File mixed search method of structured label file management system based on object element decomposition

Abstract

The application provides a file mixed search method of a structured label file management system based on object element decomposition, comprising the following steps: step S1, establishing a site, establishing a site classification label tree, and establishing the relationship between sites; step S2, determining a search space, i.e. determining the range of search sites; step S3, searching within the range of the search space, using keywords to simultaneously search for site element nodes and file label configuration names, using the combination of nodes and the presence or absence of file label configuration names to obtain a set of files; step S4, for a specified file, searching and positioning through the file name and / or the label value of the file label configuration, and the search positioning of the specified file is based on the search space determined in step S2 or the file set range locked in step S2 and step S3 or the whole site space of the system. The application is used to solve the problem that one dimension cannot completely express an object, and is suitable for the classification of objects.

Description

Technical Field

[0001] This invention relates to the field of electronic document management, specifically to a hybrid file retrieval method for a structured tag file management system based on object element decomposition. This method is applicable to the structured decomposition of objects into tree-structured tag groups, and the association of nodes of these tree-structured tag groups with files related to the objects, thereby achieving a multi-dimensional retrieval method under the file management approach. Background Technology

[0002] In the field of file management, or any resource management field, after managing various files or resources, effectively locating and retrieving the required files is a crucial task of search, especially when the volume of files reaches a certain level, where the quality of the search strategy becomes paramount. Traditional methods primarily manage files through file directories and filenames, similar to resource managers. File searches are also mainly based on filenames, combined with the directory name, file format, and file retention time. This often results in low efficiency, incomplete file searches, and even files not found. The reason is that traditional file management only describes the filename and directory; the directory only expresses information at one level, and the filename is just a short string with limited information. Furthermore, the lack of clear naming rules often makes it difficult to locate files. Patent CN112488642A discloses a cloud-based file management method based on structured tags and centered on objects. This method structurally decomposes both objects and files, associating some of the structured decomposed content of files with the structured decomposed nodes of objects. The structured content of files other than objects is stored along with the files through file structured tag configuration. This method expresses and manages many knowledge features of the file itself through the configuration of file structured tags and the structuring of objects. Therefore, a large amount of knowledge in the file is stored in a regular manner, and theoretically, the desired file can be located very accurately. As more information about the file itself is managed and stored, it is necessary to design a retrieval method based on the storage method and content of the file's structured content in order to improve the efficiency of file retrieval and location. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by providing a hybrid file retrieval method for a structured tag file management system based on object element decomposition. This method extracts and aggregates files within the structured tag file management system based on object element decomposition in a reasonable and efficient manner.

[0004] First, some concepts described in "A Cloud File Management Method Based on Structured Tags and Centered on Objects (Publication No.: CN112488642A, Publication Date: 2020-11-20)" need to be further refined.

[0005] A structured tag-based file management system based on object element decomposition can achieve functions that existing file management systems cannot. These mainly include: enabling object-based semantic file search and management through multiple tag keyword combinations; accurately locating files by searching filenames and specific tag values ​​in file tag configurations; forming multiple file directories through object tag combinations, thus automatically generating multiple file directories for each object's set of files for file organization. Files are stored and managed using a file + metadata approach. Without a good retrieval path and method, the original functionality of files already managed within the system is difficult to realize, and efficiency is not easily improved.

[0006] A structured tagging file management system based on object element decomposition decomposes objects into elements. Element decomposition involves breaking down objects into multiple dimensions to form tree-structured tag groups. These tree-structured tag groups are orthogonal to each other, representing the multi-dimensional knowledge of the object. These tree-structured tag groups, derived from object element decomposition, are used to associate files. The association principle is as follows: files are associated node by node based on their relationship to the object. During association, child nodes and parent nodes are mutually exclusive, with the later selected node taking precedence. Sibling nodes are not mutually exclusive, and child nodes carry information about their root node, with no other further constraints. Thus, files managed by this system, through the tree-structured tag groups obtained from object element decomposition, and the file management method, imbue each file with a set of tree-structured tags. This set of tree-structured tags originates from the object's element decomposition and represents a portion of the tree-structured tag group representing the object's element decomposition. Element decomposition can be carried out continuously during the process. That is, as the understanding of the object deepens or the dimensions of the content to be included in the management increase, elements can be added at any time without affecting the association of files. Files can be associated with elements, but when it is necessary to delete an element node, all files associated with the element node must be unassociated from it first.

[0007] The objects for which elements are decomposed are also classified and managed using a set of tree-structured label groups. The labels used to manage objects are mainly used for classifying the managed objects. Object classification is based on multiple dimensions, so it is also managed by associating objects with tree-structured labels of multiple dimensions. Therefore, for multi-dimensional classification problems, the object-based element decomposition management method described in "A Cloud-based File Management Method Based on Structured Tags and Centered on Objects (Publication No.: CN112488642A, Publication Date: 2020-11-20)" can be used. The core of element decomposition is the multi-dimensional classification management problem. The same method is used for object classification management and element decomposition. The difference is that object classification is performed at a more macroscopic level, and element decomposition can be viewed as a tree structure with objects as root nodes. The object's classification tag tree group can be viewed as managing objects with another higher-dimensional object as the root node. In this case, the object's element decomposition can be seen as an extension of the tree structure of the higher-dimensional object. It is evident that searching for file-related elements is actually locating the nodes of the tree structure. However, to achieve efficient file retrieval, the search space needs to be preprocessed.

