A historical document version knowledge ontology dynamic collaborative construction method and system

Abstract

The application discloses a historical document version knowledge ontology dynamic collaborative construction method and system, relates to the technical field of humanistic knowledge graph, and comprises the following steps: inputting a version feature set into a multi-node cloud collaborative processing architecture, calculating local similarity scores between versions, and outputting a version similarity matrix; utilizing a distributed conflict detection mechanism to evaluate the consistency of the version similarity matrix, marking entity mapping pairs with a confidence level lower than a confidence threshold, generating a conflict transaction log; inputting the conflict transaction log into a five-level contradiction hierarchical processing protocol, outputting arbitration results, converting the arbitration results into corresponding OWL ontology update statements, adding version trace annotations, and outputting historical document update instructions; and according to the historical document update instructions, dynamically expanding an entity relationship network of a historical document knowledge ontology, and outputting a historical document version knowledge ontology through a three-layer intelligent analysis framework. The application significantly enhances the automation and reliability of multi-version document knowledge management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of human knowledge graph, and particularly relates to a historical literature version knowledge ontology dynamic collaborative construction method and system. BACKGROUND

[0002] Under the background of the continuous development of digital humanities research and historical literature information engineering, how to realize the structured modeling and semantic unification between historical literature versions has become an important research direction in the field of knowledge engineering and semantic computing. In recent years, knowledge graph and ontology construction technology has been widely used in semantic extraction, relationship modeling and semantic reasoning of literature resources, and especially shows strong knowledge fusion capability in processing large-scale heterogeneous text resources. In the historical literature scene, in view of the entity naming difference, information addition and deletion evolution, semantic drift and other problems existing between different versions of literature, the academia has proposed some preliminary solutions based on ontology alignment, entity mapping or version comparison.

[0003] The prior art often uses simple merging rules or manual correction based methods to deal with entity mapping conflicts and version tracing and other complex semantic problems, resulting in rough processing granularity and low efficiency in version consistency evaluation and semantic change tracking. Compared with the prior art, when multiple versions of historical literature need to be constructed into knowledge ontology with traceability, dynamic evolution and conflict self-processing capability, the prior art still has optimization space in multi-node collaborative processing, semantic consistency verification and automatic ontology update expression. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a historical literature version knowledge ontology dynamic collaborative construction method to solve the efficient collaboration problem of multi-version entity conflict automatic arbitration and dynamic ontology update.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In the first aspect, the present application provides a historical literature version knowledge ontology dynamic collaborative construction method, which comprises collecting multi-source historical literature digital versions and preprocessing to generate a version structured literature set;

[0008] From the version structured literature set, the knowledge entity features in each version literature are extracted to construct a version feature set;

[0009] The version feature set is input into a multi-node cloud collaborative processing architecture, the local similarity scores between versions are calculated, and a version similarity matrix is output;

[0010] The distributed conflict detection mechanism is used for consistency evaluation of the version similar matrix, and an entity mapping pair with a confidence lower than a confidence threshold is marked to generate a conflict transaction log;

[0011] The conflict transaction log is input into a five-level contradiction hierarchical processing protocol, an arbitration result is output, converted into a corresponding OWL ontology update statement, and a version trace annotation is attached, and a historical literature update instruction is output;

[0012] According to the historical literature update instruction, the entity relationship network of the historical literature knowledge ontology is dynamically expanded, and a three-layer intelligent analysis framework is used to output the historical literature version knowledge ontology.

[0013] As a preferred scheme of the historical literature version knowledge ontology dynamic collaborative construction method, the preprocessing is performed to generate a version structured literature set, and the specific steps are as follows,

[0014] The multi-source historical literature digital version is uniformly converted to obtain a standardized literature input data set, and a character recognition operation is performed to obtain an initial recognition result.

[0015] The initial recognition result is subjected to text error correction and layout structure recovery processing to obtain a text information set, and metadata information extraction and labeling are performed to obtain a version structured literature set.

[0016] As a preferred scheme of the historical literature version knowledge ontology dynamic collaborative construction method, the preprocessing is performed to generate a version structured literature set, and the specific steps are as follows,

[0017] Named entity recognition is performed on each historical literature in the version structured literature set, five types of knowledge entities are extracted and standardized to form a standardized entity set, and subject-predicate-object triples are constructed by combining semantic role labeling and dependency syntax analysis.

[0018] The same five types of knowledge entities in different versions are matched and associated by using the subject-predicate-object triples, an aligned entity relationship graph is output, multi-dimensional feature vectors of the five types of knowledge entities are extracted, and standardization processing is performed to output a standardized feature vector set.

[0019] The standardized feature vector set is subjected to entity feature weighted average aggregation according to the literature version to construct a version-level feature matrix, and a mapping relationship between the row vectors in the version-level feature matrix and the corresponding version identifiers is established to output a version feature set.

[0020] As a preferred scheme of the historical literature version knowledge ontology dynamic collaborative construction method, the preprocessing is performed to generate a version structured literature set, and the specific steps are as follows,

[0021] The version feature set is input into the multi-node cloud collaborative processing architecture, the master node in the multi-node cloud collaborative processing architecture receives the version feature set, distributes the version feature set to the collaborative node cluster according to the version identifier, respectively calculates the Euclidean distance of each historical literature version feature vector, generates a local similarity score through a nonlinear conversion, adds a confidence weight label, and generates a local similarity matrix;

[0022] All local similarity matrices are aggregated according to the confidence weight through a distributed reduction operation to generate a global version similarity matrix, the version identifier is reversely mapped to the row and column indexes of the global version similarity matrix, and the version similarity matrix is output.

[0023] As a preferred scheme of the historical literature version knowledge ontology dynamic collaborative construction method, the specific steps of generating the conflict transaction log are as follows,

[0024] The version similarity matrix is divided into multiple entity mapping pairs by the master node, and is distributed to each collaborative node.

[0025] After receiving the entity mapping pair, each collaborative node performs version consistency verification based on the adjacency version entity consistency strategy, performs semantic consistency analysis combined with entity context semantic information, identifies semantic conflict candidate entity pairs, and forms a structure conflict candidate set and a semantic conflict candidate set.

