Method and system for constructing lake and Hunan pattern culture recognition degree evaluation network

Abstract

The invention relates to the technical field of digital evaluation of cultural heritage, and particularly discloses a lake and Hunan pattern culture recognition degree evaluation network construction method and system, and the method comprises the steps: collecting pattern ontology data, audience cognition and emotion data and pattern transmission and behavior data, and constructing a multi-source data set; constructing a network model containing pattern nodes, audience nodes and context nodes based on the multi-source data set, and synchronously defining a cognitive relationship, an emotion relationship and a co-occurrence relationship among the nodes; based on the centrality index and the dynamic emotion weighted recognition score, the cultural recognition degree of the pattern is generated; performing visual display on the network model and the evaluation result; according to the method, the network model integrating the multi-source data is constructed, when the network model is constructed, the multi-source data is quantized, the relationship among different nodes is synchronously determined, and the challenges of single data dimension, lack of relationship modeling, one-sided evaluation indexes and insufficient dynamic monitoring in the prior art are effectively handled.

Description

Technical Field

[0001] This invention relates to the field of digital assessment technology for cultural heritage, specifically a method and system for constructing a network for assessing the cultural identity of Hunan patterns. Background Technology

[0002] As an important visual carrier and material legacy of Hunan culture, Hunan patterns carry profound historical memories, regional characteristics, and national sentiments. Conducting a scientific assessment of their cultural identity is of vital importance for understanding the contemporary value of traditional culture, promoting cultural innovation and transformation, and formulating effective cultural dissemination strategies.

[0003] Currently, the field of cultural symbol identification assessment mainly relies on qualitative research methods, including sociological and anthropological approaches such as literature review, fieldwork, in-depth interviews, and focus groups. While these methods can achieve deep contextual understanding, they are limited by inherent flaws such as limited sample size, significant influence of subjective judgment, and difficulty in large-scale replication. This results in assessment results that cannot effectively quantify macro-level identification trends and dynamic evolution processes. In addition, some studies have turned to single-dimensional quantitative analysis, such as conducting statistical analysis through questionnaires or using web crawlers to conduct social media sentiment analysis, focusing only on mention frequency or simple sentiment classification. These methods artificially separate key dimensions such as cognition, emotion, and behavior, lacking a comprehensive assessment framework. Specifically, simply counting the online mention frequency of patterns cannot distinguish between cognitive breadth and emotional depth, and simple sentiment analysis ignores differences in cognitive levels, the influence of communication context, and the weight of audience influence, resulting in a one-sided assessment perspective. Therefore, how to provide a comprehensive pattern information statistical scheme with high accuracy in evaluation results is the technical problem that this invention aims to solve. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for constructing a network for evaluating the cultural identity of Hunan patterns, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method and system for constructing a network for assessing the cultural identity of Hunan patterns, the method comprising: Collect pattern sample data, audience cognition and emotion data, and pattern dissemination and behavior data to construct a multi-source dataset; A network model containing pattern nodes, audience nodes, and context nodes is constructed based on a multi-source dataset, and the cognitive relationships, emotional relationships, and co-occurrence relationships between these nodes are defined simultaneously. The centrality index of each pattern node in the network model is calculated, and the dynamic emotional weighted identification score of each pattern node is calculated based on the preset weights. The cultural identification score of the pattern is generated based on the centrality index and the dynamic emotional weighted identification score. The cultural identification score is used to quantitatively evaluate the pattern. The network model and evaluation results are visualized and updated periodically; the evaluation results include cultural identity and trend analysis results of cultural identity.

[0006] As a further aspect of the present invention: the step of collecting pattern sample body data, audience cognition and emotion data, and pattern dissemination and behavior data to construct a multi-source dataset includes: Establish a connection channel with the digital resource library to obtain the image, name, era, and symbolic information of the pattern, and construct the pattern data layer; the symbolic information is text. Build and push out survey questionnaires, receive questionnaire data from user feedback, generate public awareness, understanding, emotional attitudes and behavioral intentions regarding patterns, and build a cognitive and emotional data layer for the audience. By crawling network data from news, e-commerce platforms, and cultural applications in media data, we can obtain data on the frequency of pattern appearance, application scenarios, and public interaction, thus forming a communication and behavioral data layer. The statistical texture sample data layer, the audience cognition and emotion data layer, and the communication and behavior data layer are used to obtain a multi-source dataset.

[0007] As a further aspect of the present invention: the step of constructing a network model containing pattern nodes, audience nodes, and context nodes based on a multi-source dataset, and simultaneously defining the cognitive relationships, emotional relationships, and co-occurrence relationships between the nodes, includes: Pattern nodes are constructed based on the pattern body data layer, audience nodes are constructed based on the audience cognition and emotion data layer, and context nodes are constructed based on the communication and behavior data layer. Cognitive and emotional relationships are determined based on audience cognition and emotion data layers, and co-occurrence relationships are determined based on communication and behavior data layers. The pattern nodes are used to characterize the identity features of the pattern, including but not limited to the pattern number; The audience node is used to represent the identity characteristics of the pattern audience, including but not limited to social media user IDs and user profile tags; The context nodes are used to represent the interactive scenarios of the pattern, including but not limited to the platforms, scenarios, and derivatives where the pattern appears; The cognitive relationship is established between the audience node and the pattern node, and its relationship weight is determined by the cognitive score and mention frequency of the pattern audience. The emotional relationship is established between the audience node and the pattern node, and its relationship attributes are determined by the emotional polarity and intensity. The co-occurrence relationship is established between pattern nodes, and its weight is determined by the frequency of the two patterns appearing together in the same context.

[0008] As a further aspect of the present invention: the centrality index includes degree centrality, eigenvector centrality and betweenness centrality, wherein the degree centrality is used to measure the cognitive breadth of the pattern, the eigenvector centrality is used to measure the recognition depth of the pattern among the audience, and the betweenness centrality is used to identify patterns with a transitive role in the network.

[0009] As a further aspect of the present invention: the step of calculating the dynamic emotional weighted identification score of each pattern node based on preset weights, and generating the cultural identification degree of the pattern based on the centrality index and the dynamic emotional weighted identification score includes: For any pattern node, obtain the total number of interactive behaviors related to the pattern within a preset time window in its propagation and behavior data layer; For any interactive behavior, read its corresponding audience cognition and emotion data layer, normalize each data, and obtain the normalized emotion value of each interactive behavior. The normalized sentiment values ​​of all interactive behaviors are statistically analyzed to obtain the dynamic sentiment-weighted identification score of the pattern nodes; Cultural identity is determined based on the centrality index of pattern nodes and the dynamic emotional weighted identity score.

