Archives automatic management and control method and system based on archives robot

Abstract

This invention relates to the field of archival management technology, specifically to an automated archival management method and system based on an archival robot. The method includes: the archival robot receiving management instructions from a management terminal; semantically parsing the content of the management instructions to obtain target feature data; searching a distributed index based on the target feature data to obtain a target archival identifier code matching the target feature data; the archival robot retrieving the metadata feature set of the corresponding target entity archival from the archival storage backend according to the target archival identifier code; and matching the metadata feature set with several historical management feature template libraries to obtain a target management strategy. This invention achieves automated archival management through an archival robot, reducing manual intervention and improving the efficiency and accuracy of archival management. Through semantic parsing of management instructions and matching of historical management features, the archival robot can intelligently select a suitable management strategy.

Description

Technical Field

[0001] This invention relates to the field of archives management technology, specifically to an automatic archives management method and system based on an archives robot. Background Technology

[0002] Currently, traditional methods often rely on manual classification, organization, and management of archives, resulting in significant human resource consumption. They are also prone to human error, affecting the accuracy and consistency of the archives. Furthermore, the process of manually managing archives is cumbersome and time-consuming, especially when dealing with a large number of documents. The time required for searching and retrieving information is relatively long, which cannot meet the need for rapid information acquisition. In addition, traditional methods lack intelligent support, making it difficult to automatically adjust management strategies based on historical data and actual conditions. This leads to decision-making processes relying on personal experience, which may not be accurate enough.

[0003] Furthermore, due to the lack of classified management and corresponding control strategies for archives, the processing of sensitive information may not be standardized, increasing the risk of information leakage and misuse, making it difficult to meet the requirements of relevant laws and regulations. Moreover, traditional management methods often adopt isolated data storage methods, resulting in information dispersion and a lack of effective integration and sharing mechanisms, which limits the full utilization of archive data. In addition, in traditional methods, retrieval is usually based on simple keyword searches, lacking multi-dimensional indexes and intelligent matching mechanisms, resulting in low accuracy and efficiency in finding information. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides the following technical solution: an automatic archive management method based on an archive robot, applied to an archive management system, the archive management system including a server, a management terminal, and an archive robot, the server including an archive storage backend and an archive management backend, the method including: The file robot receives control instructions sent by the control terminal; performs semantic parsing on the content of the control instructions to obtain target feature data; and searches in a distributed index database based on the target feature data to obtain a target file identifier code that matches the target feature data. The distributed index library stores at least one feature index corresponding to the archives that are forwarded by the archive storage backend; the value of at least one feature index is the unique identifier of the archive; at least one feature index includes a keyword index, an entity type index, a time index, and a permission index. The archive robot retrieves the metadata feature set of the corresponding target entity archive from the archive storage backend according to the target archive identification code; it matches the metadata feature set with several historical control feature template libraries, and selects the corresponding target control strategy according to the obtained feature matching result set; wherein, different historical control feature template libraries correspond to different archive security levels, and the target control strategy is a hierarchical control strategy constructed by introducing a preset adjustment threshold; Based on the target control strategy, the archive robot determines the matching score between the metadata feature set and the several historical control feature template libraries; and selects the target matching score with the highest score from the matching scores. The file robot determines the target file security level corresponding to the target matching score and classifies the file to be managed into the target file security level; it then calls the file management backend to process the target entity file corresponding to the target file security level.

[0005] Preferably, the server further includes an archive database, and the method further includes: The archive storage backend generates an archive identification code corresponding to the uploaded archive; performs format conversion and metadata extraction processing on the uploaded archive to obtain a standard archive file; the uploaded archive is sent to the archive storage backend after being verified by a trusted terminal. The archive storage backend stores the standard archive file and the archive identification code of the uploaded archive into the archive database; The archive storage backend forwards the uploaded archive and its file identifier to the archive robot via bypass based on the File Transfer Protocol (FTP).

[0006] Preferably, based on the target feature data, a search is performed in a distributed index to obtain a target file identifier code that matches the target feature data, including: Retrieve the target feature index that matches the target feature data from the feature index library; The value corresponding to the target feature index is read as the target file identifier code.

[0007] Preferably, the metadata feature set is matched with several historical control feature template libraries to select the corresponding target control strategy based on the obtained feature matching result set, including: Several metadata features in the metadata feature set are sequentially matched with historical control features of the same type in the several historical control feature template libraries to obtain several target feature matching results; The matching results of the target features are classified based on the file security level to obtain a set of feature matching results. A first target control strategy is assigned to the results of complete matching in the set of feature matching results, and a second target control strategy is assigned to the results of partial matching in the set of feature matching results.