[0008] Files managed by a structured tag-based file management system based on object feature decomposition must be associated with object features and a file tag configuration. The file feature tag tree formed by associating site features and the file tag configuration constitute the file's metadata, which is also used to manage and express the file's own knowledge. Furthermore, the tree structure formed by the object features associated with the file is contained within the object's feature decomposition tree group.

[0009] File tag configuration expresses information about the file itself. Its content can overlap with the element decomposition content of an object. File tag configuration is a customizable file form, set on an object-by-object basis; that is, within an object, a configured file tag contains the tag information for that object. File tag configurations are combined using different types of tag groups, including: ordinary tags, single-value tags, multi-value tags, and enumerated tags. File tag configuration is divided into a general section and a custom section. The general section includes: theme and keywords. All uploaded and managed files have a default general tag configuration, and file tag configurations can be switched.

[0010] File attributes refer to information about the file itself, such as the file's creation time, upload time, or GIS information contained in a photograph. File attributes are only saved in the system and cannot be modified.

[0011] File management is object-centric. For each object, its elements are decomposed into a tree-structured tag group. Files associated with the object are linked using these tag groups. Files managed in this way carry a set of partial tree-structured tags based on the object's element decomposition, forming a new tree-structured tag set. File tags are configured to express the file's own information, forming a set of structured tags. Each tag group is named a tag configuration name, containing tag class, tag name, and tag value. For an object, the set of related elements is defined as a "site," including: the tree-structured tag group of the object's element decomposition, the set of files linked using these tags, the set of file tag configurations managed by the object, the set of file tag configurations associated with the files and their file tag values, and a tag tree formed by the category tags associated with the object. Multiple such sets of objects constitute multiple sites. Sites are independent spaces; the node names of the element decomposition (tree-structured tag group), file tag configuration names, and file tag names and values ​​can be repeated across different sites. Therefore, simple searches yield cumbersome and inefficient results, failing to demonstrate the advantages of a structured tag file management system based on element decomposition. Differentiating sites' definitions of search spaces, access control, and other functional extensions are crucial steps. Sites need to be named, and these names should be tailored to the characteristics of the objects being searched.

[0012] According to the method described in "A Cloud-Based File Management Method Based on Structured Tags and Centered on Objects (Publication No.: CN112488642A, Publication Date: 2020-11-20)," a collection of files is obtained by jointly searching tag names using multiple keywords, with the input keywords as the subject. This achieves object-based semantic search and management of files, thereby realizing the goal of multi-dimensional classification and management of files. It combines tags decomposed from the elements of multiple objects to obtain a collection of files. The concept of a site was previously defined; a site is a collection of files and data. The information contained in the files can be divided into two parts: one part is the element association of the file within the site, and the other part is the information within the site itself. Feature association is essentially a structured tag tree group formed by decomposing the features of an object. After association, a file has a tree-structured tag tree group based on feature nodes within a site, with a root node, which is the site name. The collection of sites can be accessed through higher-dimensional root nodes. Therefore, the feature decomposition tags of a file are actually a part of the nodes in a complete tree structure under the same root node. First, a portion of the tree structure is filtered out by site, leaving the remaining nodes within the site space. Cross-site file retrieval can be viewed as a retrieval with some of the feature tree constraints removed. This forms the file retrieval space, which can be divided into the system-wide retrieval space, the cross-site retrieval space, and the intra-site retrieval space.

[0013] The typical file browsing path involves selecting a site, then selecting site feature nodes, and finally browsing the files associated with those feature nodes. By default, only files associated with that specific node are viewed. Alternatively, you can view all files associated with the selected node, including its child nodes. This browsing method is similar to opening directories layer by layer in Windows Explorer, except that you can also select to view files in all subdirectories. When browsing feature nodes, you can select and lock a feature node while browsing other feature nodes. Selecting a feature node imposes restrictions on subsequent selected feature nodes, ensuring mutual exclusion within the same root node. When selecting multiple node features, the result is the intersection of all features, including their associated child nodes, or the union of all features, including their associated child nodes. In the intersection case, it represents the set of files under the feature node's restrictions, similar to "and" or "again" logic. In the union case, it represents all files under the node, similar to "or" or "all" logic.

[0014] File tag configuration is used in file retrieval to further filter or locate files within a specific file set. File tag configuration follows the site; within a site, there is a series of file sets. Selecting and combining certain nodes also creates a series of file sets. Based on this, file tag configuration is used to filter specific files within the set. By the file tag configuration name, the set of files associated with that tag configuration can be filtered. The tag name and tag value of the specific tag configured in the file tag configuration are used to further narrow down the scope within the file set until the file set is precisely located. The file tag value within the file tag configuration is used to search and locate specific files. The file tag configuration value is a structured description of the file content. Searching and locating by file tag value can be based on the search space under the combination of element nodes, on the site space, on the cross-site space, or on the entire system space.