[0010] As a further aspect of the present invention: the step of visualizing the network model and evaluation results, and periodically generating update instructions includes: In the network model, pattern nodes are clustered to construct pattern communities, and the network model and its clustering results are displayed based on a graph data structure. Obtain the cultural recognition degree with time label for each pattern node, fit the recognition degree curve, and insert the recognition degree curve as an attribute into the corresponding pattern node in the network model. The update frequency of the network model is determined based on the acceptance curve, and update instructions are generated based on the update frequency.

[0011] The present invention also provides a network construction system for assessing the cultural identity of Hunan patterns, the system comprising: The dataset construction module is used to collect pattern sample data, audience cognition and emotion data, and pattern dissemination and behavior data to build a multi-source dataset; The network model construction module is used to build a network model containing pattern nodes, audience nodes, and context nodes based on a multi-source dataset, and simultaneously define the cognitive relationships, emotional relationships, and co-occurrence relationships between the nodes. The network model application module is used to calculate the centrality index of each pattern node in the network model, calculate the dynamic sentiment-weighted identification score of each pattern node based on preset weights, and generate the cultural identification degree of the pattern based on the centrality index and the dynamic sentiment-weighted identification score; the cultural identification degree is used to quantitatively evaluate the pattern. The visualization module is used to visualize the network model and evaluation results, and to generate update instructions periodically; the evaluation results include cultural identity and trend analysis results of cultural identity.

[0012] As a further aspect of the present invention: the dataset construction module includes: The pattern layer establishment unit is used to establish a connection channel with the digital resource library, acquire the image, name, era, and symbolic information of the pattern, and construct the pattern body data layer; the symbolic information is text. The cognitive layer building unit is used to construct and push survey questionnaires, receive questionnaire data from user feedback, generate the public's awareness, understanding, emotional attitude and behavioral intention of the pattern, and construct the audience's cognitive and emotional data layer. The behavior layer building unit is used to crawl network data from news, e-commerce platforms and cultural applications in media data to obtain the frequency of pattern appearance, application scenarios and public interaction data, thus forming the communication and behavior data layer. The data statistics unit is used to statistically analyze the texture sample data layer, the audience cognition and emotion data layer, and the communication and behavior data layer to obtain a multi-source dataset.

[0013] As a further aspect of the present invention: the network model construction module includes: The node construction unit is used to construct pattern nodes based on the pattern body data layer, audience nodes based on the audience cognition and emotion data layer, and context nodes based on the communication and behavior data layer. The relationship determination unit is used to determine cognitive and emotional relationships based on the audience's cognitive and emotional data layer, and to determine co-occurrence relationships based on the communication and behavioral data layer. The pattern nodes are used to characterize the identity features of the pattern, including but not limited to the pattern number; The audience node is used to represent the identity characteristics of the pattern audience, including but not limited to social media user IDs and user profile tags; The context nodes are used to represent the interactive scenarios of the pattern, including but not limited to the platforms, scenarios, and derivatives where the pattern appears; The cognitive relationship is established between the audience node and the pattern node, and its relationship weight is determined by the cognitive score and mention frequency of the pattern audience. The emotional relationship is established between the audience node and the pattern node, and its relationship attributes are determined by the emotional polarity and intensity. The co-occurrence relationship is established between pattern nodes, and its weight is determined by the frequency of the two patterns appearing together in the same context.

[0014] As a further aspect of the present invention: the centrality index includes degree centrality, eigenvector centrality and betweenness centrality, wherein the degree centrality is used to measure the cognitive breadth of the pattern, the eigenvector centrality is used to measure the recognition depth of the pattern among the audience, and the betweenness centrality is used to identify patterns with a transitive role in the network.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention constructs a network model that integrates multi-source data. When constructing the network model, the multi-source data is quantified, and the relationship between different nodes is determined simultaneously. This effectively addresses the challenges of existing technologies, such as single data dimension, lack of relationship modeling, one-sided evaluation indicators, and insufficient dynamic monitoring. In addition, it is extremely easy to interface with visualization technology and can be directly displayed, resulting in a very high degree of data integration. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention.

[0017] Figure 1 The overall flowchart of the method for constructing a network to assess the cultural identity of Hunan patterns is shown.

[0018] Figure 2 The diagram shows the structure of the network construction system for assessing the cultural identity of Hunan patterns. Detailed Implementation

[0019] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0020] Figure 1 This invention provides a flowchart of a method and system for constructing a network to assess the cultural identity of Hunan patterns. In this embodiment, a method for constructing such a network includes: Step S100: Collect pattern sample data, audience cognition and emotion data, and pattern dissemination and behavior data to construct a multi-source dataset; The multi-dimensional data collection process refers to the process of acquiring pattern samples, audience cognition and emotion, and dissemination behavior information from heterogeneous data sources. It can be achieved by manually consulting historical archives to obtain the images and symbolic information of patterns as pattern sample data, by recording behavioral trajectory data of offline cultural activities through sensor devices as audience cognition and emotion data, and by acquiring online interaction data to determine pattern dissemination and behavioral data. Specific operational processes include, but are not limited to, organizing pattern chronology information in museum collection databases and using wearable devices to collect the duration of public stay and interaction frequency at exhibition sites.

[0021] Step S200: Construct a network model containing pattern nodes, audience nodes, and context nodes based on a multi-source dataset, and simultaneously define the cognitive relationships, emotional relationships, and co-occurrence relationships between the nodes; The construction of a multi-relationship network model refers to the process of establishing a topological structure that includes pattern nodes, audience nodes, and context nodes and defining the relationships between nodes. This can be achieved by using a pattern coding system to uniquely identify patterns, or by generating anonymized audience tags based on user behavior clustering. For example, pattern nodes can be associated with unique numbers in a digital resource database, and context nodes can be mapped to coordinate locations in a geographic information system. The main purpose is to achieve a structured representation of the complex relationships between cultural elements.