[0008] Preferably, based on the target control strategy, a matching score is determined between the metadata feature set and the plurality of historical control feature template libraries, including: If the target control strategy is the first target control strategy, then the archive robot determines the number of feature items in the first target historical control feature template library; the first target historical control feature template library is the library currently used for scoring calculation among the plurality of historical control feature template libraries; Calculate the first cumulative value between the preset base score and the preset adjustment threshold, and the second cumulative value between the number of feature items and the preset adjustment threshold; The ratio of the first accumulated value to the second accumulated value is used as the matching degree between the archive metadata feature and the feature item of the first target historical control feature template library; If the target control strategy is the second target control strategy, then the archive robot will use the ratio of the preset adjustment threshold to the second accumulated value as the feature matching degree between the archive metadata feature and the first target historical control feature template library.

[0009] Preferably, the target match score with the highest score is selected from the various match scores, including: Based on the target control strategy and the preset adjustment threshold, the feature item matching degree between each metadata feature in the metadata feature set and the first target historical control feature template library is calculated to obtain several feature item matching degrees; The product of the matching degrees of the aforementioned feature items is used as the matching degree score between the metadata feature set and the first target historical management feature template library.

[0010] Preferably, the method further includes: The file robot obtains the uploaded file and its file identifier code; it then extracts features based on the full text content of the uploaded file to obtain a feature set to be indexed. The file robot establishes at least one feature index for the feature set to be indexed; the value of at least one feature index is the file identifier code of the uploaded file; at least one feature index includes a keyword index, an entity type index, a time index, or a permission index; The archive robot stores at least one of the feature indexes in a distributed index library.

[0011] Preferably, at least one feature index is established for the feature set to be indexed, including: Extract the core keywords from the feature set to be indexed; Identify the entity type, creation time, and security level corresponding to the core keywords; Construct keyword indexes, entity type indexes, time indexes, and permission indexes respectively; the key of the keyword index is the core keyword, and the value of the keyword index is the file identifier code of the uploaded file; the key of the entity type index is the entity type, and the value of the entity type index is the file identifier code of the uploaded file; the key of the time index is the creation time, and the value of the time index is the file identifier code of the uploaded file; the key of the permission index is the confidentiality level, and the value of the permission index is the file identifier code of the uploaded file.

[0012] Preferably, the method further includes: If the target management interface corresponding to the security level of the target file completes the encrypted storage of the file to be managed, then the file robot will update the metadata feature set to the historical management feature template library corresponding to the security level of the target file; If the target management interface cannot identify the file to be managed, the file robot generates a target parsing interface adapted to the file to be managed based on the target management interface, and generates a new historical management feature template library based on the metadata feature set; The archive robot determines the second target historical control feature template library corresponding to the confidentiality level of the target archive; If the archive to be managed is classified into the security level of the target archive, the archive robot determines whether there are any new features in the metadata feature set that are not included in the second target historical management feature template library, so as to obtain the corresponding feature determination result; The archive robot adjusts the preset adjustment threshold based on the feature determination result; If the newly added feature exists, the archive robot adds the newly added feature to the second target historical control feature template library, and lowers the preset adjustment threshold based on the preset adjustment threshold; If the newly added feature does not exist, the archive robot increases the preset adjustment threshold based on the preset adjustment threshold.

[0013] An automated archive management system based on an archive robot, applicable to the aforementioned automated archive management method based on an archive robot, includes: The file retrieval module is configured to receive control instructions sent by the control terminal; perform semantic parsing on the content of the control instructions to obtain target feature data; and perform a retrieval in a distributed index database based on the target feature data to obtain a target file identification code that matches the target feature data. The feature matching module is configured to retrieve the metadata feature set of the corresponding target entity file from the file storage backend based on the target file identification code; match the metadata feature set with several historical control feature template libraries, and select the corresponding target control strategy based on the obtained feature matching result set; wherein, different historical control feature template libraries correspond to different file security levels, and the target control strategy is a hierarchical control strategy constructed by introducing a preset adjustment threshold; The template matching module is configured to determine the matching score between the metadata feature set and the several historical control feature template libraries based on the target control strategy; and to select the target matching score with the highest score from the matching scores. The document management module is configured to determine the security level of the target document corresponding to the target matching score, and classify the document to be managed into the security level of the target document; and call the document management backend to process the target entity document corresponding to the security level of the target document.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This invention automates the management of archives through an archive robot, reducing manual intervention and improving the efficiency and accuracy of archive management. Furthermore, by semantically parsing control instructions and matching historical control features, the archive robot can intelligently select appropriate control strategies, thereby providing more accurate decision support. Moreover, by using a distributed index library and multiple feature indexes for retrieval, the speed and accuracy of archive searches can be greatly improved, enhancing the user experience. This invention enhances the security of archives by introducing classified management and corresponding control strategies, ensuring that sensitive information is properly handled and complies with relevant laws and regulations. Furthermore, by incorporating historical control feature templates into the management process, the system can utilize past data and experience to optimize current archive management strategies, achieving knowledge accumulation and reuse. Additionally, through a matching score mechanism, the system can effectively assess the degree of matching between metadata features and historical templates, thereby selecting the most appropriate control strategy. Attached Figure Description