[0015] Based on the above description, assuming that the files have been organized and uploaded to the system using the method described in "A Cloud File Management Method Based on Structured Tags and Centered on Objects (Publication No.: CN112488642A, Publication Date: 2020-11-20)," this invention is achieved through the following technical solution:

[0016] A hybrid file retrieval method for a structured tag file management system based on object feature decomposition, characterized by comprising:

[0017] Step S1: Create a site, create a site category tag tree, and establish the relationships between sites;

[0018] The objects for element decomposition are also classified and managed using a set of tree-structured label groups. The labels used to manage the objects are mainly used for the classification of the managed objects. The object classification is managed from multiple dimensions and also uses a multi-dimensional tree-structured label to associate with the objects.

[0019] Step S2: Determine the search space, that is, determine the scope of search sites;

[0020] Determining the search space, i.e., defining the scope of search sites, uses a site classification tree structure tag tree and a site object element decomposition tree structure tag tree group to express site characteristics. The determination of the search space requires the assistance of site classification tags and site element decomposition content. By selecting site classification tags, a set of sites can be obtained. The logic of obtaining a set of sites using site classification tags is that the target tree structure tag composed of the selected classification tags is ⊆ the site's classification tag, then the site is added to this selected set. Different selections of classification tags will result in different set of sites.

[0021] Step S3: Within the search space, using keywords, simultaneously search for: site element nodes and file tag configuration names. By combining node combinations and checking for the presence or absence of file tag configuration names, a collection of files can be obtained. For searching file tag configurations, the file tag configuration name is only one method; it can be combined with the content of the file tag configuration, i.e., the set file tag names, to confirm the retrieved file tag configuration. The file tag configuration is essentially a form. For the file collection, filtering can be performed by checking if it is associated with this form.

[0022] Step S4: For the specified file to be retrieved, the search and location are performed by the file name and / or the tag value configured by the file tag. The retrieval and location of the specified file is based on the search space determined in step S2, or on the range of the file set locked in steps S2 and S3, or on the entire site space of the system.

[0023] The hybrid file retrieval method of the structured tag file management system based on object element decomposition, established through the above steps, demonstrates the convenience of file retrieval under tag-based file management. It categorizes the metadata of files within the structured tag file management system and achieves effective organization of file retrieval through switching and combining search spaces and search areas. The ability to retrieve a collection of files through keyword combinations is a function unavailable in other file management systems. The retrieval method is integrated with system topic browsing, while also providing precise file location, covering almost all needs for proactive file retrieval.

[0024] While adopting the above technical solutions, the present invention may also adopt or combine the following technical solutions:

[0025] As a preferred technical solution of the present invention, in step S1, the set of object-related elements is a site, including: a tree-structured tag tree group of object element decomposition, a set of files associated with these tags, a set of file tag configurations managed by objects, a set of file tag configurations associated with files and their file tag values, and a tag tree formed by the classification tags associated with objects.

[0026] As a preferred technical solution of the present invention, in step S1, the objects for which element decomposition is performed are also classified and managed by a set of tree-structured label groups. The labels used to manage the objects are mainly used for the classification of the managed objects. The object classification is managed by associating objects with tree-structured labels from multiple dimensions.

[0027] As a preferred technical solution of the present invention, in step S1, the site features are expressed by a site classification tree structure label tree and a site object element decomposition tree structure label tree group.

[0028] As a preferred technical solution of the present invention, in step S2, the retrieval space is determined, that is, the retrieval site range is determined. The site classification tree structure label tree and the site object element decomposition tree structure label tree group are used to express the site characteristics. The determination of the retrieval space needs to be assisted by the site classification labels and the site element decomposition content.

[0029] As a preferred technical solution of the present invention, in step S2, the retrieval space is determined, that is, the retrieval site range is determined. By selecting site classification tags, a set of sites can be obtained. The logic of obtaining the site set by using site classification tags is that the target tree structure tag composed of the selected classification tags is ⊆ the site's classification tag. Then the site is added to this selected set. Different site sets will be obtained under different classification tag selections.

[0030] As a preferred technical solution of the present invention, in step S2, the search space is determined by searching for specific sites by site name.

[0031] As a preferred technical solution of the present invention, in step S2, the retrieval space is determined, the specific site is determined by decomposing the content of the site elements, the node of the site element decomposition is located by keyword search, and the site is confirmed by the tree structure label of the node.

[0032] In a preferred embodiment of the present invention, in step S2, a search space is determined. The search space can be a single site or multiple sites. Once the sites are determined, the site space is defined, and this determination is used to narrow the search scope. Determining the search space is equivalent to determining the range of search sites. The defined range of search sites yields the element decomposition of the corresponding sites, and this element decomposition is used to retrieve files within the search space.

