A method and system for managing activities related to ancient architecture research and study

Abstract

This application provides a method and system for managing activities related to ancient architecture research and study, relating to the field of digital culture. The method includes: acquiring multi-source data on ancient architecture and basic data of research and study users; performing entity recognition and relationship extraction on the multi-source data to construct a knowledge graph of ancient architecture and form structured knowledge of ancient architecture; constructing a set of rules for the knowledge graph of ancient architecture based on the multi-source data and the knowledge graph of ancient architecture; constructing user profiles based on the basic data of research and study users and the knowledge graph of ancient architecture; the user profiles are used to characterize the user's cognitive level, interests, preferences, and learning abilities; retrieving knowledge subgraphs from the knowledge graph that match the user profiles, and generating research and study activity paths. This application, used in the management of activities related to ancient architecture research and study, solves the technical problem that existing technologies cannot meet the needs of multi-dimensional knowledge integration and application in ancient architecture research and study activities.

Description

Technical Field

[0001] This application relates to the field of digital culture, and in particular to a method and system for managing activities related to ancient architecture research and study. Background Technology

[0002] The fields of cultural heritage protection and education are increasingly incorporating digital and intelligent technologies. Big data, natural language processing, and knowledge graphs are widely used in museum tours, virtual exhibitions, and cultural dissemination, evolving the traditional digitization of cultural resources from "static display" to "dynamic interaction" and "intelligent retrieval." However, the existing digitization of ancient architecture and its associated cultural knowledge largely remains at the unstructured level of text and images, lacking a machine-readable knowledge system that can uniformly express the relationships between architectural components, historical context, and cultural semantics. This makes it difficult for users to achieve cross-level and cross-dimensional knowledge connections and understanding during in-depth cognitive and research exploration. Therefore, a method is urgently needed to address the technical problem that existing technologies cannot meet the needs of multi-dimensional knowledge integration and application in ancient architecture research activities. Summary of the Invention

[0003] This application provides a method and system for managing activities related to ancient architecture research and study, which solves the technical problem that existing technologies cannot meet the needs of ancient architecture research and study activities for comprehensive understanding and application of multi-dimensional knowledge.

[0004] To achieve the above objectives, this application adopts the following technical solution: Firstly, this paper provides a management method for activities related to ancient architecture study tours, including: acquiring multi-source data on ancient architecture and basic data on study tour users; the multi-source data on ancient architecture includes: entity data of ancient architecture, data on building structure and components, historical background data, cultural connotation data, and data on individuals; the basic data on study tour users includes: basic user attribute data, basic knowledge data, learning preference data, and interactive behavior data; performing entity recognition and relationship extraction on the multi-source data to construct an ancient architecture knowledge graph and form structured ancient architecture knowledge; constructing a set of rules for the ancient architecture knowledge graph based on the multi-source data and the ancient architecture knowledge graph; the set of rules for the ancient architecture knowledge graph includes: structural rules between ancient architecture components, historical logic rules of ancient architecture, and cultural semantic rules of ancient architecture; constructing user profiles based on the basic data of study tour users and the ancient architecture knowledge graph; the study tour user profiles are used to characterize the user's cognitive level, interest preferences, and learning ability characteristics; retrieving knowledge subgraphs from the knowledge graph that match the study tour user profiles, and generating study tour activity paths.

[0005] Based on the above technical solutions, the activity management method for ancient architecture study tours provided in this application constructs an ancient architecture knowledge graph by performing entity recognition and relationship extraction on multi-source data of ancient architecture and introducing structural rules, historical logic rules, and cultural semantic rules. This achieves a unified expression of ancient architecture-related knowledge at the structural, historical, and cultural semantic levels, effectively improving the systematicness and relevance of knowledge. Simultaneously, user profiles are constructed by combining study tour user basic data and the knowledge graph, and matching study tour activity paths are generated based on these profiles. This ensures that the study tour content matches the user's cognitive level and learning characteristics, enhancing the relevance and rationality of the study tour activities. Furthermore, during the execution of study tour activities, user interaction operations are reasoned and verified based on the knowledge graph and rule set, and user profiles are dynamically updated, forming a closed-loop mechanism for study tour guidance and user cognitive evolution. This enhances the interactivity, guidance effectiveness, and continuous optimization capabilities of study tour activities, better meeting the needs for comprehensive understanding and application of multi-dimensional knowledge in study tour scenarios.

[0006] In conjunction with the first aspect mentioned above, in one possible implementation, after generating the study tour activity path, the method further includes: obtaining the study tour activity path of the study tour user and the interactive operation information of executing the study tour activity path; reasoning about the interactive operation information based on the ancient architecture knowledge graph and the rule set of the ancient architecture knowledge graph to obtain matching reasoning results; verifying the interactive operation information based on the matching reasoning results to generate ancient architecture study tour guidance information; and updating the user profile based on the study tour activity path, interactive operation information, and ancient architecture study tour guidance information.

[0007] In conjunction with the first aspect mentioned above, one possible implementation involves entity recognition and relation extraction from multi-source data to construct an ancient architecture knowledge graph and form structured ancient architecture knowledge. This includes: preprocessing the multi-source data on ancient architecture; performing entity recognition on the preprocessed multi-source data based on a defined entity type system for the ancient architecture domain, extracting ancient architecture entities, architectural component entities, historical event entities, cultural connotation entities, and human figures; extracting relations from the identified entities based on the contextual semantic relationships between entities in the multi-source data on ancient architecture, obtaining entity-relationship pairs; using these entity-relationship pairs to reflect the structural composition relationships, historical connections, and cultural relationships of ancient architecture; and storing the ancient architecture entities and entity-relationship pairs in a structured manner to construct an ancient architecture knowledge graph and generate structured ancient architecture knowledge corresponding to the ancient architecture knowledge graph.

[0008] In conjunction with the first aspect mentioned above, one possible implementation involves performing entity recognition on preprocessed multi-source data of ancient buildings based on a predefined entity type system in the field of ancient architecture, and using a fine-tuned BERT model to achieve entity recognition.

[0009] In conjunction with the first aspect mentioned above, in one possible implementation, the fine-tuned BERT model is built on the BERT-base model. By freezing the parameters of the first set number of layers in the BERT-base model, the parameters of the second set number of layers and the classification layer parameters are fine-tuned. The fine-tuning process uses text data with entity annotations in the field of ancient architecture as training samples, and the BERT-base model is fine-tuned and trained through supervised learning to obtain the fine-tuned BERT model.

[0010] In conjunction with the first aspect mentioned above, one possible implementation involves constructing a set of rules for an ancient architectural knowledge graph based on multi-source data and the ancient architectural knowledge graph. This includes: extracting hierarchical, combination, and constraint relationships between architectural components based on ancient architectural structure and component data to determine structural rules between components; extracting formal characteristics and evolutionary constraints of ancient buildings from different periods based on historical background data to construct historical logic rules; constructing cultural semantic rules to define the mapping relationship between structural features and cultural semantics based on cultural connotation data; and binding the structural rules, historical logic rules, and cultural semantic rules with entities and entity relationships in the ancient architectural knowledge graph to obtain the set of rules for the ancient architectural knowledge graph.