[0015] Figure 1 This is a schematic flowchart of the overall method in one embodiment of the present invention; Figure 2 This is a schematic diagram of the overall system architecture in one embodiment of the present invention.

[0016] In the diagram: 1. Archive retrieval module; 2. Feature matching module; 3. Template matching module; 4. Archive management module. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1, please refer to Figure 1 This invention provides a technical solution: an automatic archive management method based on an archive robot, applied to archive management. The archive management includes a server, a management terminal, and an archive robot. The server includes an archive storage backend and an archive management backend. The method includes: S1. The archive robot receives control instructions sent by the control terminal; performs semantic parsing on the content of the control instructions to obtain target feature data; and searches in the distributed index database based on the target feature data to obtain the target archive identification code that matches the target feature data. The distributed index database stores at least one feature index corresponding to the archives that are forwarded by the archive storage backend; the value of at least one feature index is the unique identifier of the archive; at least one feature index includes a keyword index, an entity type index, a time index, and a permission index. S2. The archive robot retrieves the metadata feature set of the corresponding target entity archive from the archive storage backend based on the target archive identification code; it matches the metadata feature set with several historical control feature template libraries, and selects the corresponding target control strategy based on the obtained feature matching result set; wherein, different historical control feature template libraries correspond to different archive security levels, and the target control strategy is a hierarchical control strategy constructed by introducing a preset adjustment threshold. S3. Based on the target control strategy, the archive robot determines the matching score between the metadata feature set and several historical control feature template libraries; and selects the target matching score with the highest score from each matching score. S4. The archive robot determines the security level of the target archive corresponding to the target matching score and classifies the archives to be managed into the security level of the target archive; it calls the archive management backend to process the target entity archives corresponding to the security level of the target archive.

[0019] It should be noted that the archive robot first receives instructions from the control terminal, which may involve operations such as querying, classifying, and updating archives; The document robot performs semantic parsing on the control commands to extract key information, such as the type and characteristics of the document to be searched; this step is to convert human language into a format that computers can understand. Based on the parsed target feature data, the archive robot performs a search in a distributed index; this index stores feature information related to archives, including keywords, entity type, time and permissions, with the aim of quickly finding the archive identification code that matches the target features; Once the identifier of the target file is found, the file robot will retrieve the set of metadata features of the file from the file storage backend; these features may include information such as creation date, author, and file type. The archive robot matches the acquired metadata feature set with multiple historical control feature templates; these templates are classified according to different archive security levels (such as confidential, internal, and public); through matching, a set of feature matching results can be obtained to help determine the applicable control strategies; Based on the matching results, the archive robot selects the most suitable target control strategy; this strategy is generated after considering preset adjustment thresholds and is designed to ensure that the archives receive appropriate security control. The document robot scores the matching degree between the metadata feature set and the template; this score reflects the quality of the match; finally, the robot selects the highest score from multiple scores. The document robot determines the document security level corresponding to the target matching score and classifies the documents to be managed into this security level; then, it calls the document management backend to process the documents related to this security level. Example: Suppose a company has an employee file management system; one day, the control terminal issues a command: "Find the files of all employees who joined in 2021 and are in management positions;" The file robot receives this instruction; it parses the instruction and obtains the target feature data, such as "Year of Joining: 2021" and "Position: Management"; the robot searches in the distributed index and finds the file identification code that matches these features, such as "EMP123" and "EMP456"; based on these identification codes, the robot retrieves the corresponding employee file metadata from the file storage backend, such as name, position, and date of joining; The robot matches this metadata with historical control feature templates, which may include "management files" that define specific management requirements. Based on the matching results, the robot selects a suitable control strategy, such as "high-level protection for management files". Each matching result is scored to evaluate which one best meets the prescribed standards. The robot marks these files as "high-level" and notifies the file control backend to take corresponding measures, such as restricting access.