[0033] As a preferred technical solution of the present invention, in step S3, within the search space, keywords are used to simultaneously search for: site element nodes and file tag configuration names. By combining the nodes and the presence or absence of file tag configuration names, a set of files is obtained.

[0034] In a preferred embodiment of the present invention, in step S3, within the search space, a set of files is obtained by combining the decomposed nodes of the site elements. Within the search space, i.e., within the selected site range, the tree structure formed by the selected element nodes is used as the target for file retrieval. File retrieval includes the intersection and union of files. For the intersection of files: if the target tree structure ⊆ the tree structure formed by the elements associated with the file, then the file is selected into the file set. For the union of files, files containing the selected tree structure nodes and all their child nodes are given as the selected file set.

[0035] In a preferred embodiment of the present invention, in step S3, within the search space, the content associated with the file includes site element decomposition and file tag configuration, and whether the file tag configuration contains a distinguishable set of files. The file tag configuration name is searched using keywords to filter whether a file set containing that configuration exists.

[0036] As a preferred technical solution of the present invention, in step S3, when searching for file tag configuration, the content of file tag configuration includes file tag name and file tag data type. The file tag configuration is a form. When it is difficult to locate the specific file tag configuration name by keywords, the file tag configuration can be confirmed by the file tags contained in the file tag configuration.

[0037] As a preferred technical solution of the present invention, in step S3, file retrieval is performed by configuring the presence or absence of element nodes and file tags within the site. These can be combined with each other. That is, the site element nodes can be combined first and then the results can be filtered using the presence or absence of file tags. Alternatively, the results can be filtered first using the presence or absence of file tags and then combined using the site element nodes.

[0038] As a preferred technical solution of the present invention, in step S4, for a specific file that needs to be retrieved, the file can be located and searched using its file name keywords and the tag values ​​in the file tag configuration. The file name and the tag values ​​in the file tag configuration are efficient ways to locate the file.

[0039] As a preferred technical solution of the present invention, in step S4, for a specific file that needs to be retrieved, the search can be performed by using its file name keywords and the tag values ​​in the file tag configuration. The search location space can be the entire site space of the system, the search space determined in step S2, or the file set locked in steps S2 and S3.

[0040] As a preferred technical solution of the present invention, in step S4, the file is located and searched by its file name keywords and the tag values ​​in the file tag configuration. This is different from step S2 and step S3, which decompose the tags by the site elements associated with the file. However, the two can be combined arbitrarily to achieve multi-dimensional file retrieval. Both are file metadata, but they are distinguished during retrieval.

[0041] As a preferred technical solution of the present invention, the structured tag file management system based on object element decomposition distinguishes different objects and their element decomposition and associated files by site. The relationship between files is generated through the element decomposition of objects. When browsing site elements, associated files can be retrieved at the same time. The combination of multiple element nodes and the combination of parent and child node relationships can achieve a certain semantic search effect. At the same time, the file's own information can be used to accurately locate the file.

[0042] As a preferred technical solution of the present invention, the content structure between sites and between files is consistent. Limiting the search scope and further limiting the site elements is to gradually narrow the search space in accordance with the actual needs of browsing and searching in complex systems. The consistent structure between sites and files facilitates the design of unified permission management in the later stage. The combination of site elements realizes the automatic collection of files. It is also possible to further refer to the mutual reference of site element nodes to form a new tree structure, and realize the automatic collection of files based on this tree structure.

[0043] This invention provides a hybrid file retrieval method for a structured tag file management system based on object element decomposition, which has the following advantages and beneficial effects compared with the prior art:

[0044] 1) Group the files that decompose and associate objects and their elements into a site. A site is a relatively virtual scope. Files and element node names may be duplicated between different sites. By distinguishing them by site and using the site as the unit of search scope during retrieval, the search scope can be narrowed. The structure between sites is consistent and the same search strategy can be used, which is convenient for operation.

[0045] 2) Defining the search space defines the search scope. A unified data structure within the search scope ensures efficient switching between search spaces. File sets are dynamically presented when browsing site nodes. File set retrieval under multi-element node combinations is flexible and efficient. For massive amounts of files, the system stores the files themselves and their associated metadata. Metadata provides a precise representation of the files, and retrieval involves filtering and combining data within the metadata space—a method of metadata processing. Combined with site management, this facilitates the management and retrieval of project files.

[0046] 3) It can be used for precise location of specific files, including the retrieval of part of the structured content of a file. It has a good effect on precise file location in the management of massive files. For the retrieval of file tag configuration values, it can be used for precise file location. It can replace some functions in the advanced search of the document system. The structured tag file management system based on object element decomposition and its search method manage more and wider file types and file ranges, and the system adaptability is better.

[0047] 4) With fewer keywords, users can obtain a collection of files, locate specific files, browse the decomposition of object elements on the site, and configure file tags to identify the characteristics of the files themselves. This method is more conducive to users' exploration of a system. That is, less input can yield rich feedback, making knowledge acquisition more convenient and efficient, and enhancing interactivity.

