A geological archive structure integrity guarantee method and system based on atlas dynamic maintenance

Abstract

The application discloses a kind of based on atlas dynamic maintenance geological archive structure integrity guarantee method and system, comprising: for the peripheral associated data marked as key archive, obtain its direct and indirect connection archive flow path, analyze potential fracture point in path, obtain the risk area that can cause data isolated problem;By scanning the archive data in risk area in depth, obtain the content features and historical flow record of archive, judge whether it meets data accuracy standard, determine the non-key archive list that needs to be deleted;According to non-key archive list, analyze the operation influence range of each archive to be deleted, obtain its potential impact on overall system stability after deletion, judge if deletion will lead to information network integrity damaged, then it is removed from the deletion list;For the final determined deletion list, obtain the backup data and flow path log of archive, update archive connection atlas by dynamic maintenance mechanism, obtain new information network structure.

Description

Technical Field

[0001] This invention relates to the field of geological archives management technology, and in particular to a method and system for ensuring the structural integrity of geological archives based on dynamic map maintenance. Background Technology

[0002] Geological archives management is a crucial area for information resource protection and utilization, possessing irreplaceable value in natural resource development, disaster early warning, and scientific research. Its management efficiency and data accuracy directly impact the scientific rigor of industry decision-making. However, current geological archives management methods struggle to adapt to dynamically changing environments when faced with complex data relationships, particularly exhibiting significant shortcomings in data deletion operations.

[0003] Specifically, current methods lack a comprehensive consideration of the overall information structure when deleting data, which can easily lead to isolated data or loss of connections, thereby affecting the integrity of the archival system. This limitation is not simply due to a lack of technology, but rather to insufficient ability to uncover and dynamically maintain the deep connections between archives.

[0004] Against this backdrop, the core challenges facing geological archives management are becoming increasingly apparent. First, archive deletion operations must be conducted while ensuring data accuracy. However, determining accuracy often relies on multi-dimensional analysis of the archive content, and the current lack of unified standards and criteria means that deletion may inadvertently damage critical information. Second, maintaining the continuity of the archive's flow path during deletion is also a pressing issue. The connections between archives are intricate, and a single deletion operation may affect the entire information network, causing a break in the knowledge structure. These two factors are intertwined: the former determines the rationale for deletion, while the latter further requires that the operation not disrupt the stability of the overall system. Summary of the Invention

[0005] In order to solve the above-mentioned technical problems, the present invention provides a method and system for ensuring the structural integrity of geological archives based on dynamic maintenance of maps.

[0006] The technical solution of this invention is implemented as follows:

[0007] A method for ensuring the structural integrity of geological archives based on dynamic map maintenance includes the following steps:

[0008] By using a pre-established database of archival connections, an initial archival connection map is formed, and the importance score of each archival document in the overall information network is analyzed.

[0009] Based on the importance score in the file connection map, determine whether each file is a key node;

[0010] For the surrounding related data of the marked key archives, obtain the archive circulation paths that are directly and indirectly connected, analyze the potential breakpoints in the paths, and identify risk areas that may cause data silos.

[0011] By conducting in-depth scanning of the archival data within the risk area, the content characteristics and historical circulation records of each archival document are obtained, and it is determined whether they meet the data accuracy standards to identify a list of non-critical archives that need to be deleted.

[0012] Based on the list of non-critical files, analyze the scope of impact of each file to be deleted, obtain the potential impact of its deletion on the stability of the overall system, and determine if the deletion would damage the integrity of the information network, then remove it from the deletion list.

[0013] For the final deletion list, obtain backup data and flow path logs for each file, update the file connection graph through a dynamic maintenance mechanism, and obtain a new information network structure;

[0014] Based on the new information network structure, the continuity of the archive circulation path is analyzed, and if there are any breaks, the connection is rebuilt using the completion technology.

[0015] Furthermore, the process of obtaining its importance score in the overall information network includes:

[0016] By using a pre-defined archive association database, the flow paths and dependencies between geological archives are obtained. Using graph database technology, an initial archive connection graph is constructed to obtain the data structure of nodes and edges.

[0017] Based on the archive connection map, the node attributes and edge weights of each geological archive are extracted. The PageRank algorithm is used to calculate the preliminary importance score of each archive in the map and obtain the score distribution.

[0018] Furthermore, after obtaining the rating distribution, the process also includes obtaining a ranking of file importance, specifically including:

[0019] If the calculated preliminary importance score is lower than the preset threshold, then through multidimensional data analysis, the metadata features of the archive are obtained, and the key feature vectors are extracted using principal component analysis to obtain a feature-enhanced archive representation.

[0020] Based on the feature-enhanced file representation, combined with the flow path and dependencies, a weighted average method is used to fuse multidimensional features and graph scores to calculate a comprehensive importance score, resulting in an updated score distribution.

[0021] The updated rating distribution is used to determine the relative importance of the archives in the information network. If the rating is higher than the preset threshold, it is marked as a key archive, and a set of key archives is obtained.

[0022] Based on the key file set, we analyze the frequent patterns of its flow path, use the Apriori algorithm to extract high-frequency path rules, and obtain the optimized rule set of path dependency.

[0023] By optimizing the rule set, updating the edge weights of the archive connection graph, and recalculating the overall importance score, the final archive importance ranking is obtained.

[0024] Furthermore, the process of determining whether each file is a key node includes:

[0025] If the importance score in the file connection graph is higher than the preset threshold, the file is marked as a core node based on the score, and a set of core nodes is obtained.

[0026] Based on the core node set, a graph traversal algorithm is used to extract the node attributes of each file, thus obtaining the node attribute distribution;

[0027] By analyzing the distribution of node attributes, the correlation strength of core nodes in the knowledge structure is obtained, and the correlation strength distribution is obtained.

[0028] If the strength of a node in the association strength distribution is lower than a preset threshold, its connection weight in the file connection graph is adjusted through structural analysis to obtain an updated graph structure.

[0029] Based on the updated graph structure, the importance score of each file is recalculated using the PageRank algorithm to obtain the optimized score distribution.

[0030] By optimizing the scoring distribution, the key position of each file in the continuity of the knowledge structure is determined, and the final ranking of core nodes is obtained.

[0031] Based on the final core node sorting, a list of key archive tags is generated to determine the protection set for the continuity of the knowledge structure.

[0032] Furthermore, the process of identifying risk areas that may lead to data silos includes:

[0033] From the associated data marked as key archives, obtain the direct and indirect archive transfer paths, and use the shortest path algorithm to calculate the node sequence of each path to obtain the set of transfer paths;

[0034] Based on the set of flow paths, the node weights and connection strengths in each path are extracted, and the stability score of the path is calculated using a weighted average method to obtain the path stability distribution.

[0035] If the stability score of a path in the path stability distribution is lower than a preset threshold, then potential breakpoints in the path are identified through topological analysis to obtain a set of breakpoints.

[0036] Based on the set of breakpoints, a community detection algorithm is used to divide the risk areas in the archive transfer path, resulting in a set of areas with the risk of data isolation.

[0037] Furthermore, the process of determining the list of non-critical files to be deleted is as follows: based on the file data in the risk area, a set of abnormal files is obtained. If the distance to a file in the abnormal file set exceeds a preset threshold, it is marked as a file to be deleted, thus obtaining a list of files to be deleted.

[0038] Furthermore, the process of obtaining the distance anomaly file set includes:

[0039] Content features and historical transfer records are obtained from archival data within the risk area. Text mining methods are used to extract semantic feature vectors for each archival document, resulting in a set of semantic feature vectors.