[0026] The intersection set processing is performed on the structure conflict candidate set and the semantic conflict candidate set, the intersection entity mapping pair is obtained, and the corresponding confidence is extracted for judgment, and when the confidence is lower than the confidence threshold, the conflict entity mapping pair is marked, and the de-duplication and merging and mapping relationship integration are performed to generate the conflict transaction log.

[0027] As a preferred scheme of the historical literature version knowledge ontology dynamic collaborative construction method, the specific steps of outputting the historical literature update instruction are as follows,

[0028] The conflict transaction log is parsed into a conflict evolution graph, and is input into a five-level contradiction hierarchical processing protocol according to five conflict types of syntax, structure, version, semantics and value.

[0029] Under the five-level contradiction hierarchical processing protocol, the conflict is judged by a context weighted arbitrator based on the entity context semantic track and the semantic stability index, and a judgment result list is output.

[0030] According to the judgment result list, different judgment types are converted into corresponding OWL ontology update statements.

[0031] The OWL ontology update statement is additionally processed with node information, version identifier and conflict level annotation, and is integrated to form a historical literature update instruction.

[0032] As a preferred embodiment of the dynamic collaborative construction method for historical document version knowledge ontology described in this invention, the specific steps for outputting the historical document version knowledge ontology are as follows:

[0033] The historical document update instructions are parsed into incremental update events for entities and relationships;

[0034] Incremental update events are prioritized according to conflict level annotations and version identifiers to generate a queue of pending update events.

[0035] Based on the queue of pending update events, the entity relationship network of the historical document knowledge ontology is updated by inserting, deleting and updating the attributes of entities and relationships through each collaborative node.

[0036] The system utilizes an entity-level semantic verification layer to perform consistency verification on the updated entity relationship network, a version-level association reasoning layer to perform deep semantic reasoning and version difference analysis, and an overall structure optimization layer to perform global structure optimization, outputting a historical document version knowledge ontology.

[0037] Secondly, the present invention provides a dynamic collaborative construction system for historical document version knowledge ontology, including a document acquisition module for acquiring digital versions of historical documents from multiple sources, performing preprocessing, and generating a version structured document set.

[0038] The entity extraction module is used to extract knowledge entity features from each version of the document in the version structured document collection and construct a version feature set.

[0039] The similarity calculation module is used to input the version feature set into the multi-node cloud collaborative processing architecture, calculate the local similarity score between versions, and output the version similarity matrix.

[0040] The conflict detection module is used to perform consistency evaluation on the version similarity matrix using a distributed conflict detection mechanism, mark entity mapping pairs with confidence scores below the confidence threshold, and generate conflict transaction logs.

[0041] The conflict resolution module is used to input the conflict transaction log into the five-level conflict hierarchical resolution protocol, output the arbitration result into the corresponding OWL ontology update statement, attach version traceability annotations, and output historical document update instructions.

[0042] The ontology construction module is used to dynamically expand the entity relationship network of the historical document knowledge ontology according to the historical document update instructions, and output the historical document version knowledge ontology through a three-layer intelligent analysis framework.

[0043] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program, when executed by the processor, implements any step of the method for dynamically constructing a historical document version knowledge ontology according to the first aspect of the present application.

[0044] In a fourth aspect, the present application provides a computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements any step of the method for dynamically constructing a historical document version knowledge ontology according to the first aspect of the present application.

[0045] The present application has the following beneficial effects: through the multi-node distributed conflict detection mechanism, efficient and accurate conflict identification of historical document multi-version entities is achieved; in combination with the five-level contradiction hierarchical processing protocol and the context weighted arbitrator, intelligent adjudication and standardized dynamic updating of complex conflicts are achieved; the additional version traceability ensures the transparency of updating; at the same time, with the aid of unified preprocessing, multi-dimensional semantic feature extraction and weighted aggregation, multi-version similarity parallel computing and a three-layer intelligent analysis framework, dynamic collaborative construction and deep semantic reasoning of the historical document knowledge ontology are completed, which effectively improves the accuracy, consistency and maintainability of the knowledge system, and significantly enhances the automation and reliability of multi-version document knowledge management. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0047] Fig. 1 Flowchart of the method for dynamically constructing a historical document version knowledge ontology.

[0048] Fig. 2 Schematic diagram of the system for dynamically constructing a historical document version knowledge ontology.

[0049] Fig. 3 Flowchart for generating a version structured document set.

[0050] Fig. 4 Flowchart for generating a historical document version knowledge ontology. DETAILED DESCRIPTION

[0051] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0052] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that the present application can be practiced without the specific details set forth in this description. In other instances, well-known methods have not been described in detail in order not to unnecessarily obscure aspects of the present application.

[0053] Second, the "one embodiment" or "an embodiment" described herein refers to a specific feature, structure, or characteristic in at least one implementation of the application. The following description uses "in one embodiment" or "in an embodiment" in the sense of "in at least one implementation" as points in the specification that follow are specific implementation of various aspects of the present application, but this does not necessarily refer to the same embodiment, or exclude the existence of alternative, or additional embodiments.

[0054] Reference Figs. 1-4 For one embodiment of the present application, the embodiment provides a historical document version knowledge ontology dynamic collaborative construction method, comprising the following steps:

[0055] S1, collect multi-source historical literature digital versions, and pre-process to generate version structured literature sets.

[0056] It should be noted that the multi-source historical literature digital version includes digital historical literature copies from different institutions, carriers and periods, scanning images, OCR text, structured data and metadata information.

[0057] S1.1, uniformly format conversion is performed on the multi-source historical literature digital version to obtain a standardized literature input data set, and a character recognition operation is performed to obtain an initial recognition result.

[0058] It should be noted that the format of each literature file in the collected multi-source historical literature digital version is identified, and different formats such as literature image data, structured text files or PDF are uniformly converted into a unified image+text composite format, and numbered and structured annotated by page to obtain a standardized literature input data set. Each page in the standardized literature input data set contains a unified size of literature image content and a corresponding metadata framework.

[0059] After completing the format conversion, for the literature image content in the standardized literature input data set, first, image preprocessing operations are performed, including image denoising, tilt correction, contrast enhancement and edge sharpening steps, to improve the accuracy of subsequent character recognition and image structure clarity. An optical character recognition method is used to detect and recognize characters row by row in the pre-processed literature image area, extract the text information contained in the literature image, and combine the literature image structure to preliminarily aggregate and reorder the text area to obtain an initial recognition result.