[0020] In an optional embodiment, the server further includes a file database, and the method further includes: The archive storage backend generates an archive identification code corresponding to the uploaded archive; performs format conversion and metadata extraction processing on the uploaded archive to obtain a standard archive file; the uploaded archive is sent to the archive storage backend after being verified as a trusted terminal. The archive storage backend stores standard archive files and the archive identification codes of uploaded archives in the archive database; The archive storage backend forwards the uploaded archive and its file identifier to the archive robot via bypass based on the File Transfer Protocol (FTP).

[0021] It should be noted that when a user uploads a file, the file storage backend will generate a unique file identifier for that file; this identifier is used to identify the file for subsequent retrieval and management. Uploaded files may be in various formats (such as PDF, Word, images, etc.). In order to manage them uniformly, the file storage backend will convert these files into standard formats. In addition, the backend will also extract the file's metadata, such as file name, creation date, author, keywords, etc., for subsequent indexing and retrieval. Before uploading files, the terminal to which the file is being uploaded will be subject to security verification; this process ensures that only terminals that have passed security authentication can upload files, effectively preventing unauthorized access and data leakage. Once the archives have been verified and processed, the archive storage backend will store the standard archive file and the corresponding archive identification code in the archive database; this step ensures that all archives are in a centralized location for easy access and management later. In addition to storing the files in the database, the file storage backend also forwards the uploaded files and their file identification codes to the file robot via File Transfer Protocol (FTP). This allows the file robot to obtain newly uploaded files in real time and perform subsequent management and control operations.

[0022] In an optional embodiment, a search is performed in a distributed index based on the target feature data to obtain a target file identifier code that matches the target feature data, including: Retrieve the target feature index that matches the target feature data from the feature index library; Read the value corresponding to the target feature index as the target file identifier code.

[0023] It should be noted that target feature data refers to the characteristics of the files that the user or the file they wish to find, such as file type, creation date, author, keywords, etc.; this feature data is used to guide the retrieval in the index. The feature index is a database that stores various file features and their corresponding relationships. When target feature data is received, a search is performed in the feature index to find the target feature index that matches the target feature data. The target feature index is an entry that records a specific feature and its associated file identification code. Once a target feature index that matches the target feature data is found, the value corresponding to that index will be read; this value is the file identifier code that matches the target file feature, and this identifier code will be used to access and manage the corresponding file. Example: Suppose a library's records management system requires finding specific book records; The goal is to find all books written by "John Smith" and published in "2020"; therefore, "John Smith" and "2020" are the target feature data. After receiving this target feature data, the search begins in the feature index database; the feature index database may contain multiple entries, each of which records the book's feature information, such as author, publication year, book type, etc. After searching, a feature index entry was found, which showed that the book with "Author: John Smith, Publication Year: 2020" and its corresponding file identification code "BK20201234" was found. Read the value of this feature index "BK20201234" and use it as the target file identifier; this means that the administrator can further access, view or manage the corresponding book file through this identifier.

[0024] In an optional embodiment, a metadata feature set is matched with several historical control feature template libraries to select an appropriate target control strategy based on the obtained feature matching result set, including: Several metadata features in the metadata feature set are matched sequentially with historical control features of the same type in several historical control feature template libraries to obtain several target feature matching results; Based on the archive's security classification, the matching results of several target features are classified to obtain a set of several feature matching results; Assign a first target control strategy to the results that are completely matched in a set of feature matching results, and assign a second target control strategy to the results that are partially matched in a set of feature matching results.