[0022] Furthermore, specifically, the cognitive, emotional, and co-occurrence relationships between nodes can be quantitatively defined based on data characteristics. The weight of cognitive relationships can be calculated through knowledge test scores, the attributes of emotional relationships can be determined based on voice emotion recognition technology, and the weight of co-occurrence relationships can be statistically analyzed using time series analysis methods. The main purpose of this is to capture the dynamic interaction mechanism in the process of identity formation.

[0023] Step S300: Calculate the centrality index of each pattern node in the network model, calculate the dynamic sentiment-weighted identification score of each pattern node based on the preset weights, and generate the cultural identification degree of the pattern based on the centrality index and the dynamic sentiment-weighted identification score; the cultural identification degree is used to quantitatively evaluate the pattern. Networked identity assessment refers to the process of quantitative analysis through calculating node centrality indicators and dynamic sentiment-weighted identity scores. The centrality indicator can be implemented using graph theory algorithms. For example, degree centrality is based on the number of directly connected nodes, eigenvector centrality is solved based on the eigenvectors of the adjacency matrix, and betweenness centrality is determined by the proportion of betweennesses along the shortest path. It is mainly used to measure the functional attributes of patterns in the network structure. The dynamic sentiment-weighted identity score can be calculated based on a weighted summation model. The cognitive depth weight is allocated based on the behavioral transformation hierarchy, the sentiment intensity weight is normalized based on the results of text sentiment analysis, the time decay weight is modeled using a multinomial decay function, and the platform authority weight is determined based on an expert scoring system. It is mainly used to comprehensively reflect the dynamic impact of multiple dimensions on identity.

[0024] Step S400: Visualize the network model and evaluation results, and generate update instructions periodically; wherein, the evaluation results include cultural identity degree and trend analysis results of cultural identity degree; The construction of a visualization and dynamic monitoring platform refers to the process of graphically displaying network models and evaluation results and establishing a periodic data update process. It can be achieved by using interactive charts to present node distribution characteristics or by triggering data collection and model retraining through scheduled tasks. For example, it can use geographic heat maps to display the spread of patterns and set up a monthly data refresh mechanism to track the trend of changes in recognition. Its main purpose is to achieve intuitive interpretation of evaluation results and temporal evolution analysis.

[0025] In one example of the technical solution of this invention, the actual application process of the above content is defined. The collected data is divided into three categories: 1. Pattern information (equivalent to the pattern's "identity card"), including the pattern's image, name, formation period, and underlying meaning; 2. Audience cognition and emotional data, which essentially reflects how people perceive the pattern, whether they like it or not, and to what extent they understand it. The data sources include questionnaires, social media sentiment analysis (Weibo, etc.), and interviews. The questionnaires and interviews are all known text records, which can be directly read for the technical solution of this invention. 3. Dissemination and behavioral data, which actually means the frequency and scenarios in which the pattern appears in reality or online, including the use of the pattern on news, e-commerce and short video platforms, user interaction data, and offline observation data. By statistically analyzing these data, a network model is simultaneously determined. The network model contains nodes corresponding to the pattern. By analyzing the nodes corresponding to the pattern, each pattern can be evaluated. The evaluation is represented by two parameters: centrality index and cultural identity. Finally, the network model and the obtained centrality index and cultural identity are visualized so that viewers can intuitively understand the comprehensive information of the entire pattern.

[0026] Regarding step S100, the step of collecting pattern sample data, audience cognition and emotion data, and pattern dissemination and behavior data to construct a multi-source dataset includes: Establish a connection channel with the digital resource library to obtain the image, name, era, and symbolic information of the pattern, and construct the pattern data layer; the symbolic information is text. Build and push out survey questionnaires, receive questionnaire data from user feedback, generate public awareness, understanding, emotional attitudes and behavioral intentions regarding patterns, and build a cognitive and emotional data layer for the audience. By crawling network data from news, e-commerce platforms, and cultural applications in media data, we can obtain data on the frequency of pattern appearance, application scenarios, and public interaction, thus forming a communication and behavioral data layer. The statistical texture sample data layer, the audience cognition and emotion data layer, and the communication and behavior data layer are used to obtain a multi-source dataset.

[0027] In one example of the technical solution of this invention, images, names, dates, and symbolic information of Hunan patterns are obtained through digital resources and documents, forming a pattern data layer; the public's awareness, understanding, emotional attitudes, and behavioral intentions regarding the patterns are obtained through one or more methods such as questionnaires, social media sentiment analysis, in-depth interviews, and focus groups, forming an audience cognition and emotion data layer; and data on the frequency of appearance, application scenarios, and public interaction of the patterns are obtained by web crawling from news, e-commerce platforms, and cultural applications, combined with offline behavioral observations, forming a dissemination and behavior data layer.

[0028] In practical applications, the pattern sample data layer refers to a systematic collection of basic attribute information of patterns. This can be achieved through methods such as digital scanning of museum collections, image recognition of ancient books and documents, or retrieval from cultural heritage databases. The aim is to ensure the authority and historical continuity of pattern sample information and avoid the loss of historical dimensions caused by fragmented data. The audience cognition and emotion data layer can be understood as a continuous spectrum of data covering the public's subjective identification state. Specifically, it can be achieved by collecting structured cognitive data through online questionnaire platforms, analyzing the sentiment of social media texts using natural language processing technology, or recording behavioral intentions through focus group discussions. The aim is to overcome the limitations of single methods in terms of sample depth and breadth and dynamically capture the transformation process from surface cognition to deep emotion. The dissemination and behavior data layer specifically reflects the omnichannel behavior records of the actual application ecosystem of patterns. This can be achieved by using distributed web crawlers to crawl publicly available data from multiple platforms and supplementing it with offline data through non-intrusive behavior recorders at cultural event sites. The aim is to bridge the gap between purely online data and real interactive scenarios and ensure the spatiotemporal integrity of dissemination data and the authenticity of behavior.

[0029] Furthermore, the pattern data layer serves as the baseline node source for the network model, ensuring the uniqueness and historical accuracy of the pattern definition; the audience cognition and emotion data layer quantifies cognitive depth and emotional intensity into calculable relational weights through multimodal data fusion; and the communication and behavior data layer constructs a spatiotemporally continuous contextual network, providing a behavioral trajectory basis for dynamic monitoring. This is a hierarchical collection mechanism that enables the structured integration of multi-source heterogeneous data within a unified framework, supporting the accurate definition of cognitive, emotional, and co-occurrence relationships in the subsequent network model construction, thereby overcoming the one-sidedness of evaluation caused by a single data dimension.