[0060] S1.2, text correction and layout structure recovery processing is performed on the initial recognition result to obtain a text information set, and metadata information extraction and labeling are performed to obtain a version structured literature set.

[0061] It should be noted that the text correction processing based on dictionary comparison and language context is performed on the character content in the initial recognition result, the spelling correction and semantic matching repair are performed on the misrecognized words, the accuracy of the text content is improved, according to the position information of each text line in the standardized historical literature image page, combined with line spacing, paragraph alignment, page layout clues and the difference between title and body font, the layout structure recovery processing is performed, the original chapter structure, paragraph level and title level relationship of historical literature are reconstructed, and the text information set containing complete literature hierarchical structure is output.

[0062] After obtaining the text information set, according to the six types of metadata information in the literature content including keywords, paragraph labels, page number marks, author information, time labels and literature source identifiers, metadata information extraction operation is performed on the text information set, using entity positioning method based on rules and position relationship, six types of information elements are extracted and structured stored, and each type of metadata information is semantically annotated through label system, metadata information extraction and labeling are completed, and finally version structured literature set containing text content, structure information and metadata is output.

[0063] S2, from the version structured literature set, extract the knowledge entity features in each version literature, and construct a version feature set.

[0064] S2.1, named entity recognition is performed on each historical literature in the version structured literature set, five types of knowledge entities are extracted and standardized to form a standardized entity set, and subject-predicate-object triples are constructed combined with semantic role labeling and dependency syntax analysis.

[0065] It should be noted that named entity recognition is performed on each historical literature text in the version structured literature set, named entity recognition uses recognition method based on part-of-speech tagging and context feature rules, and scans and labels continuous word groups with semantic independence in historical literature text, and identifies five types of knowledge entities; the categories of the five types of knowledge entities include person entity, event entity, location entity, organization entity and time entity; the five types of knowledge entities identified in the original text have various expression forms, therefore, the five types of knowledge entities are normalized, the expressions with the same semantic direction are unified to standard expressions through morphological restoration, synonym dictionary mapping and entity alias matching, and a standardized entity set is generated.

[0066] After the construction of the normalized entity set, the semantic role labeling operation is combined to identify the semantic function of each component in the sentence, for example, to identify the agent, the recipient, the tool and the target, and further based on the dependency syntax, the syntactic tree structure of the sentence is parsed to identify the dependency relationship between words-main-verb relationship (such as "meeting-convene"), verb-object relationship (such as "through-resolution") and adverbial relationship (such as "rapid-development") and other syntactic relationships, and the dependency syntax path similarity is obtained, and by jointly matching the semantic role labeling result and the dependency syntax path similarity, the subject-predicate-object triple consisting of entities and behavior relationships is extracted, and each subject-predicate-object triple consists of a normalized entity as the subject, a semantic behavior word as the predicate, and another normalized entity as the object, fully depicting the semantic relationship between entities in historical documents.

[0067] S2.2, the same five types of knowledge entities in different versions are matched and associated by using the subject-predicate-object triple, an aligned entity relationship graph is output, multi-dimensional feature vectors of the five types of knowledge entities are extracted, and standardized processing is performed, and a standardized feature vector set is output.

[0068] It should be noted that the subject-predicate-object triples in each version are grouped according to the five types of knowledge entities, and the expression and semantic context features of each type of entity in different historical document versions are extracted, and the same five types of knowledge entities in different historical document versions are compared one by one, and based on the standardized expression, semantic context features, subject-predicate-object triples and dependency syntax path similarity, each pair is matched and judged; when the matching condition is met, the semantic equivalent entities are identified, and the corresponding relationship between the versions is established, and an aligned entity relationship graph composed of entity nodes and matching edges is constructed.

[0069] For each five types of knowledge entities, based on the semantic context, word frequency statistics, word vector representation, syntactic dependency structure and role in the subject-predicate-object triple in different versions, multi-dimensional semantic, syntactic, structural and contextual features are extracted, and a multi-dimensional feature vector is constructed, and for the consistency of subsequent calculations, all multi-dimensional feature vectors of the five types of knowledge entities are uniformly standardized, including value domain normalization, distribution transformation and feature compression, and finally a standardized feature vector set with consistent format and unified dimension is obtained.

[0070] It should also be noted that the matching condition refers to consistent or highly similar standardized expression, and similar context semantics, same or similar predicate relationship and similar dependency syntax path structure.

[0071] S2.3, the standardized feature vector set is aggregated by entity feature weighted average according to the literature version, the version level feature matrix is constructed, the row vector in the version level feature matrix is mapped with the corresponding version identifier, and the version feature set is output.

[0072] It should be noted that the five types of knowledge entities in the standardized feature vector set are grouped according to the literature version, ensuring that each group corresponds to all five types of knowledge entities in a historical literature version. In each group, the standardized feature vectors of all five types of knowledge entities are weighted, and the weights are allocated according to the frequency of occurrence of the five types of knowledge entities in the version literature, the importance of semantic roles, or the structural position in the subject-predicate-object triple. For example, entities with higher occurrence frequency or acting as the subject role can be given higher weights. The weighted average calculation of the standardized feature vectors of all five types of knowledge entities in the version obtains the version level feature vector representing the overall version semantic features.

[0073] The version level feature vectors of each historical literature version are arranged in sequence to form a version level feature matrix, and each row of the version level feature vector is bound with the corresponding version identifier, so that each row of the version level feature matrix has a clear version pointing property. Finally, the version feature set is output.

[0074] S3, the version feature set is input into a multi-node cloud collaborative processing architecture, the local similarity score between versions is calculated, and the version similarity matrix is output.

[0075] It should be noted that "multi-node cloud collaborative processing architecture" is a prior art in the field of distributed computing, and its general composition includes 1 master node (responsible for task scheduling and result aggregation) and multiple collaborative nodes (performing parallel computing). The data interaction between nodes is realized through distributed communication middleware (such as gRPC / MPI), and the resource manager (such as YARN / Kubernetes) is used to dynamically allocate computing resources.

