A graphical knowledge structure data processing method and device for artificial intelligence context construction, electronic equipment and storage medium

Abstract

The application discloses a kind of graphical knowledge structure data processing methods for artificial intelligence context construction.The method reads the content unit data, associated data and structure relationship data of target graphical knowledge structure from persistent storage, converts into memory object and is managed according to structure identification, establishes object positioning index based on structure identification and content unit unique identification, restores the hierarchical relationship, associated relationship or reference relationship between content units;Determine the candidate content unit object set and generate context data when searching, correlation analysis or artificial intelligence context construction;After the reference result is returned by artificial intelligence, quickly locate the target content unit object, and drive the display interface to perform jump display, local expansion, highlight display, local refresh or view adjustment.The method can improve loading efficiency, object positioning efficiency and interface linkage efficiency.

Description

Technical Field

[0001] This invention relates to the field of computer software and information processing technology, and more particularly to a method, apparatus, electronic device, and computer-readable storage medium for processing graphical knowledge structures based on artificial intelligence context construction. More specifically, this invention relates to persistent data loading, memory-resident management, object location indexing, relationship recovery, context generation, and AI result linkage display technology for graphical knowledge structures such as mind maps. Background Technology

[0002] Mind mapping applications typically persist mind map data in a database or other persistent media. When a user opens a specific mind map, the corresponding data is read into the running memory and restored into a tree structure, graph structure, or a combination thereof based on the parent-child relationship, connection relationship, or reference relationship between nodes, so that it can be displayed, edited, and interacted with in a graphical interface.

[0003] As mind mapping applications expand their functionality, systems need to support not only the display and editing of single mind maps, but also rapid switching between multiple mind maps, cross-mind map searches, cross-mind map association analysis, and context building for artificial intelligence modules. In these scenarios, existing technologies typically suffer from the following problems: First, after the target mind map data is loaded from persistent storage into memory, it often requires repeated construction of node objects, connection objects, and their relationship structures, resulting in high loading latency. Second, when multiple mind maps reside in memory simultaneously, there is a lack of a unified and efficient object management and rapid object location mechanism, leading to low object lookup efficiency and high switching costs. Third, during cross-mind map searches and artificial intelligence context building, it is often necessary to traverse a large number of node objects across multiple mind maps, resulting in high computational overhead and slow response speed. Fourth, when artificial intelligence outputs a reference result corresponding to a certain node, existing solutions struggle to quickly map this reference result to the target object in memory and drive stable interface transitions and partial refreshes, affecting interaction continuity and display stability.

[0004] Therefore, it is necessary to provide a graphical knowledge structure data processing scheme for artificial intelligence context construction to improve the loading efficiency, object location efficiency, cross-structure context construction efficiency, and linkage efficiency between artificial intelligence results and graphical interfaces. Summary of the Invention

[0005] The purpose of this invention is to provide a graphical knowledge structure data processing method, apparatus, electronic device, and storage medium for artificial intelligence context construction, so as to solve the problems of low loading efficiency of multiple graphical knowledge structures, low object positioning efficiency, high context construction overhead, and difficulty in quickly linking artificial intelligence results with interface objects in the prior art.

[0006] To achieve the above objectives, this invention provides a method for processing graphical knowledge structure data based on artificial intelligence context construction, applied in electronic devices. The graphical knowledge structure includes mind maps, tree-like knowledge structures, graph-like knowledge structures, or combinations thereof. The method includes: receiving a loading instruction for a target graphical knowledge structure; reading content unit data, associated data, and structural relationship data corresponding to the target graphical knowledge structure from persistent storage according to the loading instruction; converting the content unit data into a set of content unit objects in memory, converting the associated data into a set of associated objects in memory, and storing them in a multi-structure resident management container according to structural identifiers, so that multiple graphical knowledge structures can reside in memory in parallel; establishing an object location index structure based on the structural identifiers and the unique identifiers of each content unit to support the resident objects in memory. The system directly locates the content unit objects; it recovers the hierarchical, association, or reference relationships between the content unit objects based on the structural relationship data to form a memory organization structure corresponding to the target graphical knowledge structure; in response to search requests, association analysis requests, or artificial intelligence context construction requests, it determines a set of candidate content unit objects based on the multi-structure resident management container and the object positioning index structure, and generates context data for use by the artificial intelligence module based on the set of candidate content unit objects; in response to the reference results output by the artificial intelligence module, it locates the target content unit object through the object positioning index structure based on the structural identifier and unique identifier of the content unit contained in the reference results, and drives the display interface to perform jump display, partial expansion, highlight display, partial refresh, or view adjustment operations corresponding to the target content unit object.