[0040] If the vectors in the semantic feature vector set deviate from the preset accuracy standard, the list of non-critical files is determined by comparing the operation logs in the historical flow records.

[0041] Based on the set of semantic feature vectors, cosine similarity is used to calculate the similarity between the semantic vector of each file and the preset standard vector, and the similarity distribution is obtained.

[0042] If the similarity is lower than the preset threshold, the corresponding file is marked as a file to be reviewed, and a set of files to be reviewed is obtained;

[0043] The operation logs of each file in the set of files to be reviewed are extracted, and the operation frequency and modification cycle are calculated by time series analysis to obtain a set of operation behavior characteristics.

[0044] If the frequency of operations in the set of operational behavior characteristics is lower than a preset threshold or the modification cycle exceeds a preset range, the file is determined to be a non-critical file, and a set of non-critical files is obtained.

[0045] Based on the set of non-critical archives, the metadata attributes of each archive are extracted, and the archives are grouped according to the metadata similarity using cluster analysis to obtain the archive group set;

[0046] For each group of files, the mean of the semantic feature vectors of the files within the group is calculated to obtain the set of grouped feature vectors;

[0047] The feature vectors of each group are extracted from the set of grouped feature vectors. The vector distance calculation method is used to determine the distance distribution between the files in the group and the group feature vectors, thus obtaining the set of files with abnormal distances.

[0048] Furthermore, the process of obtaining the new information network structure includes:

[0049] The backup data of each file is extracted from the final deletion list, and the metadata and content data of the file are separated by data parsing methods to obtain the file backup set;

[0050] If the metadata in the archive backup set matches the preset sensitive data template, it is marked as a high-risk archive, and a high-risk archive set is obtained;

[0051] Extract circulation paths from high-risk archives sets, use path analysis methods to calculate the scope of archive dissemination in information networks, and determine the set of dissemination scopes;

[0052] Based on the propagation range set, the archive node and its associated edge are located in the connection graph using the graph traversal method to obtain the node association set;

[0053] If the strength of the associated edges in the node association set is lower than a preset threshold, it is marked as a weak association file, and a weak association file set is obtained.

[0054] Archive nodes are extracted from a set of weakly associated archives, and the edge weights in the connection graph are adjusted using a dynamic update mechanism to obtain the updated information network structure.

[0055] Furthermore, the process of reconstructing connections using the completion technique includes:

[0056] Node sequences are extracted from the archive transfer path, and feature vectors of the node sequences are generated using sequence parsing methods to obtain a set of node features;

[0057] If the feature vectors in the node feature set do not match the preset path pattern, they are marked as abnormal node sequences, and an abnormal sequence set is obtained.

[0058] For the set of abnormal sequences, cluster analysis is used to group the abnormal node sequences, resulting in a set of abnormal sequence groups;

[0059] By grouping abnormal sequences into sets, data mining methods are used to find potential related nodes, resulting in a set of potential nodes.

[0060] The associated nodes are extracted from the potential node set, and a low-dimensional representation of the nodes is generated using the node embedding method to obtain the node embedding set;

[0061] If the distance between the embedding vectors in the node embedding set is lower than a preset threshold, then the completion path is generated by the vector concatenation method, and the completion path set is obtained.

[0062] Based on the completed path set, the connectivity of the paths is checked using a path verification method to obtain a set of connected paths.

[0063] A geological archive structure integrity assurance system based on dynamic map maintenance includes:

[0064] The map construction and evaluation module establishes and maintains a relational database of geological archives; constructs an initial archive connection map using graph database technology; calculates the comprehensive importance score of the archives and generates a set of key archives; and marks core nodes to generate a marked list of key archives.

[0065] The risk area identification module obtains the direct and indirect connection paths of key files, calculates the stability score of the paths, identifies potential breakpoints, and delineates areas with data isolation risks through topology analysis and community detection algorithms.

[0066] The non-critical archive identification module performs a deep scan of archives within the risk area, extracts content features and historical transfer records, analyzes operation frequency and modification cycle, determines the set of non-critical archives, generates a list of archives to be deleted, and assesses the impact of deletion operations.

[0067] The graph dynamic maintenance module extracts backup data, separates metadata and content data, adjusts edge weights, updates the information network structure, and ensures network connectivity; it also updates the graph structure in real time to adapt to dynamic changes.

[0068] The path continuity assurance module extracts node sequences and identifies abnormal node sequences; it generates complete paths and optimizes file transfer paths through vector concatenation and path verification; and it calculates connectivity indicators to ensure the continuity of the knowledge structure.

[0069] Compared with the prior art, the present invention has the following advantages:

[0070] 1. This invention forms an initial archive connection map through a pre-established archive association database and scores the importance of each archive in the overall information network to determine whether it is a key node. For the surrounding related data of key archives, it obtains the direct and indirect archive flow paths, analyzes potential breakpoints in the paths, identifies risk areas that may cause data isolation problems, and performs in-depth scanning of the archive data in the risk areas to obtain the content characteristics and historical flow records of each archive, determine whether they meet the data accuracy standards, and then determine the list of non-key archives to be deleted. This process ensures the rationality of the deletion operation through multi-dimensional analysis and avoids the accidental deletion of key information.

[0071] 2. After determining the list of non-critical archives, this invention further analyzes the scope of impact of each archive to be deleted, and obtains the potential impact of its deletion on the stability of the overall system. This further ensures that the deletion operation will not cause irreparable damage to the integrity of the archive system. For the final deletion list, backup data and flow path logs of each archive are obtained, and the archive connection map is updated through a dynamic maintenance mechanism. This dynamic maintenance mechanism can adjust the archive connection relationship in a timely manner, ensuring that the structure of the archive system remains intact after the deletion operation. By analyzing the continuity of the archive flow path, if there are any breaks, the connection is reconstructed through technical means, further ensuring the continuity of the archive flow path and avoiding the problem of path breakage caused by the deletion operation.

[0072] 3. In summary, this invention ensures the rationality of deletion operations and avoids the accidental deletion of key information through steps such as importance scoring, in-depth scanning, and impact range analysis, while also taking into account data accuracy. Secondly, through dynamic maintenance mechanisms and path continuity completion, it guarantees the integrity and stability of the archival system, enabling it to adapt to the dynamic changes in complex data relationships. Finally, this technical solution not only improves the efficiency and data accuracy of geological archive management, but also provides a more scientific basis for industry decision-making in fields such as natural resource development, disaster early warning, and scientific research, and has important application value and practical significance. Attached Figure Description

[0073] Figure 1 This is a flowchart of a method for ensuring the structural integrity of geological archives based on dynamic maintenance of maps, as described in Example 1.

[0074] Figure 2 This is a system framework diagram of a geological archive structure integrity assurance system based on dynamic map maintenance, as shown in Example 2. Detailed Implementation

[0075] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. Example

[0076] like Figure 1 As shown, this embodiment provides a method for ensuring the structural integrity of geological archives based on dynamic map maintenance, including the following steps:

[0077] By using a pre-established database of archival connections, an initial archival connection map is formed, and the importance score of each archival document in the overall information network is analyzed.

[0078] Based on the importance score in the file connection map, determine whether each file is a key node;

[0079] For the surrounding related data of the marked key archives, obtain the archive circulation paths that are directly and indirectly connected, analyze the potential breakpoints in the paths, and identify risk areas that may cause data silos.

[0080] By conducting in-depth scanning of the archival data within the risk area, the content characteristics and historical circulation records of each archival document are obtained, and it is determined whether they meet the data accuracy standards to identify a list of non-critical archives that need to be deleted.