[0076] S3.1, the version feature set is input into a multi-node cloud collaborative processing architecture, the master node in the multi-node cloud collaborative processing architecture receives the version feature set, distributes it evenly to the collaborative node cluster according to the version identifier, calculates the Euclidean distance of each historical literature version feature vector, generates a local similarity score through nonlinear conversion, and adds a confidence weight label to generate a local similarity matrix.

[0077] It should be noted that the master node receives the complete version feature set, divides each version level feature vector in the version feature set according to the version identifier, and distributes the divided version level feature vectors evenly to the collaborative node cluster.

[0078] After receiving version-level feature vectors from different historical document versions, the collaborative nodes calculate the Euclidean distance between the version feature vectors to quantify the degree of feature differences between versions. The expression is as follows:

[0079] ;

[0080] In the formula, It is the first The eigenvector of the reference version and the eigenvector of the _th reference version Euclidean distance between the feature vectors of each target version It is the first The first reference version feature vector of the feature vector Each version-level feature vector dimension feature value, It is the first The first target version feature vector Each version-level feature vector dimension feature value, It is an index variable of the version-level feature vector dimension. It is the number of dimensions of the version-level feature vector. It is the index variable of the reference version's feature vector. It is the index variable of the target version feature vector.

[0081] After calculation, the collaborative nodes use a nonlinear transformation function, such as a transformation method based on exponential decay or Gaussian kernel function, to convert the Euclidean distance into a local similarity score, expressed as:

[0082] ;

[0083] In the formula, It is the first The eigenvector of the reference version and the eigenvector of the _th reference version Local similarity scores between the feature vectors of each target version It is an adjustment parameter for nonlinear transformation, used to control the similarity decay rate.

[0084] Based on the quality indicators and data integrity of the version-level feature vectors, a corresponding confidence weight label is added to each local similarity score to reflect the reliability of the similarity score. Finally, each collaborative node outputs a local similarity matrix containing local similarity scores and confidence weight labels.

[0085] It should also be noted that the confidence weight is calculated by comprehensively evaluating indicators such as the completeness of the version feature vector, data quality, and text recognition accuracy. It reflects the reliability of the local similarity score and is used to enhance the credibility labeling of each element in the similarity matrix.

[0086] The quality index of the version-level feature vector is obtained by checking the integrity of the content of the version-level feature vector, mainly evaluating whether the feature vector completely contains the dimension information of the five types of knowledge entities (person, event, place, organization, and time), and investigating the rationality of the distribution of each dimension feature and its correspondence with the semantics of the original text.

[0087] The data integrity of the version-level feature vector is determined by checking the integrity of the original literature data. The completeness of the literature metadata field, the accuracy of the text recognition, and the coverage of the matching of the five types of knowledge entities between different versions are mainly evaluated. These checks verify the integrity and reliability of the data source from different angles.

[0088] S3.2, aggregate all local similarity matrices by confidence weight through distributed reduction operation, generate global version similarity matrix, and map version identifiers to global version similarity matrix row and column indexes in reverse, output version similarity matrix.

[0089] It should be noted that the local similarity matrices output by multiple collaborative nodes are summarized according to the corresponding version identifiers, and the local similarity scores between the same version pairs are weighted and averaged according to the confidence labels of each local similarity score, ensuring that high-confidence similarity scores have greater influence in aggregation.

[0090] Through distributed reduction operation, all local similarity matrix data on the collaborative nodes are aggregated into a unified global version similarity matrix. Each element in the global version similarity matrix represents the global similarity score between the corresponding two historical literature versions. According to the mapping relationship between the version identifier and the matrix index, the version identifier is accurately mapped to the row and column index positions of the global version similarity matrix, ensuring that the version identifiers corresponding to the matrix rows and columns are unique and correct. Finally, the version similarity matrix containing all global similarity scores between historical literature versions and version identifier mapping information is output.

[0091] S4, use a distributed conflict detection mechanism to evaluate the consistency of the version similarity matrix, mark entity mapping pairs with confidence lower than the confidence threshold, and generate a conflict transaction log.

[0092] S4.1, use the master node to divide the version similarity matrix into multiple entity mapping pairs and distribute them to each collaborative node.

[0093] It should be noted that the master node receives the complete version similarity matrix, parses the version pair relationship corresponding to each element in the version similarity matrix into an entity mapping pair, i.e., a matching pair of five types of knowledge entities between two historical literature versions, and according to the task allocation strategy, reasonably divides all entity mapping pairs to ensure that the divided entity mapping pair set is evenly distributed and easy to process collaboratively.

[0094] After the division is completed, the master node separately issues the different batches of entity mapping pairs to the multiple collaborative nodes according to the division results, to ensure that each collaborative node receives entity mapping pairs that can support subsequent version consistency verification and semantic analysis work.

[0095] It should also be noted that the task allocation strategy is an intelligent scheduling mechanism in which the master node dynamically selects the optimal allocation scheme (uniform allocation or load balancing) according to the real-time load state (CPU / memory / network) of the collaborative nodes and the task characteristics (computational complexity / data volume / dependency relationship).

[0096] S4.2, After receiving the entity mapping pairs, each collaborative node performs version consistency verification based on the adjacent version entity consistency strategy, and performs semantic consistency analysis combined with the entity context semantic information, to identify semantic conflict candidate entity pairs, forming a structural conflict candidate set and a semantic conflict candidate set.

[0097] It should be noted that each collaborative node uses the adjacent version entity consistency strategy for the received entity mapping pairs, compares the adjacency relationships and attribute consistency of the five types of knowledge entities in different historical literature versions, verifies the structural consistency of the five types of knowledge entities between versions, and performs semantic consistency analysis combined with the context semantic information extracted from the five types of knowledge entities in the version structured literature set, to evaluate the similarity and potential contradictions of entity semantic content, and identify semantic conflict candidate entity pairs.

[0098] Through structural consistency verification, it is compared whether the direct associated entities (such as the subject-object relationship in triples) of the five types of knowledge entities in different versions remain consistent, and it is checked whether the attributes (such as the position of a person, the time of an event, etc.) of the five types of knowledge entities in different versions are recorded the same; when it is found that the associated relationships of the five types of knowledge entities between versions are broken (such as the absence of a key predicate connection in a version) or the core attributes are inconsistent (such as the conflicting record of a participant in the same event), the semantic conflict candidate entity pairs are classified into the structural conflict candidate set.