[0007] In addition to the above-mentioned methods and technical solutions, the present invention also provides a graphical knowledge structure data processing device for artificial intelligence context construction. The device includes functional modules for executing the steps of the aforementioned methods to realize persistent data reading of graphical knowledge structures, object construction, multi-structure residency management, object location index establishment, relationship recovery, context generation, artificial intelligence result linkage, and incremental update. In a preferred embodiment, the device includes a persistent storage access module 101, an object construction module 102, a multi-structure residency management module 103, an object location index module 104, a relationship recovery module 105, a context generation module 106, a result linkage module 107, and an incremental update module 108.

[0008] The present invention also provides an electronic device, including at least one processor, at least one memory, and a computer program stored in the at least one memory and executable by the at least one processor, wherein the processor executes the computer program to implement the steps of the aforementioned method.

[0009] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the aforementioned method.

[0010] Compared with existing technologies, the present invention has at least the following beneficial effects: It enables parallel residency of multiple graphical knowledge structures through a multi-structure residency management container, reducing the overhead of repeated loading during structure switching and cross-structure analysis; it improves object lookup efficiency by enabling rapid direct location of target content unit objects through an object positioning index structure; it improves loading efficiency by rapidly converting persistent data into a memory organization structure that can be used for display and processing through a relationship recovery mechanism; it improves the efficiency of artificial intelligence context construction through candidate content unit filtering and structured context generation for artificial intelligence; and it improves the accuracy of reference result bounce, local refresh efficiency, and display stability through a linkage mechanism between artificial intelligence results and the interface. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the module structure of a graphical knowledge structure data processing system built for artificial intelligence context in one embodiment of the present invention.

[0012] Figure 2 This is a schematic diagram illustrating the process of loading graphical knowledge structure data, constructing objects, restoring relationships, and managing residency in one embodiment of the present invention.

[0013] Figure 3 This is a schematic diagram of an object location index structure according to one embodiment of the present invention.

[0014] Figure 4 This is a schematic diagram of the process for generating context data for an artificial intelligence module in one embodiment of the present invention.

[0015] Figure 5 This is a schematic diagram illustrating the process of AI-driven graphical interface execution of jump display, partial expansion, and partial refresh in one embodiment of the present invention.

[0016] Figure 6 This is a flowchart illustrating the incremental update process in one embodiment of the present invention. Detailed Implementation

[0017] The present invention will now be described in further detail with reference to the accompanying drawings and preferred embodiments. It should be understood that the following embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the present invention. Various modifications and substitutions can be made by those skilled in the art without departing from the concept of the present invention, and all such modifications and substitutions should fall within the scope of protection of the present invention.

[0018] In one embodiment, the present invention is applied to an electronic device, which may be a mobile phone, tablet computer, laptop computer, desktop computer, or other device with a processor, memory, and graphical display interface. The electronic device runs a graphical knowledge structure application for managing at least one graphical knowledge structure. The graphical knowledge structure is preferably a mind map, but is not limited to a mind map; it may also be a tree-like knowledge structure, a graph-like knowledge structure, a hierarchical knowledge structure, or a combination thereof.

[0019] like Figure 1 As shown, in a preferred embodiment, the system includes at least: a persistent storage access module 101, an object construction module 102, a multi-structure residency management module 103, an object location index module 104, a relationship recovery module 105, a context generation module 106, a result linkage module 107, and an incremental update module 108. These modules can be implemented by software or by a combination of software and hardware.

[0020] In one embodiment, each graphical knowledge structure has a corresponding structure identifier, preferably a globally unique identifier. Each content unit has a corresponding content unit unique identifier. For mind mapping scenarios, each mind map has a corresponding mapID, and each node has a corresponding node unique identifier. Persistent storage stores content unit data, associated data, and structural relationship data corresponding to the target graphical knowledge structure. The persistent storage is preferably a database; in a preferred implementation, an object persistent database can be used, and in a further preferred implementation, Core Data can be used, but the present invention is not limited to a specific database type.

[0021] like Figure 2 As shown, when a user clicks on a target mind map in the application's mind map list interface, the system receives a loading instruction. The persistent storage access module, based on the structure identifier in the loading instruction, reads the content unit data, associated data, and structural relationship data corresponding to the target graphical knowledge structure from the database. The content unit data may include at least one of the following fields: title, body text, type information, position parameters, style parameters, creation time, update time, and unique identifier; the associated data may include at least one of the following: main connection data, subordinate connection data, and cross-structure reference data; the structural relationship data may include at least one of the following: parent content unit identifier, child content unit identifier, hierarchical order, branch order, and dependency information.