[0081] Based on the list of non-critical files, analyze the scope of impact of each file to be deleted, obtain the potential impact of its deletion on the stability of the overall system, and determine if the deletion would damage the integrity of the information network, then remove it from the deletion list.

[0082] For the final deletion list, obtain backup data and flow path logs for each file, update the file connection graph through a dynamic maintenance mechanism, and obtain a new information network structure;

[0083] Based on the new information network structure, the continuity of the archive circulation path is analyzed, and if there are any breaks, the connection is rebuilt using the completion technology.

[0084] Furthermore, the process of obtaining its importance score in the overall information network includes:

[0085] By using a pre-defined archive association database, the flow paths and dependencies between geological archives are obtained. Using graph database technology, an initial archive connection graph is constructed to obtain the data structure of nodes and edges.

[0086] Based on the archive connection map, the node attributes and edge weights of each geological archive are extracted. The PageRank algorithm is used to calculate the preliminary importance score of each archive in the map and obtain the score distribution.

[0087] Specifically, the system connects to the archive association database, queries the metadata of each geological archive, including information such as type, creation time, modification time, keywords and project to which it belongs, and performs standardization processing on the extracted metadata, such as uniformly formatting timestamps and performing word segmentation on keywords.

[0088] Analyze the flow paths between files to determine direct and indirect connections. Assign higher weights (e.g., 1.0) to directly connected files, and decrease the weights for indirectly connected files according to the path length (e.g., 0.5 for path length 2, 0.25 for path length 3, etc.). Calculate the similarity between two files using text similarity analysis (e.g., cosine similarity) and include it as part of the edge weights. Finally, sum the weights to obtain the final edge weights.

[0089] Initialize the importance score of each node to 1, set the damping coefficient d=0.85, and the number of iterations N=100;

[0090] In each iteration, update the importance score of each node:

[0091]

[0092] Where PR(i) is the importance score of node i, Bi is the set of nodes pointing to node i, L(j) is the number of outgoing edges of node j, and N is the total number of nodes in the graph.

[0093] After each iteration, calculate the change in importance scores for all nodes. If the change is less than a set threshold (e.g., 0.001), stop the iteration.

[0094] Output the final importance score for each node, forming a score distribution.

[0095] For example, the process of obtaining its importance score in the overall information network is as follows:

[0096] By using a pre-established archive association database, the circulation path and dependencies of geological archives can be obtained. First, archive metadata, such as archive number, creation time, associated project number, and circulation record, needs to be extracted from the database.

[0097] Assuming the database contains 1000 geological archives, each archive has fields such as "Archive ID", "Project ID", "Transfer Timestamp", and "Associated Archive ID", using graph database technology (such as Neo4j), an initial connection graph is constructed with archives as nodes and transfer relationships as edges. For example, archive A1 transfers to A2, and A2 depends on A3, forming edges A1→A2 and A2→A3. The weights are based on the transfer frequency, and the weight of A1→A2 is set to 0.8 (8 transfers / 10 total transfers).

[0098] Subsequently, multi-dimensional data analysis was performed on each document, including the number of times the document was cited, content complexity (assuming it was calculated using text entropy, with an entropy value ranging from 0 to 8, and A1's entropy value being 5.2), and item relevance (A1 is associated with 3 items, with a weight of 3 / total number of items 10 = 0.3). The PageRank algorithm was used to evaluate the importance of the document in the graph. After calculation, A1's PageRank value was 0.65, indicating its high importance. The importance score was a combination of the number of citations (40%), entropy value (30%), and relevance (30%).

[0099] After all archive scores are normalized (range 0-1), an importance ranking is generated, with A1 ranking in the top 10%. This process creates a logically rigorous archive connection map and importance assessment, supporting subsequent archive management optimization.

[0100] Furthermore, after obtaining the rating distribution, the process also includes obtaining a ranking of file importance, specifically including:

[0101] If the calculated preliminary importance score is lower than the preset threshold, then through multidimensional data analysis, the metadata features of the archive are obtained, and the key feature vectors are extracted using principal component analysis to obtain a feature-enhanced archive representation.

[0102] Based on the feature-enhanced file representation, combined with the flow path and dependencies, a weighted average method is used to fuse multidimensional features and graph scores to calculate a comprehensive importance score, resulting in an updated score distribution.

[0103] The updated rating distribution is used to determine the relative importance of the archives in the information network. If the rating is higher than the preset threshold, it is marked as a key archive, and a set of key archives is obtained.

[0104] Based on the key file set, we analyze the frequent patterns of its flow path, use the Apriori algorithm to extract high-frequency path rules, and obtain the optimized rule set of path dependency.

[0105] By optimizing the rule set, updating the edge weights of the archive connection graph, and recalculating the overall importance score, the final archive importance ranking is obtained.

[0106] Specifically, a preliminary importance score is calculated for each file. If the score is lower than a preset threshold (e.g., 0.5), multidimensional data analysis is used to obtain the file's metadata features, including but not limited to file type, creation time, modification time, keywords, and project to which it belongs. Principal component analysis (PCA) is then used to extract key feature vectors to obtain a feature-enhanced file representation.

[0107] Based on the feature-enhanced archive representation, combined with the archive's circulation path and dependencies, a weighted average method is used to fuse multidimensional features and graph scores; weight coefficients α, β, and γ are set, corresponding to citation count, content complexity (entropy value), and item relevance, respectively;

[0108] Calculate the overall importance score:

[0109] PR'(i) = α × number of citations + β × entropy + γ × item relevance + (1 − α − β − ​​γ) × PR(i), to obtain the updated rating distribution;

[0110] The relative importance of the archive in the information network is determined by the updated rating distribution PR'(i).

[0111] Step 2: Set a threshold θ2 (e.g., 0.7). If PR'(i) > θ2, mark the file as a critical file and add it to the critical file set.

[0112] Based on the key archive set, analyze the frequent patterns of its circulation path;

[0113] The Apriori algorithm is used to extract high-frequency path rules, resulting in an optimized rule set for path dependence. The minimum support (min_sup) and minimum confidence (min_conf) are set, and the Apriori algorithm is used to generate frequent itemsets and association rules.

[0114] By optimizing the rule set, the edge weights of the file connection graph are updated. For each path P, the edge weight wij is adjusted according to the optimized rule set, where wij = wij × rule weight.

[0115] Recalculate the overall importance score PR''(i):

[0116]

[0117] PR''(i) is the final importance score of node i, Bi is the set of nodes pointing to node i, L(j) is the number of outgoing edges of node j, N is the total number of nodes in the graph, and d is the damping coefficient (e.g., 0.85).

[0118] The final ranking of the importance of the archives is obtained.

[0119] For example, suppose there is a database containing 1,000 geological archives, each archive having fields such as "archive ID", "project ID", "transfer timestamp" and "associated archive ID";

[0120] After calculating the initial importance score for each file, it was found that the score PR(A1) of file A1 was 0.45, which was lower than the threshold θ1=0.5;

[0121] Obtain the metadata characteristics of file A1:

[0122] File type: Exploration report

[0123] Creation time: 2023-01-01

[0124] Keywords: geological exploration, mineral analysis

[0125] Project: Project X

[0126] Principal Component Analysis: Key feature vectors are extracted using PCA to obtain a feature-enhanced archival representation.