[0099] Through semantic consistency analysis, the context descriptions (such as co-occurrence words, modifiers, etc.) of the five types of knowledge entities in different versions of text are compared, logical contradictions in semantic expressions (such as entity A being the superior of entity B in one version, but becoming a peer relationship in another version) are identified; when there are irreconcilable conflicts in the semantic roles of the five types of knowledge entities in different versions, they are included in the semantic conflict candidate set.

[0100] It should also be noted that potential contradictions include attribute contradictions (such as inconsistent records of a person's position in different versions), relationship contradictions (such as opposite descriptions of event causal relationships in different versions), and timing contradictions (such as differences in the time records of the same event in different versions).

[0101] S4.3, the intersection set of the structural conflict candidate set and the semantic conflict candidate set is processed, the intersection entity mapping pair is obtained, and the corresponding confidence is extracted for judgment. When the confidence is lower than the confidence threshold, it is marked as a conflict entity mapping pair.

[0102] It should be noted that all entity mapping pairs in the structural conflict candidate set and the semantic conflict candidate set are uniformly encoded, the encoding rule is "entity type + version ID + entity ID" (such as "person_V1_E005"), a hash table is constructed based on the encoding index to quickly locate the completely matched entity mapping pairs in the two sets, and the accuracy of the intersection result is ensured through a double verification mechanism - both entity ID and conflict type are required to be strictly matched (such as "time attribute inconsistency" in structural conflict requires "time description contradiction" in semantic conflict), and finally the intersection entity mapping pair list that meets the structural conflict and semantic conflict conditions is output.

[0103] For each intersection entity mapping pair, the local similarity score and the corresponding confidence weight are extracted from the version similarity matrix, the performance of the intersection entity mapping pair in the conflict detection stage (such as structural conflict intensity classification 1-3 levels, semantic conflict severity score 0.1-1.0) is combined, the output confidence is dynamically adjusted through a weighted formula as a judgment basis, and the confidence is compared. If the confidence is lower than the confidence threshold, the intersection entity mapping pair is marked as a conflict entity mapping pair.

[0104] It should also be noted that the confidence threshold is set by analyzing the integrity of the version feature vector, the accuracy of text recognition, and the stability of similarity calculation, combined with historical data experience and actual task requirements.

[0105] S4.4, using the master node, the conflict entity mapping is de-duplicated and merged, and the mapping relationship is integrated to generate a conflict transaction log.

[0106] It should be noted that the master node receives the conflict entity mapping output by multiple collaborative nodes, de-duplicates the conflict entity mapping, merges the repeatedly appearing entity mapping pairs, eliminates redundant information, and then integrates the mapping relationship of the de-duplicated entity mapping pairs according to the corresponding version identifier and entity identifier, to ensure that the version-to-version correspondence of all conflict entity mapping pairs is complete and has no conflict duplication.

[0107] During the integration process, the attribute information and confidence value in the mapping relationship are uniformly updated to ensure the consistency and accuracy of the mapping relationship, and finally the conflict entity mapping set that is de-duplicated, merged and has complete mapping relationship is organized to generate a structured conflict transaction log.

[0108] S5.1, parse the conflict transaction log into a conflict evolution graph, and map it into the five-level conflict resolution protocol according to the five levels of conflict types: syntax, structure, version, semantics, and value.

[0109] It should be noted that the five-level conflict resolution protocol is a conflict resolution mechanism in the construction of historical literature knowledge graph, and the core composition includes five levels: the syntax layer (L1) handles surface problems such as character encoding and spelling errors; the structure layer (L2) repairs triplets broken or property conflicts; the version layer (L3) calibrates the time or content deviation caused by literature revision; the semantic layer (L4) resolves context logical contradictions (such as inconsistent wording or narrative perspective); and the value layer (L5) finalizes core disputes involving historical evaluation sensitivity.

[0110] Technical implementation of the five-level conflict resolution protocol:

[0111] Conflict detection: based on the version identifier, confidence and evolution characteristics (such as context semantic vector) of the entity mapping pair, automatically classified into five levels.

[0112] Layered arbitration: L1-L3 are automatically processed by algorithm (such as Neo4j graph alignment), L4 uses "semantic similarity + historical attribute weight" hybrid scoring, and L5 is automatically ruled by the historical value decision engine (based on authoritative historical data rule base).

[0113] Closed-loop feedback: when converting the ruling result into an OWL statement, add conflict level annotations (such as "L4 semantic conflict confidence 0.72"), and generate test cases through the test case generator (such as time sequence consistency verification).

[0114] The role of the five-level conflict resolution protocol:

[0115] Precise disambiguation: through five-level resolution from the syntax layer (L1) to the value layer (L5), ensure the historical compliance and academic rigor of the knowledge ontology.

[0116] Dynamic traceability: all ruling results carry version identifiers and processing node information, supporting "entity-conflict-correction" full-link tracking, providing an audit baseline for subsequent ontology expansion. Its innovation lies in incorporating the unique historical literature evaluation sensitivity assessment into the standardized processing flow.

[0117] S5.1, parse the conflict transaction log into a conflict evolution graph, and map it into the five-level conflict resolution protocol according to the five levels of conflict types: syntax, structure, version, semantics, and value.

[0118] It should be noted that each conflict entity mapping pair in the conflict transaction log is taken as a basic unit, and each item is analyzed according to the version identifier, entity content, conflict type and confidence information, and the relevant content including the conflict source, conflict relationship, conflict entity pair and evolution path is extracted to construct the evolution relationship chain of the conflict entity in multiple historical literature versions.

[0119] Taking each conflict entity pair with historical evolution characteristics as a conflict evolution node, connecting the version evolution information and the structure change, semantic change or context transfer relationship on the time axis, a conflict evolution graph containing the structure transition and semantic evolution process between entities is formed.

[0120] According to the information marked in the conflict transaction log, the conflict entity mapping pair is type-mapped based on five dimensions of syntax difference, structure change, version difference, semantic conflict and historical value, and the corresponding conflict evolution graph is divided into the corresponding conflict type level in the five-level contradiction hierarchical processing protocol.

[0121] It should also be noted that the syntax difference: by comparing the lexical form (such as traditional / simplified conversion) and syntactic structure (such as passive / active voice difference) of the five types of knowledge entities in different versions, the surface language contradictions such as spelling and word order are identified.

