CN117349251A

CN117349251A - Data resource management method, electronic device and storage medium

Info

Publication number: CN117349251A
Application number: CN202311410524.3A
Authority: CN
Inventors: 张光卫; 吴振宇; 郭传洲
Original assignee: Beijing Daxiang Intelligent Technology Co ltd
Current assignee: Beijing Daxiang Intelligent Technology Co ltd
Priority date: 2023-10-27
Filing date: 2023-10-27
Publication date: 2024-01-05

Abstract

The invention discloses a data resource management method, electronic equipment and a storage medium, wherein the data resource management method comprises the following steps: responding to the user to select the node on the semantic knowledge graph to store multi-mode data, and calculating the shortest path from the selected node to the root node, wherein the node of the semantic knowledge graph comprises the root node on logic; determining a semantic path based on the names of the nodes on the shortest path; and storing the multi-mode data based on the semantic path. According to the method, the semantic knowledge graph with the root node is used for carrying out interface interaction with the user, and the data is stored through the mapping relation between the semantic knowledge graph and the actual storage position, so that the user can store the data more quickly and conveniently, the subsequent searching of the data is also more quickly and conveniently, and the use experience of the user in storing and searching the file is greatly improved.

Description

Data resource management method, electronic device and storage medium

Technical Field

The embodiment of the application relates to the technical field of data resource management, in particular to a data resource management method, electronic equipment and a storage medium.

Background

In the related art, data can be generally divided into structured data and unstructured data. Wherein structured data may be displayed in rows, columns, and relational databases, typically including numbers, dates, and strings, representing about 20% of enterprise data, it is more convenient to use legacy solutions to protect and manage data because structured storage generally requires less memory. Unstructured data, which may not be displayed in rows, columns, and relational databases, typically includes pictures, audio, video, word processing documents, mail, data sheets, etc., occupies about 80 a of enterprise data, requires more storage space, and is difficult to protect and manage with legacy solutions. Structured data already has a mature full life cycle processing framework, mature database theory and database storage technology, mature data statistics, analysis and visualization technology stacks, and a systematic data mining theory and algorithm.

In data collection, structured data is related to business, data is collected according to the existing schema, the data enters a database, unstructured data is generated through a sensor, the data volume is large, and the data is stored in a file form. In the data storage, the structured data is a mature relational database, the relation among the data is clear, the unstructured data is usually stored in a file storage system, the files are completely independent, and the relation among the files cannot be expressed. In the aspect of data management, structured data can be easily managed by formulating data standards, data dictionaries, metadata, data models and the like, and unstructured data has no unified management standard. In analysis and mining technologies, structured data has mature theoretical technical support, such as probability statistics, classification algorithms, clustering algorithms and the like, and unstructured data can only discover the value of multi-modal data through an AI algorithm. In the data service mode, the structured data can be used for data service in the forms of report forms, data set inquiry, statistical analysis results and the like, and the non-structured data usually trains an AI model through historical data and returns identification or prediction results.

Unstructured data management is very difficult due to the above problems with unstructured data. In the related art, everything is a file searching tool, which has a light software volume, a professional searching function and an ultra-fast searching speed, and can help you to find a desired file quickly. However, everying can only retrieve files, but cannot retrieve pictures. The elastomer search is a distributed, open-source search analysis engine that supports a variety of data types including text, numeric, geographic, structured, unstructured, but lacks organized semanticalization of labels.

Disclosure of Invention

The embodiment of the invention provides a data resource management method, a data resource management model training method, a device and electronic equipment, which are used for at least solving one of the technical problems.

In a first aspect, an embodiment of the present invention provides a data resource management method, including: responding to the node selected by a user on the semantic knowledge graph to store multi-mode data, and calculating the shortest path from the selected node to a root node, wherein the node of the semantic knowledge graph comprises the root node; determining a semantic path based on the names of the nodes on the shortest path; and storing the multi-mode data based on the semantic path.

In a second aspect, an embodiment of the present invention provides an electronic device, including: the system comprises at least one processor and a memory communicatively connected with the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any one of the data resource management methods of the present invention.