[0127] By combining the feature-enhanced representation of archive A1 and the path dependency relationship of the flow, a weighted average method is used to fuse multidimensional features and map scores;

[0128] Citations: 5

[0129] Entropy value: 5.2

[0130] Project relevance: 0.3

[0131] Weighting coefficients: α=0.4, β=0.3, γ=0.3

[0132] Calculate the overall importance score:

[0133] PR'(A1)==0.4×5+0.3×5.2+0.3×0.3+(1−0.4−0.3−0.3)×0.45=2+1.56+0.09+0.045=3.695,

[0134] The ratings are normalized to the range [0,1], assuming the normalized rating PR'(A1) = 0.65;

[0135] The relative importance of file A1 is determined by the updated rating distribution PR'(A1)=0.65; if the rating of file A1 is higher than the threshold θ2=0.7, it is marked as a key file and added to the key file set.

[0136] Analyzing the flow paths of files in the key file set reveals that file A1 frequently depends on files A2 and A3; high-frequency path rules are extracted to obtain an optimized rule set for path dependencies, such as the rule "file A1 → file A2 → file A3";

[0137] Minimum support min_sup=0.5;

[0138] Minimum confidence level min_conf=0.7;

[0139] Update the edge weights of the archive connection graph based on the optimization rule set.

[0140] For the path "File A1→File A2→File A3", assume the rule weight is 1.2.

[0141] Adjust edge weights:

[0142] wA1A2=0.8×1.2=0.96, wA2A3=0.7×1.2=0.84;

[0143] Recalculate the overall importance score PR''(A1):

[0144]

[0145] Assumption =0.7, =0.65, L(A2)=2, L(A3)=1, calculated as follows =0.85015, normalize the score to the range [0,1]. Assume the normalized score is... =0.75;

[0146] Through the above steps, the ranking of the importance of the archives was optimized, resulting in the final ranking of the archives' importance, which provides a more accurate basis for decision-making in geological archives management.

[0147] Furthermore, the process of determining whether each file is a key node includes:

[0148] If the importance score in the file connection graph is higher than the preset threshold, the file is marked as a core node based on the score, and a set of core nodes is obtained.

[0149] Based on the core node set, a graph traversal algorithm is used to extract the node attributes of each file, thus obtaining the node attribute distribution;

[0150] By analyzing the distribution of node attributes, the correlation strength of core nodes in the knowledge structure is obtained, and the correlation strength distribution is obtained.

[0151] If the strength of a node in the association strength distribution is lower than a preset threshold, its connection weight in the file connection graph is adjusted through structural analysis to obtain an updated graph structure.

[0152] Based on the updated graph structure, the importance score of each file is recalculated using the PageRank algorithm to obtain the optimized score distribution.

[0153] By optimizing the scoring distribution, the key position of each file in the continuity of the knowledge structure is determined, and the final ranking of core nodes is obtained.

[0154] Based on the final core node sorting, a list of key archive tags is generated to determine the protection set for the continuity of the knowledge structure.

[0155] Specifically, if the importance score PR(i) of a file is greater than the set threshold θ, then the file is marked as a core node;

[0156] Generate a core node set C. Based on the core node set C, use a graph traversal algorithm (such as depth-first search or breadth-first search) to extract the node attributes of each file and obtain the node attribute distribution, including file type, creation time, keywords, and project to which it belongs.

[0157] By analyzing the distribution of node attributes, the association strength of core nodes in the knowledge structure is analyzed, and the association strength S(i) of each core node is calculated:

[0158]

[0159] Where N(i) is the set of nodes directly connected to node i, and wij is the edge weight between node i and node j;

[0160] Set a threshold for association strength θS (e.g., 0.5). If the association strength S(i) of a node is less than θS, then adjust its connection weight in the file connection graph through structural analysis: wij = wij × adjustment factor, where the adjustment factor can be set according to specific needs, such as 1.2 or 0.8.

[0161] Based on the updated graph structure, the importance score of each document was recalculated using the PageRank algorithm:

[0162]

[0163] PR(i) is the importance score of node i, Bi is the set of nodes pointing to node i, L(j) is the number of outgoing edges of node j, N is the total number of nodes in the graph, and d is the damping coefficient (e.g., 0.85).

[0164] By optimizing the scoring distribution, the key position of each file in the continuity of the knowledge structure is determined, and the final core node ranking is generated.

[0165] Based on the final core node sorting, a list of key archive tags is generated to determine the protection set for the continuity of the knowledge structure.

[0166] For example, suppose there is a database containing 1,000 geological archives, each archive having fields such as "archive ID", "project ID", "transfer timestamp" and "associated archive ID";

[0167] After calculating the importance score for each file, it was found that the score PR(A1) of file A1 was 0.8, which was higher than the threshold θ=0.7. Therefore, file A1 was added to the core node set C.

[0168] Using a breadth-first search algorithm, extract the node attributes of file A1:

[0169] File type: Exploration report

[0170] Creation time: 2023-01-01

[0171] Keywords: geological exploration, mineral analysis

[0172] Project Affiliation: Project X;

[0173] Calculate the association strength S(A1) of file A1:

[0174] Assuming that file A1 is directly connected to files A2 and A3, with edge weights wA1A2=0.8 and wA1A3=0.7 respectively, then the association strength S(A1)=wA1A2+wA1A3=1.5;

[0175] Set the association strength threshold θS=0.5. The association strength S(A1) of file A1 is 1.5, which is higher than the threshold and no adjustment is needed.

[0176] Assuming the association strength S(A4) of file A4 is 0.4, which is below the threshold, adjust its connection weight: wA4A5=0.6×1.2=0.72;

[0177] Based on the updated map structure, the importance score PR(A1) of file A1 is recalculated:

[0178]

[0179] Assuming PR(A2)=0.7, PR(A3)=0.65, L(A2)=2, L(A3)=1, then we calculate PR(A1)=0.85015, and the normalized score PR(A1)=0.85;

[0180] Based on the optimized rating distribution, the final core node ranking is generated. File A1 has a rating PR(A1) = 0.85, ranking in the top 10%.

[0181] Based on the final core node sorting, a list of marked key files is generated, with file A1 marked as non-deletable;

[0182] Protection set: Determine that file A1 belongs to the protection set for knowledge structure continuity;

[0183] Through the above steps, the core nodes were identified and marked, ensuring the continuity and stability of the knowledge structure and providing a more accurate basis for decision-making in geological archives management.

[0184] Furthermore, the process of identifying risk areas that may lead to data silos includes:

[0185] From the associated data marked as key archives, obtain the direct and indirect archive transfer paths, and use the shortest path algorithm to calculate the node sequence of each path to obtain the set of transfer paths;

[0186] Based on the set of flow paths, the node weights and connection strengths in each path are extracted, and the stability score of the path is calculated using a weighted average method to obtain the path stability distribution.

[0187] If the stability score of a path in the path stability distribution is lower than a preset threshold, then potential breakpoints in the path are identified through topological analysis to obtain a set of breakpoints.

[0188] Based on the set of breakpoints, a community detection algorithm is used to divide the risk areas in the archive transfer path, resulting in a set of areas with the risk of data isolation.

[0189] Extract the node attributes of each risk region from the region set, calculate the connection strength distribution between nodes, and obtain the association strength set of risk regions.

[0190] If the connection strength of a certain region in the association strength set is lower than the preset threshold, the connection structure of the file flow path is adjusted by reallocating node weights to obtain the optimized path structure.

[0191] Based on the optimized path structure, the stability of each path and the connection strength of risk areas are calculated to obtain the final optimized scheme for the archive transfer path.