[0011] In conjunction with the first aspect mentioned above, one possible implementation involves constructing a user profile based on the basic data of study tour participants and a knowledge graph of ancient architecture. This includes: constructing a structural cognitive vector of users regarding the structural hierarchy, component functions, and spatial relationships of ancient architecture based on the knowledge base data of study tour participants; constructing temporal understanding features of users regarding the historical evolution and era characteristics of ancient architecture based on the interactive behavior data of study tour participants during study tour activities; constructing user understanding preference features regarding the cultural semantics of ancient architecture based on the interpretation choices, path selections, and questioning behaviors of study tour participants during study tour activities; and integrating the structural cognitive vector, temporal understanding features, and cultural understanding preference features to determine the study tour participant profile.

[0012] In conjunction with the first aspect mentioned above, one possible implementation involves fusing structural cognitive vectors, temporal understanding features, and cultural understanding preference features to determine the profile of study tour users. This includes: determining the fusion weights of structural cognitive vectors, temporal understanding features, and cultural understanding preference features based on the degree of correlation among structural rules, historical logic rules, and cultural semantic rules in the ancient architecture knowledge graph rule set; weighting and fusing the structural cognitive vectors, temporal understanding features, and cultural understanding preference features according to the fusion weights to obtain an understanding state vector; and mapping the understanding state vector to the understanding state identifiers of study tour users in the structural understanding dimension, historical understanding dimension, and cultural understanding dimension to determine the profile of study tour users.

[0013] In conjunction with the first aspect mentioned above, one possible implementation involves reasoning about interactive operation information based on an ancient architecture knowledge graph and a set of rules for that knowledge graph, to obtain matching reasoning results. This includes: locating the target ancient architecture entity and building component entity corresponding to the interactive operation in the ancient architecture knowledge graph based on the objects in the interactive operation information, and obtaining the structural, historical, and cultural relationships directly associated with the target entity to construct a local knowledge subgraph; determining the structural rules, historical logic rules, and cultural semantic rules in the ancient architecture knowledge graph rule set corresponding to the local knowledge subgraph based on the operation type in the interactive operation information; and performing a line-by-line matching judgment between the results in the interactive operation information and the structural rules, historical logic rules, and cultural semantic rules to obtain matching reasoning results.

[0014] Secondly, an activity management system for ancient architecture research and study is provided. The system includes: a data acquisition module, a knowledge graph construction module, a rule extraction module, a user profile construction module, and a research and study path planning module. The system comprises several modules: a data acquisition module for acquiring multi-source data on ancient architecture and basic data on study tour users; multi-source data on ancient architecture includes: entity data, structural and component data, historical background data, cultural connotation data, and biographical data; basic data on study tour users includes: basic user attribute data, basic knowledge data, learning preference data, and interactive behavior data; a knowledge graph construction module for entity recognition and relationship extraction from multi-source data to construct a knowledge graph of ancient architecture and form structured knowledge about ancient architecture; a rule extraction module for constructing a set of rules for the knowledge graph of ancient architecture based on multi-source data and the knowledge graph of ancient architecture; the rule set for the knowledge graph of ancient architecture includes: structural rules between components of ancient architecture, historical logic rules of ancient architecture, and cultural semantic rules of ancient architecture; a user profile construction module for constructing and updating user profiles based on the basic data of study tour users and the knowledge graph of ancient architecture; the study tour user profile is used to characterize the user's cognitive level, interests, and learning ability; and a study tour path planning module for retrieving knowledge subgraphs from the knowledge graph that match the study tour user profiles and generating study tour activity paths.

[0015] This application provides a method and system for managing activities related to ancient architecture research and study. By uniformly acquiring, identifying entities, and extracting relationships from multi-source data on ancient architecture, a knowledge graph of ancient architecture is constructed, forming structured knowledge. Based on this, structural rules, historical logic rules, and cultural semantic rules are introduced, achieving unified modeling and structured expression of knowledge related to ancient architecture across multiple dimensions, including component structure, historical evolution, and cultural connotations. This effectively solves the problems of scattered knowledge and weak correlation in existing technologies. Simultaneously, by combining basic user data and the ancient architecture knowledge graph, user profiles are constructed and dynamically updated, ensuring that the generation of research and study activity paths is aligned with user needs. Matching users' cognitive levels, interests, and learning abilities improves the relevance and adaptability of study tour organization. Furthermore, during the execution of study tours, the system infers and verifies user interaction information based on knowledge graphs and their rule sets, generating corresponding study tour guidance information. This forms a closed-loop mechanism for the continuous evolution of study tour guidance and user profiles, thereby enhancing interactivity, guidance effectiveness, and overall management efficiency during the study tour process. It can better meet the actual needs for comprehensive understanding and application of multi-dimensional knowledge in ancient architecture study tour scenarios, solving the technical problem that existing technologies cannot meet the needs of comprehensive understanding and application of multi-dimensional knowledge in ancient architecture study tours.

[0016] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0017] Figure 1 A system architecture diagram of an activity management system for ancient architecture research and study provided in this application embodiment; Figure 2 A flowchart illustrating an activity management method for ancient architecture research and study provided in this application embodiment; Figure 3 A flowchart illustrating another activity management method for ancient architecture research and study provided in this application embodiment; Figure 4A flowchart illustrating another activity management method for ancient architecture research and study provided in this application embodiment; Figure 5 A flowchart illustrating another activity management method for ancient architecture research and study provided in this application embodiment; Figure 6 This is a flowchart illustrating another activity management method for ancient architecture research and study provided in this application embodiment. Detailed Implementation

[0018] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0019] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0020] The activity management method for ancient architecture research and study provided in this application embodiment can be applied to, for example... Figure 1 In the activity management system for ancient architecture research and study shown, such as Figure 1 As shown, the system includes: a data acquisition module 101, a knowledge graph construction module 102, a rule extraction module 103, a user profile construction module 104, and a study tour path planning module 105; Among them, the data acquisition module 101 is used to acquire multi-source data on ancient buildings and basic data of study tour users; the multi-source data on ancient buildings includes: data on ancient building entities, data on building structure and components, historical background data, cultural connotation data, and data on people; the basic data on study tour users includes: basic user attribute data, basic knowledge data, learning preference data, and interactive behavior data. The knowledge graph construction module 102 performs entity recognition and relation extraction on multi-source data to construct an ancient architecture knowledge graph and form structured ancient architecture knowledge. The rule extraction module 103 constructs a set of rules for the ancient architecture knowledge graph based on multi-source data and the ancient architecture knowledge graph. The set of rules for the ancient architecture knowledge graph includes: structural rules between ancient architectural components, historical logic rules of ancient architecture, and cultural semantic rules of ancient architecture. User profile building module 104 builds and updates user profiles based on basic data of study tour users and ancient architecture knowledge graph; study tour user profiles are used to characterize users’ cognitive level, interest preferences and learning ability characteristics. The study tour path planning module 105 retrieves knowledge subgraphs from the knowledge graph that match the study tour user profile and generates study tour activity paths.