[0025] It should be noted that the metadata feature set contains important information related to the archives, such as the archive's creation date, author, content summary, and archive classification. These features are used to guide the selection of subsequent control strategies. The historical control feature template library is a database that stores past control strategies and their corresponding features. These features typically include different types of archive classification, confidentiality level, and storage location. The features in the metadata feature set (such as the author and classification of the archive) are matched one by one with the features of the same type in the historical control feature template library; this matching process will generate several target feature matching results, which represent the relationship between the current archive features and the historical control features. The matching results are categorized according to the classification level of the file (e.g., public, internal, confidential); for example, if a file is classified as "confidential", then the results that match the characteristics of this file will be classified as confidential. For results that are a complete match in the feature matching result set, a first-target control policy will be assigned; this policy may require stricter access control or encryption measures. For results that are a partial match, a second-target control policy will be assigned, which may mean moderate control, such as periodic review or restriction of certain access permissions. Example: Suppose a company has an internal records management system for managing sensitive financial reports; The metadata characteristics of a certain financial report include: Author: Zhang Wei; Creation Date: March 1, 2026; Summary: 2025 Financial Summary; Classification: Confidential; There is a historical control feature template library, which includes: Feature Template 1: Author: Li Ming, Classification: Confidential; Feature Template 2: Author: Zhang Wei, Classification: Confidential; Feature Template 3: Author: Wang Fang, Classification: Public; The metadata features of the financial report were matched with historical control feature templates; the results are as follows: Feature template 2 was matched (complete match) because the author and security level are the same; Feature templates 1 and 3 were not matched (partial match). Since the matching results show that the characteristics of the financial report fully match the historical control template 2, it was decided to assign it to the first target control strategy; this may mean that the report requires a high level of security measures, such as stricter access control and encryption.

[0026] In an optional embodiment, based on the target control strategy, a matching score is determined between the metadata feature set and several historical control feature template libraries, including: If the target control strategy is the first target control strategy, then the archive robot determines the number of feature items in the first target historical control feature template library; the first target historical control feature template library is the library currently used for scoring calculation among several historical control feature template libraries; Calculate the first cumulative value between the preset base score and the preset adjustment threshold, and the second cumulative value between the number of feature items and the preset adjustment threshold; The ratio of the first accumulated value to the second accumulated value is used as the matching degree between the archive metadata features and the feature items of the first target historical control feature template library; If the target control strategy is the second target control strategy, the archive robot will use the ratio of the preset adjustment threshold to the second accumulated value as the feature matching degree between the archive metadata feature and the feature template library of the first target historical control feature.

[0027] It should be noted that different control strategies are selected as needed; there are generally two strategies: the first-target control strategy, which involves stricter control measures; and the second-target control strategy, which involves relatively lenient control measures. The first-target historical control feature template library is the feature template library used when calculating scores; this library contains multiple historical control feature templates, each of which has several feature items, such as author, security level, type, etc. If the first-target control strategy is currently being used, the number of feature items in the first-target historical control feature template library will be determined; these feature items will be used for score calculation. First accumulated value: Calculation result based on the preset base score and adjustment threshold; Second accumulated value: Calculation result based on the number of feature items and the preset adjustment threshold; Depending on the target control strategy, different methods will be used to calculate the matching degree: for the first target control strategy, the matching degree score will be the ratio of the first accumulated value to the second accumulated value; for the second target control strategy, the matching degree score will be the ratio of the adjustment threshold to the second accumulated value.

[0028] In an optional embodiment, selecting the target match score with the highest score from the match scores includes: Based on the target control strategy and the preset adjustment threshold, the feature matching degree between each metadata feature in the metadata feature set and the first target historical control feature template library is calculated to obtain several feature matching degrees. The product of the matching degrees of several feature items is used as the matching degree score between the metadata feature set and the first target historical control feature template library.

[0029] It should be noted that selecting a specific target control strategy (such as the first target control strategy) will affect the subsequent matching degree calculation; Set an adjustment threshold, which will be used to evaluate the matching degree between each metadata feature and the historical control feature template library; the metadata feature set is the archive metadata features to be evaluated, such as: author; creation date; security level; the first target historical control feature template library contains multiple historical control feature templates, each template also has similar feature items; for example: feature template 1: author: Li Ming, security level: confidential; feature template 2: author: Zhang Wei, security level: confidential; feature template 3: author: Wang Fang, security level: public; For each feature in the metadata feature set, its matching degree with the feature items in the first target historical control feature template library is calculated according to a preset adjustment threshold; this usually includes comparing whether the feature content is consistent or similar. After obtaining the matching scores of all feature items, these scores are multiplied to obtain the final matching score; this score reflects the overall matching degree between the entire metadata feature set and the historical control feature template library.

[0030] In an optional embodiment, the method further includes: The file robot obtains the uploaded file and its file identification code; it extracts features based on the full text content of the uploaded file to obtain a feature set to be indexed. The archive robot establishes at least one feature index for the feature set to be indexed; the value of at least one feature index is the archive identifier code of the uploaded archive; at least one feature index includes a keyword index, entity type index, time index, or permission index; The archive robot stores at least one feature index in a distributed index library.