[0192] Specifically, from the associated data marked as key archives, the direct and indirect connection archive transfer paths are obtained, and the shortest path algorithm (such as Dijkstra's algorithm or Floyd's algorithm) is used to calculate the node sequence of each path to obtain the set of transfer paths P;

[0193] Based on the set of flow paths P, the node weights wi and connection strengths sij of each path are extracted, and the stability score of the path is calculated using a weighted average method. :

[0194]

[0195] in, It is the k-th path. It is a path The number of edges in the array, where wi is the weight of node i and sij is the connection strength between node i and node j;

[0196] Set a path stability threshold θS (e.g., 0.5). If a path's stability score is... If <θS, then potential breakpoints in the path are identified through topological analysis, resulting in a set of breakpoints B;

[0197] Based on the set of breakpoints B, community detection algorithms (such as the Louvain algorithm or the Girvan-Newman algorithm) are used to delineate risk areas in the archive transfer path, resulting in a set of regions R with the risk of data isolation.

[0198] Extract the node attributes of each risk region from the region set R, calculate the connection strength distribution between nodes, and obtain the association strength set of the risk regions. :

[0199]

[0200] Where r is the r-th risk region, |r| is the number of edges in risk region r, and sij is the connection strength between node i and node j;

[0201] Set a threshold θR for the association strength of risk areas (e.g., 0.3). If the connection strength of a certain area... If the value is less than θR, the connection structure of the file transfer path is adjusted by reallocating node weights.

[0202] Adjust node weights wi:

[0203] wi = wi × adjustment factor, where the adjustment factor can be set according to specific needs, such as 1.2 or 0.8;

[0204] Based on the optimized path structure, the stability of each path and the connection strength of risk areas are recalculated to obtain the final optimized scheme for the archive transfer path.

[0205] For example, suppose there is a database containing 1,000 geological archives, each archive having fields such as "archive ID", "project ID", "transfer timestamp" and "associated archive ID";

[0206] Obtain the direct and indirect link paths of file A1 from the associated data marked as critical files.

[0207] Set of flow paths: Assume the set of paths P={P1,P2}, where P1={A1→A2→A3} and P2={A1→A4→A5};

[0208] Calculate the stability score of path P1:

[0209] Node weights: wA1=0.8, wA2=0.7, wA3=0.6;

[0210] Connection strength: sA1A2=0.9, sA2A3=0.8;

[0211] Stability score: S(P1) = 1 / 2 × (0.8 × 0.9 + 0.7 × 0.8) = 0.64;

[0212] S(P1) = 0.64, S(P2) = 0.55;

[0213] The path stability threshold θS = 0.6 is set, and the stability score S(P2) = 0.55 of path P2 is lower than the threshold.

[0214] Potential breakpoints in path P2 are identified through topological analysis, assuming the breakpoint is A4. Based on breakpoint A4, a community detection algorithm is used to divide risk areas, assuming the risk area set R={R1}, where R1 contains files A4 and A5.

[0215] Calculate the association strength of risk region R1:

[0216] Connection strength: sA4A5=0.4;

[0217] Association strength: =1 / 1 × 0.4 = 0.4;

[0218] Set the risk region association strength threshold θR=0.5, and the connection strength of risk region R1. =0.4 is below the threshold, adjust the weight of file A4: wA4=0.6×1.2=0.72;

[0219] Based on the optimized path structure, recalculate the stability of path P2:

[0220] Adjusted node weights: wA4 = 0.72;

[0221] Connection strength: sA1A4=0.8, sA4A5=0.4;

[0222] The new stability score is calculated as: S(P2) = 0.464;

[0223] The stability score of path P2 improved to 0.464, but it is still below the threshold of 0.6, requiring further optimization or other measures.

[0224] Through the above steps, the identification of risk areas and the optimization of the archive transfer path were completed, ensuring the stability of the archive connection map and the continuity of the knowledge structure, and providing a more accurate decision-making basis for geological archive management.

[0225] Furthermore, the process of determining the list of non-critical files to be deleted is as follows: based on the file data in the risk area, a set of abnormal files is obtained. If the distance to a file in the abnormal file set exceeds a preset threshold, it is marked as a file to be deleted, thus obtaining a list of files to be deleted.

[0226] Furthermore, the process of obtaining the distance anomaly file set includes:

[0227] Content features and historical transfer records are obtained from archival data within the risk area. Text mining methods are used to extract semantic feature vectors for each archival document, resulting in a set of semantic feature vectors.

[0228] If the vectors in the semantic feature vector set deviate from the preset accuracy standard, the list of non-critical files is determined by comparing the operation logs in the historical flow records.

[0229] Based on the set of semantic feature vectors, cosine similarity is used to calculate the similarity between the semantic vector of each file and the preset standard vector, and the similarity distribution is obtained.

[0230] If the similarity is lower than the preset threshold, the corresponding file is marked as a file to be reviewed, and a set of files to be reviewed is obtained;

[0231] The operation logs of each file in the set of files to be reviewed are extracted, and the operation frequency and modification cycle are calculated by time series analysis to obtain a set of operation behavior characteristics.

[0232] If the frequency of operations in the set of operational behavior characteristics is lower than a preset threshold or the modification cycle exceeds a preset range, the file is determined to be a non-critical file, and a set of non-critical files is obtained.

[0233] Based on the set of non-critical archives, the metadata attributes of each archive are extracted, and the archives are grouped according to the metadata similarity using cluster analysis to obtain the archive group set;

[0234] For each group of files, the mean of the semantic feature vectors of the files within the group is calculated to obtain the set of grouped feature vectors;

[0235] The feature vectors of each group are extracted from the set of grouped feature vectors. The vector distance calculation method is used to determine the distance distribution between the files in the group and the group feature vectors, thus obtaining the set of files with abnormal distances.

[0236] For example, suppose there is a database containing 1,000 geological archives, each archive having fields such as "archive ID", "project ID", "transfer timestamp" and "associated archive ID";

[0237] Obtain the content characteristics and historical transfer records of archive A1 from the archive data within the risk area;

[0238] Content features: keywords, abstract, main text, etc.;

[0239] Historical process flow records: process flow timestamps, operation logs, etc.;

[0240] The semantic feature vector of file A1 is extracted using text mining methods, resulting in a set of semantic feature vectors;

[0241] Semantic feature vector: Suppose the semantic feature vector of file A1 is vA1=[0.1,0.2,0.3,0.4];

[0242] If the semantic feature vector of file A1 deviates from the preset accuracy standard, the list of non-critical files is determined by comparing the operation log in the historical transfer record. The operation log records that file A1 has not been modified in the past 6 months.

[0243] The similarity between the semantic vector of file A1 and the preset standard vector is calculated using cosine similarity. The preset standard vector is:

[0244] vstd=[0.2,0.3,0.4,0.5];

[0245] Similarity:

[0246]

[0247] The similarity distribution was found to be similarity(A1) = 0.97;

[0248] Set the similarity threshold θsim=0.7. File A1 has a similarity of 0.97 > θsim, so it is not marked as a file to be reviewed.

[0249] In another scenario, assuming file A2 has a similarity of 0.6, it is marked as a file to be reviewed.

[0250] The operation log of file A2 was extracted from the set of files to be reviewed. The operation frequency and modification cycle were calculated using time series analysis. The operation frequency of file A2 in the past 6 months was once a month, and the modification cycle of file A2 was every 4 months.

[0251] If the operation frequency threshold θfreq = 2 times / month and the modification cycle threshold θcycle = 3 months is set, and the operation frequency of file A2 is 1 time / month, which is lower than θfreq, and the modification cycle of file A2 is 4 months, which exceeds θcycle, then file A2 is judged as a non-critical file.

[0252] Based on the set of non-critical archives, the metadata attributes of archive A2 are extracted. Cluster analysis is used to group the archives according to the metadata similarity. The metadata attributes include archive type, creation time, project to which they belong, etc. Assuming that archives A2 and A3 are assigned to the same group, the mean of the semantic feature vector of the archives in each group is calculated to obtain the set of group feature vectors.