[0030] Specifically, in a preferred embodiment of the technical solution of this invention, in the research on the "Hunan Embroidery Butterfly and Flower" pattern, firstly, high-definition images of the pattern, Qing Dynasty origin documents, and traditional symbolic explanations are obtained from the digital resource database of the Hunan Provincial Museum, forming a pattern data layer. Secondly, a hybrid survey method is adopted. On the one hand, a sentiment analysis API is deployed on the Weibo platform to capture public comments on related topics. On the other hand, an online questionnaire is distributed to 500 users to assess their cognitive understanding of the pattern, and 20 cultural enthusiasts are interviewed in depth to record their behavioral intentions, forming an audience cognition and sentiment data layer. At the same time, customized web crawlers are used to monitor the pattern application data of products on mainstream e-commerce platforms, and behavioral observation points are set up at the Changsha Intangible Cultural Heritage Exhibition to record the frequency of audience interaction, forming a dissemination and behavior data layer. After cleaning, these data are input into the network model, so that the relationship between pattern nodes, audience nodes, and context nodes can be supported by multi-dimensional data.

[0031] Regarding step S200, the step of constructing a network model containing pattern nodes, audience nodes, and context nodes based on a multi-source dataset, and simultaneously defining the cognitive relationships, emotional relationships, and co-occurrence relationships among the nodes, includes: Pattern nodes are constructed based on the pattern body data layer, audience nodes are constructed based on the audience cognition and emotion data layer, and context nodes are constructed based on the communication and behavior data layer. Cognitive and emotional relationships are determined based on audience cognition and emotion data layers, and co-occurrence relationships are determined based on communication and behavior data layers. The pattern nodes are used to characterize the identity features of the pattern, including but not limited to the pattern number; The audience node is used to represent the identity characteristics of the pattern audience, including but not limited to social media user IDs and user profile tags; The context nodes are used to represent the interactive scenarios of the pattern, including but not limited to the platforms, scenarios, and derivatives where the pattern appears; The cognitive relationship is established between the audience node and the pattern node, and its relationship weight is determined by the cognitive score and mention frequency of the pattern audience. The emotional relationship is established between the audience node and the pattern node, and its relationship attributes are determined by the emotional polarity and intensity. The co-occurrence relationship is established between pattern nodes, and its weight is determined by the frequency of the two patterns appearing together in the same context.

[0032] In one example of the technical solution of this invention, the application process of multi-source datasets is described. It essentially constructs a holistic data network to reflect the multi-source datasets. The pattern node refers to a specific, unique traditional or innovative Hunan pattern. A pattern node is a cultural symbol entity with a unique identifier, which can be implemented by combining the pattern name with a regional code or date identifier. For example, "Hunan embroidery-fish pattern-1890" can be used to identify pattern variations from a specific historical period. The purpose is to ensure that each pattern exists as an irreplaceable independent unit in the model, avoiding data confusion caused by pattern duplication or ambiguity, thereby accurately representing the cultural uniqueness of the pattern. The audience node refers to a specific social media user ID or aggregated user profile tags. A point can be understood as a quantifiable behavioral subject unit, which can be implemented using original social platform user IDs or profile tags aggregated based on behavioral characteristics (such as "Generation Z Guochao enthusiasts"). Its purpose is to adapt to the needs of audience data granularity, while retaining the traceability of individual behavior and supporting the aggregation and analysis of group characteristics, thereby enhancing the flexibility and representativeness of audience data in cognitive and emotional relationship modeling. The context node refers to the platform, scene, or derivative product of the pattern. The context node is specifically the concrete environmental carrier of fingerprint pattern dissemination, which can be realized as e-commerce platforms, social media platforms, or physical cultural scenes (such as museum exhibitions). Its purpose is to transform abstract context into an operable concrete entity, providing a clear basis for the dynamic calculation of co-occurrence and emotional relationships.

[0033] Specifically, the above measures ensure the independent representation of cultural symbols in the model by limiting pattern nodes to unique entities. Simultaneously, the granular adaptation design of audience nodes preserves the originality of micro-level behavioral data while supporting the aggregation of macro-level group analysis. The concretization of context nodes allows for precise quantification of the communication environment. These three elements collectively construct a clearly structured and entity-defined network model framework, enabling pattern, audience, and context elements to form computable relational units within the model. This resolves the network construction inaccuracies caused by ambiguous node representations and provides a reliable data foundation for subsequent centrality index calculations and dynamic sentiment-weighted identification scores.

[0034] Furthermore, the cognitive relationship is established between the audience node and the pattern node, and its relationship weight is based on the audience's cognitive score or mention frequency. The cognitive relationship refers to the structured representation of the audience's cognitive degree of the pattern. It can be achieved by statistically analyzing the accuracy of pattern name or meaning in a questionnaire survey, or by normalizing the frequency of keyword mentions on social media platforms. The purpose is to transform subjective cognitive breadth into objective behavioral data that can be repeatedly measured. The emotional relationship is established between the audience node and the pattern node, and its relationship attributes include emotional polarity and intensity. The emotional relationship can be understood as a quantitative dimension of the audience's emotional response. It can be achieved by using natural language processing technology to classify the emotional polarity (such as positive, negative, neutral) and intensity (such as weak, medium, strong) of the comment text, with the aim of distinguishing the directionality and depth of emotional identification. The co-occurrence relationship is established between pattern nodes, and its relationship weight is based on the frequency of the two patterns appearing together in the same context. The co-occurrence relationship is specifically the closeness of the association between fingerprint patterns in cultural practice. It can be achieved by statistically analyzing the co-occurrence frequency of pattern combinations in the description of cultural derivative products, or by recording and analyzing the simultaneous appearance of patterns in offline cultural exhibition scenarios, with the aim of capturing the implicit cultural synergy between patterns.

