A social group member dynamic management method and system

Abstract

The application discloses a kind of social group member dynamic management method and system, it is related to social network analysis field, including, the online interaction data and offline co-occurrence data of social group member are acquired, and are washed, the online interaction data and offline co-occurrence data after washing are converted into structured data set;Structured data set is imported into graph database to obtain initial weight, and dynamic attenuation weight is calculated, and basic relationship network is constructed;The edge weight matrix in basic relationship network is converted into quantum bit entanglement state, quantum bit entanglement state is measured to form quantum probability amplitude, quantum entanglement determination threshold is set, and quantum relationship thermodynamic diagram is generated by Grover algorithm.The application realizes the accurate management and risk prevention and control of social group member relationship by the technical scheme that quantum computation is combined with dynamic entropy analysis, the edge weight matrix of basic relationship network is converted into quantum bit entanglement state, and quantum relationship thermodynamic diagram is generated using quantum measurement and Grover algorithm.

Description

Technical Field

[0001] This invention relates to the field of social network analysis, and in particular to a method and system for dynamic management of members of social groups. Background Technology

[0002] In recent years, the technology for dynamic management of social group members has gradually evolved from traditional rule-based and statistical methods to data-driven and intelligent approaches. Among existing technologies, social network analysis methods based on graph databases and machine learning have become mainstream. This paper proposes a dynamic relationship network construction method based on the Neo4j graph database and time-decaying weights. This method constructs a time-sensitive weighted relationship network by collecting online interaction data and offline co-occurrence data of members, and uses a community discovery algorithm to identify key subgroups. It quantifies the uncertainty of member behavior through dynamic entropy values ​​and combines this with a propagation model to predict risk diffusion paths. This approach has been validated in scenarios such as e-commerce communities and open-source communities, and can achieve basic risk monitoring and intervention functions.

[0003] While existing technologies enable dynamic management of social group members, there is still room for improvement in the accuracy and real-time performance of implicit relationship mining. Specifically, traditional methods rely on explicit interaction data (such as chat frequency and co-occurrence duration) to construct relationship networks, making it difficult to capture potential connections between members through indirect interactions. Community segmentation based on the Louvain algorithm can only identify explicit groups, while Grover's algorithm in quantum computing has been proven to have a secondary speedup advantage in unstructured searches. However, existing social network management technologies have not yet effectively integrated quantum computing with classical graph analysis. Traditional propagation force models use static threshold intervention, which cannot adapt to the dynamic changes in social networks, resulting in a high false blocking rate. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method for dynamic management of social group members, which solves the problems of insufficient accuracy in identifying implicit relationships and lag in the response of static thresholds in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for dynamic management of members of a social group, which includes acquiring online interaction data and offline co-occurrence data of members of the social group, cleaning them, and converting the cleaned online interaction data and offline co-occurrence data into a structured dataset.

[0008] The structured dataset is imported into a graph database to obtain initial weights, and the dynamically decaying weights are calculated to construct a basic relational network.

[0009] The edge weight matrix in the basic relation network is transformed into entangled states of qubits. The entangled states of qubits are measured to form quantum probability amplitudes. A quantum entanglement determination threshold is set, and a quantum relation heatmap is generated using the Grover algorithm.

[0010] A list of strongly connected member pairs is marked in the quantum relationship heatmap, the dynamic entropy value of the social group members is calculated, a risk monitoring dataset is formed, the propagation force value is calculated based on the quantum relationship heatmap and the risk monitoring dataset, the herd immunity threshold is obtained, and a graded blocking strategy is implemented.

[0011] As a preferred embodiment of the dynamic management method for social group members described in this invention, the method includes: acquiring online interaction data and offline co-occurrence data of social group members, cleaning them, and converting the cleaned online interaction data and offline co-occurrence data into a structured dataset, comprising the following steps:

[0012] Collect members' group chat messages, private chat records, and file collaboration logs, and extract the sender ID, receiver ID, interaction time, and content type fields to obtain online interaction data;

[0013] Physical contact events between members are detected in group activity venues, and the IDs of the parties involved, start time, and duration are recorded to obtain offline co-occurrence data.

[0014] The online interaction data is deduplicated and invalid sessions are filtered out, and the cleaned online interaction data is output.

[0015] Signal strength filtering and short-time contact removal are performed on the offline co-occurrence data to obtain cleaned offline co-occurrence data;

[0016] The cleaned online interaction data and offline co-occurrence data are mapped and merged according to a unified standard to obtain a structured dataset.