[0048] 5) A structured tag file management system based on object element decomposition, which forms multiple tree-structured tag groups based on object element decomposition, is a better means to solve the classification problem of files and other objects. It is used to solve the problem that a single dimension, i.e., a single tree structure, cannot fully express an object. It is suitable for object classification. The technical solution proposed in this paper is a superior solution for managing objects with a tree structure decomposition in multiple dimensions. In the structured tag file management system based on object element decomposition, the classification of files and sites uses the same method. This technical solution has universality. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the processing flow of the hybrid file retrieval method of the structured tag file management system based on object element decomposition provided in an embodiment of the present invention.

[0050] Figure 2 The new tree structure formed after decomposing the elements of the file association object into tree nodes is illustrated in the diagram of file metadata storage.

[0051] Figure 3 This is a diagram illustrating the site's category tags.

[0052] Figure 4 This is a schematic diagram showing the classification nodes and element decomposition of sites associated with a certain survey and design project.

[0053] Figure 5a This is an example of file association management in a structured tag file management system based on object element decomposition.

[0054] Figure 5b This is a tree structure label diagram of the element decomposition of the site element nodes associated with the document.

[0055] Figure 6 This is a schematic diagram of the main interface for browsing the website.

[0056] Figure 7 A diagram showing the display of a file set selected for multiple nodes. Detailed Implementation

[0057] To enable those skilled in the art to better understand the technical solutions of the present invention, the preferred embodiments of the present invention are described below in conjunction with specific examples. However, it should be understood that the accompanying drawings are for illustrative purposes only and should not be construed as limiting the present patent.

[0058] See attached document Figures 1-7 The embodiments of the present invention provide a hybrid file retrieval method for a structured tag file management system based on object element decomposition, comprising the following steps:

[0059] Step S1: Create a site and build a site category tag tree.

[0060] Step S2: Determine the search space, that is, determine the scope of search sites.

[0061] Step S3: Within the search space, use keywords to simultaneously search for: site element nodes and file tag configuration names. By combining the nodes and checking for the presence or absence of file tag configuration names, a set of files can be obtained.

[0062] Step S4: For the specified file, simultaneous search and location are performed using the filename and the tag values ​​configured in the file tag configuration. The search and location of the specified file can be based on the search scope and file set defined in steps S2 and S3, or it can be based on the entire site space of the system. This enables effective retrieval of associated files in the structured tag file management system based on object element decomposition.

[0063] First, let's explain the website and its category tag tree, such as... Figure 3 As shown, a tree-structured tag system is established for classifying sites. Site classification is primarily based on the type of object decomposition, with different dimensions used to create a set of tree-structured tags for managing site types. For example, based on possible engineering types, sites can be categorized as water conservancy, construction, and new energy. Water conservancy can be further divided into hydropower, water conservancy, and shipping, while new energy can be further divided into photovoltaic and wind power. Alternatively, they can be categorized as knowledge-based, operational, and lifestyle-based. Knowledge-based sites can be further divided into standards, cases, and data; operational sites into pre-operation, product promotion, and operation management; and lifestyle-based sites into project activities and corporate activities. Based on size, they can be categorized as large, medium, and small. This results in a set of mutually orthogonal tree structures used to manage object classification. This is similar to object element decomposition, where the object element decomposition is centered on the object itself, while the object classification tag tree group uses the object as the management content, which can be considered a higher-level element decomposition.

[0064] For object categorization management, objects are associated with corresponding object category tag nodes. The association principle is that parent and child nodes are mutually exclusive. Specifically, for example... Figure 3 The object classification uses a tree-structured tag group. When an object is associated with a tree-structured tag, if it is first associated with "Operation Management" and then with "Product Promotion," both tags are associated. If it is first associated with "Operation Management" and then with "Operations Category," only "Operations Category" is associated. If it is associated with "Operations Category" and then with "Operation Management," only "Operation Management" is associated. Of course, the complete description of this node should actually be "Operations Category - Operation Management." If it is associated with both "Operation Management" and "Product Promotion," and then with "Operations Category," only "Operations Category" is associated. The association principle is consistent with the node decomposition of file association elements, that is, the same principle of node decomposition of file association object elements described in "A Cloud File Management Method Based on Structured Tags and Centered on Objects (Publication No.: CN112488642A, Publication Date: 2020-11-20)."

[0065] An object's elements are decomposed into knowledge representations of the object itself. These elements are further decomposed into a tree-structured tagging system, and the files associated with the nodes of this decomposed element system are extensions of the object's knowledge. For example... Figure 4 As shown, the main content of an object consists of three parts: the object's category tags, the object's element decomposition, and the file space associated with the object's element decomposition nodes. The files associated with the object's element decomposition nodes are dynamic annotations of object knowledge, while the object's element decomposition and category tags are relatively static annotations of the object. This part is used to retrieve and locate specific objects.

