel identification corresponding to the node addressing request, including:

[0028] Constructing an identification type feature library, wherein the identification type feature library is composed based on a mapping relationship between identification type features and personnel identification types;

[0029] generating an identification data feature based on the node addressing request;

[0030] The identification data feature is matched with the identification type feature to obtain the personnel identification type corresponding to the node addressing request.

[0031] Furthermore, the construction of the identification type feature library includes:

[0032] Collect identification data;

[0033] Extracting identification data features, the identification data features including: at least one of vocabulary occurrence frequency, part-of-speech distribution, syntactic structure, encoding rules, data length, ratio of numbers to letters, and specific character combination patterns;

[0034] The identification data features are embedded into a prompt template of a large language model, and the identification data features are classified using the large language model.

[0035] Further, matching the identification data feature with the identification type feature to obtain the personnel identification type corresponding to the node addressing request includes:

[0036] Based on a similarity calculation algorithm, a similarity value between the identification data feature and the identification type feature is calculated; wherein the similarity calculation algorithm includes: a cosine similarity algorithm or a Euclidean distance algorithm;

[0037] Setting a similarity threshold;

[0038] The similarity value is compared with the similarity threshold to obtain the personnel identification type corresponding to the node addressing request.

[0039] Furthermore, before the identity resolution request initiated by the client, it also includes:

[0040] The client determines whether there is a resolution result corresponding to the identity resolution request in the cache;

[0041] If the resolution result corresponding to the identity resolution request exists in the cache, the resolution result is pulled from the cache;

[0042] If the resolution result corresponding to the identity resolution request does not exist in the cache, the client sends the identity resolution request to the resolution node V i .

[0043] Furthermore, the root node is collaboratively maintained by multiple servers; wherein the multiple servers implement load balancing based on the load balancing strategy of Nginx reverse proxy, and the load balancing strategy based on Nginx reverse proxy includes:

[0044] Add the weight attribute to the reverse proxy address and configure the server weight according to the attribute value of the weight attribute;

[0045] Add the ip_hash attribute to the reverse proxy address and implement client allocation based on the ip_hash attribute;

[0046] The least_conn attribute is added to the reverse proxy address, and the node addressing request is assigned to the server with the least number of current connections.

[0047] A digital identity node parsing device for scholars, the device comprising:

[0048] An identification type acquisition module, used to initiate a node addressing request to a root node based on an identification resolution request initiated by a client, so that the root node identifies a personnel identification type corresponding to the node addressing request;

[0049] An identification resolution module, used to select a resolution node V according to the personnel identification type j , so that the parsing node V j Generate a resolution result of the identity resolution request; wherein the resolution node V j They are used to parse the personnel identification type q respectively;

[0050] The analysis result returning module is used to return the analysis result to the client.

[0051] A digital identity node parsing system for scholars, the system comprising:

[0052] The client is used to generate an identity resolution request;

[0053] Parsing Node V i , used to initiate a node addressing request to the root node based on the identity resolution request initiated by the client; select a resolution node V according to the personnel identification type j , so that the parsing node V j Generate a resolution result of the identity resolution request; wherein the resolution node V i and the parsing node V j Respectively used to parse personnel identification type p and personnel identification type q; and return the parsing results to the client;

[0054] The root node is used to identify the personnel identification type corresponding to the node addressing request.

[0055] Furthermore, the system further includes: an API gateway, wherein the API gateway is used to:

[0056] Resolve node V based on the information in the node addressing request i Perform identity verification;

[0057] In the parsing node V iIf the identity authentication is passed, the API gateway forwards the node addressing request to the root node.

[0058] Compared with the prior art, the present invention has at least the following beneficial effects.

[0059] 1) The device can break down the barriers between multiple heterogeneous personnel identification systems in the academic field and realize the interconnection and interoperability of cross-platform data. This will greatly facilitate the activities of scholars on multiple academic platforms and improve the efficiency and convenience of academic exchanges.

[0060] 2) Provide a unified solution for cross-industry and cross-field personnel identification resolution applications, and reduce the complexity and coupling between the system and other academic platforms;

[0061] 3) The device’s built-in heterogeneous recognition engine and efficient parsing algorithm can significantly improve data processing speed and accuracy, reduce manual intervention costs, and improve overall operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a system architecture diagram of the present invention.