In a third aspect, embodiments of the present invention provide a storage medium having stored therein one or more programs including execution instructions that can be read and executed by an electronic device (including, but not limited to, a computer, a server, or a network device, etc.) for performing any of the above-mentioned data resource management methods of the present invention.

In a fourth aspect, embodiments of the present invention also provide a computer program product comprising a computer program stored on a storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform any of the data resource management methods described above.

According to the method, the semantic knowledge graph with the root node is used for carrying out interface interaction with the user, and the data is stored through the mapping relation between the semantic knowledge graph and the actual storage position, so that the user can store the data more quickly and conveniently, the subsequent searching of the data is also more quickly and conveniently, and the use experience of the user in storing and searching the file is greatly improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flowchart of a method for managing data resources according to an embodiment of the present invention;

FIG. 2 is a schematic diagram illustrating mapping between nodes and distributed storage according to an embodiment of the present invention;

FIG. 3 is a specific example of a data storage scheme provided by an embodiment of the present invention;

FIG. 4 is a diagram of a solution combined with a conventional interface manner (API and SQL statement) according to an embodiment of the present invention;

FIG. 5 is a specific example of a data storage scheme provided by an embodiment of the present invention;

FIG. 6 is a schematic diagram of a hierarchical structure of a specific example according to an embodiment of the present invention;

FIG. 7 is a schematic diagram of labeling modes according to an embodiment of the present invention;

fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Referring to fig. 1, a flowchart of a data resource management method according to an embodiment of the invention is shown.

As shown in fig. 1, in step 101, in response to a user selecting a node on a semantic knowledge graph to store multi-modal data, a shortest path from the selected node to a root node is calculated;

in step 102, determining a semantic path based on the names of the nodes on the shortest path;

in step 103, the multimodal data is stored based on the semantic path.

In this embodiment, for step 101, when the user needs to store multi-modal data, the operation may be directly performed on the semantic knowledge graph. The nodes of the semantic knowledge graph comprise root nodes, all the nodes are developed from the root nodes, for example, only one root node is arranged on the semantic knowledge graph just before the beginning, new labels exist when new data are stored by a user, the labels can be used for all the nodes forming the semantic knowledge graph, each node can have a path to reach the root node, the shortest path to the root node can be calculated if the root node exists, and the labels of all father nodes of the node can be directly marked when any node is selected to store the data. When a user selects one node in the semantic knowledge graph to store multi-mode data, the data resource management device can calculate the shortest distance from the node selected by the user to the root node of the semantic knowledge graph. The root node belongs to a logically specified root node, is not a root node of a knowledge graph, is not a root node, and can be designated as a root node by a user, or can be newly built as a root node, wherein the root node is used for calculating the shortest path. It should be noted that the knowledge graph may be designed by the user according to his own organization structure, and the more reasonable the knowledge graph is designed, the faster the data is stored and fetched. Further, the knowledge graph can be replaced by a semantic web, and the application is not limited herein. Wherein, the knowledge graph and the semantic net are understood by both a robot and a human.

Then, for step 102, a semantic path may be determined according to the names of the nodes on the shortest path, for example, the semantic path may be obtained by sequentially splicing the nodes on the shortest path, where the semantic path may be represented as "root.

Finally, for step 103, after the semantic path is obtained, the distributed data storage platform may be called according to the semantic path to query the actual storage location corresponding to the node, and then the multi-mode data that the user needs to store may be stored in the actual storage location. The above process is invisible to the user, and the user only needs to know that the data is stored at a certain position on the semantic knowledge graph, and the form of the semantic knowledge graph is very friendly to the user.

In a further alternative embodiment, the storing the multimodal data based on the semantic path includes: and calling a distributed data storage platform to store the multi-mode data based on the semantic path, wherein the bottommost layer of the distributed data storage platform is a distributed storage system, the middle layer is a mapping relation between a label set and a file, the uppermost layer is a corresponding relation between a semantic knowledge graph and the label set, and the label set is a set of names of all nodes. According to the scheme, the data are semantically stored, so that the isolation of the physical file and a semantic path is realized, and an organized semantically network is manufactured by using the labels of the file. More convenient inquiry and storage can be realized. It should be noted that, the mapping relationship is established through a network, and the application does not limit the distributed storage system, and may support label-based storage, or may not support label-based storage, which is not described herein. Specifically, a tag set can be obtained based on a semantic path, and then a mapping relationship is formed with a storage path.