[0066] like Figure 4As shown, this represents the element decomposition of an object, which is a set of structured tree-structured tags used to associate files associated with the object and manage files accordingly. The space comprised of the object's element decomposition, associated files, and the file tag configurations within the files themselves constitutes a site. A site is essentially a collection of files and their structured tags managed based on the object. In this system, the object, element decomposition, and structured file tag configurations are metadata added to the files for management purposes. They are stored and managed in a file + metadata manner. A file may appear in multiple sites; therefore, a site is a unique form within this system, a virtual space containing a set of tree-structured tags, a series of files, and their corresponding structured file tag configurations. The stored content essentially represents the relationships and expressions between files. Therefore, a site is object element decomposition + category tags + associated files + file tag configurations.

[0067] For files associated with and managed in this system, such as Figures 5a-5b The data shown primarily includes one or more tree-structured tags composed of nodes representing the elements of the associated objects, and a collection of file tag configurations for the file's own content. The tags in the file tag configuration are those described in "A Cloud-Based File Management Method Based on Structured Tags and Centered on Objects (Publication No.: CN112488642A, Publication Date: 2020-11-20)," which are composed of type tags and used to express specific information about the file itself. Here, a data structure is used to display the node ID. The node ID specifies the uniqueness of the node, while the node name is only for convenient naming and understanding in business contexts. For a file, this means storing such a node ID. The site itself is a virtual structure. A file contains multiple tree-structured tag groups. These tags originate from the decomposition of object features, meaning they contain information about the object. Therefore, by examining the associated tag nodes, file data can reveal related information about the object, such as a category tag tree. The file's metadata also hides some object information, such as site information. Thus, the tree-structured tags in the file can be viewed as sibling nodes under the root node of a tree-structured tag. The decomposition nodes associated with the file contain a significant amount of information. Therefore, a retrieval hierarchy is needed to narrow down the file search scope. The first step is to define the site space... Figures 5a-5b The example shown demonstrates a narrowing of the file search scope, limiting the search space to a specific set of files. In other words, it filters out a portion of the files' feature decomposition node label trees, thus restricting the scope of the next search step. Furthermore, the intersection of files selected through multiple nodes involves a comparison of their tree structures, which will be explained in the specific search methods below.

[0068] Let's look at the main file retrieval and browsing methods of this system:

[0069] (1) Site-based browsing exploration.

[0070] A site is a unit for managing files within a system. While a site itself is a virtual abstraction, it represents a set of files and related knowledge. Site-based browsing and exploration is used when one is unfamiliar with the entire system and its managed files, or when there is no specific target. Before browsing the site's internal content, one can identify it by its name and category tags. By switching between these tags, one can view the collection of sites and gain a general understanding of the site's structure. Once inside a site, its elements and files are displayed. Figure 6 As shown, this is the main interface after entering the site. The left side shows the decomposition of the site's elements, which is a set of tree-structured labels. The right side shows the file collection, which changes as the node on the left is selected.

[0071] When browsing the site, the list of file collections on the right switches according to the node selection on the left. Normally, selecting a feature decomposition node displays all files associated with that node, excluding files associated with its child nodes. In this case, the node being browsed is the target node. The search matches the last node of the file's associated nodes with that node; that is, the file is displayed based on the node it's associated with. This is the conventional browsing and searching method, similar to finding files in a folder—you see the files wherever you put them.

[0072] When browsing site feature nodes, if multiple nodes are selected, the selected feature nodes can be arranged into a tree-like label structure, such as... Figure 7 As shown, after selecting a node, a target tree structure is generated. This tree structure is compared with the tree structure formed by nodes in the file. If the target tree structure is contained within the tree structure formed by the nodes associated with the file, then the file is selected. For example, in a complex site, selecting "Contact Form" and "Water Diversion Specialty" represents all contact forms for "Water Diversion Specialty." These files may be associated with a variety of nodes, but they all share a common feature: they contain the target tree structure. Here, "containment" means that the child node wraps the parent node. Site-based browsing exploration involves the continuous combination of site element nodes, resulting in corresponding changes in the set of files. This leads to a collection of files based on site node browsing. These files are labeled based on site elements and are integrated with the entire virtual space of the site, which is a key feature of this system.

[0073] (2) Fuzzy exploration based on keywords

[0074] Based on keyword-driven fuzzy search, as mentioned earlier, in this system, when the search space (i.e., the site scope) is not limited, the searchable content mainly falls into two categories: the site itself and files. There is overlap between sites and files, namely the site's element nodes. The search content includes: from a site perspective, site category tags, site name, and site element node name; from a file perspective, the site element node name associated with the file, file name, file tag configuration name, tag name in the file tag configuration, and tag value in the file tag configuration.

[0075] Without limiting the search space, a system-wide search using keywords requires organizing the following: site category tags, site nodes, filenames, and file tag configurations. This results in numerous and difficult-to-organize results, explaining why limiting the search space is necessary first. File tag configurations, as expressions of file characteristics, should be used when searching for specific files. Therefore, the primary search targets are sites, nodes, and files. As mentioned earlier, nodes overlap with sites and files. For keyword searches, the search can be limited to searching only sites, only nodes, only files, only files, and also searching for file tag configurations.