[0021] To address the technical problem that existing technologies cannot meet the needs of multi-dimensional knowledge integration and application in ancient architecture study tours, this application provides an activity management method for ancient architecture study tours. The method includes: acquiring multi-source data on ancient architecture and basic data on study tour users; the multi-source data includes: ancient architecture entity data, architectural structure and component data, historical background data, cultural connotation data, and personnel data; the basic data on study tour users includes: user basic attribute data, knowledge base data, learning preference data, and interactive behavior data; performing entity recognition and relationship extraction on the multi-source data to construct an ancient architecture knowledge graph and form structured ancient architecture knowledge; constructing a set of rules for the ancient architecture knowledge graph based on the multi-source data and the ancient architecture knowledge graph; the set of rules for the ancient architecture knowledge graph includes: structural rules between ancient architecture components, historical logic rules of ancient architecture, and cultural semantic rules of ancient architecture; constructing user profiles based on the basic data of study tour users and the ancient architecture knowledge graph; the study tour user profiles are used to characterize the user's cognitive level, interests, and learning ability characteristics; retrieving knowledge subgraphs from the knowledge graph that match the study tour user profiles, and generating study tour activity paths.

[0022] Figure 2 A flowchart illustrating the activity management method for ancient architecture research and study provided in this application embodiment is shown below. Figure 2 As shown, the method includes: S201. Obtain multi-source data on ancient buildings and basic data on research and study users.

[0023] Among them, multi-source data on ancient buildings refers to multi-type information data collected and organized around specific ancient building objects, including at least: ancient building entity data, building structure and component data, historical background data, cultural connotation data, and people data; basic data of study tour users refers to a structured set of information used to characterize the learning characteristics and behavioral characteristics of study tour participants, including at least: basic user attribute data, basic knowledge data, learning preference data, and interactive behavior data.

[0024] In one possible implementation, data on ancient building entities, building structures and components, historical background, cultural connotations, and figures are collected from document databases, museum digital archives, 3D modeling systems, and manual annotation systems. Data from different sources are then formatted, time-labeled, and cleaned. Simultaneously, basic data on study tour users is obtained through user registration information, learning assessment records, historical study tour records, and interactive behavior logs, and these are organized into a unified data structure.

[0025] It should be noted that structured or semi-structured storage methods are preferred for multi-source data during the acquisition process. For unstructured text or image data, intermediate data representations that can be used for subsequent analysis are generated through preprocessing to avoid the impact of data noise on subsequent processing.

[0026] It should be noted that the multi-source data on ancient buildings and basic data on research and study users collected and processed in this application strictly comply with the provisions on personal privacy protection in relevant laws and regulations such as the "Personal Information Protection Law of the People's Republic of China" and the "Cybersecurity Law of the People's Republic of China." All data has undergone de-identification and anonymization preprocessing and does not contain any information that can directly or indirectly identify a natural person, such as name, mobile phone number, ID card number, home address, bank account information, or other private data. The data collection process strictly follows the principle of minimum necessity, collecting only information directly related to user needs analysis, such as publicly available product reviews, de-identified visitor flow statistics, and scene images without personal portraits, without collecting any additional irrelevant private content. Furthermore, the data processed in this application is only used for technical research and business optimization purposes such as user needs insight, product optimization, and service improvement, and will never be used for any unauthorized commercial promotion, targeted marketing, data selling, or other acts that infringe on personal rights. In addition, the data storage and processing process has adopted security protection measures such as encrypted transmission and access control to further prevent the risk of data leakage and ensure that the entire data processing process is compliant and secure, without infringing on the privacy rights and other legitimate rights of any natural person.

[0027] As an example, the system inputs raw data related to ancient architecture, registration information of study tour participants, pre-visit questionnaires, and historical study tour logs. Geometric and physical data such as the location, dimensions, and materials of the ancient buildings are collected through laser scanning and drone aerial photography. Structural data such as the type, specifications, and connection methods of building components are extracted through drawing analysis and expert annotation. Historical data such as the initial construction date, repair process, and related events of the ancient buildings are compiled through official historical documents and local chronicles. Cultural data such as religious symbolism and environmental layout embodied in the ancient buildings are explored through intangible cultural heritage lists and expert interpretations. Data on builders and repairers are extracted through biographies and historical archives. Simultaneously, basic attribute data (grade, user ID, etc.) is collected through user registration forms, basic knowledge data through pre-visit knowledge tests, learning preference data through preference selection questionnaires, and interactive behavior data through study tour platform logs. Finally, a structured multi-source dataset of ancient architecture and a basic dataset of study tour participants are output.

[0028] S202. Perform entity recognition and relation extraction on multi-source data, construct a knowledge graph of ancient buildings, and form structured knowledge of ancient buildings.

[0029] Among them, entity recognition refers to the process of extracting core entities related to ancient buildings with specific meanings from multi-source data; relation extraction refers to the process of mining the relationships between different entities; ancient building knowledge graph refers to a visualized knowledge network constructed with entities as nodes and entity relationships as edges; structured ancient building knowledge refers to a knowledge set with a fixed format formed based on the transformation of knowledge graph.

[0030] In one possible implementation, a multi-source dataset of ancient architecture is input. First, textual data undergoes word segmentation, stop word removal, and synonym normalization preprocessing. Numerical geometric data undergoes outlier removal, unit unification, and format standardization preprocessing. Then, based on a defined entity type system for the ancient architecture domain, including ancient building entities, building component entities, historical event entities, cultural connotation entities, and figure entities, a BERT pre-trained model is used for entity recognition to extract various core entities. Next, based on the contextual semantic relationships between entities, dependency parsing combined with remote supervision is used to extract relations, obtaining entity relation pairs reflecting structural composition, historical connections, and cultural mapping. Finally, entities and relation pairs are stored in a graph database in the form of triples to construct an ancient architecture knowledge graph. Based on the graph, structured ancient architecture knowledge documents such as entity association tables and component structure tables are generated, outputting the ancient architecture knowledge graph and structured ancient architecture knowledge.

[0031] It should be noted that after entity recognition and relation extraction, manual sampling verification is required, with a sampling ratio of no less than 10%, to ensure that the accuracy of entity recognition and relation extraction is controllable.

[0032] Based on the above steps, this process transforms scattered, unstructured multi-source data on ancient architecture into a structured, visualized knowledge network, achieving systematic integration and efficient organization of knowledge about ancient architecture. Through entity recognition, relation extraction, and graph construction, fragmented data is transformed into systematic knowledge, providing a structured knowledge carrier for subsequent rule set construction and knowledge subgraph retrieval.

[0033] S203. Based on multi-source data and ancient building knowledge graph, construct a set of rules for ancient building knowledge graph.

[0034] The ancient architecture knowledge graph rule set includes: structural rules between ancient architectural components, historical logic rules of ancient architecture, and cultural semantic rules of ancient architecture. Structural rules refer to the rules that define the hierarchical, combination, and constraint relationships between ancient architectural components; historical logic rules refer to the rules that define the formal characteristics and evolutionary sequence of ancient architecture in different periods; and cultural semantic rules refer to the rules that define the mapping relationship between the structural characteristics and cultural connotations of ancient architecture.

[0035] In one possible implementation, firstly, based on architectural structure and component data, the hierarchical affiliation, combination method, and dimensional ratio constraints of components are extracted through expert knowledge analysis and linear regression fitting to form structural rules. Secondly, based on historical background data, the formal characteristics and evolutionary sequence constraints of ancient buildings from different dynasties are extracted through historical document analysis and time series data mining to form historical logic rules. Thirdly, based on cultural connotation data, the mapping relationship between architectural structure and cultural symbols and aesthetic concepts is extracted through expert interpretation and semantic analysis to form cultural semantic rules. Finally, the three types of rules are bound to entities and relationships in the knowledge graph, and rule ID attributes are added to entities and relationships to output a set of ancient architectural knowledge graph rules.