[0031] It should be noted that when a user uploads a file, the file robot first obtains the file's content and a unique file identifier (ID); this identifier is used to uniquely identify the file for subsequent retrieval and management. The archive robot analyzes the full text of uploaded archives and extracts key features. These features typically include keywords, entities (such as names, locations, and organizations), time information, and permission information. These extracted features form a "feature set to be indexed". The document retrieval robot will create at least one feature index based on the feature set to be indexed; the feature index is a structure designed to improve document retrieval efficiency. The feature index value includes the file identifier code of the uploaded file, so that feature information associated with the identifier code can be quickly found; the feature index may include the following types: Keyword index: records important keywords in the file, which facilitates retrieval based on keywords; Entity type index: classifies and indexes entities in the document (such as company names, place names, etc.); Time index: records date or time information related to the file content; Permission index: marks the access permissions of the file to ensure that only authorized users can access it; The document robot stores the generated feature index in a distributed index; the distributed index supports efficient data storage and fast retrieval, ensuring that the required documents can be found quickly when needed.

[0032] In an optional embodiment, building at least one feature index for the feature set to be indexed includes: Extract the core keywords from the feature set to be indexed; Identify the entity type, creation time, and security level corresponding to the core keywords; Construct keyword indexes, entity type indexes, time indexes, and permission indexes respectively; the key of the keyword index is the core keyword, and the value of the keyword index is the file identifier of the uploaded file; the key of the entity type index is the entity type, and the value of the entity type index is the file identifier of the uploaded file; the key of the time index is the creation time, and the value of the time index is the file identifier of the uploaded file; the key of the permission index is the confidentiality level, and the value of the permission index is the file identifier of the uploaded file.

[0033] It should be noted that the archive robot extracts important core keywords from the set of features to be indexed. These keywords are usually the essence of the archive content and can reflect the main theme and important information of the document. After extracting the core keywords, we will further identify other information corresponding to these keywords, including: Entity type: refers to the specific object related to the keyword, such as a person's name, place, organization, etc.; Creation time: the creation date or time information of the file; Confidentiality level: the confidentiality level of the file, such as "public", "confidential" or "top secret". Based on the extracted core keywords and identified information, four types of indexes are constructed: Keyword Index: using the core keyword as the key and the uploaded file's file identifier as the value; this allows for quick retrieval of relevant files based on keywords; Entity Type Index: using the identified entity type as the key and the file identifier as the value; this index allows for file searching based on specific entity types; Time Index: using the creation time as the key and the file identifier as the value; this allows users to search for files by date; Permission Index: using the security level as the key and the file identifier as the value; users can restrict access to specific files based on the security level.

[0034] In an optional embodiment, the method further includes: If the target management interface corresponding to the security level of the target file completes the encrypted storage of the file to be controlled, the file robot will update the metadata feature set to the historical control feature template library corresponding to the security level of the target file. If the target management interface cannot identify the file to be managed, the file robot generates a target parsing interface that is compatible with the file to be managed based on the target management interface, and generates a new historical management feature template library based on the metadata feature set. The document robot identifies the second target historical control feature template library corresponding to the confidentiality level of the target document; If the archives to be managed are classified into the target archives security level, the archive robot determines whether there are any new features in the metadata feature set that are not included in the second target historical management feature template library, so as to obtain the corresponding feature determination results; The archive robot adjusts the preset adjustment threshold based on the feature determination results; If new features are added, the archive robot will add the new features to the second target historical control feature template library and reduce the preset adjustment threshold based on the preset adjustment threshold. If no new features are added, the archive robot will increase the preset adjustment threshold based on the preset adjustment threshold.

[0035] It should be noted that if the management interface corresponding to the classification level (such as confidential or top secret) of the target file to be managed can successfully complete the encrypted storage of the file, the file robot will update the metadata feature set of the file (including keywords, entity type, etc.) to the historical management feature template library related to the classification level of the file; this update is helpful for subsequent file management and retrieval. If the target management interface cannot identify the file to be managed, the file robot will generate a new target parsing interface adapted to the file based on the existing information; this means that a new processing path will be created for files that cannot be processed directly to ensure that the files can still be effectively managed; at the same time, the file robot will generate a new historical management feature template library based on the metadata feature set to record and manage the features of this type of file. The document robot will then determine a second target historical control feature template library corresponding to the target document's security classification; this library will contain feature information related to the target document's security classification. After classifying the files to be managed into the target file security level, the file robot will check whether there are any new features in the metadata feature set that are not included in the second target historical management feature template library; this step is to ensure that all relevant features are identified and included in management in a timely manner. Based on the results of the feature inspection, the archive robot will adjust the preset adjustment threshold accordingly: if new features are found, the archive robot will add these features to the second target historical control feature template library and lower the preset adjustment threshold; this means that it will be more sensitive when processing future archives and be able to better identify new features; if no new features are found, the archive robot will raise the preset adjustment threshold, thereby making feature recognition more lenient and promoting the efficiency of subsequent archive management.