[0122] Structure change: analyze the role change (such as subject to object) and attribute field increase and decrease (such as the difference in the version of the position of a person) of the five types of knowledge entities in the cross-version triple, and detect the inconsistency of the knowledge graph structure.

[0123] Version difference: according to the time axis and revision background of the literature version, the expression difference of the five types of knowledge entities caused by version change (such as deleting the description of a specific event in the 1935 version in the 1951 version) is located.

[0124] Semantic conflict: by comparing the sentiment tendency (such as praise or derogatory modifier) and behavior logic (such as "support" vs "oppose") of the context of the five types of knowledge entities, the deep semantic contradiction is determined.

[0125] Historical value: based on the historical evaluation system (such as the authority of historical materials and the degree of social influence), the loss and gain of the core value of the conflict entity to the literature is evaluated.

[0126] S5.2, under the five-level contradiction hierarchical processing protocol, based on the entity context semantic trajectory and semantic stability index, the conflict is arbitrated by the context weighted arbitrator, and the arbitration result list is output.

[0127] It should be noted that under the five-level contradiction hierarchical processing protocol, for the input conflict evolution graph, the context semantic trajectory of the conflict entity mapping pair in the historical literature version is first extracted, including the semantic change path, context collocation word and syntax dependency feature of the five types of knowledge entities in different historical literature versions, and combined with the version evolution order and semantic change trend of the five types of knowledge entities in the time dimension, a complete semantic trajectory expression is formed. At the same time, based on the confidence information and semantic change amplitude attached in the conflict transaction log, the semantic stability index of each conflict entity mapping pair is calculated, which is used to measure the consistency degree of the semantic direction of the conflict entity in different historical literature versions, and the expression is:

[0128] ;

[0129] In the formula, represents the semantic stability index of the conflict entity mapping pair , and represents the context semantic vector extracted from the conflict entity mapping pair in the th historical literature version, represents the context semantic vector extracted from the conflict entity mapping pair in the th historical literature version, represents the index variable of the historical literature version, represents the total number of historical literature versions.

[0130] After obtaining the entity context semantic trajectory and semantic stability index, the context weighted arbitrator is used for arbitration. The context weighted arbitrator performs weighted scoring on the conflict entity mapping pair according to the frequency of the conflict entity mapping pair appearing in different contexts, version semantic continuity, semantic stability index and semantic evolution direction, and determines the reservation priority and version direction of the conflict entity mapping pair in the historical literature knowledge ontology according to the scoring result, and finally outputs the arbitration result list containing the arbitration result of all conflict entity mapping pairs.

[0131] S5.3, according to the arbitration result list, convert different arbitration types into corresponding OWL ontology update statements.

[0132] It should be noted that each conflict entity mapping pair in the arbitration result list is classified, and according to the arbitration type output by the five-level contradiction hierarchical processing protocol, the specific type corresponding to the ontology update operation is output, including five basic ontology operations of adding entity, deleting entity, updating attribute, adding relationship and deleting relationship.

[0133] For the arbitration result judged as an added entity, a new description statement with a clear category label and a unique identifier in semantics is constructed.

[0134] For the decision result of deleting an entity, an invalidation statement with a marked state is generated to indicate that the entity is no longer used.

[0135] For the decision result of modifying an attribute, a corresponding update expression content is generated according to the type of the attribute, and a new attribute value and format specification are marked.

[0136] For the decision result of adding a relationship, a structural description of the relationship between the two entities involved is generated according to the behavior or logical connection between the two entities.

[0137] For the decision result of deleting a relationship, an explanatory statement reflecting the cancellation of the original relationship is constructed.

[0138] After all the conversion operations are completed, each conversion result is represented as an ontology update statement with complete semantic structure and syntax consistent with the OWL 2 specification.

[0139] S5.4, additional processing node information, version identifier and conflict level annotation are added to the OWL ontology update statement, and historical document update instructions are integrated.

[0140] It should be noted that each OWL ontology update statement generated in the previous refinement step is used as the basis for the additional annotation information in three dimensions, where the processing node information is used to indicate the specific collaborative node identifier for performing the ontology update operation, the version identifier is used to record the historical document version number corresponding to the current ontology update statement, and the conflict level annotation is used to mark the conflict entity mapping pair involved in the current update statement belongs to the conflict level type in the five-level contradiction hierarchical processing protocol.

[0141] Based on all the OWL ontology update statements with additional information, the update statements are merged and integrated according to entity structure consistency and time sequence continuity, and historical document update instructions with complete attribute annotations and traceable version information are generated.

[0142] It should be noted that the master node automatically extracts the node identifier for performing the update operation and binds it with the operation timestamp to obtain the processing node information based on the collaborative node allocation table recorded when the task is distributed during the OWL statement generation stage.

[0143] S6, according to the historical document update instructions, dynamically expand the entity relationship network of the historical document knowledge ontology, and output the historical document version knowledge ontology through a three-layer intelligent analysis framework.

[0144] It should be noted that the three-layer intelligent analysis framework is composed of an entity-level semantic verification layer, a version-level association reasoning layer and an overall structure optimization layer.

[0145] Entity-level semantic verification layer: Based on existing ontology verification tools (such as Pellet reasoner), through the extension of historical literature specific rule base (such as the logic constraints of the position of the person), the attribute consistency check is realized.

[0146] Version-level association reasoning layer: Using time sequence knowledge graph technology combined with cross-version semantic trajectory analysis, the evolution law of five types of knowledge entities in different versions is mined.

[0147] Overall structure optimization layer: Using complex network optimization algorithm (such as community detection) to adjust the topology structure, and adding historical literature specific index to guide optimization.

[0148] S6.1, the historical literature update instruction is parsed into entity and relationship incremental update event.

[0149] It should be noted that the entity description structure, attribute update content or entity relationship expression content contained in each OWL ontology update statement extracted from the historical literature update instruction is identified, the subject, predicate and object triple elements in the statement are identified, and the update operation type is distinguished, including adding, modifying or deleting entity, updating attribute value, or adding association relationship between entities; According to the processing node information, version identifier and conflict level annotation added in the update statement, the version time sequence position and semantic conflict intensity involved in the update operation are determined, which is used to guide the priority judgment of the subsequent incremental update event; The multiple update statements parsed in the same historical literature update instruction are aggregated as a structural unit according to the entity identifier, and the multiple attribute updates or relationship updates of the same subject entity are integrated to generate a set of entity and relationship incremental update events with clear structure and clear semantics.