[0036] In addition to the steps described above, this step can endow the ancient architecture knowledge graph with logical constraints, enabling in-depth interpretation and reasoning of ancient architecture knowledge, and improving the accuracy of subsequent interactive operation verification.

[0037] S204. Construct user profiles based on basic data of study tour users and knowledge graph of ancient architecture.

[0038] Among them, the study tour user profile is used to depict the user's cognitive level, interest preferences, and learning ability characteristics.

[0039] In one possible implementation, based on user knowledge base data, a structural cognition vector is constructed by normalizing pre-test scores, encompassing structural hierarchy cognition, component function cognition, and spatial relationship cognition. Then, based on user interaction behavior data, a temporal comprehension feature is constructed through weighted calculations of historical content dwell time, number of correct answers, and number of interactions on historical topics. Next, based on user interpretation choices, path selections, and questioning behaviors, a cultural comprehension preference feature is constructed by calculating the percentage of cultural content selections, including symbolic cultural preference, aesthetic cultural preference, and environmental cultural preference. Finally, the fusion weights of the three types of features are determined through a set of rules for the ancient architecture knowledge graph, and a weighted fusion is performed to obtain a comprehension state vector. The values ​​of each dimension of the vector are mapped to comprehension state identifiers according to thresholds, outputting a study tour user profile.

[0040] Among them, the structural cognition vector refers to the vector that describes the user's mastery of the structural hierarchy, component function and spatial relationship of ancient buildings; the temporal understanding feature refers to the feature value that describes the user's ability to understand the historical evolution and era characteristics of ancient buildings; the cultural understanding preference feature refers to the feature vector that describes the user's preference direction for the cultural semantics of ancient buildings; and the understanding state vector refers to the user's comprehensive understanding state vector obtained by integrating multi-dimensional features.

[0041] Based on the above steps, this step can achieve accurate quantification and multi-dimensional characterization of the characteristics of study tour participants, providing a core basis for the generation of personalized study tour paths and improving the adaptability of study tour activities.

[0042] S205. Retrieve knowledge subgraphs from the knowledge graph that match the user profile of the study tour, and generate study tour activity paths.

[0043] Among them, the knowledge subgraph refers to a local knowledge network retrieved from the ancient architecture knowledge graph that is highly matched with the user profile; the study tour route refers to an on-site study tour route formed based on the knowledge subgraph mapping, which includes study tour nodes, activity content, and time allocation.

[0044] In one possible implementation, search conditions are generated based on the user's understanding state identifier and preferred topics. Starting with the core entities corresponding to the preferred topics, a breadth-first search algorithm is used to traverse the knowledge graph, filtering out entities that do not match the user's understanding state, and constructing an initial knowledge subgraph. Then, the association density of the initial subgraph is calculated by the ratio between the number of relations and the number of entities, and the subgraph with the highest association density is retained as the final matching knowledge subgraph. Next, the entities in the subgraph are mapped to on-site study nodes. According to the principle of cognitive progression, the Dijkstra algorithm is used to optimize the node order, and study activity content and duration are assigned to each node to output a personalized study activity path.

[0045] It should be noted that the number of entities in the knowledge subgraph needs to be dynamically adjusted according to the duration of the study tour to avoid overloading or underloading of the path content. At the same time, the path planning needs to take into account both cognitive patterns and the convenience of on-site visits.

[0046] As an example, targeting users with a deep understanding of culture and basic knowledge of structure, we retrieved cultural semantic knowledge subgraphs centered on dragon and phoenix carvings and symbols of imperial power from the Palace Museum's knowledge graph. These subgraphs were then mapped to study routes at the South Gate entrance of the Palace Museum, the Taihe Hall Square, under the eaves of the Taihe Hall, and inside the Taihe Hall. We then allocated activity content and corresponding durations for historical background introduction, overall form observation, basic explanation of brackets, and interpretation of dragon and phoenix carving culture.

[0047] Based on the above steps, this process can achieve a precise match between ancient architectural knowledge and the needs of study tour participants, generate personalized study tour paths that match the users' cognitive level and interests, and improve the relevance and effectiveness of study tour activities.

[0048] Based on the above technical solutions, through the collaborative processing of multi-source data on ancient buildings and data from study tour participants, the structured organization of knowledge, and the generation of personalized paths, a systematic management of study tour activities from data acquisition and knowledge modeling to path generation is achieved. This enables the study tour process to be carried out within a reasonably structured and logically clear knowledge system, which helps to improve the depth of understanding and participation in study tour activities and solves the technical problem that existing technologies cannot meet the needs of multi-dimensional knowledge integration and application in ancient building study tour activities.

[0049] In one possible implementation, combining Figure 2 ,like Figure 3 As shown, after S205, the activity management method for ancient building research and study provided in this application embodiment further includes the following S301 to S304: S301. Obtain the study tour activity path of the study tour user and the interactive operation information for executing the study tour activity path.

[0050] Interactive operation information refers to various operation data generated by study tour users during the execution of study tour activities, including interactive objects, operation types, operation results, etc.

[0051] In one possible implementation, the study tour activity path is invoked, and relying on devices such as study tour apps and smart guide terminals, the user's operational behavior at each study tour node is collected in real time. The interactive objects such as the study tour nodes visited and the knowledge content viewed by the user are recorded. The operation types such as viewing, asking questions, answering, selecting options, and staying are distinguished. The operation results such as the correctness rate of answers, the content of question text, the selected option tags, and the duration of stay at nodes are statistically analyzed. At the same time, the study tour activity path information currently executed by the user is associated, the path identifier, the node sequence number and the interactive operation data are bound together, and the format is standardized and outlier filtering is performed. Finally, a structured interactive operation dataset containing path information is output.

[0052] Based on the above steps, this step can capture users' research and study behavior feedback in real time, providing real and effective data support for subsequent interactive reasoning and profile updates, and realizing dynamic tracking of the research and study process.

[0053] S302. Based on the knowledge graph of ancient architecture and the rule set of the knowledge graph of ancient architecture, reason about the interactive operation information to obtain the matching reasoning result.

[0054] Among them, the local knowledge subgraph refers to the sub-network composed of entities and relationships directly related to user interaction objects extracted from the ancient architecture knowledge graph; the matching reasoning result refers to the conclusion that the user interaction operation conforms to the knowledge graph rule set, including three types: conforming, not conforming, and partially conforming.

[0055] In one possible implementation, based on the objects in the interactive operation information, the target ancient building entity and building component entity corresponding to the interactive operation are located in the ancient building knowledge graph. The structural, historical, and cultural relationships directly associated with the target entity are obtained through a graph query algorithm, and a local knowledge subgraph is constructed. Based on the operation type in the interactive operation information, the structural rules, historical logic rules, and cultural semantic rules in the ancient building knowledge graph rule set corresponding to the local knowledge subgraph are determined. The consistency between the interactive operation result and the retrieved rules is checked one by one. The matching degree is calculated through a preset inference function. Based on the matching degree threshold, the result is determined to be consistent, inconsistent, or partially consistent. A matching inference result set containing the matching type and basis is output.

[0056] Preferably, the matching inference function satisfies the following formula:

[0057]

[0058] Where R represents the result of the interactive operation. Let be the i-th matching rule, n be the total number of matching rules, and Result be the final matching inference result.