[0017] As a preferred embodiment of the dynamic management method for social group members described in this invention, the method includes the following steps: importing a structured dataset into a graph database to obtain initial weights, calculating dynamically decaying weights, and constructing a basic relationship network.

[0018] Import the cleaned structured dataset into the Neo4j graph database, use Cypher statements to create member nodes and interaction edges, and output the initial weights;

[0019] Calculate the dynamic decay weights based on the initial weights, update the edge weights, and output the time-sensitive weighted relationship graph;

[0020] A basic relational network is constructed by pruning low-weight edges and removing invalid connections in dynamically decaying weights.

[0021] As a preferred embodiment of the dynamic management method for social group members described in this invention, the method includes the following steps: converting the edge weight matrix in the basic relationship network into entangled quantum bit states, and measuring the entangled quantum bit states to form quantum probability amplitudes.

[0022] The edge weight matrix is ​​derived from the basic relational network, normalized, and then converted into quantum rotation gate angles using the Qiskit library. Finally, CNOT and RY gates are used to convert the edges into entangled states of qubits.

[0023] The quantum probability amplitude is obtained by measuring the entangled state of a qubit on a quantum processor.

[0024] As a preferred embodiment of the dynamic management method for social group members described in this invention, the following steps are included: setting a quantum entanglement determination threshold and generating a quantum relationship heatmap using the Grover algorithm.

[0025] The threshold for determining quantum entanglement is set based on the quantum probability amplitude;

[0026] Based on the entanglement determination threshold, a label reflecting the strong connection conditions of social networks is constructed. The quantum probability amplitude is iteratively amplified using the Grover algorithm to generate a quantum relationship heatmap.

[0027] As a preferred embodiment of the dynamic management method for social group members described in this invention, the method includes the following steps: marking a list of strongly connected member pairs in a quantum relationship heatmap, calculating the dynamic entropy values ​​of social group members, and forming a risk monitoring dataset.

[0028] The quantum relation heatmap is analyzed to filter quantum probability amplitudes and output a list of strongly connected member pairs.

[0029] Calculate the dynamic entropy value of social group members based on a list of strongly connected member pairs;

[0030] Combine the dynamic entropy value with the list of strongly connected members to calculate the risk score and output a member-risk score table;

[0031] Based on the member-risk scoring table and dynamic entropy values, the risk monitoring dataset is obtained by real-time classification according to the hierarchical rules.

[0032] As a preferred embodiment of the dynamic management method for social group members described in this invention, the method includes the following steps: calculating the transmissibility value based on the quantum relationship heatmap and risk monitoring dataset to obtain the herd immunity threshold, and implementing a tiered blocking strategy.

[0033] The strong connections with probability amplitudes are extracted from the quantum relation heatmap and a member-neighbor mapping table is output.

[0034] Based on the risk monitoring dataset, the propagation force value is calculated by combining the time series of dynamic entropy values ​​with a list of strongly connected member pairs.

[0035] Based on the transmissibility value, the herd immunity threshold is obtained through an incremental learning method;

[0036] Based on the herd immunity threshold, a tiered blocking strategy is implemented.

[0037] Secondly, the present invention provides a dynamic management system for members of a social group, including a cleaning module, which acquires online interaction data and offline co-occurrence data of members of the social group, cleans them, and converts the cleaned online interaction data and offline co-occurrence data into a structured dataset.

[0038] The basic relation network module imports the structured dataset into the graph database to obtain initial weights, calculates the dynamically decaying weights, and constructs the basic relation network.

[0039] The quantum relationship heatmap module transforms the edge weight matrix in the basic relationship network into entangled states of qubits, measures the entangled states of qubits to form quantum probability amplitudes, sets a quantum entanglement determination threshold, and generates a quantum relationship heatmap using the Grover algorithm.

[0040] The tiered blocking strategy module marks a list of strongly connected member pairs in the quantum relationship heatmap, calculates the dynamic entropy value of social group members, forms a risk monitoring dataset, calculates the propagation force value based on the quantum relationship heatmap and the risk monitoring dataset, obtains the herd immunity threshold, and executes the tiered blocking strategy.

[0041] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the dynamic management method for members of social groups as described in the first aspect of the present invention.

[0042] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the dynamic management method for members of a social group as described in the first aspect of the present invention.