[0022] The object construction module converts content unit data read from persistent storage into a collection of content unit objects in memory, and converts associated data into a collection of associated objects in memory. In a preferred embodiment, different types of objects are created according to the content unit type, such as ordinary node objects, parenthesis objects, radial objects, and tag objects, and main connection objects and subordinate connection objects are created according to the associated data. By classifying objects by type during the loading stage, traversal of irrelevant objects can be reduced during subsequent display, search, context generation, and partial refresh.

[0023] In a preferred but not limited implementation, the system uses the following mapping to manage different types of object collections: thoughts[mapID] manages the collection of ordinary node objects; brackets[mapID] manages the collection of parenthesis objects; radials[mapID] manages the collection of radial objects; links[mapID] manages the collection of main link objects; and subLinks[mapID] manages the collection of subordinate link objects. The above variable names are merely examples of preferred implementations, and the present invention is not limited to using these variable names or the exact same container format.

[0024] The system allows multiple graphical knowledge structures to reside in memory simultaneously, supporting multi-structure switching, cross-structure search, and joint construction of AI-oriented contexts. To this end, the system sets up a multi-structure residency management container, preferably a mapping structure with structure identifiers as keys. For each structure identifier, at least one of the following is maintained: a set of content unit objects, a set of main associated objects, a set of subordinate associated objects, a set of special type content objects, and a set of auxiliary display objects. Preferably, the multi-structure residency management module also records the status information of each residency structure, including but not limited to the most recent access time, whether it is currently participating in AI context construction, whether it is currently visible on the interface, its current memory usage status, and whether it is in a high-priority cache state. Based on the above status information, the system can perform at least one of the following controls on each graphical knowledge structure: retention, delayed release, degraded caching, reloading, or partial reconstruction.

[0025] like Figure 3As shown, to improve the efficiency of locating target objects under multi-structure parallel residency conditions, this embodiment establishes an object location index structure. Preferably, the object location index structure is a multi-level index structure, including at least: a first-level index with the structure identifier as the key, and a second-level index with the unique identifier of the content unit as the key; the value of the second-level index is used to represent the memory reference, object handle, or object address identifier of the corresponding content unit object. In a preferred example matching your current implementation, the following index can be established: id2obj[mapID][nodeID] = corresponding node object. Wherein, mapID is the unique identifier of the mind map, and nodeID is the unique identifier string of the node. Based on this index, after artificial intelligence returns a node reference result, or after a node is hit in a cross-mind map search, the corresponding memory object can be directly located without traversing all node objects in thoughts[mapID]. Furthermore, type indexes, hierarchical depth indexes, time indexes, visibility indexes, and popularity indexes can also be established to assist in the selection of context candidate nodes and local refresh.

[0026] The relationship restoration module restores the hierarchical, associative, and referential relationships between content unit objects based on structural relationship data, forming an organizational structure in memory that can be displayed, edited, searched, and processed by artificial intelligence. For mind map scenarios, the relationship restoration process includes at least: restoring parent-child reference relationships between node objects based on parent node identifiers; restoring the arrangement logic in the tree structure based on hierarchical and branching order; restoring connection relationships between node objects based on main connection records; and restoring subordinate reference relationships across levels, branches, or structures based on subordinate connection records. In a preferred embodiment, relationship restoration is performed in stages. First, the main hierarchical relationships are restored to quickly form the main framework for display and interaction; then, the main connection relationships are restored; and finally, subordinate relationships and cross-structure reference relationships are restored.

[0027] like Figure 4 As shown, when the system receives a search request, an association analysis request, or an artificial intelligence context construction request, the context generation module determines a set of candidate content unit objects based on a multi-structure resident management container and an object positioning index structure. The candidate set can be determined based on at least one of the following factors: currently selected content unit, recently accessed content unit, content unit within the currently visible area, content unit matching the search keywords, content unit with a parent-child relationship to the current content unit, content unit with a connection or subordinate relationship to the current content unit, content unit belonging to a specified structure set, content unit meeting a preset hierarchical depth range, and content unit meeting a preset popularity or importance threshold.

[0028] In a preferred embodiment, the context generation module extracts at least one of the following information from the candidate content unit object set to form structured context data for use by the artificial intelligence module: content unit title, body text, summary, hierarchical path, parent node information, child node information, association path information, belonging structure identifier, content unit unique identifier, and jumpable reference identifier. Preferably, the jumpable reference identifier includes at least the structure identifier and the content unit unique identifier. In further embodiments, hierarchical path, target area parameters, expansion parameters, or highlight parameters may also be added.