[0059] As an example, when a user answers that the ratio of the size of the dougong (bracket set) to the building bay width in the Tang Dynasty is 0.3, the target entity dougong is located and a local knowledge subgraph is constructed. The structural rule coefficient k∈[0.15,0.2] for the ratio of the size of the dougong to the building bay width is retrieved. After matching calculation, the result is determined to be inconsistent, and the reasoning basis is marked as the ratio coefficient of the answer exceeds the limit of the rule.

[0060] Based on the above steps, this step can achieve intelligent reasoning about users' learning behavior based on knowledge graphs and rule sets, accurately determine users' knowledge mastery, and provide a basis for decision-making in generating subsequent guidance information.

[0061] S303. Based on the matching reasoning results, verify the interactive operation information and generate ancient architecture study tour guidance information.

[0062] Among them, the ancient architecture study guide information refers to the prompts generated based on the matching reasoning results, which are used to guide users to correct cognitive biases or deepen their understanding of knowledge. It includes three types: encouraging guidance, corrective guidance, and supplementary guidance.

[0063] In one possible implementation, the results are categorized based on their reasoning type. For results that match the criteria, encouraging guidance information is generated, including positive feedback and advanced knowledge guidance. For results that do not match the criteria, corrective guidance information is generated, including error correction and basic knowledge explanation. For results that partially match the criteria, comprehensive guidance information is generated, including partial affirmation and supplementary knowledge. Next, supplementary knowledge related to the interactive object is retrieved from the ancient architecture knowledge graph. The guidance information is then combined with the knowledge content and polished using natural language generation technology to ensure that the expression is easy to understand. Finally, the guidance information is bound to the user's current study node and interactive operation to output personalized ancient architecture study guidance information.

[0064] As an example, if a user's answer regarding the incorrect proportion coefficient of the dougong (bracket set) does not conform to the reasoning result, a corrective guidance message is generated: "Incorrect. According to the structural rules of ancient architecture, the proportion coefficient of the dougong size to the building bay should be between 0.15 and 0.2. Due to their large size, the proportion coefficient of Tang Dynasty dougong is often close to 0.2. Click to view detailed comparison data of dynasties." This message is pushed in real time after the user submits their answer.

[0065] Based on the above steps, this process can provide users with real-time and accurate guidance for study tours, correct cognitive biases in a timely manner, deepen users' understanding of ancient architecture, and improve the quality of study tours.

[0066] S304. Update user profiles based on study tour activity paths, interactive operation information, and ancient architecture study tour guidance information.

[0067] In one possible implementation, the feature values ​​of the user profile are adjusted based on the results of interactive operations. For correctly answered structured questions, the corresponding dimension value of the structure cognition vector is increased; for operations involving multiple selections of cultural content, the corresponding dimension value of the cultural understanding preference feature is increased; for historical nodes with longer dwell times, the temporal understanding feature value is increased; then, based on the type and content of the guidance information, the user's comprehension weakness dimension is specifically adjusted; next, based on the updated feature values, the comprehension status identifier is remapped according to the threshold classification standard (e.g., updated from basic cognition to intermediate comprehension); finally, the updated user profile is synchronized to the database and cache system, overwriting the original profile data, and the updated study tour user profile is output.

[0068] Based on the above steps, this step enables dynamic optimization of user profiles, making them more accurate in reflecting users' real-time study and research status.

[0069] This application embodiment collects user study tour path execution and interaction operation information in real time, relies on ancient architecture knowledge graph and rule set to complete the reasoning and verification of interactive behavior, generates accurate personalized guidance information, and dynamically iterates user profiles accordingly. This not only realizes real-time and refined management of the study tour process and promotes the upgrading of user profiles to dynamic and accurate, but also promotes users to move from superficial visits to in-depth knowledge understanding, while strengthening the intelligence and scalability of the entire management method.

[0070] In one possible implementation of the embodiments of this application, combined with Figure 2 ,like Figure 4 As shown, the above S202 can be specifically implemented through the following S401 to S404, which are explained in detail below: S401. Perform data preprocessing on multi-source data of ancient buildings.

[0071] One possible implementation involves inputting multi-source raw data on ancient buildings and performing preprocessing such as word segmentation, outlier removal, and data alignment.

[0072] S402. Based on the established entity type system in the field of ancient architecture, perform entity recognition on the preprocessed multi-source data of ancient architecture, and extract ancient building entities, building component entities, historical event entities, cultural connotation entities, and human entity entities.

[0073] In one possible implementation, a preprocessed multi-source dataset of ancient buildings and a preset entity type system for the ancient building domain are input. A BERT entity recognition model that has been fine-tuned and validated in the domain is called. Standardized text and structured data are converted according to the model input format. After being input into the model, the model outputs the type labels and text boundaries of each entity in the data. The model output results are initially screened to remove candidate entities with confidence scores below a preset threshold. The same entity with different descriptions is normalized, and finally, a precise set of ancient building entities, building component entities, historical event entities, cultural connotation entities, and figure entities is output.

[0074] It should be noted that a fine-tuned BERT model is used for entity recognition. The fine-tuned BERT model is built upon the BERT-base model. The parameters of the first set number of layers in the BERT-base model are frozen, while the parameters of the second set number of layers and the classification layer parameters are fine-tuned. The fine-tuning process uses text data with entity annotations in the field of ancient architecture as training samples. The BERT-base model is trained using supervised learning to obtain the fine-tuned BERT model. The first and second preset number of layers can be adjusted according to the actual downstream task. In this embodiment, the parameters of the first 8 layers are frozen, while the parameters of the last 4 layers and the classification layer parameters are fine-tuned using the cross-entropy loss function. This embodiment does not limit the specific implementation of this method.

[0075] Preferably, the confidence level of entity recognition Satisfy the following formula:

[0076] Based on the above steps, this step can efficiently and accurately extract core entities in the field of ancient architecture, identify the core nodes of the knowledge graph, and provide core elements for subsequent relationship extraction and graph construction.

[0077] S403. Based on the contextual semantic relationships between entities in the multi-source data of ancient buildings, perform relation extraction on the identified entities to obtain entity relation pairs.

[0078] Among them, entity relationship pairs are used to reflect the structural composition relationships, historical connections, and cultural relationships of ancient buildings.

[0079] In one possible implementation, based on the set of ancient building entities and preprocessed multi-source data of ancient buildings, a preset relation extraction model is invoked. The model outputs the relationship type and corresponding confidence level between entity pairs based on the contextual semantic features of the entities. A confidence level screening threshold is set for the model output results to remove candidate relation pairs with low confidence. Then, the misjudgment of relation types is corrected by manual sampling verification (sampling ratio ≥10%), with a focus on distinguishing easily confused relations, such as construction and repair, inclusion and subordination. Finally, a precise set of entity relation pairs covering the three core relation types is output.

[0080] It should be noted that the preset relation extraction model is a dedicated model optimized for entity relation recognition in the field of ancient architecture, based on the BERT model for entity recognition. The main architecture retains the core of the BERT-base encoder, while adding a feature fusion layer and a relation classification layer. The input layer adapts to the contextual semantic input format of entity pairs, extracts global features and entity local features, concatenates them for dimensionality reduction, and then outputs three types of relation labels and confidence scores through a Softmax classification layer: structural composition, historical association, and cultural relationship. The model's training reuses the pre-trained weights of the entity recognition model, fine-tuning the parameters of the six layers based on 50,000 labeled relations in the field of ancient architecture, optimizing with the cross-entropy loss function, and undergoing validation on the test set and secondary fine-tuning with easily confused relation samples to finally obtain a high-precision relation extraction model.