[0043] The beneficial effects of this invention are as follows: By combining quantum computing with dynamic entropy analysis, it achieves precise management and risk control of relationships among members of social groups. It transforms the edge weight matrix of the basic relationship network into entangled states of qubits, and uses quantum measurement and Grover's algorithm to generate a quantum relationship heatmap. This breakthrough solves the technical problem that traditional methods struggle to identify implicit strong connections. Based on the quantum relationship heatmap, it calculates dynamic entropy and propagation force values, and constructs a scientific risk assessment system by quantifying the characteristics of member behavior changes and propagation potential. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 A flowchart for the dynamic management method of social group members;

[0046] Figure 2 A schematic diagram of a dynamic management system for members of a social organization;

[0047] Figure 3 Flowchart for a structured dataset;

[0048] Figure 4 A schematic diagram for extracting quantum probability amplitude. Detailed Implementation

[0049] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0050] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0051] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0052] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a method for dynamic management of members of a social group, comprising the following steps:

[0053] S1. Obtain online interaction data and offline co-occurrence data of social group members, clean them, and convert the cleaned online interaction data and offline co-occurrence data into a structured dataset.

[0054] S1.1 Collect members' group chat messages, private chat records, and file collaboration logs, and extract the sender ID, receiver ID, interaction time, and content type fields to obtain online interaction data.

[0055] Furthermore, we collected group chat records, private chat records, and file collaboration logs from members of social groups. From the raw data, we extracted four key fields: sender ID, receiver ID, interaction timestamp, and content type. The sender ID and receiver ID are used to identify the identities of the two parties involved in the interaction. The interaction timestamp records the time when the action occurred. The content type distinguishes between interaction forms such as text, images, or files. The extraction process uses regular expression matching and log parsing technology to ensure the completeness and accuracy of field extraction, resulting in the obtained online interaction data.

[0056] S1.2 Detect physical contact events between members in group activity venues, record the IDs of the parties involved, the start time, and the duration of the contact to obtain offline co-occurrence data.

[0057] Furthermore, RFID readers are deployed in group activity venues to detect physical contact events between members. When the signal strength of the terminal devices of two members reaches a preset threshold, it is determined to be a valid contact. The system records three core fields: the unique ID of both members, the contact start time stamp, and the duration. The IDs of both parties are processed using an encrypted hash algorithm to ensure privacy and security. The start time stamp is accurate to the millisecond level. The obtained offline co-occurrence data and online interaction data maintain a corresponding field structure, including structured information such as the IDs of both parties, the contact time, and the duration.

[0058] S1.3 Deduplicatize and filter invalid sessions from the online interaction data, and output the cleaned online interaction data.

[0059] Furthermore, the collected online interaction data undergoes deduplication. Completely duplicate records are identified and deleted based on a combination of sender ID, receiver ID, and interaction timestamp. Regular expression matching and keyword filtering techniques are used to identify invalid sessions, including non-member-to-member valid interactions such as advertising information and notifications. The deduplicated and filtered online interaction data retains the original data structure, including sender ID, receiver ID, interaction time, and content type fields, but ensures that each record represents a genuine and valid interaction between members. The cleaning process uses a hash algorithm to accelerate duplicate record detection and an invalid content feature library to improve filtering accuracy, outputting the cleaned online interaction data.

[0060] S1.4. Perform signal strength filtering and short-term contact removal on the offline co-occurrence data to obtain cleaned offline co-occurrence data.

[0061] Furthermore, signal strength filtering is performed on the collected offline co-occurrence data. Invalid contact records with insufficient signal strength are removed based on RFID signal strength thresholds, and short-term contact events are removed based on duration thresholds. Contact records with a duration that meets the effective interaction standard are retained. Signal strength filtering uses the received signal strength indication value as the judgment basis, and duration filtering is based on the difference between the contact start time and the end time. The cleaned offline co-occurrence data retains the IDs of the contacting parties, the start time, and the duration field, but ensures that each record is a real and effective physical contact behavior between members. The filtering process uses a sliding window algorithm to optimize processing efficiency, resulting in cleaned offline co-occurrence data.

[0062] S1.5. The cleaned online interaction data and offline co-occurrence data are mapped and merged according to a unified standard to obtain a structured dataset.