[0035] Specifically, the above technical solution quantifies the audience's cognitive coverage of patterns through cognitive relationships, and its weights are based on verifiable behavioral data (such as the proportion of questionnaires that accurately interpret the meaning or the number of mentions on social media), avoiding the arbitrariness of traditional qualitative methods. At the same time, emotional relationships expand simple emotional classification into a multi-dimensional psychological identification characterization through the dual attributes of emotional polarity and intensity, providing a basic input for dynamic emotional weighting. In addition, co-occurrence relationships construct the association between patterns based on the frequency of their co-occurrence in actual contexts, effectively identifying collaborative patterns in cultural networks. Combining these three relationships, a structured multi-dimensional relationship network is formed, enabling the interaction between patterns, audiences, and contextual elements to be systematically modeled, thereby supporting the accurate execution of subsequent centrality index calculation and dynamic identification assessment.

[0036] Through the above scheme, the present invention makes the definition of the relationship between nodes operable and quantifiable, and can be used as an attribute when constructing a network model.

[0037] As a preferred embodiment of the technical solution of the present invention, the centrality index of step S300 includes degree centrality, eigenvector centrality and betweenness centrality, wherein the degree centrality is used to measure the cognitive breadth of the pattern, the eigenvector centrality is used to measure the recognition depth of the pattern among the audience, and the betweenness centrality is used to identify patterns with a transitive role in the network.

[0038] Step S300 accomplishes two tasks: first, calculating the centrality index; and second, determining the dynamic emotional weighted identification score. Combining these two results in the cultural identification degree. The above content defines the centrality index, which includes: degree centrality, eigenvector centrality, and betweenness centrality. Degree centrality measures the cognitive breadth of a pattern. Degree centrality measures the number of direct connections between pattern nodes, which can be achieved by counting the number of direct edges between pattern nodes and audience or context nodes. Its purpose is to quantify the extent to which the pattern is widely known and recognized by the public. Eigenvector centrality... The eigenvector centrality is used to measure the depth of recognition of a pattern among its influential audience. Eigenvector centrality is an indicator that measures the ability of a pattern node to spread its influence in the network. It can be implemented using an eigenvector iteration calculation method based on the adjacency matrix. Its purpose is to assess the deep level of recognition of a pattern among key audience groups. The median centrality is used to identify patterns that play a cultural hub role in the network. Median centrality is an indicator that measures the role of a pattern node as a bridge for information or cultural flow. It can be implemented by calculating the frequency of the node's occurrence in all shortest paths. Its purpose is to identify core patterns that promote communication between different cultural groups.

[0039] Specifically, degree centrality objectively reflects the cognitive reach of a pattern by counting the number of direct connections between pattern nodes; eigenvector centrality considers not only the number of connections but also the influence weights of the connected objects, thus accurately capturing the depth of recognition of the pattern among high-value audiences; and betweenness centrality focuses on the bridging role of patterns in the network structure, revealing key patterns that connect different cultural contexts by analyzing the betweenness frequency of nodes in the shortest path. The application of these three indicators is synergistic, enabling cultural identity assessment to simultaneously cover the breadth, depth, and structural function levels, rather than relying on a traditional single indicator.

[0040] Furthermore, the step of calculating the dynamic emotional weighted identification score of each pattern node based on preset weights, and generating the cultural identification degree of the pattern based on the centrality index and the dynamic emotional weighted identification score includes: For any pattern node, obtain the total number of interactive behaviors related to the pattern within a preset time window in its propagation and behavior data layer; For any interactive behavior, read its corresponding audience cognition and emotion data layer, normalize each data, and obtain the normalized emotion value of each interactive behavior. The normalized sentiment values ​​of all interactive behaviors are statistically analyzed to obtain the dynamic sentiment-weighted identification score of the pattern nodes; Cultural identity is determined based on the centrality index of pattern nodes and the dynamic emotional weighted identity score.

[0041] Regarding the dynamic sentiment-weighted identification score, the technical solution of this invention provides a calculation model, called the performance prediction model, as follows: ; in, Pattern In time The normalized dynamic recognition score has a range of [0,1], and the larger the value, the higher the recognition. : within the time window Inside, with patterns Total number of related interactions; : No. Normalized sentiment value of each interactive behavior, It is obtained by substituting the original sentiment data (such as a 1-5 rating or a sentiment score of -1 to +1) into the maximum-minimum normalization; : No. Normalized cognitive depth weights for each interactive behavior The assignment rule is as follows: interactions that represent deep cognition or behavioral transformation (such as accurately interpreting the meaning or purchasing derivative products) are given a higher weight than superficial cognitive interactions (such as browsing or liking). : No. Normalized sentiment intensity weights for each interactive behavior; The assignment rule is as follows: interactions with strong emotional expression (such as those containing words with strong emotions) are given a higher weight than interactions with neutral emotions. : No. Each interactive behavior in time Normalized time decay weights, ; : No. The normalized authority weight of the platform where the interaction occurs. The assignment rule is as follows: platforms with high authority (such as national media and authoritative academic journals) are given higher weight than ordinary personal social media platforms.

[0042] In this embodiment, the normalized dynamic consensus score It is a comprehensive quantitative indicator of consensus based on fingerprint data at a specific point in time. It can be calculated using a multi-dimensional weighted sentiment score average, aiming to provide a standardized benchmark for assessing consensus; normalized sentiment score. This can be understood as mapping raw sentiment data to values ​​on a uniform scale. Specifically, this can be achieved by using the max-min normalization method to process sentiment rating data from different sources, aiming to eliminate evaluation bias caused by differences in data scale; cognitive depth weighting. This refers to the weighting coefficients set based on differences in cognitive levels reflected in interactive behaviors. These can be implemented using weighting rules based on behavior type classification, aiming to distinguish between surface cognition and deep identification; emotional intensity weighting. This refers to weighting coefficients set based on the intensity of emotional expression. These can be implemented using weighting methods based on the intensity of emotional vocabulary, aiming to capture the impact of emotional depth on cultural identity; time decay weighting. This refers to a dynamic weighting that considers the impact of time on recognition, which can be implemented using various time decay function models. Its purpose is to reflect the characteristics of cultural identity evolving over time; platform authority weighting. This refers to the weighting coefficient set based on the reliability of the data source platform. It can be implemented using weighting rules based on the platform's reputation rating, with the aim of increasing the weight of reliable data sources.