[0253] Assuming the mean semantic feature vector of the files within a group is vgroup=[0.2,0.3,0.4,0.5], the feature vector of each group is extracted from the set of group feature vectors, and the distance distribution between the files within the group and the group feature vector is determined by the vector distance calculation method.

[0254] Vector distance:

[0255]

[0256] Right now

[0257] Set the distance threshold θdist=0.5. If the distance of file A2 is 0.2<θdist, it will not be marked as a file with abnormal distance.

[0258] Assuming file A4 has a distance of 0.6, which exceeds the threshold θdist=0.5, it is marked as a file to be deleted, and file A4 is added to the list of files to be deleted;

[0259] Through the above steps, the list of non-critical archives that need to be deleted was determined, ensuring the stability of the archive connection map and the continuity of the knowledge structure, and providing a more accurate basis for decision-making in geological archive management.

[0260] Furthermore, the process of obtaining the new information network structure includes:

[0261] The backup data of each file is extracted from the final deletion list, and the metadata and content data of the file are separated by data parsing methods to obtain the file backup set;

[0262] If the metadata in the archive backup set matches the preset sensitive data template, it is marked as a high-risk archive, and a high-risk archive set is obtained;

[0263] Extract circulation paths from high-risk archives sets, use path analysis methods to calculate the scope of archive dissemination in information networks, and determine the set of dissemination scopes;

[0264] Based on the propagation range set, the archive node and its associated edge are located in the connection graph using the graph traversal method to obtain the node association set;

[0265] If the strength of the associated edges in the node association set is lower than a preset threshold, it is marked as a weak association file, and a weak association file set is obtained.

[0266] File nodes are extracted from a weakly associated set of files, and the edge weights in the connection graph are adjusted using a dynamic update mechanism to obtain the updated information network structure.

[0267] For the updated information network structure, topology analysis is used to calculate network connectivity and determine the stability of the new information network structure.

[0268] For example, suppose there is a database containing 1,000 geological archives, each archive having fields such as "archive ID", "project ID", "transfer timestamp" and "associated archive ID";

[0269] The backup data of files A1 and A2 were extracted from the final deletion list. The backup data of files A1 and A2 includes metadata and content data.

[0270] If the metadata of files A1 and A2 is checked and found to match the preset sensitive data template, then file A1 is marked as a high-risk file.

[0271] Extract the circulation path of file A1 from the high-risk file set. The propagation range of file A1 includes files A2, A3 and A4. Use the breadth-first search method to locate file A1 and its associated edges in the connection graph. The associated edges of file A1 include A1→A2, A1→A3 and A1→A4.

[0272] Set the associated edge strength threshold θedge=0.5, and the associated edge strengths: wA1A2=0.6, wA1A3=0.4, wA1A4=0.3;

[0273] Files A3 and A4 have an edge strength below the threshold and are therefore marked as weakly associated files.

[0274] Extract the nodes of files A3 and A4 from a weakly related set of files.

[0275] Dynamic update mechanism: Adjust the edge weights of files A3 and A4: wA1A3=0.4×1.2=0.48; wA1A4=0.3×1.2=0.36. The updated information network structure: the new edge weights are wA1A3=0.48 and wA1A4=0.36.

[0276] The connectivity of the updated information network structure is calculated using topology analysis.

[0277] Calculate network connectivity metrics, such as the number of connected components and the size of the largest connected component.

[0278] If the connectivity index meets the preset conditions (e.g., the largest connected component contains more than 80% of the nodes), then the new information network structure is judged to be stable.

[0279] Through the above steps, the generation and stability assessment of the new information network structure were completed, ensuring the stability of the archive connection map and the continuity of the knowledge structure, and providing a more accurate basis for decision-making in geological archive management.

[0280] Furthermore, the process of reconstructing connections using the completion technique includes:

[0281] Node sequences are extracted from the archive transfer path, and feature vectors of the node sequences are generated using sequence parsing methods to obtain a set of node features;

[0282] If the feature vectors in the node feature set do not match the preset path pattern, they are marked as abnormal node sequences, and an abnormal sequence set is obtained.

[0283] For the set of abnormal sequences, cluster analysis is used to group the abnormal node sequences, resulting in a set of abnormal sequence groups;

[0284] By grouping abnormal sequences into sets, data mining methods are used to find potential related nodes, resulting in a set of potential nodes.

[0285] The associated nodes are extracted from the potential node set, and a low-dimensional representation of the nodes is generated using the node embedding method to obtain the node embedding set;

[0286] If the distance between the embedding vectors in the node embedding set is lower than a preset threshold, then the completion path is generated by the vector concatenation method, and the completion path set is obtained.

[0287] Based on the completed path set, the connectivity of the paths is checked using a path verification method to obtain a set of connected paths;

[0288] If the path coverage in the connected path set is higher than a preset threshold, it is marked as a stable path, and a stable path set is obtained.

[0289] Path dependency features are extracted from the set of stable paths, and the dependency weights between paths are calculated using association rule mining methods to obtain a set of path dependency weights.

[0290] If the weight values ​​in the path-dependent weight set meet the preset conditions, the knowledge structure is updated by adjusting the weights to obtain the updated knowledge structure.

[0291] To address the updating of knowledge structures, network analysis methods are used to calculate the connectivity indices of the structures, resulting in a set of connectivity indices.

[0292] If the index values ​​in the connectivity index set are higher than the preset threshold, it is determined that the continuity of the knowledge structure is maintained.

[0293] By using a set of connectivity indicators and a path optimization method to adjust the node order of the path, an optimized path set is obtained.

[0294] For example, suppose we have a database containing 1,000 geological archives, each with fields such as "archive ID", "project ID", "transfer timestamp" and "associated archive ID";

[0295] Extract the node sequence P={A1→A2→A3} from the file transfer path;

[0296] The feature vectors of the node sequence are generated using the sequence parsing method, resulting in the node feature sets vA1=[0.1,0.2,0.3], vA2=[0.2,0.3,0.4], and vA3=[0.3,0.4,0.5].

[0297] The feature vectors in the node feature set are matched with the preset path pattern. If vA3 does not match the preset path pattern, then A3 is marked as an abnormal node sequence.

[0298] Cluster analysis is used to group the abnormal node sequences, resulting in a set of abnormal sequence groups. It is assumed that A3 and A4 are grouped into the same group.

[0299] By grouping anomalous sequences into sets, data mining methods are used to find potential related nodes. It is assumed that A5 and A6 are potentially related to A3 and A4.

[0300] The node embedding method is used to generate low-dimensional representations of nodes, resulting in the following node embedding sets: vA3=[0.3,0.4,0.5], vA5=[0.4,0.5,0.6], vA6=[0.5,0.6,0.7];

[0301] Calculating the distance between embedding vectors in the node embedding set reveals that the distance between vA3 and vA5 is less than the threshold of 0.5.

[0302] The complete path P new={A3→A5} is generated by vector concatenation.

[0303] The connectivity of the completed path P new is checked using a path verification method, and the connectivity of P new is confirmed after verification.

[0304] If the path coverage in the connected path set is higher than 80%, it is marked as a stable path, and Pnew is marked as a stable path.

[0305] Path dependency features were extracted from the stable path set, and the dependency weights between paths were calculated using the association rule mining method, resulting in w A3A5 = 0.8.

[0306] If the weight value in the path dependency weight set is greater than 0.7, the knowledge structure is updated by adjusting the weight. In the updated knowledge structure, the weight of path A3→A5 is 0.8.

[0307] The connectivity index for updating the knowledge structure was calculated using network analysis methods, and the calculated connectivity index was 0.85.