[0063] Furthermore, the cleaned online interaction data and offline co-occurrence data are standardized. The sender ID and contact ID are uniformly mapped to the member ID field, and the interaction time and contact start time are uniformly mapped to the timestamp field. The content type field of the online interaction data and the duration field of the offline co-occurrence data are retained separately as feature identifiers to distinguish interaction types. The field mapping process adopts a unified encoding rule and time zone conversion to ensure that data from different sources are comparable in terms of time and identity. The merged structured dataset contains five core fields: member ID, interaction object ID, timestamp, interaction type, and duration, which fully records the interaction behavior of social group members across all online and offline channels. The data merging adopts a timestamp-based merge sorting algorithm to maintain the temporal consistency of interaction events, resulting in the final structured dataset.

[0064] S2. Import the structured dataset into the graph database to obtain the initial weights, calculate the dynamic decay weights, and construct the basic relation network.

[0065] S2.1. Import the cleaned structured dataset into the Neo4j graph database, use Cypher statements to create member nodes and interaction edges, and output the initial weights.

[0066] Furthermore, the cleaned structured dataset is then imported into the Neo4j graph database in batches.

[0067] The system loads data and uses the CREATE statement in the Cypher query language to create nodes representing members of a social group. Each node contains a member ID attribute. Based on the interaction records in the structured dataset, the MERGE command in the Cypher statement is used to create directed relationship edges between member nodes. The relationship type is distinguished according to the content type of online interaction data and the duration of offline co-occurrence data. The initial weight calculation comprehensively considers three dimensions: interaction frequency, duration, and interaction type. Text, image, and file types of online interaction data are assigned different weight coefficients. The duration of offline co-occurrence data is converted into weight values ​​according to a linear relationship, generating the initial weights.

[0068] S2.2 Calculate dynamically decaying weights based on initial weights, update edge weights, and output time-sensitive weighted relationships.

[0069] System diagram.

[0070] Specifically, the expression is,

[0071] ;

[0072] in, Current time t The dynamic decay weight is below. As the initial weights, The attenuation coefficient is... The time difference between the current time and the last interaction. t This is the current time.

[0073] Furthermore, when calculating the dynamic decay weight based on the initial weight, a time decay function is used to dynamically adjust the initial weight. The decay coefficient is set differently according to the interaction type; online interaction data (text, images, and files) corresponds to different decay rates. The duration of offline co-occurrence data is converted into initial weights, and decay parameters are set according to physical contact characteristics. The dynamic decay weight calculation process is implemented in the Neo4j graph database using mathematical operations via Cypher statements. The time-sensitive weighted relationship graph reflects the latest strength of interaction relationships among members of a social group through dynamic decay weights, preserving the complete topological structure of member nodes and relationship edges. The weight update process employs a batch processing optimization strategy, outputting a time-sensitive weighted relationship graph.

[0074] S2.3. By pruning low-weight edges to remove invalid connections in the dynamically decaying weights, a basic relational network is constructed.

[0075] Furthermore, pruning is performed on the time-weighted relation graph. Low-weight edges are filtered out based on a threshold set by dynamically decaying weights. Interaction edges whose dynamic decay weights are lower than the threshold are identified as invalid connections and removed from the network. The pruning threshold is dynamically adjusted according to the size of the social group and the interaction density to ensure that the core relation structure is preserved while eliminating noisy connections. The pruning process is implemented in the Neo4j graph database through the WHERE condition of the Cypher statement, which filters the weight attribute of relation edges. The basic relation network retains the complete topology of member nodes and valid relation edges.

[0076] It should be noted that the dynamic decay weights achieve the required strong connections, and the degree distribution of the basic relation network nodes after pruning is more consistent with the characteristics of real social networks, eliminating the interference of sparse connections on subsequent quantum computing; the construction process and the dynamic decay weight calculation stage form a continuous processing chain, ensuring the simplicity and effectiveness of the network structure and generating the basic relation network.

[0077] S3. Transform the edge weight matrix in the basic relational network into entangled states of qubits, and measure the entangled states of qubits to form quantum probability amplitudes.

[0078] S3.1. Derive the edge weight matrix from the basic relational network, normalize the edge weight matrix, convert the normalized edge weight matrix into quantum rotation gate angles using the Qiskit library, and convert it into entangled states of qubits using CNOT gates and RY gates.