[0076] When searching only sites, the search first starts with the site name. For site category tags, the relationship is similar to that of site element decomposition tags for files. You need to locate the specific node in the category tag tree with the keyword, and then find the sites that match the site name and keyword. Clicking on a site category tag will switch the display of related sites based on the selected category tag. Keyword search no longer provides feedback; it becomes a mode for viewing sites by category. Clicking on a specific site takes you to the site space. Keywords are displayed as matching sites based on site names. Selecting a specific site takes you to the site space to further browse the site's element decomposition. If the site is not the target site, you return and search again. You can also further limit the keyword search to only sites, only nodes, or only files.

[0077] Searching for nodes only involves matching keywords across all feature decomposition nodes across all sites. To understand the hierarchical relationship of a node, you need to pinpoint its specific tag name within the feature decomposition tree structure. Without specifying a site, the search is cross-site. Once a target node is identified, clicking it will take you to a node browsing page within that site, essentially entering a site with nodes consistent with the browsing site. You can then further search for files and other nodes, but this search is limited to the site space. Alternatively, you can use keywords to search for files, which will then be limited to the file set appearing under the selected node. After finding a node using the node-only search, confirming entry or selecting the node will take you to the site; exiting the site allows you to search again.

[0078] If you only search for files, you can search by filename using keywords, or further refine your file search by configuring file tags.

[0079] (3) Node search mode for known sites and target files

[0080] Given a site and target file node search, based on some understandings of the system, such as the intersection of files under the selection of multiple feature nodes (representing "that is," "have," or "simultaneously"), and child nodes under a parent node representing the "child node names" of the parent node "under" or "within" the parent node; the union of files under the selection of multiple feature nodes represents "all" files under multiple nodes; the presence or absence of file tag configurations can be understood as files of a specific "type" in the file set, the combination of multiple file tag configurations is "OR," and the absence of file tag configurations is "NOT." Therefore, based on this syntax, the corresponding set of files in the system can be realized.

[0081] For example, under a specific site, searching for all "Work Contact Forms" under "Water Diversion Specialty" and "Southern Water Transmission Line," as shown in the image, clicking the corresponding node will display a series of files. This series of files represents all "Work Contact Forms" for "Water Diversion Specialty" and "Southern Water Transmission Line." Deselecting "Southern Water Transmission Line" will then display all "Work Contact Forms" for "Water Diversion Specialty," increasing the file set size. The default is file intersection; switching to file union will display all files associated with the "Water Diversion Specialty" and "Work Contact Form" nodes, including their child nodes. The file set will then contain all files related to "Water Diversion Specialty" and "Work Contact Forms." Essentially, files associated with the "Work Contact Form" node are mostly document-type files. However, due to file characteristics, different forms (file tag configurations) are used to represent different file types. For example, "Work Contact Forms" are further divided into "Owner Work Contact Forms" and "Supervision Work Contact Forms." File tag configurations can be used to filter files. As mentioned earlier, file metadata consists of two parts: one is the site's feature node association information, and the other is the expression and storage of the file's own structural content, namely file tag configurations. A file is currently associated with one file tag configuration in the system. The relationship between file tag configurations and files is relatively simple, unlike the tree-like structure of feature nodes; it's simply a matter of presence or absence. Furthermore, selecting a file tag configuration allows for adding file set filtering conditions, further filtering within the already selected file set. Adding a file tag configuration as a filtering condition filters for files containing that specific file tag configuration. Multiple file tag configurations are filtered using an "OR" condition, which selects files with the target file tag configuration from the file set. A "NOT" condition can also be used, selecting files other than those specified by a particular file tag configuration. Therefore, after selecting the "Water Diversion Specialty" and "Work Contact Form" element nodes as described above, and then selecting the "Owner's Work Contact Form" file tag configuration, it means that the "Owner's Work Contact Form" is one of the "Work Contact Forms" under "Water Diversion Specialty". If "Non-"Supervision Work Contact Form" is selected, it means all "Work Contact Forms" under "Water Diversion Specialty" except for "Supervision Work Contact Forms". Of course, "Owner" and "Supervisor" may exist in the organizational structure of the site element decomposition. You can also select the corresponding "Owner" element node to get the file set of "Work Contact Forms" related to "Water Diversion Specialty" sent or received by "Owner". It can be seen that the combination of node elements and file tag configuration can create subtle semantics through various combinations and obtain corresponding file results, which is convenient for use in various search scenarios. In other words, it can achieve the purpose of searching for files in multiple ways, which is flexible and variable, and the method is not unique, but it is easy to obtain.The selection of file feature configurations and file nodes will change the file set accordingly. File feature configurations and feature nodes are spatially independent and therefore interact and combine without any specific order; only the final selection state matters. This will not be elaborated further here. Of course, when locating file label configurations, the file label configuration name alone may not be sufficient to express or distinguish them. It is necessary to combine the file label settings within the file label configuration to confirm the file label configuration.