[0036] Example 2, please refer to Figure 2 This invention provides a technical solution: an automated archive management system based on an archive robot, applicable to the aforementioned automated archive management method based on an archive robot, comprising: The document retrieval module 1 is configured to receive control instructions sent by the control terminal; perform semantic parsing on the content of the control instructions to obtain target feature data; and perform retrieval in a distributed index database based on the target feature data to obtain the target document identification code that matches the target feature data. Feature matching module 2 is configured to retrieve the metadata feature set of the corresponding target entity file from the file storage backend based on the target file identification code; match the metadata feature set with several historical control feature template libraries, and select the corresponding target control strategy based on the obtained feature matching result set; wherein, different historical control feature template libraries correspond to different file security levels, and the target control strategy is a hierarchical control strategy constructed by introducing a preset adjustment threshold; Template matching module 3 is configured to determine the matching degree score between the metadata feature set and several historical control feature template libraries based on the target control strategy; and to select the target matching degree score with the highest score from each matching degree score. The document management module 4 is configured to determine the security level of the target document corresponding to the target matching score, and classify the documents to be managed into the security level of the target document; and call the document management backend to process the target entity documents corresponding to the security level of the target document.

[0037] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A method for automatic management and control of archives based on an archive robot, characterized in that, Applied to an archive management system, the archive management system includes a server, a control terminal, and an archive robot, the server includes an archive storage backend and an archive control backend, and the method includes: The file robot receives control instructions sent by the control terminal; performs semantic parsing on the content of the control instructions to obtain target feature data; and searches in a distributed index database based on the target feature data to obtain a target file identifier code that matches the target feature data. The distributed index library stores at least one feature index corresponding to the archives that are forwarded by the archive storage backend; the value of at least one feature index is the unique identifier of the archive; at least one feature index includes a keyword index, an entity type index, a time index, and a permission index. The archive robot retrieves the metadata feature set of the corresponding target entity archive from the archive storage backend according to the target archive identification code; it matches the metadata feature set with several historical control feature template libraries, and selects the corresponding target control strategy according to the obtained feature matching result set; wherein, different historical control feature template libraries correspond to different archive security levels, and the target control strategy is a hierarchical control strategy constructed by introducing a preset adjustment threshold; Based on the target control strategy, the archive robot determines the matching score between the metadata feature set and the several historical control feature template libraries; and selects the target matching score with the highest score from the matching scores. The file robot determines the target file security level corresponding to the target matching score and classifies the file to be managed into the target file security level; it then calls the file management backend to process the target entity file corresponding to the target file security level.

2. The method for automatic management and control of archives based on an archive robot according to claim 1, characterized in that, The server also includes an archive database, and the method further includes: The archive storage backend generates an archive identification code corresponding to the uploaded archive; performs format conversion and metadata extraction processing on the uploaded archive to obtain a standard archive file; the uploaded archive is sent to the archive storage backend after being verified by a trusted terminal. The archive storage backend stores the standard archive file and the archive identification code of the uploaded archive into the archive database; The archive storage backend forwards the uploaded archive and its file identifier to the archive robot via bypass based on the File Transfer Protocol (FTP).

3. The method for automatic management and control of archives based on an archive robot according to claim 2, characterized in that, Based on the target feature data, a search is performed in a distributed index to obtain a target file identifier code that matches the target feature data, including: Retrieve the target feature index that matches the target feature data from the feature index library; The value corresponding to the target feature index is read as the target file identifier code.

4. The method for automatic management and control of archives based on an archive robot according to claim 3, characterized in that, The metadata feature set is matched with several historical control feature template libraries to select the corresponding target control strategy based on the obtained feature matching result set, including: Several metadata features in the metadata feature set are sequentially matched with historical control features of the same type in the several historical control feature template libraries to obtain several target feature matching results; The matching results of the target features are classified based on the file security level to obtain a set of feature matching results. A first target control strategy is assigned to the results of complete matching in the set of feature matching results, and a second target control strategy is assigned to the results of partial matching in the set of feature matching results.