[0308] If the index value in the connectivity index set is greater than 0.8, it is determined that the continuity of the knowledge structure is maintained.

[0309] Using a set of connectivity indicators, the node order of the path is adjusted by a path optimization method. The adjusted path is {A1→A2→A3→A5}.

[0310] Through the above steps, the process of completing the connection reconstruction technology was completed, ensuring the integrity of the archival connection map and the continuity of the knowledge structure, and providing a more accurate basis for decision-making in geological archives management.

[0311] This method also includes:

[0312] By performing a comprehensive verification of the reconnected information network, obtaining the association status of each file, determining whether data isolation issues still exist, and obtaining the final archival system integrity report.

[0313] Furthermore, the process of obtaining the final archival system integrity report includes:

[0314] The association feature set of archive nodes is obtained from the information network, and the association strength between nodes is calculated using graph analysis methods to obtain the association strength set;

[0315] If the strength value in the association strength set is lower than the preset threshold, it is marked as a potential isolated data node, and a potential isolated node set is obtained;

[0316] For a set of potentially isolated nodes, clustering analysis is used to group the nodes and generate a set of node groups;

[0317] Context information of isolated data nodes is extracted from the node group set, and low-dimensional embeddings of nodes are generated using vector representation methods to obtain a node embedding set.

[0318] If the distance between the embedding vectors in the node embedding set is lower than a preset threshold, a supplementary connection path is generated by the path reconstruction method to obtain a supplementary path set.

[0319] Based on the supplementary path set, the connectivity of the paths is verified using network connectivity analysis methods to obtain the connected path set;

[0320] Path coverage features are extracted from the set of connected paths, and feature matching methods are used to determine the integrity of the archive system, thus obtaining the integrity assessment results.

[0321] For example, suppose there is a database containing 1,000 geological archives, each archive having fields such as "archive ID", "project ID", "transfer timestamp" and "associated archive ID";

[0322] The associated feature sets of files A1, A2, and A3 are obtained from the reconnected information network, and wA1A2=0.6, wA2A3=0.4, and wA1A3=0.2 are calculated.

[0323] Set the association strength threshold θassoc=0.3. If the association strength w A1A3=0.2<θassoc for file A3, it is marked as a potential isolated node.

[0324] Cluster analysis is used to group the potentially isolated node A3. Assuming that A3 and A4 are grouped into the same group, the contextual information of A3 and A4 is extracted from the node group set to generate low-dimensional embeddings, vA3=[0.2,0.3,0.4] and vA4=[0.3,0.4,0.5].

[0325] Set the distance threshold between embedded vectors θdist=0.5. The calculated distance between v A3 and v A4 is 0.3, which is lower than the threshold. Therefore, the supplementary path A3→A4 is generated.

[0326] The connectivity of the supplementary path A3→A4 was verified using network connectivity analysis, and the path was confirmed to be connected.

[0327] Path coverage features are extracted from the set of connected paths, and feature matching methods are used to determine the integrity of the archive system, confirming that the archive system is complete and there are no data isolation issues.

[0328] Based on the integrity assessment results, a final archival system integrity report is generated, which records in detail the association status of each document, confirming the integrity of the archival system and the absence of data silos.

[0329] Through the above steps, the overall verification of the reconnected information network was completed, ensuring the integrity of the archival system and the continuity of the knowledge structure, and providing a more accurate basis for decision-making in geological archives management. Example

[0330] like Figure 2 As shown, this embodiment provides a geological archive structure integrity assurance system based on dynamic map maintenance, including:

[0331] The map construction and evaluation module establishes and maintains a relational database of geological archives; constructs an initial archive connection map using graph database technology; calculates the comprehensive importance score of the archives and generates a set of key archives; marks core nodes and generates a marked list of key archives.

[0332] The risk area identification module obtains the direct and indirect connection paths of key files, calculates the stability score of the paths, identifies potential breakpoints, and delineates areas with data isolation risks through topology analysis and community detection algorithms.

[0333] The non-critical archive identification module performs a deep scan of archives within the risk area, extracts content features and historical transfer records, analyzes operation frequency and modification cycle, determines the set of non-critical archives, generates a list of archives to be deleted, and assesses the impact of deletion operations.

[0334] The graph dynamic maintenance module extracts backup data, separates metadata and content data, adjusts edge weights, updates the information network structure, and ensures network connectivity; it also updates the graph structure in real time to adapt to dynamic changes.

[0335] The path continuity assurance module extracts node sequences and identifies abnormal node sequences; it generates complete paths and optimizes file transfer paths through vector concatenation and path verification; and it calculates connectivity indicators to ensure the continuity of the knowledge structure.

[0336] Furthermore, the graph construction and evaluation module automatically backs up the current graph state each time the archive connection graph is updated, so that it can be quickly restored in case of problems; records all operation logs during the graph construction and update process to facilitate subsequent auditing and problem investigation; monitors performance indicators such as construction time and memory usage during the graph construction and evaluation process to ensure efficient system operation; and provides a visual interface for the graph to display archive nodes and connection relationships, so that users can intuitively understand the graph structure.

[0337] Furthermore, when the risk area identification module identifies a potential risk area, it automatically sends an early warning notification to the administrator, indicating possible data isolation issues; it dynamically adjusts the identification criteria for risk areas based on real-time data to ensure the accuracy and timeliness of the identification results; it generates a detailed report that records the identification process and results of the risk areas, including specific information on potential breakpoints and risk areas; and it allows the administrator to manually mark or adjust risk areas, increasing the system's flexibility.

[0338] Furthermore, the non-critical file identification module records the identification and deletion operations of non-critical files, including operation time, operator, and reason for operation; when generating a list of files to be deleted, it provides a detailed deletion impact assessment report, including the impact on related files and knowledge structures; before performing the deletion operation, it requires administrator confirmation to ensure the correctness and security of the operation; and it provides a recovery mechanism for accidentally deleted files to ensure that data can be quickly recovered after an accidental operation.

[0339] Furthermore, the dynamic maintenance module of the graph automatically backs up the current graph state each time the graph is updated to ensure data security; automatically optimizes the graph structure according to the system operation to improve query and update efficiency; monitors the connectivity and stability of the graph in real time to promptly identify and fix potential problems; and provides a user feedback mechanism to allow users to report problems in the graph or make suggestions for improvement.

[0340] Furthermore, when the path continuity assurance module detects path connectivity issues, it automatically sends an early warning notification to the administrator, prompting for necessary measures; provides path optimization suggestions to help the administrator improve the connectivity and stability of the graph; provides a visual display of the path, allowing users to intuitively understand the path connectivity and optimization effects; and allows users to customize path connectivity rules, increasing the system's flexibility and adaptability.

[0341] The specific embodiments of the invention have been described in detail above, but these are merely examples. The invention is not limited to the specific embodiments described above. Those skilled in the art should understand that the embodiments and descriptions in the specification are only illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for guaranteeing the integrity of a geological archive structure based on dynamic maintenance of a map, characterized in that, The method comprises the following steps: An initial archive connection graph is formed by a pre-established archive association database, and the importance score of each archive in the overall information network is analyzed; Whether each archive is a key node is determined according to the importance score in the archive connection graph; For the peripheral associated data of the key archive, the archive flow path directly and indirectly connected therewith is obtained, potential breaking points in the path are analyzed, and a risk area that may cause data isolation is obtained; The content features and historical flow records of each archive are obtained by deep scanning of the archive data in the risk area, it is judged whether the archive meets the data accuracy standard, and a non-key archive list to be deleted is determined; According to the non-key archive list, the operation influence range of each archive to be deleted is analyzed, the potential impact of the archive on the overall system stability after deletion is obtained, and if the deletion will cause damage to the integrity of the information network, the archive is removed from the deletion list; For the final deletion list, the backup data and flow path log of each archive are obtained, the archive connection graph is updated through a dynamic maintenance mechanism, and a new information network structure is obtained; According to the new information network structure, the continuity of the archive flow path is analyzed, and if there is a breaking point, the connection is reconstructed.