[0150] S6.2, the priority of the incremental update event is sorted according to the conflict level annotation and version identifier, and the update event queue to be processed is generated.

[0151] It should be noted that the conflict level annotation and version identifier information in each incremental update event are read. Secondly, according to the hierarchical relationship of the conflict level annotation, the incremental update events are sorted from high to low according to the conflict severity, to ensure that the update events with high conflict level are processed first; At the same time, combined with the version identifier, the incremental update events with the same conflict level are sorted according to the version time sequence, to ensure that the time sequence of version update is respected, and the sorted incremental update events are arranged in order to form the update event queue to be processed.

[0152] S6.3, according to the update event queue to be processed, the entity and relationship insertion, deletion and attribute update operation of the entity relationship network of the historical literature knowledge ontology are carried out through each cooperative node, and the updated entity relationship network is obtained.

[0153] It should be noted that the entity and relationship change information in the incremental update event is read item by item, and each collaborative node performs corresponding operations on the entity relationship network of the historical literature knowledge ontology according to the operation type described in the incremental update event: for newly added entities or relationships, perform insertion operation and add to the entity relationship network; for entities or relationships that need to be deleted, perform deletion operation and remove the corresponding nodes or edges from the entity relationship network; for attribute update content, replace or supplement according to attribute type and new attribute value to ensure the accuracy and integrity of the attribute information.

[0154] Each operation is strictly performed in the order in the update event queue, ensuring the order and version consistency of the update process. After the sequential processing of all incremental update events, an updated state entity relationship network containing the latest entity, relationship and attribute information is generated.

[0155] S6.4, the updated state entity relationship network is verified by the entity level semantic verification layer, deep semantic reasoning and version difference analysis are performed through the version level association reasoning layer, global structure optimization is performed by the overall structure optimization layer, and the historical literature version knowledge ontology is output.

[0156] It should be noted that the entity level semantic verification layer checks each entity and corresponding attribute in the updated state entity relationship network one by one, verifies whether the attribute value of the entity conforms to the predefined semantic standard through semantic rules and constraint conditions, identifies contradictions or unreasonable phenomena between the structured field contents of the five types of knowledge entities, excludes duplicate or conflicting entity information, ensures the complete and accurate semantic expression of the five types of knowledge entities, and thus improves the overall consistency and credibility of knowledge expression.

[0157] The version level association reasoning layer combines the evolution of entity relationships in each version of historical literature, constructs semantic association chains between versions, and identifies semantic evolution paths between mapped entities based on semantic levels, attribute value changes, relationship structure adjustments and other rules, such as attribute value replacement, relationship addition or deletion; then according to semantic constraints, it judges whether these changes have logical conflicts or meaning deviations, thereby identifying semantic differences and potential conflicts between versions, mining knowledge increments or correction contents brought by version updates, and supporting in-depth understanding and semantic consistency guarantee of historical literature versions.

[0158] The overall structure optimization layer analyzes the structural characteristics of the entity relationship network from a global perspective, adjusts the connection relationship between entities using network topology optimization method, eliminates redundant relationships, balances the connection degree of entity nodes, optimizes the hierarchical structure and path length of knowledge ontology, improves query efficiency and reasoning performance, and finally generates a historical literature version knowledge ontology with reasonable structure and excellent performance.

[0159] It should be noted that by analyzing the existing entity types, attribute structures and upper and lower semantic relationships in historical literature, combining the field definition rules, attribute value ranges and entity constraint conditions in the version ontology, a unified semantic standard is formed.

[0160] The embodiment also provides a historical literature version knowledge ontology dynamic collaborative construction system, including: a literature collection module, used for collecting multi-source historical literature digital versions, and performing preprocessing to generate a version structured literature set;

[0161] An entity extraction module is used for extracting knowledge entity features in each version literature from the version structured literature set to construct a version feature set;

[0162] A similarity calculation module is used for inputting the version feature set into a multi-node cloud collaborative processing architecture, calculating local similarity scores between versions, and outputting a version similarity matrix;

[0163] A conflict detection module is used for performing consistency evaluation on the version similarity matrix by using a distributed conflict detection mechanism, marking entity mapping pairs with a confidence lower than a confidence threshold, and generating a conflict transaction log;

[0164] A contradiction processing module is used for inputting the conflict transaction log into a five-level contradiction hierarchical processing protocol, outputting an arbitration result converted into a corresponding OWL ontology update statement, and attaching a version trace annotation, and outputting a historical literature update instruction;

[0165] An ontology construction module is used for dynamically extending an entity relationship network of the historical literature knowledge ontology according to the historical literature update instruction, and outputting the historical literature version knowledge ontology through a three-layer intelligent analysis framework.

[0166] The embodiment also provides a computer device suitable for the historical literature version knowledge ontology dynamic collaborative construction method, including: a memory and a processor; the memory is used for storing computer executable instructions, and the processor is used for executing the computer executable instructions to realize the historical literature version knowledge ontology dynamic collaborative construction method proposed in the above embodiment.

[0167] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected by a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is configured to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved by WIFI, an operator network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0168] The embodiment also provides a storage medium having a computer program stored thereon, the program being executed by a processor to implement the method for dynamically constructing a history document version knowledge ontology proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disk.

[0169] To sum up, the present application realizes efficient and accurate conflict identification of multi-version entities of history documents by the multi-node distributed conflict detection mechanism, realizes intelligent arbitration and standardized dynamic update of complex conflicts by combining the five-level contradiction hierarchical processing protocol and the context weighted arbitrator, and guarantees the transparency of update by the additional version traceability. Meanwhile, the present application completes the dynamic collaborative construction and deep semantic reasoning of the history document knowledge ontology by means of the unified preprocessing, multi-dimensional semantic feature extraction and weighted aggregation, multi-version similarity parallel calculation and three-layer intelligent analysis framework, effectively improves the accuracy, consistency and maintainability of the knowledge system, and significantly enhances the automation and reliability of multi-version document knowledge management.