[0081] S404. Store the ancient building entities and their relationships in a structured manner to construct an ancient building knowledge graph and generate structured ancient building knowledge corresponding to the ancient building knowledge graph.

[0082] Among them, structured storage refers to the use of graph databases and other adapted related data storage methods to organize and store entities (nodes) and relationships (edges) in a standardized format; structured ancient architecture knowledge refers to knowledge documents with a fixed format based on knowledge graph transformation.

[0083] In one possible implementation, triples are used to model entities and relations; the triple data is imported into a graph database, and information such as name, type, and attributes are added to entity nodes, and information such as type and confidence is added to relation edges to construct a knowledge graph of ancient architecture; structured documents such as entity association tables, component structure tables, and historical and cultural association tables are generated based on the graph data; the graph is then validated for completeness, and missing entities and relations are supplemented, ultimately outputting the knowledge graph of ancient architecture and structured knowledge of ancient architecture.

[0084] It should be noted that the storage of knowledge graphs requires optimization of the index design of nodes and edges to improve subsequent retrieval efficiency. Knowledge graphs can be optimized based on the updating and enrichment of data, and can also be further applied to fields other than ancient architecture; however, this application does not limit this application.

[0085] As an example, the entities and relationships of the Forbidden City are modeled as RDF triples, imported into Neo4j to build a knowledge graph of the Forbidden City, generating an entity association table containing: entity ID, entity name, relationship type, and associated entity ID, supplementing the missing bracket, subordinate, and roof relationships, and outputting a complete knowledge graph and structured knowledge.

[0086] Based on the above steps, this step enables the visualization and networking of ancient architectural knowledge, integrating fragmented entities and relationships into a systematic knowledge system, which facilitates subsequent rule construction and knowledge retrieval.

[0087] This application's embodiments, through standardized preprocessing of multi-source data on ancient buildings, accurate entity recognition using domain-adaptive models, extraction of three types of core relationships based on contextual semantics, and then graph-based storage and structured knowledge generation of entities and relationships, not only transform scattered and heterogeneous original data on ancient buildings into a systematic and visualized knowledge network, solving the problems of fragmented and inefficiently reusable knowledge of ancient buildings, but also provide a core knowledge carrier for subsequent rule set construction, user profile matching, and personalized study path generation. At the same time, the unified entity and relationship definition standard also improves the migration capability and application flexibility of the entire management method in different ancient building scenarios.

[0088] In one possible implementation of the embodiments of this application, combined with Figure 2 ,like Figure 5 As shown, the above S203 can be specifically implemented through the following S501 to S504, which are explained in detail below: S501. Based on the data of ancient building structure and components, extract the hierarchical relationship, combination relationship and constraint relationship between building components, and determine the structural rules between ancient building components.

[0089] Among them, hierarchical relationship refers to the subordinate relationship between components; combination relationship refers to the assembly and matching relationship between components; constraint relationship refers to the limiting conditions such as size and material of component combination.

[0090] In one possible implementation, the system inputs data on the structure and components of ancient buildings. First, it uses expert knowledge to sort out the hierarchical affiliation and assembly methods of the components, and preliminarily extracts the hierarchical and combination relationships. Then, it performs statistical analysis on the component dimensions in the survey data, uses linear regression to fit the correlation between the component dimensions and the main building dimensions, and determines the dimensional constraints. Finally, it supplements the material constraints by combining material testing data. The extracted relationships and constraints are organized into rule clauses, reviewed and corrected by domain experts, and finally outputs the structural rules for ancient building components.

[0091] It should be noted that structural rules need to differentiate between different building types to avoid compatibility issues caused by generalizing rules; dimensional constraints need to specify the applicable historical period, as component size standards differ across dynasties.

[0092] Based on the steps outlined above, this step clarifies the core relationships and constraints between components of the ancient building, providing a logical basis for verifying the structural rationality of subsequent interactive operations. By combining expert knowledge with data fitting, the structural rules of the components are precisely extracted.

[0093] S502. Based on historical background data of ancient buildings, extract the formal characteristics and evolutionary constraints of ancient buildings in different periods, and construct the historical logic rules of ancient buildings.

[0094] Among them, formal characteristics refer to the typical features of ancient buildings in different historical periods in terms of roof slope, component style, layout, etc.; evolutionary constraints refer to the restrictions on the form of ancient buildings as they develop in history.

[0095] In one possible implementation, based on historical background data of ancient buildings, the data is first classified according to historical dynasties, the core formal features of ancient buildings in each dynasty are extracted, and a mapping table between dynasties and formal features is established; the evolution of formal features is sorted out through time series analysis, and the evolution order and non-reversible constraints are clarified; the mapping table and evolution constraints are organized into rule texts, and finally the historical logic rules of ancient buildings are output.

[0096] It should be noted that the rules of historical logic need to take into account the influence of regional differences on form. Ancient buildings in different regions during the same dynasty may have variations in form; evolutionary constraints need to indicate the sudden changes in form caused by special historical events.

[0097] Based on the steps outlined above, this step clarifies the differences in form and evolution of ancient buildings across different historical periods, providing a logical basis for verifying the historical rationality of subsequent interactive operations. Through dynastic classification and chronological analysis, it precisely extracts the rules of historical logic.

[0098] S503. Based on the data of ancient architectural cultural connotations, construct cultural semantic rules to define the mapping relationship between the structural features of ancient buildings and cultural semantics.

[0099] Among them, cultural semantics refers to the cultural connotations such as religious symbolism, environmental culture, aesthetic thought, and hierarchical system carried by the structural features of ancient buildings; mapping relationship refers to the fixed association between specific structural features and corresponding cultural semantics.

[0100] In one possible implementation, the data on the cultural connotations of ancient buildings is input. First, cultural descriptions related to the structure of ancient buildings are extracted from the data, and a database of correspondences between structural features and cultural descriptions is established. Then, the core semantics of the cultural descriptions are extracted through semantic analysis. The unique or many-to-one mapping relationship between structural features and core semantics is clarified, and ambiguous mappings are eliminated. The mapping relationship is organized into rule clauses, reviewed and corrected by experts in the cultural field, and finally, the semantic rules of ancient building culture are output.

[0101] It should be noted that cultural semantic rules need to distinguish the differences in different cultural contexts. The same structural feature may have semantic differences in different regions and ethnic cultures. It is also necessary to mark the historical evolution of cultural semantics, as the cultural semantics of some structural features will change with the times.

[0102] Based on the above steps, this process clarifies the connection between the structure and cultural connotation of ancient buildings, providing a logical basis for verifying the cultural semantic rationality of subsequent interactive operations and matching user cultural preferences. Through semantic analysis and expert review, cultural semantic rules are accurately extracted, providing core cultural logical support for the construction of the rule set.

[0103] S504. Bind the structural rules, historical logic rules, and cultural semantic rules to the entities and entity relationships in the ancient architecture knowledge graph to obtain the rule set of the ancient architecture knowledge graph.

[0104] Among them, rule binding refers to establishing associations between the three types of rules and the corresponding entity nodes and relation edges in the knowledge graph, thus endowing the graph with logical constraint capabilities; the ancient architecture knowledge graph rule set refers to a structured rule system that integrates the three types of rules and associates them with the graph.