2. The method according to claim 1, wherein, The process of obtaining the importance score of each archive in the overall information network comprises: The flow path and dependency relationship between the geological archives are obtained through a pre-established archive association database, an initial archive connection graph is constructed by using a graph database technology, and the data structure of nodes and edges is obtained; According to the archive connection graph, the node attributes and edge weights of each geological archive are extracted, the preliminary importance score of each archive in the graph is calculated, and a score distribution is obtained.

3. The method according to claim 2, wherein, After obtaining the score distribution, the archive importance ranking is obtained, specifically comprising: If the calculated preliminary importance score is lower than a preset threshold, the metadata features of the archive are obtained through multidimensional data analysis, the key feature vector is extracted, and the archive representation is enhanced by features; According to the archive representation enhanced by features, the flow path and dependency relationship are combined, the multidimensional features and graph scores are fused, the comprehensive importance score is calculated, and an updated score distribution is obtained; Through the updated score distribution, the relative importance of the archive in the information network is judged, if the score is higher than a preset threshold, the archive is marked as a key archive, and a key archive set is obtained; According to the key archive set, the frequent pattern of the flow path is analyzed, the high-frequency path rule is extracted, and an optimization rule set of path dependence is obtained; Through the optimization rule set, the edge weight of the archive connection graph is updated, the comprehensive importance score is recalculated, and the final archive importance ranking is obtained.

4. The method of claim 1, wherein, The process of determining whether each archive is a key node comprises: If the importance score in the archive connection graph is higher than a preset threshold, the archive is marked as a core node through score judgment, and a core node set is obtained; According to the core node set, the node attributes of each archive are extracted, and a node attribute distribution is obtained; Through the node attribute distribution, the association strength of the core node in the knowledge structure is analyzed, and an association strength distribution is obtained; If the strength of a node in the association strength distribution is lower than a preset threshold, the connection weight of the node in the archive connection graph is adjusted through structural analysis to obtain an updated graph structure; According to the updated graph structure, the importance score of each archive is recalculated to obtain an optimized score distribution; Through the optimized score distribution, the key position of each archive in the knowledge structure continuity is determined to obtain a final core node ranking; According to the final core node ranking, a marking list of key archives is generated to determine a protection set of the knowledge structure continuity.

5. The method of claim 1, wherein, The process of obtaining the risk area that may cause the data isolation problem includes: From the associated data marked as the key archive, a direct connection and an indirect connection archive flow path are obtained, a node sequence of each path is calculated, and a flow path set is obtained; According to the flow path set, the node weight and the connection strength in each path are extracted, the stability score of the path is calculated, and a path stability distribution is obtained; If the stability score of a path in the path stability distribution is lower than a preset threshold, a potential breaking point in the path is identified through topological analysis to obtain a breaking point set; According to the breaking point set, the risk area in the archive flow path is divided to obtain a region set with a data isolation risk.

6. The method of claim 1, wherein, The process of determining the list of non-key archives to be deleted is: according to the archive data in the risk area, a distance abnormal archive set is obtained, if the distance of the archive in the distance abnormal archive set exceeds a preset threshold, the archive is marked as a to-be-deleted archive to obtain a to-be-deleted archive list.

7. The method according to claim 6, wherein, The process of obtaining the distance abnormal archive set includes: From the archive data in the risk area, content features and historical flow records are obtained, semantic feature vectors of each archive are extracted, and a semantic feature vector set is obtained; According to the semantic feature vector set, the similarity between the semantic vector of each archive and a preset standard vector is calculated by using the cosine similarity to obtain a similarity distribution; If the similarity is lower than a preset threshold, the corresponding archive is marked as a to-be-reviewed archive to obtain a to-be-reviewed archive set; From the to-be-reviewed archive set, the operation log of each archive is extracted, the operation frequency and the modification period are calculated, and an operation behavior feature set is obtained; If the operation frequency in the operation behavior feature set is lower than a preset threshold or the modification period exceeds a preset range, the archive is determined as a non-key archive to obtain a non-key archive set; According to the non-key archive set, the metadata attribute of each archive is extracted, and the archives are grouped according to the metadata similarity to obtain an archive grouping set; For each group of archives, the mean value of the semantic feature vectors of the archives in the group is calculated to obtain a grouping feature vector set; From the grouping feature vector set, the feature vector of each group is extracted, the distance distribution between the archives in the group and the grouping feature vector is judged, and a distance abnormal archive set is obtained.

8. The method of claim 1, wherein, The process of obtaining the new information network structure includes: From the finally determined deletion list, the backup data of each archive is extracted, the metadata and the content data of the archive are separated, and an archive backup set is obtained; If the metadata in the archive backup set matches a preset sensitive data template, the archive is marked as a high-risk archive to obtain a high-risk archive set; Extract the flow path from the high-risk archive set, calculate the propagation range of the archive in the information network, and determine the propagation range set; According to the propagation range set, locate the archive node and its associated edge in the connection graph, and obtain the node association set; If the association edge in the node association set is lower than the preset threshold, it is marked as a weakly associated archive, and a weakly associated archive set is obtained; Extract the archive node from the weakly associated archive set, adjust the edge weight in the connection graph using a dynamic updating mechanism, and obtain the updated information network structure.

9. The method of claim 1, wherein, The process of the completion technology to reconstruct the connection includes: Extract the node sequence from the archive flow path, generate the feature vector of the node sequence, and obtain the node feature set; If the feature vector in the node feature set does not match the preset path pattern, it is marked as an abnormal node sequence, and an abnormal sequence set is obtained; For the abnormal sequence set, group the abnormal node sequences to obtain an abnormal sequence grouping set; Through the abnormal sequence grouping set, find the potential associated nodes to obtain a potential node set; Extract the associated nodes from the potential node set, generate the low-dimensional representation of the nodes, and obtain a node embedding set; If the distance between the embedding vectors in the node embedding set is lower than the preset threshold, a completed path is generated, and a completed path set is obtained; According to the completed path set, check the connectivity of the path to obtain a connected path set.

10. A geological archive structure integrity assurance system based on atlas dynamic maintenance, characterized in that, It includes: A graph construction and evaluation module establishes and maintains the association database of geological archives; An initial archive connection graph is constructed using graph database technology; Calculate the comprehensive importance score of the archive to generate a key archive set; Label the key nodes to generate a label list of key archives; A risk area identification module obtains the direct and indirect connection paths of key archives, calculates the stability score of the path, identifies potential breaking points, and divides the area with data isolation risk through topology analysis and community detection algorithm; A non-key archive identification module scans the archives in the risk area in depth, extracts content features and historical flow records, analyzes operation frequency and modification period, determines a non-key archive set, generates a list of archives to be deleted, and evaluates the impact range of the deletion operation; A graph dynamic maintenance module extracts backup data, separates metadata and content data; Adjust the edge weight and update the information network structure to ensure network connectivity; Real-time update of graph structure to adapt to dynamic changes; A path continuity guarantee module extracts node sequences, identifies abnormal node sequences, generates completed paths through vector splicing and path verification, optimizes archive flow paths, and calculates connectivity indicators to ensure the continuity of the knowledge structure.

Images