According to the scheme, unstructured data Schema modeling is completed by utilizing the semantic knowledge graph, and the data resources are virtually hung on the entity of the knowledge graph, so that the user can store and inquire conveniently. The strong knowledge expression capability of the knowledge graph is combined with the distributed storage system, so that mapping of the entity node or attribute node of the knowledge graph and the physical data resource is realized. A node may map to either one or a group of multimodal files or to an SQL query of a relational database, as the application is not limited in this respect.

The mapping between the nodes and the distributed storage space can be specifically shown with reference to fig. 2. Further, the mapping between the data resource and the semantic expression thereof is realized through the semantic path between the node and the distributed storage space, which can be shown in fig. 3. The above procedure can be expressed as: f (semantic path) < semantic space of data resource > → < data resource >, the semantic space of data resource is expressed by a knowledge graph, and the RDF (Resource Description Framework ) specification is followed, so that both 'people' and 'machines' can understand. Most knowledge maps use RDF to describe various resources in the world and are saved to the knowledge base in triples. RDF is a resource description language that is affected by many aspects of metadata standards, framework systems, object oriented languages, etc. to describe various network resources, and appears to provide a standard data description framework for people to publish structured data on the Web. The RDF language is used to facilitate the creation of human-machine readable files on a network that can be automatically processed by a machine.

With further reference to fig. 4, a scenario in which embodiments of the present application incorporate conventional interface approaches (API and SQL statements) is illustrated.

With continued reference to FIG. 5, one specific example of a data storage scheme of the present application is shown.

As shown in fig. 5, the above procedure is illustrated by way of one example: assume that 10 photos are deposited on the < graduation > attribute of the < Zhang Sanj > entity to the semantic knowledge-graph as shown, the procedure is as follows:

firstly, inquiring by using SPARQL language to obtain < graduation > attribute of < Zhang Sany >;

then, calculating the shortest path of the attribute to the root node;

then, the names of all nodes on the shortest path are spliced to form a semantic path of root, unit, line 1, line 2, line 1, line three, graduation;

and finally, calling a distributed data storage platform data insertion interface (str < - > tags, str < - > files), wherein tags are root units, 1 row, 1 to 2 row, 1 class, three classes, graduation, and files are file names to be uploaded.

It is essential that 10 photo files are not stored in the semantic knowledge graph, but rather are stored in a location of the distributed data storage platform, and the specific storage path is not a concern for the user.

Referring to FIG. 6, in a further alternative embodiment, the distributed storage from the user's interface to the underlying layer may further include the following layers: the AI application support layer, the open service layer, the corpus semantic processing labeling layer, the semantic management layer based on the knowledge graph, the index layer (ES), the semantic file access, the data access and the distributed storage.

In some optional embodiments, the nodes of the semantic knowledge graph further include non-resource nodes and resource nodes, and the calculating the shortest path from the selected node to the root node in response to the user selecting the node on the semantic knowledge graph to store the multi-modal data includes: responding to the user to select the node on the semantic knowledge graph to store data, and judging whether the node is a resource node or not; if the node is a resource node, calculating the shortest path from the node to the root node; if the node is a non-resource node, inquiring whether the user mounts a resource node on the non-resource node for storing the multi-mode data; and in response to the user confirmation, mounting the resource node on the non-resource node, calculating the shortest path from the mounted resource node to the root node. Therefore, the data storage operation can be directly performed on the resource node (such as the attribute node in the foregoing embodiment), and the data storage operation can be performed on the non-resource node after the resource node is mounted on the non-resource node. It should be noted that any node may be used as a resource node. The above classification of nodes may be performed according to different needs of the user. In fact, it is possible that 90% of the nodes are non-resource nodes, and no data is located thereon, so that the semantic path can be more complete, and the finer the score is, the more advantageous the subsequent storage and searching is.