5. The method for automatic management and control of archives based on an archive robot according to claim 4, characterized in that, Based on the target control strategy, a matching score is determined between the metadata feature set and the several historical control feature template libraries, including: If the target control strategy is the first target control strategy, then the archive robot determines the number of feature items in the first target historical control feature template library; the first target historical control feature template library is the library currently used for scoring calculation among the plurality of historical control feature template libraries; Calculate the first cumulative value between the preset base score and the preset adjustment threshold, and the second cumulative value between the number of feature items and the preset adjustment threshold; The ratio of the first accumulated value to the second accumulated value is used as the matching degree between the archive metadata feature and the feature item of the first target historical control feature template library; If the target control strategy is the second target control strategy, then the archive robot will use the ratio of the preset adjustment threshold to the second accumulated value as the feature matching degree between the archive metadata feature and the first target historical control feature template library.

6. The method for automatic management and control of archives based on an archive robot according to claim 5, characterized in that, The target match score with the highest score is selected from all the match scores, including: Based on the target control strategy and the preset adjustment threshold, the feature item matching degree between each metadata feature in the metadata feature set and the first target historical control feature template library is calculated to obtain several feature item matching degrees; The product of the matching degrees of the aforementioned feature items is used as the matching degree score between the metadata feature set and the first target historical management feature template library.

7. The method for automatic management and control of archives based on an archive robot according to claim 6, characterized in that, The method further includes: The file robot obtains the uploaded file and its file identifier code; it then extracts features based on the full text content of the uploaded file to obtain a feature set to be indexed. The file robot establishes at least one feature index for the feature set to be indexed; the value of at least one feature index is the file identifier code of the uploaded file; at least one feature index includes a keyword index, an entity type index, a time index, or a permission index; The archive robot stores at least one of the feature indexes in a distributed index library.

8. The method for automatic management and control of archives based on an archive robot according to claim 7, characterized in that, Establish at least one feature index for the feature set to be indexed, including: Extract the core keywords from the feature set to be indexed; Identify the entity type, creation time, and security level corresponding to the core keywords; Construct keyword indexes, entity type indexes, time indexes, and permission indexes respectively; the key of the keyword index is the core keyword, and the value of the keyword index is the file identifier code of the uploaded file; the key of the entity type index is the entity type, and the value of the entity type index is the file identifier code of the uploaded file; the key of the time index is the creation time, and the value of the time index is the file identifier code of the uploaded file; the key of the permission index is the confidentiality level, and the value of the permission index is the file identifier code of the uploaded file.

9. The method for automatic management and control of archives based on an archive robot according to claim 8, characterized in that, The method further includes: If the target management interface corresponding to the security level of the target file completes the encrypted storage of the file to be managed, then the file robot will update the metadata feature set to the historical management feature template library corresponding to the security level of the target file; If the target management interface cannot identify the file to be managed, the file robot generates a target parsing interface adapted to the file to be managed based on the target management interface, and generates a new historical management feature template library based on the metadata feature set; The archive robot determines the second target historical control feature template library corresponding to the confidentiality level of the target archive; If the archive to be managed is classified into the security level of the target archive, the archive robot determines whether there are any new features in the metadata feature set that are not included in the second target historical management feature template library, so as to obtain the corresponding feature determination result; The archive robot adjusts the preset adjustment threshold based on the feature determination result; If the newly added feature exists, the archive robot adds the newly added feature to the second target historical control feature template library, and lowers the preset adjustment threshold based on the preset adjustment threshold; If the newly added feature does not exist, the archive robot increases the preset adjustment threshold based on the preset adjustment threshold.

10. An automated archive management system based on an archive robot, applicable to the automated archive management method based on an archive robot as described in any one of claims 1-9, characterized in that, include: The document retrieval module is configured to receive control commands sent by the control terminal; The instruction content of the control command is semantically parsed to obtain target feature data; based on the target feature data, a search is performed in a distributed index to obtain a target file identifier code that matches the target feature data; The feature matching module is configured to retrieve the metadata feature set of the corresponding target entity file from the file storage backend based on the target file identification code; match the metadata feature set with several historical control feature template libraries, and select the corresponding target control strategy based on the obtained feature matching result set; wherein, different historical control feature template libraries correspond to different file security levels, and the target control strategy is a hierarchical control strategy constructed by introducing a preset adjustment threshold; The template matching module is configured to determine the matching score between the metadata feature set and the several historical control feature template libraries based on the target control strategy; and to select the target matching score with the highest score from the matching scores. The document management module is configured to determine the security level of the target document corresponding to the target matching score, and classify the document to be managed into the security level of the target document; and call the document management backend to process the target entity document corresponding to the security level of the target document.

Images