[0043] Specifically, the above scheme constructs a dynamically weighted recognition assessment model by multiplying normalized sentiment values ​​by multiplied with multi-dimensional weights and averaging the results. The model first acquires interactive behavior data related to the pattern and calculates a normalized sentiment value for each behavior. Then, it determines corresponding weight coefficients based on the cognitive depth, emotional intensity, timing, and platform authority of the behavior. Finally, it multiplies these weights by the sentiment values ​​and averages them to obtain the dynamic recognition score of the pattern at a specific point in time. This design allows the assessment results to simultaneously reflect multiple dimensions such as cognitive depth, emotional intensity, temporal evolution, and platform reliability, thereby achieving a comprehensive quantification of cultural recognition.

[0044] To facilitate understanding of the above, an example is provided below: When assessing the cultural recognition of the traditional Hunan embroidery pattern "Two Fish Playing with Lotus," the system collects relevant interactive behavior data from social media platforms. For user comments such as "This pattern has a wonderful meaning, I bought the same scarf," the system identifies it as a deep cognitive behavior (accurately interpreting the meaning and purchasing derivative products), assigning it a high cognitive depth weight. Simultaneously, it analyzes the intensity of emotional words; if positive words such as "wonderful" are included, a high emotional intensity weight is assigned. Considering that the behavior occurred recently and originated from an e-commerce platform, a moderate time decay weight and a high platform authority weight are assigned. Finally, these weights are multiplied by the normalized sentiment value and averaged to calculate the dynamic recognition score of the pattern at the current time.

[0045] As a preferred embodiment of the present invention, the normalized time decay weight The calculation is based on the exponential decay model. ,in An adjustable attenuation coefficient ( ), Used to control the sensitivity of identity evolution over time; For the current time, For behavior The time of occurrence.

[0046] Among them, the exponential decay model is a mathematical model that describes the exponential decline of cultural identity over time. It can be implemented using a natural exponential function, aiming to more accurately simulate the objective law of the natural decline of cultural identity over time, avoiding evaluation distortion caused by linear decay or fixed weights; decay coefficient. This can be understood as an adjustable parameter controlling the decay rate. It can be set to a real value greater than zero, and its purpose is to flexibly adjust the sensitivity of time decay according to the actual monitoring scenario, such as reducing it during cultural hot events. The value is used to prolong the lasting effect of high-impact behaviors, or to increase the value in long-term trend analysis. To quickly mitigate the interference of outdated data; current time and the time of occurrence of the behavior This refers to the specific timestamps of the system evaluation and the occurrence of historical interactions. These timestamps can be obtained through the system clock or timestamp records. The purpose is to accurately calculate time differences to quantify the time decay effect and ensure that the time impact of each interaction is accurately incorporated into the evaluation system.

[0047] Specifically, by defining the time decay weight as an exponential decay model, the impact of historical interactions on current cultural identity decays exponentially with time, thus more realistically reflecting the dynamic evolution of cultural identity. (Decayation coefficient) The introduction of this feature allows the system to adjust the decay rate according to actual needs, when When the value is small, the decay process is gradual, and the impact of historical behavior lasts longer, making it suitable for scenarios requiring long-term memory retention, such as cultural hot topics; when... When the value is large, the decay process is steep, and the system focuses more on recent behavior, making it suitable for rapidly changing communication environments. The above scheme actually improves flexibility, enabling the determination process of dynamic emotion-weighted identity score to adapt to the dynamic needs of different cultural communication environments. This allows the cultural identity assessment process to truly reflect the time evolution effect, which is equivalent to introducing a time parameter, improving the parameter's sensitivity to time, and thus more accurately capturing the temporal trend of cultural identity change.

[0048] Regarding step S400, the step of visualizing the network model and evaluation results, and periodically generating update instructions includes: In the network model, pattern nodes are clustered to construct pattern communities, and the network model and its clustering results are displayed based on a graph data structure. Obtain the cultural recognition degree with time label for each pattern node, fit the recognition degree curve, and insert the recognition degree curve as an attribute into the corresponding pattern node in the network model. The update frequency of the network model is determined based on the acceptance curve, and update instructions are generated based on the update frequency.

[0049] In one example of the technical solution of this invention, a network model display process is introduced based on an existing network model. A community discovery algorithm is used to perform cluster analysis on the network model to identify different pattern cultural identity clusters. The community discovery algorithm refers to a computational method used to identify tightly connected subgroups in a complex network. It can be implemented using the Louvain algorithm, spectral clustering algorithm, or Infomap algorithm. Its purpose is to automatically divide subgroups with strong internal correlations and weak external correlations based on the topological structure of connections between nodes. Cluster analysis can be understood as a technique for grouping patterns based on the similarity of relationships between nodes in the network model. Its purpose is to merge pattern sets with similar identity characteristics into independent clusters, thereby revealing the group structure of cultural identity. Displaying the network model while simultaneously displaying the clustering results allows viewers to better understand the group characteristics of the patterns.

[0050] Furthermore, the cultural recognition degree with time tags for each pattern node is obtained, and a recognition degree curve is fitted. At this point, the changes in the pattern can be reflected based on existing data. The recognition degree curve is used as an attribute and inserted into the corresponding pattern node in the network model. This allows viewers to better understand the overall data of the pattern. In addition, the update frequency of the network model is determined based on the recognition degree curve, and update instructions are generated based on the update frequency. Generally, the derivative curve of the recognition degree curve can be obtained, and the absolute value of the derivative curve can be taken. The sum of the absolute values ​​of the derivative curve over a period of time (the time difference between the current acquisition and the previous acquisition) is obtained periodically. The update frequency is determined based on the proportionality of the sum. This means that the more obvious the change, the higher the update frequency. Update instructions are generated based on the update frequency to ensure the real-time performance of the network model. In addition, the update instructions execute steps S100 to S300 to update the network model and evaluation indicators, realizing the time-series trend analysis of cultural recognition.

[0051] In one example of the technical solution of this invention, by inputting the topological structure of the cognitive, emotional, and co-occurrence relationships formed between pattern nodes, audience nodes, and context nodes in the network model into the community discovery algorithm, the algorithm automatically divides the community based on the tightness of the connections between nodes, thereby identifying pattern sets with similar identity characteristics. This process makes full use of the implicit patterns of network relationship data, avoids the limitations of relying solely on node attributes, and enables the analysis to capture the formation mechanism and boundaries of different cultural identity communities. Since the network model fully preserves the multi-dimensional relationships between pattern nodes, audience nodes, and context nodes, the community discovery algorithm can comprehensively consider cognitive breadth, emotional intensity, and contextual relevance, thereby integrating the originally isolated node centrality index and dynamic emotional weighted identity score into the community dimension, forming a three-dimensional analysis of the cultural identity structure.