[0079] Furthermore, an edge weight matrix is ​​extracted from the basic relational network. The matrix element values ​​represent the dynamic decay weights between member nodes. The edge weight matrix is ​​subjected to maximum and minimum value normalization to linearly map the weight values ​​to a fixed interval. The normalized edge weight matrix values ​​are converted into rotation angle parameters of the quantum rotation gate RY gate using the Qiskit quantum computing library. The CNOT gate is used to create entanglement relationships between qubits in the quantum circuit. Combined with the RY gate rotation operation, the edge weight information is encoded into the quantum state. The conversion process maintains the topology of the basic relational network. Each matrix element corresponds to the entanglement strength of a specific pair of qubits, generating entangled qubit states.

[0080] S3.2 Measure the entangled state of the qubits on the quantum processor to obtain the quantum probability amplitude.

[0081] Furthermore, the prepared entangled states of qubits are loaded into a quantum processor to perform measurement operations. Pauli Z-basis measurements are run on each qubit, and the measurement results are used to reconstruct the quantum density of states matrix using quantum state tomography. Quantum probability amplitudes are extracted, and the probability amplitude calculation is based on the statistical results of multiple repeated executions of the quantum circuit. The number of measurements ensures convergence of the results. The obtained quantum probability amplitudes reflect the strength of the quantized correlation between member nodes in the basic relational network and maintain a topological correspondence with the original edge weight matrix. Quantum error correction coding is used in the measurement process to reduce the influence of noise and ensure the accuracy of probability amplitude extraction.

[0082] S4. Set the quantum entanglement determination threshold and generate a quantum relationship heatmap using the Grover algorithm.

[0083] S4.1. Set the quantum entanglement determination threshold based on the quantum probability amplitude.

[0084] Furthermore, based on the distribution characteristics of quantum probability amplitudes obtained from quantum processor measurements, the correlation strength distribution law of entangled states of qubits is analyzed. The quantum entanglement determination threshold is set according to the statistical characteristics of quantum probability amplitudes, taking into account the distribution density and dispersion of probability amplitudes. The quantum entanglement determination threshold setting process adopts the quantile statistical method, selecting specific quantiles as the boundary values ​​to distinguish between strong and weak entanglement. The quantum entanglement determination threshold maintains a corresponding relationship with the dynamic decay characteristics of edge weights in the basic relational network, reflecting the actual interaction strength among members of the social group. After the threshold is determined, it is used to screen bit pairs with significant quantum correlations.

[0085] S4.2 Based on the entanglement determination threshold, construct a label reflecting the strong connection conditions of the social network, and use the Grover algorithm to iteratively amplify the quantum probability amplitude to generate a quantum relationship heatmap.

[0086] Furthermore, the Grover algorithm is used to perform iterative amplification operations on the quantum processor to amplify the quantum probability amplitude of the marked strongly connected qubit pairs in multiple rounds. During the iteration process, the diffusion operator is used to enhance the amplitude of the target state while suppressing the amplitude of the non-target state. After the optimal number of iterations, the amplified quantum state is measured to obtain the enhanced quantum probability amplitude distribution. The amplified quantum probability amplitude is then mapped back to the node connection relationship of the basic relation network to generate a quantum relation heatmap.

[0087] It should be noted that the node positions in the quantum relationship heatmap correspond to the member nodes in the basic relationship network, and the connection strength is represented by the amplified quantum probability amplitude. The Grover algorithm processing and the quantum entanglement determination threshold setting stage form a coherent technical chain, ensuring an effective conversion from quantum information processing to classical network visualization. The generated quantum relationship heatmap intuitively displays the quantum correlation strength among members of a social group.

[0088] S5. Mark the list of strongly connected member pairs in the quantum relationship heatmap, calculate the dynamic entropy value of the social group members, and form a risk monitoring dataset.

[0089] S5.1 Analyze the quantum relationship heatmap to filter the quantum probability amplitude and output a list of strongly connected member pairs.

[0090] Furthermore, the distribution of quantum probability amplitudes of each node connection in the quantum relationship heatmap is analyzed, and connections exceeding the strength threshold are extracted according to preset screening conditions. The selected strong connections are mapped back to member node pairs in the basic relationship network, and the member ID and corresponding quantum probability amplitude are recorded. The generation process of the strong connection member pair list adopts a graph traversal algorithm to ensure that no connection that meets the conditions is missed. The output strong connection member pair list contains member ID pairs and normalized quantum probability amplitudes, reflecting the closest interaction relationships in the social group.