The semantic path can be displayed to a user in an interface way, the user can store the data by dragging the data to a certain node, and the user can also realize addition, deletion and/or modification of the semantic path through interface interaction. These semantic paths are visible to the user, exhibit clearer relationships, are more user friendly, and visually render the semantic paths more intuitive, nor does the user care about the specific storage path. This approach makes the storage semantical, virtualized, normalized, and machine-learning oriented, more friendly to subsequent data processing, such as data statistics, storage and lookup of meeting disciplines, and so forth.

In some specific examples, the semantic path may be defined by using an organization structure and/or a business structure, for example, the semantic path may be defined by using the organization structure and/or the business structure of a certain bank through data management of the certain bank, for example, zhang san is an employee of a certain department of the bank, and then data of Zhang san may be stored on corresponding nodes in the semantic path defined by the corresponding organization structure, so that the semantic path is consistent with the organization structure, and the user stores more standard, and searches for data more conveniently and quickly. Similarly, the personal data may be stored according to a common organization structure or a common business structure of the user, which is not described herein.

In other alternative embodiments, the semantic knowledge graph has two relationship spaces of a tree structure including only vertically oriented relationships and a mesh structure including vertically oriented relationships of the tree structure and horizontal relationships between different branches. By layering the relationships, a tree structure with only vertical relationships and a mesh structure with both vertical and horizontal relationships can be obtained. Further, the relationship attribute values may also be used to represent different relationship spaces, such as 0 representing a mesh structure and 1 representing a tree structure, so that operations may be performed in different relationship spaces by selecting different relationship attribute values when performing different operations. The tree structure is used for forming a semantic path and positioning, a unique semantic path can be formed, the network structure is used for inquiring, the semantic path of the inquiring is not limited to the stored semantic path, and the inquiring result can be more sufficient.

In a further alternative embodiment, the method further comprises: responding to the query instruction of the user, querying nodes based on node names in the query instruction, and calculating the shortest path between the queried nodes and the root node, wherein query operation is performed in the network structure, and data acquisition is performed in the tree structure; determining a semantic path based on a character string composed of names of nodes in the shortest path; and acquiring a file list based on the semantic path and displaying the file list to the user. More semantically related content can be found by searching in the network structure at the time of searching. In some specific examples, one or more tags may be entered at the time of the query, and the user may be supported to directly select a tag given by the system, so that the search may be completed more quickly. The data are displayed in an organized knowledge graph mode, so that the query is more convenient.

In some specific examples, a user may input one or more labels during a query, then find one or more nodes in the mesh structure that contain the user input labels, and the nodes may be presented to the user as a result of the query, so that the user may select from the nodes, after the user selects a node, uniquely locate the semantic path of the node in the tree structure based on the shortest path from the node to the root node, then find the corresponding node through the semantic path, and present the file list of the node to the user.

In other optional embodiments, the adding operation, the deleting operation and the modifying operation are performed in the tree structure, and before the storing of the multimodal data in response to the user selecting a node on the semantic knowledge graph, the method further includes: and constructing a semantic knowledge graph based on the adding operation, the deleting operation and the modifying operation of the user on other nodes except the root node. The user can maintain the underlying data in a manner of maintaining the semantic knowledge graph through operations such as adding, deleting, checking and the like, for example, the user deletes a certain resource node, and the data of the actual storage position is deleted accordingly, which is not described in detail herein. The add, delete, modify operations all involve changes in the semantic path.

In some optional embodiments, the multimodal data has a user-defined tag, and the tag set formed by the user-defined tag and the semantic path can be used as a tag in a query instruction to find a node containing the tag. So that a node can be found by multiple tags, the two tags do not affect each other.

The user-defined label can be obtained by labeling the data in advance before the data is stored by a user, but the labeling is not mandatory, so that the user is allowed to label himself, and the labeling is not performed in advance and is not influenced.

When a user stores data, for example, the data is dragged to a node for storage, the stored data is automatically labeled according to a semantic path in the storage process, and a label set formed by the names of the nodes on the shortest path becomes a label of the data.

The two labeling methods may be overlapped, for example, the two labels may be overlapped in a union mode, and the user-defined label may be called an additional label, a personalized label, or the like, which is not limited herein. The labels have no relation, and the subsequent retrieval can be more refined if more labels exist. The labels formed by the semantic paths can be called semantic labels or node labels, and the labels are automatically added based on the labels of the node oriented knowledge graph.