[0063] Figure 2 It is a flowchart of the digital identity node resolution method for scholars. DETAILED DESCRIPTION

[0064] The present invention is further described in detail below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention but not to limit the scope of the present invention.

[0065] The present invention aims to build an efficient and scalable scholar identity resolution system, which can realize the automatic recognition and resolution of multiple heterogeneous personnel identifications through a two-level node (root node and child node) architecture. The system aims to reduce the direct dependence on the data of each platform and improve the flexibility and low coupling of the system to other academic platforms.

[0066] The system architecture of the present invention is as follows Figure 1 As shown, it includes the following aspects.

[0067] 1. Root node: As the core control and management center of the system, it is responsible for receiving scholar identity information (such as personnel identifiers, etc.) input by users. The root node has a built-in heterogeneous recognition engine, which can realize the recognition and matching of input information based on the encoding specifications of various types of identifiers. It can identify the types of personnel identifiers that may be contained in the input information, and intelligently select the corresponding child node for the next step of processing based on the recognition results.

[0068] 2. Subnode: Responsible for parsing specific types of personnel identification. Each subnode is an independent processing unit that encapsulates the parsing logic and algorithm for a specific identification type. The subnode receives the parsing request from the user, initiates a parsing node addressing request to the root node, implements identification parsing based on the returned node information, and returns the parsing result to the user.

[0069] 3. Protocol layer: Supports multiple transmission protocols such as HTTP, HTTPS, TCP, and UDP. Supporting multiple transmission protocols ensures the reliability and stability of data transmission, meets the needs of different types of applications, and provides flexibility and scalability.

[0070] 4. API Gateway: Provides authentication and authorization mechanisms to ensure that only legitimate users can access services. Supports multiple client-server communication protocols, such as HTTP / HTTPS, REST, etc., and can convert between these protocols.

[0071] 5. Heterogeneous recognition.

[0072] 1) Build an identification feature library.

[0073] (1) Collect various identification data in advance, extract features, and build an identification feature library. Based on Natural Language Processing (NLP)

[0074] Technology and data mining algorithms are used to perform deep feature extraction on the preprocessed identification data.

[0075] (2) For text-based identification data, analyze its vocabulary composition and count the frequency of occurrence of different words, part-of-speech distribution, and other information. For example, the appearance of certain specific industry terms or product model keywords may indicate the field or type to which the identification belongs. At the same time, consider the syntactic structure of the text, such as sentence length, the use of specific punctuation marks (such as the hyphen separation form that often appears in serial numbers), and other features.

[0076] (3) Starting from the overall structural characteristics of the data, extracting characteristics such as encoding rules, data length, the ratio of numbers to letters,

[0077] Characteristic information such as specific character combination patterns (such as fixed abbreviations at the beginning of certain logos representing specific companies or organizations). For example, ID card numbers have fixed lengths and number combination rules, while the product's Universal Product Code (UPC) has its own unique coding structure and check digit algorithm.

[0078] 2) Model matching and type determination.

[0079] (1) Match the extracted identification data features with the various identification type feature libraries pre-trained and constructed within the large model. The feature library is established by learning and analyzing a large amount of identification data of different types. The following is an example of the large model template:

[0080] I will provide you with a string of identifiers and you will need to identify the identifier type that best matches

[0081] rule:

[0082] Please output in JSON format

[0083] example:

[0084] Input:0142-3923-1994-553

[0085] Output: {{"identifier":"0142-3923-1994-553","type":"ORCID"}

[0086] -----------------------------------------

[0087] Input:{input}

[0088] -----------------------------------------

[0089] Output

[0090] {

[0091] "identifier":"4924-1929-4934-4920",

[0092] "type":"ORCID"

[0093] }

[0094] (2) Using pattern matching algorithms and similarity calculation algorithms, calculate the similarity between the identification data features and the standard feature models in each feature library. For example, algorithms such as cosine similarity and Euclidean distance can be used to measure the similarity between feature vectors.

[0095] (3) Set a similarity threshold. When the similarity between the features of the identification data and a feature library of a certain identification type exceeds the threshold, the identification data is determined to belong to the corresponding identification type. For example, if the similarity between the features of an identification data and the ScopusAuthor ID identification feature library is as high as 0.85 (the threshold is set to 0.8), and the similarity between the features of the identification data and the ResearcherID identification feature library is as high as 0.88 (the threshold is set to 0.8), then the identification is determined to be a ResearcherID identification.