[0091] S5.2 Calculate the dynamic entropy value of social group members based on the list of strongly connected member pairs.

[0092] Specifically, the expression is,

[0093] ;

[0094] in, For members i In time t The dynamic entropy value, For the first k Weight coefficients of class behavior, The time decay coefficient, For the time when the behavior occurs, k Member row index.

[0095] S5.3 Combine the dynamic entropy value and the strongly connected member list to calculate the risk score and output the member-risk score table.

[0096] Specifically, the expression is,

[0097] ;

[0098] in, For members i Risk score, For members i and neighbors j The quantum probability amplitude between For members i The degree of the node. For the neighbors j The dynamic entropy value.

[0099] Furthermore, based on the quantum probability amplitude and member node degree in the strongly connected member pair list, combined with the time series characteristics of the dynamic entropy value, a comprehensive risk score is calculated for each member of the social group. The risk score calculation process uses a weighted summation formula, integrating the quantum probability amplitude, node degree, and dynamic entropy value according to preset weight coefficients. The calculation process traverses all member nodes in the basic relationship network to ensure that each member receives a corresponding risk score. The generated member-risk score table contains two fields: member ID and risk score, sorted in descending order of score value. The risk score calculation maintains data correlation with the dynamic entropy value and the generation stage of the strongly connected member pair list, ensuring the comprehensiveness and objectivity of the score results. The member-risk score table serves as a core component of the risk monitoring dataset.

[0100] S5.4. Based on the member-risk scoring table and dynamic entropy value, classify in real time according to the hierarchical rules to obtain the risk monitoring dataset.

[0101] Furthermore, based on the time-varying characteristics of risk scores and dynamic entropy values ​​in the member-risk scoring table, members of social groups are divided into different risk levels. The classification rules comprehensively consider the absolute value of the risk score and the trend changes of the dynamic entropy value, and highlight members with persistently high risk and rapidly increasing risk. The risk level classification process adopts real-time streaming processing to ensure that the newly generated member-risk scoring table and dynamic entropy value can update the classification results in a timely manner. The generated risk monitoring dataset includes member ID, risk score, dynamic entropy value and risk level.

[0102] It should be noted that data is stored in groups according to risk level; the classification process is synchronized with the generation of the member-risk scoring table and the calculation of dynamic entropy value to ensure the timeliness and accuracy of the risk monitoring dataset. The risk monitoring dataset fully records the history of risk status changes of social group members, providing a classification basis for subsequent calculation of transmissibility value and determination of herd immunity threshold. The entire classification process retains the original data characteristics of the member-risk scoring table and dynamic entropy value, and only adds risk level labels to achieve structured organization of data.

[0103] S6. Calculate the propagation force value based on the quantum relationship heatmap and risk monitoring dataset to obtain the herd immunity threshold, and implement a tiered blocking strategy.

[0104] S6.1 Extract strong connections of probability amplitude from the quantum relation heatmap and output a member-neighbor mapping table.

[0105] Furthermore, the distribution of quantum probability amplitudes of each node connection in the quantum relation heatmap is analyzed to identify strong connections exceeding a preset threshold. These strong connections are then mapped back to member node pairs in the basic relation network, recording the ID of each member and its neighboring members, as well as their corresponding quantum probability amplitudes. The member-neighbor mapping table generation process uses a breadth-first search algorithm to traverse all node connections in the quantum relation heatmap, ensuring the complete extraction of all strong connections that meet the conditions. The output member-neighbor mapping table includes the member ID, a list of neighboring member IDs, and their corresponding quantum probability amplitudes.

[0106] S6.2 Based on the risk monitoring dataset, read the time series of dynamic entropy values ​​and combine them with the list of strongly connected member pairs to calculate the propagation force value.

[0107] Specifically, the expression is,

[0108] ,

[0109] in, For members i The propagation power value, The average dynamic entropy value of all members of the current group. N The total number of samples.

[0110] S6.3. Based on the propagation power value, the herd immunity threshold is obtained through incremental learning.

[0111] Specifically, the expression is,

[0112] ;

[0113] in, This serves as the herd immunity threshold for the next time step. The current herd immunity threshold. This represents the average actual propagation power. For the purpose of dissemination, This is the learning rate.

[0114] S6.4. Implement a tiered blocking strategy based on the herd immunity threshold.