[0029] like Figure 5 As shown, when outputting processing results, the AI ​​module outputs at least one reference result item in addition to the text result. Each reference result item includes at least: a structure identifier and a unique identifier for the content unit. Further, the reference result item may also include at least one of the following: content unit title, hierarchical path, target display area parameters, expansion parameters, highlight parameters, and view compensation parameters. After receiving the above reference result items, the result linkage module locates the target content unit object through the object positioning index structure and drives the display interface to perform at least one of the following operations: scrolling, panning, or scaling the target object to the currently visible area; expanding its parent branch when the target object is in a collapsed branch; highlighting the target object; performing local rearrangement or local redrawing on the local area where the target object is located; and performing view offset compensation based on the updated display position of the target object to reduce interface jumps. For mind mapping scenarios, the target node object can be located first using mapID and nodeID, and then the canvas panning, scaling, local expansion, and highlighting strategies can be determined based on the object's position parameters, collapsed state, and layout dependencies in the canvas.

[0030] like Figure 6 As shown, when the target graphical knowledge structure data in persistent storage changes, the system does not need to completely unload and reload the entire target graphical knowledge structure. Instead, it identifies the affected content unit objects and related objects through the incremental update module, and only updates the data fields, relationship fields, and index items of the affected objects, while performing a partial refresh on the associated local display areas. For example, when adding a node, only the new node object is constructed and its parent-child relationships, index items, and necessary connection objects are established; when deleting a node, only the node object and its corresponding relationships and indexes are removed; when the node position, text, or style changes, only the node object and the local display areas affected by it are updated; when the connection relationship changes, only the corresponding connection objects and display paths are updated.

[0031] The following is a preferred embodiment that matches your existing software architecture to further illustrate the feasibility of the invention. In this embodiment, mind map data is stored in a Core Data database; each mind map has a unique identifier mapID; ordinary nodes, parenthetical objects, radial objects, main connection objects, and subordinate connection objects are managed by different object arrays; when multiple mind maps reside in memory simultaneously, each mind map object set is managed separately using a dictionary with mapID as the key; a direct access index for node objects is established using id2obj[mapID][nodeID]; during cross-mind map searches or AI context generation, candidate nodes are filtered from multiple resident object sets corresponding to mapIDs; after the AI ​​module returns the reference result, it jumps back to the target mind map and target node based on mapID and nodeID, and drives the canvas to perform partial expansion, highlighting, and view adjustment. It should be noted that the Core Data, thoughts, brackets, radials, links, subLinks, id2obj, etc. mentioned above are only specific examples in the preferred implementation, and the invention is not limited to using the same naming, the same data container, or the same software framework.

[0032] Although the present invention has been described in conjunction with preferred embodiments, those skilled in the art can make various modifications, equivalent substitutions, and improvements to the invention without departing from its spirit and essence; and where there are no contradictions, the technical features of the various embodiments can be combined with each other. Therefore, the present invention is not limited to the specific embodiments disclosed in the specification, and all technical solutions falling within the scope of protection of the claims should be included in the protection scope of the present invention.

Claims

1. A method for processing graphical knowledge structures constructed in an artificial intelligence context, applied in electronic devices, wherein the graphical knowledge structure includes mind maps, tree-like knowledge structures, graph-like knowledge structures, or combinations thereof, characterized in that, include: Receive a loading instruction for a target graphical knowledge structure; according to the loading instruction, read the content unit data, associated data, and structural relationship data corresponding to the target graphical knowledge structure from persistent storage; The content unit data is converted into a set of content unit objects in memory, and the associated data is converted into a set of associated objects in memory. These are then stored in a multi-structure residency management container according to their structural identifiers, enabling multiple graphical knowledge structures to reside in memory in parallel. An object location index structure is established based on the structural identifiers and the unique identifiers of each content unit to support direct location of content unit objects residing in memory. The hierarchical, associative, or referential relationships between the content unit objects are recovered based on the structural relationship data to form the memory organization structure corresponding to the target graphical knowledge structure. In response to search requests, association analysis requests, or artificial intelligence context construction requests, a set of candidate content unit objects is determined based on the multi-structure residency management container and the object location index structure. Context data for use by the artificial intelligence module is generated based on the set of candidate content unit objects. In response to the reference results output by the artificial intelligence module, the target content unit object is located through the object location index structure based on the structural identifiers and unique identifiers of the content units contained in the reference results. The display interface is then driven to perform jump display, partial expansion, highlighting, partial refresh, or view adjustment operations corresponding to the target content unit object.