[0096] After the type is determined, the recognition result is output to the parsing node module so that the corresponding parsing process can be carried out according to the identification type, thereby realizing efficient operation and accurate data processing of the entire heterogeneous identification system.

[0097] 6. Security and privacy protection between nodes.

[0098] (1) Data encryption: Encryption technology is used during data transmission and storage to ensure the security of user data.

[0099] (2) Access control: Implement strict access control policies to limit unauthorized access and protect user privacy.

[0100] 7.Performance optimization.

[0101] (1) Cache mechanism: Each time a user performs identity resolution, the cache is pulled first. If the cache does not exist, the resolution logic is executed. After the resolution is completed, the result is stored in the cache. The cache validity period is set to 30 minutes. Caching technology is used to reduce repeated resolution and improve system response speed.

[0102] (2) Load balancing: The present invention combines the deployment server situation and reasonably configures the load balancing strategy based on the Nginx reverse proxy to realize dynamic adjustment of resource allocation and ensure the stable operation of the system in high concurrency scenarios.

[0103] Among them, the reasonable configuration of the balancing strategy based on Nginx reverse proxy is as follows:

[0104] ① Polling, the default strategy.

[0105] ② Weighted round-robin: add the weight attribute to the reverse proxy address and configure the weight according to the attribute value.

[0106] The larger the weight value, the higher the frequency of proxying to this root node server.

[0107] ③IP hash, add the ip_hash attribute in the reverse proxy address and perform hash operation on the client IP.

[0108] The result is then mapped to a server in the root node's server list.

[0109] ④ Least connections: add the least_conn attribute in the reverse proxy address and assign the request to the root node server with the least number of current connections.

[0110] Figure 2 It is a flow chart of a method for resolving digital identity nodes of scholars, and its implementation process includes the following steps.

[0111] 1. The user initiates an identity resolution request to resolution node 1, and the node sends a node addressing request to the root node with the identity requested by the user.

[0112] 2. Root node: As the core control and management center of the system, it is responsible for receiving identifiers entered by users (such as scholar identifiers, etc.).

[0113] 3. The root node has a built-in heterogeneous recognition engine, which can realize the recognition and matching of input information based on the encoding specifications of various types of identification, return the data type, and then identify the type of personnel identification that may be contained in the input information. According to the identified identification type, it will go to the corresponding child node for the next step of processing.

[0114] 4. Parsing node 1 returns the target parsing node information after identifying the root node. Verify whether the current parsing node is directly processed. If it is not the current parsing node, it will initiate a parsing request to the template parsing node again.

[0115] 5. Parsing Node 2: Responsible for parsing specific types of personnel identification. Each parsing sub-node is an independent processing unit that encapsulates the parsing logic and algorithm for a specific identification type. The sub-node receives parsing requests from other parsing nodes, performs parsing operations, and returns the parsing results to Parsing Node 1.

[0116] 6. According to the analysis result returned by the analysis node 2, if the address is a URL, the server redirects it to the analysis target address.

[0117] 7. The user's browser jumps to the resolution page corresponding to the identifier.

[0118] Implementation case: On an academic platform, a user publishes a paper, in which author A fills in ORCID and author B fills in BRID. After the article is published, the author hopes to be able to open the homepage of the corresponding identity through the electronic journal. The academic platform uniformly configures the identity resolution as the resolution address of the secondary node and resolves the identities of authors A and B.

[0119] (1) The user clicks to jump to the electronic journal of the academic platform and configures the secondary node resolution address;

[0120] (2) The secondary node receives the resolution request and first checks the local cache. If it does not exist, it forwards the request to the root node, performs heterogeneous identification, and returns the best matching resolvable node. Based on the return information of the root node, it requests the target resolution node to perform the resolution request again;

[0121] (3) After the analysis is completed, the identification mapping relationship is cached in the local node and the corresponding person's homepage is opened.

[0122] Although the specific embodiments of the present invention are disclosed for the purpose of illustration, the purpose is to help understand the content of the present invention and implement it accordingly, those skilled in the art will understand that various substitutions, changes and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. Therefore, the present invention should not be limited to the content disclosed in the best embodiment, and the scope of the present invention is subject to the scope defined in the claims.