[0105] In one possible implementation, labels are added to structural rules, historical logic rules, and cultural semantic rules to indicate the applicable entity types and relationship types; in the Neo4j graph database, rule ID attributes are added to the corresponding entity nodes and relationship edges to establish the association mapping between entities and rules, and between relationships and rules; the three types of rules are imported into the rule engine, and the linkage logic between rules and graph queries is configured; the accuracy of rule binding is verified, and misbinding and omission issues are corrected, and finally, the set of ancient architecture knowledge graph rules is output.

[0106] It should be noted that rule binding must ensure accurate one-to-one or one-to-many associations to avoid binding a rule to unrelated entities; the linkage logic of the rule engine needs to optimize response efficiency to ensure the real-time performance of subsequent reasoning.

[0107] Based on the above steps, this step can endow the ancient architecture knowledge graph with logical constraints, realize in-depth interpretation and reasoning judgment of knowledge, enhance the application value of the knowledge graph, construct a set of rules that are deeply integrated with the knowledge graph, and provide core logical basis for subsequent interactive reasoning and user profile integration.

[0108] This application's embodiments construct structural rules for building components, historical logic rules for ancient buildings, and semantic mapping rules for structure and culture in a layered manner. Then, these three types of rules are precisely bound to entities and relationships in the knowledge graph. This not only endows the knowledge graph with logical reasoning and verification capabilities, solving the shortcomings of traditional knowledge graphs that only have association storage functions but lack semantic constraints, but also provides a clear logical basis for judging the rationality of subsequent user interaction operations and generating personalized guidance information. At the same time, the layered design of the three types of rules and the graph binding mode can flexibly adapt to different regions and types of ancient building study scenarios, improving the scenario adaptability and scalability of the entire management method.

[0109] In one possible implementation of the embodiments of this application, combined with Figure 2 ,like Figure 6 As shown, the above S204 can be specifically implemented through the following S601 to S604, which are explained in detail below: S601. Based on the knowledge base data of study tour users, construct a structural cognitive vector of users' understanding of the structural hierarchy, component functions and spatial relationships of ancient buildings.

[0110] Among them, the knowledge base data refers to the user's knowledge reserve data related to ancient building structures obtained through pre-tests, questionnaires and other methods; the structural cognition vector refers to the multi-dimensional vector that quantifies the user's mastery of the core dimensions of ancient building structures, with dimensions corresponding to structural levels, component functions and spatial relationships.

[0111] In one possible implementation, the knowledge base data of the study tour participants is input, and three core dimensions of structural cognition are first determined: structural hierarchy cognition, component function cognition, and spatial relationship cognition. The test questionnaire scores for each dimension are normalized to obtain the cognition level for each dimension. The cognition levels of the three dimensions are then used as vector elements to construct a structural cognition vector, with the vector form being... ,in, For the understanding of structural hierarchy, To improve the understanding of component functions, This represents the spatial relationship cognition level. The vectors are validated for validity, outliers are removed, and the final output is the user's structural cognition vector.

[0112] Based on the above steps, this step can quantify users' level of understanding of ancient building structures, clarify users' strengths and weaknesses in structural knowledge, and provide core elements of structural cognition for subsequent user profile integration.

[0113] S602. Based on the interactive behavior data of study tour users during study tour activities, construct the temporal understanding features of users' understanding of the historical evolution and era characteristics of ancient buildings.

[0114] Interactive behavior data refers to the records of users' interactions with history-related content during historical study activities, including dwell time, number of questions asked, and answer status; temporal understanding features refer to single-dimensional feature values ​​that quantify users' ability to understand the historical evolution of ancient buildings and the stylistic features of different eras.

[0115] In one possible implementation, based on the interactive behavior data of study tour users, including the duration of historical content viewing, the number of historical related questions, the number of correct answers to historical questions, and the number of interactions on historical topics, core influencing indicators are first selected: the proportion of historical content viewing time, the accuracy rate of historical questions, and the proportion of interactions on historical topics. Weights are assigned to each indicator based on expert experience. The specific values ​​of each indicator are calculated and normalized. The time-series understanding feature value is calculated by weighted summation, and finally, the user's time-series understanding feature is output.

[0116] S603. Based on the interpretation choices, path selections, and questioning behaviors of study tour participants in their interactive activities, construct user understanding preferences for the semantics of ancient architectural culture.

[0117] Among them, interpretation selection refers to the user's choice of interpretation options for the cultural connotation of ancient buildings; path selection refers to the user's choice of study paths related to cultural themes; and cultural semantic understanding preference features refer to a multi-dimensional vector that quantifies the degree of user preference for different cultural semantic types.

[0118] In one possible implementation, based on information such as cultural interpretation selection records, cultural theme path selection records, and cultural related question records from the interactive behavior data of study tour users, three core dimensions of cultural semantic preference are first determined: symbolic cultural preference, aesthetic cultural preference, and environmental cultural preference; the number of times users select content in each cultural dimension is counted, and the proportion of the number of times each dimension is selected to the total number of times cultural content is selected is calculated; the three proportions are used as vector elements to construct cultural semantic understanding preference features.

[0119] S604. Integrate structural cognitive vectors, temporal understanding features, and cultural understanding preference features to determine the profile of study tour users.

[0120] Among them, the research and study user profile refers to mapping the understanding state vector to specific understanding state identifiers, including user characteristic descriptions such as basic cognition, intermediate understanding, and in-depth mastery.

[0121] In one possible implementation, based on structural cognitive vectors, temporal understanding features, cultural understanding preference features, and a set of rules for ancient architecture knowledge graphs, the fusion weights of each feature are first determined by rule importance scores, with the weights being normalized from the average importance scores of the three types of rules. The three features are then weighted and fused based on the fusion weights to obtain a user understanding state vector. An understanding state threshold is set, and the values ​​of each dimension of the understanding state vector are mapped to corresponding understanding state identifiers. Finally, the understanding state identifiers of each dimension are integrated with the user's basic information to output a profile of the study tour user.

[0122] As an example, the rule importance score is calculated by weighting the rule usage rate (the proportion of each rule's frequency of use in research and reasoning) and the expert importance score (out of 5). The weight proportions are 0.6 for rule usage rate and 0.4 for expert importance score. The relevance score of a single rule is calculated, and then the relevance scores of the three rule types are normalized to obtain the fusion weight. For example, if the relevance score of structural rule is 0.4, historical logic rule is 0.3, and cultural semantic rule is 0.3, then the fusion weights are 0.4, 0.3, and 0.3, respectively.

[0123] As an example, we set the comprehension state thresholds as follows: [0, 0.3) represents basic cognition, [0.3, 0.7) represents intermediate comprehension, and [0.7, 1.0] represents deep mastery.

[0124] This application's embodiments extract users' structural cognitive vectors, temporal understanding features, and cultural understanding preference features in a hierarchical manner. Then, they combine these features with weights determined by the rules set of the ancient architecture knowledge graph for weighted fusion. This not only achieves an accurate portrayal of users' knowledge of ancient architecture, historical understanding, and cultural interests, solving the shortcomings of traditional user profiles that are single-dimensional and subjective, but also provides a core decision-making basis for the generation of personalized study tour paths through the user profiles formed by feature fusion and threshold mapping, effectively improving the matching degree between study tour content and user needs.