[0082] (4) Precise file search

[0083] Precise file searching occurs when there's a clear target and a need to locate a specific file. Typical file management systems primarily express a file's content through its filename. Document systems, on the other hand, can utilize keywords and abstract information. This information is structured data, a general-purpose table. In this system, this content is represented by file tag configurations and their values. As mentioned earlier, file metadata includes associated site element nodes and file tag configuration values. Precise file searching can be achieved by combining filenames, site element nodes, and file tag configuration values. For example, entering keywords could be a number or a string. Numbers can be searched within the file tag configuration value space, while strings can be used to search for filenames or site nodes.

[0084] The above are the main browsing and retrieval methods, all of which involve four steps. More retrieval strategies can be combined as needed, making this a relatively universal retrieval method for documents based on element decomposition and management. Furthermore, multiple organizational spaces and limited search scopes can be switched, allowing for more combined interactions with fewer keywords, thus improving system interactivity. It also enables the expression and organization of data managed in complex systems in a clear and concise manner, thereby improving management and usage efficiency.

[0085] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications or equivalent changes made to the above embodiments based on the technical essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A hybrid file retrieval method for a structured tag file management system based on object element decomposition, characterized in that: The method includes the following steps: Step S1: Create a site, create a site category tag tree, and establish the relationships between sites; The objects for element decomposition are classified and managed by a set of tree-structured label groups. The labels used to manage the objects are mainly used for the classification of the managed objects. The object classification is managed by associating objects with tree-structured labels from multiple dimensions. Step S2: Determine the search space, that is, determine the scope of search sites; Determining the search space, i.e., defining the scope of search sites, uses site classification tree structure label trees and site object element decomposition tree structure label tree groups to express site characteristics. The determination of the search space requires the assistance of site classification labels and site element decomposition content. By selecting site classification labels, a set of sites is obtained. The logic is that the target tree structure label composed of the selected classification labels is ⊆ the site's classification label, then the site is added to this selected set. Different selections of classification labels will result in different site sets. Step S3: Within the search space, use keywords to simultaneously search for: site element nodes and file tag configuration names. By combining the nodes and checking for the presence or absence of file tag configuration names, a set of files can be obtained. Among these, for the search of file tag configurations, the file tag configuration name is just one method. Combining the content of the file tag configuration, that is, the set file tag name, confirms the file tag configuration to be searched. The file tag configuration is a form. Step S4: For the specified file to be searched, the search location is performed by the file name and / or the tag value configured by the file tag. The search location of the specified file is based on the search space determined in step S2, or on the range of the file set locked in steps S2 and S3, or on the entire site space of the system. In step S2, the search space is determined, the specific site is determined by decomposing the content of the site elements, the node of the site element decomposition is located by keyword search, and the site is confirmed by the tree structure label of the node. In step S2, the search space is determined. After determining the sites, the site space is determined. Determining the site space is used to narrow the search scope. Determining the search space means determining the search site range. The determined search site range is the element decomposition of the corresponding site. The element decomposition is used for the files within the search space. In step S3, within the search space, a set of files is obtained by combining the decomposition nodes of the site elements. Within the search space, i.e. the selected site range, the tree structure formed by the selected element nodes is used as the target to retrieve the files. The file retrieval includes the intersection and union of the files. When files intersect: if the target tree structure ⊆ the tree structure formed by the elements associated with the file, then the file is selected into the file set; When selecting the union of files, the files of the selected tree structure nodes and all their child nodes will be included in the selected file set.

2. The file hybrid retrieval method of the structured tag file management system based on object element decomposition according to claim 1, characterized in that: In step S3, within the search space, the content associated with the file includes site element decomposition and file tag configuration, and whether the file tag configuration is a distinguishable file set; by searching for the file tag configuration name by keyword, it is possible to filter out whether there is a file set with that file configuration.

3. The file hybrid retrieval method of the structured tag file management system based on object element decomposition according to claim 1, characterized in that: In step S3, when searching for file tag configurations, the content of the file tag configuration includes the file tag name and the file tag data type. The file tag configuration is a form. When it is difficult to locate the specific file tag configuration name by keywords, the file tag configuration is confirmed by the file tags contained in the file tag configuration.

4. The file hybrid retrieval method of the structured tag file management system based on object element decomposition according to claim 1, characterized in that: In step S3, file retrieval is performed by configuring the presence or absence of feature nodes and file tags within the site. These are combined, either by first combining site feature nodes and then filtering the results using the presence or absence of file tags, or by first filtering using the presence or absence of file tags and then combining the results using site feature node combinations.

5. The file hybrid retrieval method of the structured tag file management system based on object element decomposition according to claim 1, characterized in that: In step S4, for a specific file that needs to be retrieved, the file is located and searched using its filename keywords and the tag values ​​in the file tag configuration. The filename and the tag values ​​in the file tag configuration are efficient ways to locate the file.

6. The file hybrid retrieval method of the structured tag file management system based on object element decomposition according to claim 1, characterized in that: In step S4, the file is located and searched by its file name keywords and the tag values ​​in the file tag configuration. This is different from step S2 and step S3, which decompose the tags by the site elements associated with the file. However, the two can be combined arbitrarily to achieve multi-dimensional file retrieval. Both are file metadata, but they are distinguished during the retrieval.

Images