Claims

1. A method for analyzing digital identity nodes of scholars, characterized in that: Applied to a parsing node V i , the method comprising: Based on the identity resolution request initiated by the client, a node addressing request is initiated to the root node, so that the root node identifies the personnel identity type corresponding to the node addressing request; Select a parsing node V according to the personnel identification type j , so that the parsing node V j Generate a resolution result of the identity resolution request; wherein the resolution node V i and the parsing node V j They are used to parse personnel identification type p and personnel identification type q respectively; The parsing result is returned to the client.

2. The method according to claim 1, characterized in that The initiating a node addressing request to the root node includes: Parsing Node V i Send the node addressing request to the API gateway The API gateway resolves the node V based on the information in the node addressing request. i Perform identity verification; In the parsing node V i If the identity authentication is passed, the API gateway forwards the node addressing request to the root node.

3. The method according to claim 1, characterized in that The root node identifies the type of person identification corresponding to the node addressing request, including: Constructing an identification type feature library, wherein the identification type feature library is composed based on a mapping relationship between identification type features and personnel identification types; generating an identification data feature based on the node addressing request; The identification data feature is matched with the identification type feature to obtain the personnel identification type corresponding to the node addressing request.

4. The method according to claim 3, characterized in that The step of constructing the identification type feature library includes: Collect identification data; Extracting identification data features, the identification data features including: at least one of vocabulary occurrence frequency, part-of-speech distribution, syntactic structure, encoding rules, data length, ratio of numbers to letters, and specific character combination patterns; The identification data features are embedded into a prompt template of a large language model, and the identification data features are classified using the large language model.

5. The method according to claim 3, characterized in that: Matching the identification data feature with the identification type feature to obtain the personnel identification type corresponding to the node addressing request includes: Based on a similarity calculation algorithm, a similarity value between the identification data feature and the identification type feature is calculated; wherein the similarity calculation algorithm includes: a cosine similarity algorithm or a Euclidean distance algorithm; Setting a similarity threshold; The similarity value is compared with the similarity threshold to obtain the personnel identification type corresponding to the node addressing request.

6. The method according to any one of claims 1 to 5, characterized in that: Before the identity resolution request initiated by the client, it also includes: The client determines whether there is a resolution result corresponding to the identity resolution request in the cache; If the resolution result corresponding to the identity resolution request exists in the cache, the resolution result is pulled from the cache; If the resolution result corresponding to the identity resolution request does not exist in the cache, the client sends the identity resolution request to the resolution node V i .

7. The method according to any one of claims 1 to 5, characterized in that: The root node is collaboratively maintained by multiple servers; wherein the multiple servers implement load balancing based on the load balancing strategy of Nginx reverse proxy, and the load balancing strategy based on Nginx reverse proxy includes: Add the weight attribute to the reverse proxy address and configure the server weight according to the attribute value of the weight attribute; Add the ip_hash attribute to the reverse proxy address and implement client allocation based on the ip_hash attribute; The least_conn attribute is added to the reverse proxy address, and the node addressing request is assigned to the server with the least number of current connections.

8. A digital identity node analysis device for scholars, characterized in that: The device comprises: An identification type acquisition module, used to initiate a node addressing request to a root node based on an identification resolution request initiated by a client, so that the root node identifies a personnel identification type corresponding to the node addressing request; An identification resolution module, used to select a resolution node V according to the personnel identification type j , so that the parsing node V j Generate a resolution result of the identity resolution request; wherein the resolution node V j They are used to parse the personnel identification type q respectively; The analysis result returning module is used to return the analysis result to the client.

9. A digital identity node resolution system for scholars, characterized in that: The system comprises: The client is used to generate an identity resolution request; Parsing Node V i , used to initiate a node addressing request to the root node based on the identity resolution request initiated by the client; select a resolution node V according to the personnel identification type j , so that the parsing node V j Generate a resolution result of the identity resolution request; wherein the resolution node V i and the parsing node V j Respectively used to parse personnel identification type p and personnel identification type q; and return the parsing results to the client; The root node is used to identify the personnel identification type corresponding to the node addressing request.

10. The system according to claim 9, characterized in that The system further includes: an API gateway, wherein the API gateway is used to: Resolve node V based on the information in the node addressing request i Perform identity verification; In the parsing node V i If the identity authentication is passed, the API gateway forwards the node addressing request to the root node.

Images