2. The method according to claim 1, characterized in that, The multi-structure residency management container includes at least one mapping structure with a structure identifier as the key. The mapping structure is used to manage at least one of the following: a set of content unit objects, a set of main associated objects, a set of subordinate associated objects, a set of special type content objects, and a set of auxiliary display objects corresponding to each graphical knowledge structure.

3. The method according to claim 1, characterized in that, The object location index structure is a multi-level index structure, including at least: a first-level index with the structure identifier as the key, and a second-level index with the unique identifier of the content unit as the key; the value of the second-level index is used to represent the memory reference, object handle or object address identifier of the corresponding content unit object, so as to realize the direct location of the target content unit object.

4. The method according to claim 1, characterized in that, The step of restoring the hierarchical, association, or reference relationships between the content unit objects based on the structural relationship data includes: restoring the parent-child reference relationship based on the parent content unit identifier; restoring the connection relationship between content unit objects based on the main association record; and restoring the subordinate reference relationship across levels, branches, or structures based on the subordinate association record.

5. The method according to claim 1, characterized in that, The process of determining the candidate content unit object set includes filtering the content unit objects residing in memory based on at least one of the following: search keywords, currently selected content unit, recently accessed content unit, content unit in the current display area, content unit hierarchy depth, connection relationship between content units, cross-structure association relationship, access frequency, or preset importance threshold.

6. The method according to claim 1, characterized in that, The generation of context data for use by the artificial intelligence module includes extracting at least one of the following from the candidate content unit object set: content unit text, summary content, hierarchical path, parent node information, child node information, associated path information, belonging structure identifier, content unit unique identifier, and jumpable reference identifier, and combining the extraction results into structured context data.

7. The method according to claim 1, characterized in that, When the persistent data corresponding to the target graphical knowledge structure changes, the method further includes: determining the affected content unit objects and associated objects, performing incremental updates on the affected content unit objects and associated objects, and synchronously updating the object positioning index structure and related relationship structure, without having to completely unload and reload the entire target graphical knowledge structure.

8. A graphical knowledge structure data processing device for constructing artificial intelligence contexts, characterized in that, include: The persistent storage access module receives a loading instruction for a target graphical knowledge structure and reads content unit data, associated data, and structural relationship data corresponding to the target graphical knowledge structure from persistent storage according to the loading instruction. The object construction module converts the content unit data into a set of content unit objects in memory, converts the associated data into a set of associated objects in memory, and stores them in a multi-structure residency management container according to the structure identifier, so that multiple graphical knowledge structures can reside in memory in parallel. The multi-structure residency management module performs residency management on each set of structure objects in the multi-structure residency management container. The object positioning index module establishes an object positioning index structure based on the structure identifier and the unique identifier of each content unit to support direct positioning of content unit objects residing in memory. The relationship recovery module recovers the hierarchical, associated, or reference relationships between the content unit objects according to the structural relationship data to form the memory organization structure corresponding to the target graphical knowledge structure. The context generation module is used to respond to search requests, association analysis requests, or artificial intelligence context construction requests, determine a set of candidate content unit objects based on the multi-structure dwelling management container and the object positioning index structure, and generate context data for use by the artificial intelligence module based on the set of candidate content unit objects; the result linkage module is used to respond to the reference results output by the artificial intelligence module, locate the target content unit object through the object positioning index structure based on the structure identifier and content unit unique identifier contained in the reference results, and drive the display interface to perform jump display, partial expansion, highlight display, partial refresh, or view adjustment operations corresponding to the target content unit object; The incremental update module is used to determine the affected content unit objects and related objects when the persistent data corresponding to the target graphical knowledge structure changes, perform incremental updates on the affected content unit objects and related objects, and synchronously update the object positioning index structure and related relationship structure.

9. The apparatus according to claim 8, characterized in that, The object location indexing module establishes a multi-level index structure, which includes at least: a first-level index with the structure identifier as the key, and a second-level index with the unique identifier of the content unit as the key; the value of the second-level index is used to represent the memory reference, object handle, or object address identifier of the corresponding content unit object.

10. An electronic device, characterized in that, include: At least one processor; at least one memory; And a computer program stored in the at least one memory and executable by the at least one processor; wherein, when the at least one processor executes the computer program, it implements the method as described in any one of claims 1 to 7.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

Images