Furthermore, the semantic tags can also automatically form machine learning tags, and the storage mode of the embodiment of the application automatically generates semantic tags and classifications, so that the method is beneficial to the subsequent data mining and is beneficial to the innate support of artificial intelligence.

In some alternative embodiments, the tag set formed by the multi-modal data and the semantic path may be used for training a model, so that the model can predict the semantic path of the multi-modal data, and thus the semantic path may be recommended by a machine learning manner, that is, by a machine learning manner, predicting under which node a certain data should be placed. Specifically, data stored after manually selecting the nodes (the data is self-tagged) can be used as training data for machine learning, and then a model capable of predicting/recommending the storage nodes can be trained.

The data stored by the method can be used as an AI (artificial intelligence) data base, the storage mode naturally supports the application of machine learning, the data and the model can be fused together better, and the model is the data condensation. Each time the scheme of the application is used to store one data, the data can be used as a training sample of the model.

In a further alternative embodiment, different models may be manually installed under different branches, so that the process of storing data becomes training data of the models, such as text classification models, image recognition models, etc., can be applied. The inventor considers that the model also belongs to data, the model can be mounted on the node, different data sets can be designated as training data of the model, and the model and the data are fused, so that the model can be mounted under different branches for training and iterative updating, new data are stored under each branch, and the training data of the model can be used for iterative updating of the model.

In some optional embodiments, before the storing of the multimodal data in response to the user selecting a node on the semantic knowledge graph, calculating a shortest path from the selected node to the root node, the method further includes: and labeling the multi-mode data uploaded by the user, wherein the labeling mode comprises manual labeling, semi-automatic labeling and artificial intelligent automatic labeling. In a further alternative embodiment, the manually labeling includes: responding to the user to mark the multi-mode data, and displaying a label for marking, wherein the label is the name of each node in the semantic knowledge graph; in response to the user selecting at least one tag, the tag is assigned to the multimodal data. The above shows various ways of user pre-labeling that do not affect the semantic tags when stored.

In particular, as shown in fig. 7, multimodal data may be annotated, such as speech, images, and text. Different data can be marked by selecting a label during marking, and the label can be newly built, so that the method is not limited.

AI (artificial intelligence) automatic labeling may also be used, or AI algorithms may be used to assist intelligent labeling, without limitation herein.

In other embodiments, embodiments of the present invention further provide a non-volatile computer storage medium storing computer-executable instructions that are capable of performing the data resource management method of any of the method embodiments described above;

as one embodiment, the non-volatile computer storage medium of the present invention stores computer-executable instructions configured to:

responding to the node selected by a user on the semantic knowledge graph to store multi-mode data, and calculating the shortest path from the selected node to a root node, wherein the node of the semantic knowledge graph comprises the root node;

determining a semantic path based on the names of the nodes on the shortest path;

and storing the multi-mode data based on the semantic path.

The non-transitory computer readable storage medium may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the data resource management device, and the like. Further, the non-volatile computer-readable storage medium may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, the non-transitory computer readable storage medium optionally includes memory remotely located with respect to the processor, the remote memory being connectable to the data resource management device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The embodiments of the present invention also provide a computer program product comprising a computer program stored on a non-volatile computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform any of the data resource management methods described above.

Fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, as shown in fig. 8, where the device includes: one or more processors 810, and a memory 820, one processor 810 being illustrated in fig. 8. The apparatus of the data resource management method may further include: an input device 830 and an output device 840. Processor 810, memory 820, input device 830, and output device 840 may be connected by a bus or other means, for example in fig. 8. Memory 820 is the non-volatile computer-readable storage medium described above. The processor 810 executes various functional applications of the server and data processing, i.e., implements the method of data resource management of the method embodiments described above, by running non-volatile software programs, instructions, and modules stored in the memory 820. The input device 830 may receive input numeric or character information and generate key signal inputs related to user settings and function control of the communication compensation device. The output device 840 may include a display device such as a display screen.

The product can execute the method provided by the embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method. Technical details not described in detail in this embodiment may be found in the methods provided in the embodiments of the present invention.