[0170] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for dynamically constructing a historical document version knowledge ontology, characterized in that: comprising, collecting multi-source historical literature digital versions, and preprocessing to generate version structured literature sets; extracting knowledge entity features in each version literature from the version structured literature sets to construct a version feature set; inputting the version feature set into a multi-node cloud collaborative processing architecture, calculating the local similarity scores between versions, and outputting a version similarity matrix, the specific steps are as follows, inputting the version feature set into the multi-node cloud collaborative processing architecture, receiving the version feature set by the master node in the multi-node cloud collaborative processing architecture, distributing it evenly to the collaborative node cluster according to the version identifier, calculating the Euclidean distance of each historical literature version feature vector respectively, generating local similarity scores through nonlinear conversion, adding confidence weight annotation, and generating a local similarity matrix; aggregating all local similarity matrices according to the confidence weight through distributed reduction operation to generate a global version similarity matrix, and mapping the version identifier back to the global version similarity matrix row and column index, outputting the version similarity matrix; using a distributed conflict detection mechanism, the consistency of the version similarity matrix is evaluated, and the entity mapping pairs with a confidence lower than the confidence threshold are marked, and a conflict transaction log is generated, the specific steps are as follows, using the master node, the version similarity matrix is divided into multiple entity mapping pairs and distributed to each collaborative node; after receiving the entity mapping pairs, each collaborative node performs version consistency verification based on the adjacency version entity consistency strategy, and performs semantic consistency analysis combined with entity context semantic information to identify semantic conflict candidate entities, forming a structural conflict candidate set and a semantic conflict candidate set; performing intersection processing on the structural conflict candidate set and the semantic conflict candidate set to obtain the intersection entity mapping pairs, and extracting the corresponding confidence for judgment, when the confidence is lower than the confidence threshold, it is marked as a conflict entity mapping pair, and the de-duplication and mapping relationship integration are performed to generate a conflict transaction log; inputting the conflict transaction log into a five-level contradiction hierarchical processing protocol, outputting the arbitration result to convert it into corresponding OWL ontology update statements, and adding version trace annotations, outputting historical literature update instructions; according to the historical literature update instructions, dynamically extending the entity relationship network of the historical literature knowledge ontology, and through a three-layer intelligent analysis framework, outputting the historical literature version knowledge ontology, the specific steps are as follows, parse the historical literature update instructions into entity and relationship incremental update events; according to the conflict level annotation and the version identifier, the incremental update events are prioritized to generate a to-be-processed update event queue; according to the to-be-processed update event queue, the entity relationship network of the historical literature knowledge ontology is operated by each collaborative node to insert, delete and attribute update the entity and relationship, and the updated entity relationship network is obtained; using the entity-level semantic verification layer to verify the consistency of the updated entity relationship network, through the version-level association reasoning layer to perform deep semantic reasoning and version difference analysis, using the overall structure optimization layer to perform global structure optimization, and outputting the historical literature version knowledge ontology.

2. The method of claim 1, wherein the method further comprises: the preprocessing to generate version structured literature sets, the specific steps are as follows, The unified format conversion is performed on the multi-source historical literature digital version to obtain a standardized literature input data set, and a character recognition operation is performed to obtain an initial recognition result; Text correction and layout structure recovery processing are performed on the initial recognition result to obtain a text information set, and metadata information extraction and labeling are performed to obtain a version structured literature set.

3. The method of claim 1, wherein the historical document version knowledge ontology dynamic collaborative construction method is characterized in that: The version feature set is constructed, and the specific steps are as follows, Named entity recognition is performed on each historical literature in the version structured literature set, five types of knowledge entities are extracted and standardized to form a standardized entity set, and subject-predicate-object triples are constructed by combining semantic role labeling and dependency syntax analysis; The same five types of knowledge entities in different versions are matched and associated by using the subject-predicate-object triples, an aligned entity relationship graph is output, multi-dimensional feature vectors of the five types of knowledge entities are extracted, and standardization processing is performed to output a standardized feature vector set; The entity features in the standardized feature vector set are weighted and averaged according to the literature version to construct a version-level feature matrix, and a mapping relationship between the row vectors in the version-level feature matrix and the corresponding version identifiers is established to output the version feature set.

4. The method of claim 1, wherein the historical document version knowledge ontology dynamic collaborative construction method is characterized by: The historical literature update instruction is output, and the specific steps are as follows, The conflict transaction log is parsed into a conflict evolution graph, and a five-level contradiction hierarchical processing protocol is mapped according to the grammar, structure, version, semantics and value five-level conflict types; Under the five-level contradiction hierarchical processing protocol, based on the entity context semantic trajectory and semantic stability index, the conflict is ruled by the context weighted arbitrator, and a ruling result list is output; According to the ruling result list, different ruling types are converted into corresponding OWL ontology update statements; The OWL ontology update statement is attached with node information, version identifier and conflict level annotation, and is integrated to form a historical literature update instruction.

5. A system for dynamically constructing a historical document version knowledge ontology, based on any one of the historical document version knowledge ontology dynamic construction methods of claims 1-4, characterized in that: It includes The literature collection module is used for collecting multi-source historical literature digital versions, and pre-processing is performed to generate a version structured literature set; The entity extraction module is used for extracting knowledge entity features in each version literature from the version structured literature set to construct a version feature set; The similarity calculation module is used for inputting the version feature set into a multi-node cloud collaborative processing architecture to calculate the local similarity score between versions and output a version similarity matrix; The conflict detection module is used for using a distributed conflict detection mechanism to evaluate the consistency of the version similarity matrix, marking entity mapping pairs with a confidence level lower than a confidence threshold, and generating a conflict transaction log; The contradiction processing module is used for inputting the conflict transaction log into a five-level contradiction hierarchical processing protocol, outputting an arbitration result into a corresponding OWL ontology update statement, and attaching a version traceability annotation to output a historical literature update instruction; The ontology construction module is used for dynamically expanding the entity relationship network of the historical literature knowledge ontology according to the historical literature update instruction, and outputting the historical literature version knowledge ontology through a three-layer intelligent analysis framework.

6. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that: The processor executes the computer program to realize the steps of the historical literature version knowledge ontology dynamic collaborative construction method of any one of claims 1-4.

7. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by a processor to realize the steps of the historical document version knowledge ontology dynamic collaborative construction method of any one of claims 1-4.

Images