[0115] Furthermore, based on the calculated herd immunity threshold, members of social groups are divided into different intervention levels. High-risk members are subject to a complete blocking strategy, restricting all their online and offline interactions. Medium-risk members are subject to partial blocking, allowing necessary contact but restricting high-frequency interactions. Low-risk members maintain normal interactions but are subject to enhanced monitoring. The tiered blocking strategy is implemented by adjusting the connection weights in the basic relationship network, with the connection weights of high-risk members set to zero and those of medium-risk members halved. The blocking measures strictly correspond to the risk levels in the risk monitoring dataset, ensuring that the intervention intensity matches the risk level. The connection status in the member-neighbor mapping table is updated in real time during the strategy execution process, and the blocking effect is fed back to the propagation force calculation stage to form a closed loop. The implementation of the tiered blocking strategy is based on the comprehensive analysis results of the quantum relationship heatmap and the risk monitoring dataset, maximizing the maintenance of normal group operation while controlling the spread of risk. The blocking intensity adapts to the dynamic adjustment of the herd immunity threshold.

[0116] This embodiment also provides a dynamic management system for social group members, including: a cleaning module, which acquires online interaction data and offline co-occurrence data of social group members, cleans them, and converts the cleaned online interaction data and offline co-occurrence data into a structured dataset;

[0117] The basic relation network module imports the structured dataset into the graph database to obtain initial weights, calculates the dynamically decaying weights, and constructs the basic relation network.

[0118] The quantum relationship heatmap module transforms the edge weight matrix in the basic relationship network into entangled states of qubits, measures the entangled states of qubits to form quantum probability amplitudes, sets a quantum entanglement determination threshold, and generates a quantum relationship heatmap using the Grover algorithm.

[0119] The tiered blocking strategy module marks a list of strongly connected member pairs in the quantum relationship heatmap, calculates the dynamic entropy value of social group members, forms a risk monitoring dataset, calculates the propagation force value based on the quantum relationship heatmap and the risk monitoring dataset, obtains the herd immunity threshold, and executes the tiered blocking strategy.

[0120] This embodiment also provides a computer device applicable to the dynamic management method for members of social groups, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the dynamic management method for members of social groups as proposed in the above embodiment.

[0121] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0122] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the method for dynamic management of social group members as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0123] In summary, this invention achieves precise management and risk control of relationships among members of social groups through a technical solution combining quantum computing and dynamic entropy analysis. It transforms the edge weight matrix of the basic relationship network into entangled states of qubits, and uses quantum measurement and Grover's algorithm to generate a quantum relationship heatmap. This breakthrough solves the technical problem of traditional methods being unable to identify implicit strong connections. Based on the quantum relationship heatmap, dynamic entropy and propagation force values ​​are calculated, and a scientific risk assessment system is constructed by quantifying the characteristics of changes in member behavior and propagation potential.

[0124] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for dynamic management of members of a social group, characterized in that: include, Acquire online interaction data and offline co-occurrence data of social group members, clean them, and convert the cleaned online interaction data and offline co-occurrence data into a structured dataset; The structured dataset is imported into a graph database to obtain initial weights, and the dynamically decaying weights are calculated to construct a basic relational network. The edge weight matrix in the basic relation network is transformed into entangled states of qubits. The entangled states of qubits are measured to form quantum probability amplitudes. A quantum entanglement determination threshold is set, and a quantum relation heatmap is generated using the Grover algorithm. The list of strongly connected member pairs is marked in the quantum relationship heatmap, the dynamic entropy value of the social group members is calculated, a risk monitoring dataset is formed, the propagation force value is calculated based on the quantum relationship heatmap and the risk monitoring dataset, the herd immunity threshold is obtained, and a graded blocking strategy is implemented. By marking a list of strongly connected member pairs in a quantum relationship heatmap, calculating the dynamic entropy values ​​of members of the social group, and forming a risk monitoring dataset, Includes the following steps, The quantum relation heatmap is analyzed to filter quantum probability amplitudes and output a list of strongly connected member pairs. Calculate the dynamic entropy value of social group members based on a list of strongly connected member pairs; Specifically, the expression is, ； in, For members i In time t The dynamic entropy value, For the first k Weight coefficients of class behavior, The time decay coefficient, For the time when the behavior occurs, k Index the member rows; Combine the dynamic entropy value with the list of strongly connected members to calculate the risk score and output the member-risk; Based on the member-risk scoring table and dynamic entropy values, the risk monitoring dataset is obtained through real-time classification according to hierarchical rules. Based on the quantum relationship heatmap and risk monitoring dataset, the transmissibility value is calculated to obtain the herd immunity threshold. A tiered containment strategy is then implemented, including the following steps: The strong connections with probability amplitudes are extracted from the quantum relation heatmap and a member-neighbor mapping table is output. Based on the risk monitoring dataset, the propagation force value is calculated by combining the time series of dynamic entropy values ​​with a list of strongly connected member pairs. Specifically, the expression is, ； in, For members i The propagation power value, The average dynamic entropy value of all members of the current group. N The total number of samples, For members i and neighbors j The quantum probability amplitude between For the neighbors j The dynamic entropy value; Based on the transmissibility value, the herd immunity threshold is obtained through an incremental learning method; Based on the herd immunity threshold, a tiered blocking strategy is implemented.