[0052] As a preferred embodiment of the present invention, the generation process of the above-mentioned update instruction is actually the establishment of an automated data pipeline. An automated data pipeline refers to a data acquisition and transmission process that does not require manual intervention. It can be implemented using message queue middleware or a distributed stream processing framework. The purpose is to ensure the continuous and stable flow of multi-source data from the acquisition end to the processing end. The periodic execution of steps S100 to S300 can be understood as a time-cycle-based end-to-end evaluation process triggering mechanism. Specifically, it can be configured through a task scheduling system with a fixed time interval execution strategy. Its purpose is to maintain the synchronization between the evaluation model and the latest data. Updating the network model and evaluation indicators is specifically an iterative process of network structure and quantitative indicators, such as using incremental updates or full reconstruction. Its purpose is to eliminate evaluation bias caused by data obsolescence. The realization of the time-series trend analysis of cultural identity can be understood as the identification of evolutionary patterns based on historical data sequences. Specifically, it can be implemented through time series decomposition algorithms. Its purpose is to extract the long-term variation characteristics of identity.

[0053] Figure 2 A structural diagram of a network construction system for assessing the cultural identity of Hunan patterns is shown. In a preferred embodiment of the technical solution of the present invention, a network construction system for assessing the cultural identity of Hunan patterns is also provided. The system 10 includes: The dataset construction module 11 is used to collect pattern sample data, audience cognition and emotion data, and pattern dissemination and behavior data to construct a multi-source dataset. The network model construction module 12 is used to construct a network model containing pattern nodes, audience nodes and context nodes based on a multi-source dataset, and simultaneously define the cognitive relationship, emotional relationship and co-occurrence relationship between the nodes. The network model application module 13 is used to calculate the centrality index of each pattern node in the network model, calculate the dynamic emotional weighted identity score of each pattern node based on preset weights, and generate the cultural identity degree of the pattern based on the centrality index and the dynamic emotional weighted identity score; the cultural identity degree is used to quantitatively evaluate the pattern. The visualization module 14 is used to visualize the network model and evaluation results, and to generate update instructions periodically; wherein, the evaluation results include cultural identity and trend analysis results of cultural identity.

[0054] Furthermore, the dataset construction module 11 includes: The pattern layer establishment unit is used to establish a connection channel with the digital resource library, acquire the image, name, era, and symbolic information of the pattern, and construct the pattern body data layer; the symbolic information is text. The cognitive layer building unit is used to construct and push survey questionnaires, receive questionnaire data from user feedback, generate the public's awareness, understanding, emotional attitude and behavioral intention of the pattern, and construct the audience's cognitive and emotional data layer. The behavior layer building unit is used to crawl network data from news, e-commerce platforms and cultural applications in media data to obtain the frequency of pattern appearance, application scenarios and public interaction data, thus forming the communication and behavior data layer. The data statistics unit is used to statistically analyze the texture sample data layer, the audience cognition and emotion data layer, and the communication and behavior data layer to obtain a multi-source dataset.

[0055] Specifically, the network model construction module 12 includes: The node construction unit is used to construct pattern nodes based on the pattern body data layer, audience nodes based on the audience cognition and emotion data layer, and context nodes based on the communication and behavior data layer. The relationship determination unit is used to determine cognitive and emotional relationships based on the audience's cognitive and emotional data layer, and to determine co-occurrence relationships based on the communication and behavioral data layer. The pattern nodes are used to characterize the identity features of the pattern, including but not limited to the pattern number; The audience node is used to represent the identity characteristics of the pattern audience, including but not limited to social media user IDs and user profile tags; The context nodes are used to represent the interactive scenarios of the pattern, including but not limited to the platforms, scenarios, and derivatives where the pattern appears; The cognitive relationship is established between the audience node and the pattern node, and its relationship weight is determined by the cognitive score and mention frequency of the pattern audience. The emotional relationship is established between the audience node and the pattern node, and its relationship attributes are determined by the emotional polarity and intensity. The co-occurrence relationship is established between pattern nodes, and its weight is determined by the frequency of the two patterns appearing together in the same context.

[0056] Furthermore, the centrality index includes degree centrality, eigenvector centrality, and betweenness centrality. The degree centrality is used to measure the cognitive breadth of the pattern, the eigenvector centrality is used to measure the recognition depth of the pattern among the audience, and the betweenness centrality is used to identify patterns that play a transitive role in the network.

[0057] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for constructing a network to assess the cultural identity of Hunan patterns, characterized in that, The method includes: Collect pattern sample data, audience cognition and emotion data, and pattern dissemination and behavior data to construct a multi-source dataset; A network model containing pattern nodes, audience nodes, and context nodes is constructed based on a multi-source dataset, and the cognitive relationships, emotional relationships, and co-occurrence relationships between these nodes are defined simultaneously. The centrality index of each pattern node in the network model is calculated, and the dynamic emotional weighted identification score of each pattern node is calculated based on the preset weights. The cultural identification score of the pattern is generated based on the centrality index and the dynamic emotional weighted identification score. The cultural identification score is used to quantitatively evaluate the pattern. The network model and evaluation results are visualized and updated periodically; the evaluation results include cultural identity and trend analysis results of cultural identity.

2. The method for constructing a network for assessing the cultural identity of Hunan patterns according to claim 1, characterized in that, The steps for collecting pattern sample data, audience cognitive and emotional data, and pattern dissemination and behavioral data to construct a multi-source dataset include: Establish a connection channel with the digital resource library to obtain the image, name, era, and symbolic information of the pattern, and construct the pattern data layer; the symbolic information is text. Build and push out survey questionnaires, receive questionnaire data from user feedback, generate public awareness, understanding, emotional attitudes and behavioral intentions regarding patterns, and build a cognitive and emotional data layer for the audience. By crawling network data from news, e-commerce platforms, and cultural applications in media data, we can obtain data on the frequency of pattern appearance, application scenarios, and public interaction, thus forming a communication and behavioral data layer. The statistical texture sample data layer, the audience cognition and emotion data layer, and the communication and behavior data layer are used to obtain a multi-source dataset.