As one embodiment, the electronic device is applied to a data resource management apparatus, and includes: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to enable the at least one processor to:

and storing the multi-mode data based on the semantic path.

The electronic device of the embodiments of the present application exist in a variety of forms including, but not limited to:

(1) A mobile communication device: such devices are characterized by mobile communication capabilities and are primarily aimed at providing voice, data communications. Such terminals include smart phones (e.g., iPhone), multimedia phones, functional phones, and low-end phones, among others.

(2) Ultra mobile personal computer device: such devices are in the category of personal computers, having computing and processing functions, and generally also having mobile internet access characteristics. Such terminals include: PDA, MID, and UMPC devices, etc., such as iPad.

(3) Portable entertainment device: such devices may display and play multimedia content. Such devices include audio, video players (e.g., iPod), palm game consoles, electronic books, and smart toys and portable car navigation devices.

(4) The server is similar to a general computer architecture in that the server is provided with high-reliability services, and therefore, the server has high requirements on processing capacity, stability, reliability, safety, expandability, manageability and the like.

(5) Other electronic devices with data interaction function.

The apparatus embodiments described above are merely illustrative, wherein elements illustrated as separate elements may or may not be physically separate, and elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on such understanding, the foregoing technical solutions may be embodied essentially or in part in the form of a software product, which may be stored in a computer-readable storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the various embodiments or methods of some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data resource management method, comprising:

responding to the user to select the node on the semantic knowledge graph to store multi-mode data, and calculating the shortest path from the selected node to the root node, wherein the node of the semantic knowledge graph comprises the root node on logic;

and storing the multi-mode data based on the semantic path.

2. The method of claim 1, wherein the storing the multimodal data based on the semantic path comprises:

and calling a distributed data storage platform to store the multi-mode data based on the semantic path, wherein the bottommost layer of the distributed data storage platform is a distributed storage system, the middle layer is a mapping relation between a label set and a file, the uppermost layer is a corresponding relation between a semantic knowledge graph and the label set, and the label set is a set of names of all nodes.

3. The method of claim 1, wherein the nodes of the semantic knowledge graph further comprise non-resource nodes and resource nodes, and wherein the computing the shortest path from the selected node to the root node in response to the user selecting a node on the semantic knowledge graph for storage of multi-modal data comprises:

responding to the user to select the node on the semantic knowledge graph to store data, and judging whether the node is a resource node or not;

if the node is a resource node, calculating the shortest path from the node to the root node;

if the node is a non-resource node, inquiring whether the user mounts a resource node on the non-resource node for storing the multi-mode data;

and in response to the user confirmation, mounting the resource node on the non-resource node, calculating the shortest path from the mounted resource node to the root node.

4. The method of claim 1, wherein the semantic knowledge graph has two relationship spaces of a tree structure including only vertically oriented relationships and a mesh structure including vertically oriented relationships of the tree structure and horizontal relationships between different branches.

5. The method of claim 4, wherein the method further comprises:

responding to the query instruction of the user, querying nodes based on node names in the query instruction, and calculating the shortest path between the queried nodes and the root node, wherein query operation is performed in the network structure, and data acquisition is performed in the tree structure;

determining a semantic path based on a character string composed of names of nodes in the shortest path;

and acquiring a file list based on the semantic path and displaying the file list to the user.

6. The method of claim 4, wherein the adding, deleting and modifying operations are performed in the tree structure, and wherein prior to the storing of multimodal data in response to a user selecting a node on a semantic knowledge graph, the method further comprises:

and constructing a semantic knowledge graph based on the adding operation, the deleting operation and the modifying operation of the user on other nodes except the root node.

7. The method of claim 1, wherein the method further comprises:

and using the tag set formed by the multi-modal data and the semantic path for training a model so that the model can predict the semantic path of the multi-modal data.

8. The method of claim 7, wherein the multimodal data is provided with user-defined labels, and the user-defined labels and the label set formed by the semantic path can be used as labels in a query instruction to find nodes containing the labels.

9. An electronic device, comprising: at least one processor, and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the method of any one of claims 1 to 8.

10. A storage medium having stored thereon a computer program, which when executed by a processor performs the steps of the method according to any of claims 1 to 8.