2. The method for dynamic management of social group members as described in claim 1, characterized in that: We acquire online interaction data and offline co-occurrence data of members of social groups, clean them, and then convert the cleaned online interaction data and offline co-occurrence data into a structured dataset. Includes the following steps, Collect members' group chat messages, private chat records, and file collaboration logs, and extract the sender ID, receiver ID, interaction time, and content type fields to obtain online interaction data; Physical contact events between members are detected in group activity venues, and the IDs of the parties involved, start time, and duration are recorded to obtain offline co-occurrence data. The online interaction data is deduplicated and invalid sessions are filtered out, and the cleaned online interaction data is output. Signal strength filtering and short-time contact removal are performed on the offline co-occurrence data to obtain cleaned offline co-occurrence data; The cleaned online interaction data and offline co-occurrence data are mapped and merged according to a unified standard to obtain a structured dataset.

3. The method for dynamic management of social group members as described in claim 2, characterized in that: The structured dataset is imported into a graph database to obtain initial weights, and dynamically decaying weights are calculated to construct a basic relational network, including the following steps: Import the cleaned structured dataset into the Neo4j graph database, use Cypher statements to create member nodes and interaction edges, and output the initial weights; Calculate the dynamic decay weights based on the initial weights, update the edge weights, and output the time-sensitive weighted relationship graph; A basic relational network is constructed by pruning low-weight edges and removing invalid connections in dynamically decaying weights.

4. The method for dynamic management of social group members as described in claim 3, characterized in that: The process of transforming the edge weight matrix in the basic relational network into entangled states of qubits, and then measuring the entangled states of qubits to form quantum probability amplitudes, includes the following steps. The edge weight matrix is ​​derived from the basic relational network, normalized, and then converted into quantum rotation gate angles using the Qiskit library. Finally, CNOT and RY gates are used to convert the edges into entangled states of qubits. The quantum probability amplitude is obtained by measuring the entangled state of a qubit on a quantum processor.

5. The method for dynamic management of social group members as described in claim 4, characterized in that: Setting a threshold for quantum entanglement and generating a quantum relationship heatmap using the Grover algorithm includes the following steps. The threshold for determining quantum entanglement is set based on the quantum probability amplitude; Based on the entanglement determination threshold, a label reflecting the strong connection conditions of social networks is constructed. The quantum probability amplitude is iteratively amplified using the Grover algorithm to generate a quantum relationship heatmap.

6. A dynamic management system for social group members, based on the dynamic management method for social group members as described in any one of claims 1 to 5, characterized in that: include, The cleaning module acquires online interaction data and offline co-occurrence data of social group members, cleans them, and converts the cleaned online interaction data and offline co-occurrence data into a structured dataset. The basic relation network module imports the structured dataset into the graph database to obtain initial weights, calculates the dynamically decaying weights, and constructs the basic relation network. The quantum relationship heatmap module transforms the edge weight matrix in the basic relationship network into entangled states of qubits, measures the entangled states of qubits to form quantum probability amplitudes, sets a quantum entanglement determination threshold, and generates a quantum relationship heatmap using the Grover algorithm. The tiered blocking strategy module marks a list of strongly connected member pairs in the quantum relationship heatmap, calculates the dynamic entropy value of social group members, forms a risk monitoring dataset, calculates the propagation force value based on the quantum relationship heatmap and the risk monitoring dataset, obtains the herd immunity threshold, and executes the tiered blocking strategy.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the dynamic management method for members of social groups as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the dynamic management method for members of social groups as described in any one of claims 1 to 5.

Images