[0125] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0126] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

Claims

1. A management method for activities related to ancient architecture research and study, characterized in that, include: Acquire multi-source data on ancient buildings and basic data on study tour users; The multi-source data on ancient buildings includes: physical data of ancient buildings, data on building structure and components, historical background data, cultural connotation data, and data on people; the basic data of study tour users includes: basic user attribute data, basic knowledge data, learning preference data, and interactive behavior data. Entity recognition and relationship extraction are performed on the multi-source data to construct a knowledge graph of ancient architecture and form structured knowledge of ancient architecture. Based on the multi-source data and the ancient architecture knowledge graph, a set of rules for the ancient architecture knowledge graph is constructed; the set of rules for the ancient architecture knowledge graph includes: structural rules between ancient architectural components, historical logic rules of ancient architecture, and cultural semantic rules of ancient architecture. User profiles are constructed based on the basic data of the study tour users and the ancient architecture knowledge graph; the study tour user profiles are used to characterize the users' cognitive level, interest preferences, and learning ability characteristics. Retrieve knowledge subgraphs from the knowledge graph that match the profile of the study tour user, and generate study tour activity paths.

2. The method according to claim 1, characterized in that, After generating the study tour activity path, the method further includes: Obtain the study tour activity paths of study tour users and the interactive operation information for executing study tour activities; Based on the knowledge graph of ancient architecture and the rule set of the knowledge graph of ancient architecture, the interactive operation information is inferred to obtain the matching inference result; Based on the matching reasoning results, the interactive operation information is verified, and ancient architecture study tour guidance information is generated; The user profile is updated based on the study tour activity path, the interactive operation information, and the ancient architecture study tour guidance information.

3. The method according to claim 1, characterized in that, The process of performing entity recognition and relation extraction on the multi-source data to construct an ancient architecture knowledge graph and form structured ancient architecture knowledge includes: Data preprocessing is performed on the multi-source data of the ancient buildings; Based on the established entity type system in the field of ancient architecture, entity recognition is performed on the preprocessed multi-source data of ancient architecture to extract entities of ancient architecture, entities of building components, entities of historical events, entities of cultural connotations, and entities of people. Based on the contextual semantic relationships between entities in multi-source data of ancient buildings, the identified entities are subjected to relation extraction to obtain entity relation pairs; the entity relation pairs are used to reflect the structural composition relationships, historical association relationships and cultural relationships of ancient buildings. The ancient building entities and their relationships are stored in a structured manner to construct an ancient building knowledge graph, and structured ancient building knowledge corresponding to the ancient building knowledge graph is generated.

4. The method according to claim 3, characterized in that, The system is based on a set entity type system for ancient architecture. It performs entity recognition on preprocessed multi-source data of ancient architecture and uses a fine-tuned BERT model to achieve entity recognition.

5. The method according to claim 4, characterized in that, The fine-tuned BERT model is built on the BERT-base model. By freezing the parameters of the first set number of layers in the BERT-base model, the parameters of the second set number of layers and the classification layer parameters are fine-tuned. The fine-tuning process uses text data with entity annotations in the field of ancient architecture as training samples. The BERT-base model is fine-tuned and trained through supervised learning to obtain the fine-tuned BERT model.

6. The method according to claim 1, characterized in that, The step of constructing a set of rules for an ancient architecture knowledge graph based on the multi-source data and the ancient architecture knowledge graph includes: Based on the data of ancient building structure and components, the hierarchical relationship, combination relationship and constraint relationship between building components are extracted to determine the structural rules between ancient building components. Based on historical background data of ancient buildings, we extract the formal characteristics and evolutionary constraints of ancient buildings in different periods and construct the historical logic rules of ancient buildings. Based on data on the cultural connotations of ancient buildings, we construct cultural semantic rules to define the mapping relationship between the structural features of ancient buildings and their cultural semantics. The structural rules, historical logic rules, and cultural semantic rules are bound to the entities and entity relationships in the ancient architecture knowledge graph to obtain the ancient architecture knowledge graph rule set.

7. The method according to claim 1, characterized in that, The process of constructing user profiles based on the basic data of the study tour users and the ancient architecture knowledge graph includes: Based on the knowledge base data of study tour users, construct a structural cognitive vector of users’ understanding of the structural hierarchy, component functions and spatial relationships of ancient buildings; Based on the interactive behavior data of study tour participants during study tour activities, we construct the temporal understanding features of users' understanding of the historical evolution and era characteristics of ancient buildings. Based on the interpretation choices, path selections, and questioning behaviors of study tour participants during their interactive activities, we construct the characteristics of users' understanding preferences for the semantics of ancient architectural culture. By fusing the structural cognitive vector, temporal understanding features, and cultural understanding preference features, a profile of study tour users is determined.

8. The method according to claim 7, characterized in that, The process of fusing the structural cognitive vector, temporal understanding features, and cultural understanding preference features to determine the profile of study tour users includes: By determining the degree of correlation between structural rules, historical logic rules, and cultural semantic rules in the ancient architecture knowledge graph rule set, the fusion weights of the structural cognitive vector, temporal understanding features, and cultural understanding preference features are determined. Based on the fusion weights, the structural cognition vector, temporal understanding features, and cultural understanding preference features are weighted and fused to obtain the understanding state vector; The understanding state vector is mapped to the understanding state identifiers of study tour users in the dimensions of structural understanding, historical understanding, and cultural understanding, thereby determining the profile of study tour users.

9. The method according to claim 2, characterized in that, The process of reasoning about the interactive operation information based on the ancient architecture knowledge graph and its rule set to obtain matching reasoning results includes: Based on the objects in the interactive operation information, locate the target ancient building entity and building component entity corresponding to the interactive operation in the ancient building knowledge graph, and obtain the structural relationship, historical relationship and cultural relationship directly associated with the target entity to construct a local knowledge subgraph; Based on the operation type in the interactive operation information, determine the structural rules, historical logic rules, and cultural semantic rules in the ancient architecture knowledge graph rule set corresponding to the local knowledge subgraph; The results in the interactive operation information are matched and judged one by one with the structural rules, historical logic rules and cultural semantic rules to obtain the matching reasoning results.

10. An activity management system for ancient architecture research and study, characterized in that, The system includes: a data acquisition module, a knowledge graph construction module, a rule extraction module, a user profile construction module, and a study tour path planning module; The data acquisition module is used to acquire multi-source data on ancient buildings and basic data on study tour users. The multi-source data on ancient buildings includes: physical data of ancient buildings, data on building structures and components, historical background data, cultural connotation data, and data on people. The basic data on study tour users includes: basic user attribute data, basic knowledge data, learning preference data, and interactive behavior data. The knowledge graph construction module performs entity recognition and relationship extraction on the multi-source data to construct an ancient architecture knowledge graph and form structured ancient architecture knowledge. The rule extraction module constructs a set of rules for the ancient architecture knowledge graph based on the multi-source data and the ancient architecture knowledge graph. The set of rules for the ancient architecture knowledge graph includes: structural rules between ancient architectural components, historical logic rules of ancient architecture, and cultural semantic rules of ancient architecture. The user profile building module constructs and updates user profiles based on the basic data of the study tour users and the ancient architecture knowledge graph; the study tour user profile is used to characterize the user's cognitive level, interest preferences and learning ability characteristics. The study tour path planning module retrieves knowledge subgraphs from the knowledge graph that match the study tour user profile and generates study tour activity paths.

Images