3. The method for constructing a network for assessing the cultural identity of Hunan patterns according to claim 1, characterized in that, The steps of constructing a network model containing pattern nodes, audience nodes, and context nodes based on a multi-source dataset, and simultaneously defining the cognitive relationships, emotional relationships, and co-occurrence relationships among the nodes, include: Pattern nodes are constructed based on the pattern body data layer, audience nodes are constructed based on the audience cognition and emotion data layer, and context nodes are constructed based on the communication and behavior data layer. Cognitive and emotional relationships are determined based on audience cognition and emotion data layers, and co-occurrence relationships are determined based on communication and behavior data layers. The pattern nodes are used to characterize the identity features of the pattern, including but not limited to the pattern number; The audience node is used to represent the identity characteristics of the pattern audience, including but not limited to social media user IDs and user profile tags; The context nodes are used to represent the interactive scenarios of the pattern, including but not limited to the platforms, scenarios, and derivatives where the pattern appears; The cognitive relationship is established between the audience node and the pattern node, and its relationship weight is determined by the cognitive score and mention frequency of the pattern audience. The emotional relationship is established between the audience node and the pattern node, and its relationship attributes are determined by the emotional polarity and intensity. The co-occurrence relationship is established between pattern nodes, and its weight is determined by the frequency of the two patterns appearing together in the same context.

4. The method for constructing a network for assessing the cultural identity of Hunan patterns according to claim 1, characterized in that, The centrality metrics include degree centrality, eigenvector centrality, and betweenness centrality. The degree centrality measures the cognitive breadth of a pattern, the eigenvector centrality measures the depth of recognition of a pattern among the audience, and the betweenness centrality is used to identify patterns that serve a transitive role in a network.

5. The method for constructing a network for assessing the cultural identity of Hunan patterns according to claim 4, characterized in that, The steps of calculating the dynamic emotional weighted identification score of each pattern node based on preset weights, and generating the cultural identification degree of the pattern based on the centrality index and the dynamic emotional weighted identification score include: For any pattern node, obtain the total number of interactive behaviors related to the pattern within a preset time window in its propagation and behavior data layer; For any interactive behavior, read its corresponding audience cognition and emotion data layer, normalize each data, and obtain the normalized emotion value of each interactive behavior. The normalized sentiment values ​​of all interactive behaviors are statistically analyzed to obtain the dynamic sentiment-weighted identification score of the pattern nodes; Cultural identity is determined based on the centrality index of pattern nodes and the dynamic emotional weighted identity score.

6. The method for constructing a network for assessing the cultural identity of Hunan patterns according to claim 5, characterized in that, The steps of visualizing the network model and evaluation results, and periodically generating update instructions include: In the network model, pattern nodes are clustered to construct pattern communities, and the network model and its clustering results are displayed based on a graph data structure. Obtain the cultural recognition degree with time label for each pattern node, fit the recognition degree curve, and insert the recognition degree curve as an attribute into the corresponding pattern node in the network model. The update frequency of the network model is determined based on the acceptance curve, and update instructions are generated based on the update frequency.

7. A network construction system for assessing the cultural identity of Hunan patterns, characterized in that, The system includes: The dataset construction module is used to collect pattern sample data, audience cognition and emotion data, and pattern dissemination and behavior data to build a multi-source dataset; The network model construction module is used to build a network model containing pattern nodes, audience nodes, and context nodes based on a multi-source dataset, and simultaneously define the cognitive relationships, emotional relationships, and co-occurrence relationships between the nodes. The network model application module is used to calculate the centrality index of each pattern node in the network model, calculate the dynamic sentiment-weighted identification score of each pattern node based on preset weights, and generate the cultural identification degree of the pattern based on the centrality index and the dynamic sentiment-weighted identification score; the cultural identification degree is used to quantitatively evaluate the pattern. The visualization module is used to visualize the network model and evaluation results, and to generate update instructions periodically; the evaluation results include cultural identity and trend analysis results of cultural identity.

8. The Hunan pattern culture identity assessment network construction system according to claim 7, characterized in that, The dataset construction module includes: The pattern layer establishment unit is used to establish a connection channel with the digital resource library, acquire the image, name, era, and symbolic information of the pattern, and construct the pattern body data layer; the symbolic information is text. The cognitive layer building unit is used to construct and push survey questionnaires, receive questionnaire data from user feedback, generate the public's awareness, understanding, emotional attitude and behavioral intention of the pattern, and construct the audience's cognitive and emotional data layer. The behavior layer building unit is used to crawl network data from news, e-commerce platforms and cultural applications in media data to obtain the frequency of pattern appearance, application scenarios and public interaction data, thus forming the communication and behavior data layer. The data statistics unit is used to statistically analyze the texture sample data layer, the audience cognition and emotion data layer, and the communication and behavior data layer to obtain a multi-source dataset.

9. The Hunan pattern culture identity assessment network construction system according to claim 7, characterized in that, The network model construction module includes: The node construction unit is used to construct pattern nodes based on the pattern body data layer, audience nodes based on the audience cognition and emotion data layer, and context nodes based on the communication and behavior data layer. The relationship determination unit is used to determine cognitive and emotional relationships based on the audience's cognitive and emotional data layer, and to determine co-occurrence relationships based on the communication and behavioral data layer. The pattern nodes are used to characterize the identity features of the pattern, including but not limited to the pattern number; The audience node is used to represent the identity characteristics of the pattern audience, including but not limited to social media user IDs and user profile tags; The context nodes are used to represent the interactive scenarios of the pattern, including but not limited to the platforms, scenarios, and derivatives where the pattern appears; The cognitive relationship is established between the audience node and the pattern node, and its relationship weight is determined by the cognitive score and mention frequency of the pattern audience. The emotional relationship is established between the audience node and the pattern node, and its relationship attributes are determined by the emotional polarity and intensity. The co-occurrence relationship is established between pattern nodes, and its weight is determined by the frequency of the two patterns appearing together in the same context.

10. The Hunan pattern culture identity assessment network construction system according to claim 7, characterized in that, The centrality metrics include degree centrality, eigenvector centrality, and betweenness centrality. The degree centrality measures the cognitive breadth of a pattern, the eigenvector centrality measures the depth of recognition of a pattern among the audience, and the betweenness centrality is used to identify patterns that serve a transitive role in a network.

Images