A social event early warning method, device and medium based on complex network

Abstract

This invention discloses a social event early warning method based on complex networks, comprising: acquiring historical social events occurring in each geographic grid within a region to be warned; constructing an event synchronization network to characterize the synchronicity of social events among geographic grids using a complex network method based on the lag time between each historical social event; identifying the spatial and temporal scales of the telecorrelation of social events based on the event synchronization network; using the event synchronization network at the temporal scale as the optimal event synchronization network and the spatial scale as the neighborhood radius of sample points in a clustering algorithm to perform spatial clustering of geographic grids with social event synchronicity in the optimal event synchronization network; identifying the spatial and temporal patterns of the telecorrelation of social events within the region to be warned based on each cluster and the network characteristics of the optimal event synchronization network; and providing early warning and / or resource allocation suggestions for future social events based on the spatial and temporal patterns.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data analysis and prediction technology, and in particular to a method, device and medium for early warning of social events based on complex networks. Background Technology

[0002] With the rapid development of globalization and information technology, long-distance interactions (telecommunications) are widely present in the propagation and evolution of natural and social events, influenced by the flow of material resources and the dissemination of information on a global scale.

[0003] In existing technologies, linear methods such as correlation analysis (regression equations, third-order partial correlation coefficients) and principal component analysis are mainly used to analyze the propagation patterns of events, thereby providing decision support for early warning or resource allocation for special events. However, the propagation patterns of social events are very complex, often containing a large amount of nonlinear information. Traditional methods struggle to account for this information, limiting the scientific validity of early warning or resource allocation recommendations. Summary of the Invention

[0004] This invention provides a method, device, and medium for early warning of social events based on complex networks. By using complex network methods, it reveals the nonlinear laws governing the propagation of social events, providing more rational decision support for early warning or resource allocation of social events.

[0005] In a first aspect, embodiments of the present invention provide a social event early warning method based on complex networks, including: Obtain historical social events occurring in each geographic grid within the area to be warned; Based on the lag time between various historical social events, an event synchronization network is constructed using complex network methods to characterize the synchronicity of social events between geographical grids. Based on the event synchronization network, identify the spatial and temporal scales of teleconnections of social events; The event synchronization network at the time scale is taken as the optimal event synchronization network, and the spatial scale is taken as the neighborhood radius of the sample points in the clustering algorithm. Spatial clustering is performed on the geographic grids with social event synchronization in the optimal event synchronization network. Based on the characteristics of each cluster and the optimal event synchronization network, spatial and temporal patterns of telecorrelation of social events within the area to be warned are identified. Based on the spatial and temporal patterns, early warnings and / or resource allocation suggestions for future social events are made.

[0006] Secondly, embodiments of the present invention provide a social event early warning system based on complex networks, comprising: The acquisition module is used to acquire historical social events that occurred in each geographic grid within the area to be warned. The network construction module is used to construct an event synchronization network to characterize the synchronicity of social events between geographic grids based on the lag time between various historical social events and using complex network methods. The scale identification module is used to identify the spatial and temporal scales of social events based on the event synchronization network. The clustering module is used to select the event synchronization network at the time scale as the optimal event synchronization network, and the spatial scale as the neighborhood radius of the sample points in the clustering algorithm, and to perform spatial clustering on the geographic grids with social event synchronization in the optimal event synchronization network. The pattern recognition module is used to identify the spatial and temporal patterns of telecorrelation of social events within the area to be warned, based on the network characteristics of each cluster and the optimal event synchronization network. The early warning module is used to provide early warnings and / or resource allocation suggestions for future social events based on the spatial and temporal patterns.

[0007] Thirdly, embodiments of the present invention also provide an electronic device, the electronic device comprising: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the social event early warning method based on complex networks as described in any embodiment.

[0008] In summary, this invention provides a social event early warning method based on complex networks, utilizing complex network theory to reveal the teleconnection phenomenon between social events. Based on this, it identifies the structural features of teleconnections through network characteristics and extracts teleconnection clusters using the DBSCAN clustering algorithm, identifying their interaction patterns and temporal evolution characteristics, thus providing decision support for early warning and resource allocation of social events. Specifically, this method first quantifies the synchronization intensity of social events using event synchronization methods and constructs a directed synchronization network using complex network methods. Subsequently, considering the natural laws of material resource flow and information dissemination, which should exhibit characteristics of clustering in small-scale regions and teleconnection only possible in large-scale regions, this method quantifies the spatiotemporal threshold of teleconnections based on the synchronization network, identifying the spatiotemporal scale of teleconnection analysis. Finally, through methods such as deriving network features and DBSCAN clustering, it identifies the structural features (hubs and directions) and interaction patterns of teleconnections, thereby revealing the nonlinear teleconnection law between social events. This law eliminates the interference of uncertain factors in the data and retains the universally stable part under the influence of natural laws, providing more information for early warning and resource allocation of social events. Attached Figure Description

[0009] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0010] Figure 1 This is a flowchart of a social event early warning method based on complex networks provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of an event synchronization method provided in an embodiment of the present invention; Figure 3 This is a distance distribution map showing significant synchronization of social events provided by an embodiment of the present invention; Figure 4 This is an interannual variation diagram of the teleconnection index of each cluster provided in an embodiment of the present invention; Figure 5 This is a flowchart of another social event early warning method based on complex networks provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of a social event early warning system based on a complex network provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0012] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0013] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0014] Figure 1 This is a flowchart of a social event early warning method based on complex networks provided by an embodiment of the present invention. This method is suitable for providing early warnings of the propagation and evolution trends of social events and is executed by electronic devices. Figure 1 As shown, the method specifically includes: S110. Obtain historical social events that occurred in each geographic grid within the area to be warned.

[0015] The areas requiring early warning here typically refer to large-scale regions, such as global or national areas. Social events within large-scale regions are more likely to exhibit teleconnection characteristics between distant geographic grids, making them more suitable for analyzing and predicting the propagation trends of events using the method in this embodiment.

[0016] The social events discussed here specifically refer to those influenced by the flow of material resources and the dissemination of information, such as public health events (e.g., peak outbreaks of disease), cybersecurity incidents targeting infrastructure (e.g., cyberattacks on oil and gas pipelines), and public opinion events arising from the flow of goods (e.g., conflicts caused by food crises). The spread and evolution of these social events are reflections of the natural laws governing the flow of material resources and the dissemination of information in human activities, and inherently follow these natural laws. The purpose of this embodiment is to use data analysis to uncover the natural laws inherent in the data, thereby providing decision support for early warning and resource allocation for special events.

[0017] Based on the above concepts, this embodiment first obtains the same type of social events that have occurred in the area to be warned within a certain historical period (such as all being public health events, or all being cybersecurity events based on infrastructure, etc.), and distributes them to each geographic grid according to their location as the data source for the entire method.

[0018] S120. Based on the lag time between various historical social events, construct an event synchronization network using complex network methods to characterize the synchronicity of social events between geographical grids.

[0019] This step utilizes complex network theory and combines it with event synchronization methods to construct a complex network by quantifying the synchronization strength between events, which serves as the basis for subsequent identification of the telecorrelation spatiotemporal scales of social events.

[0020] In one specific embodiment, the process includes the following steps: Step 1: Based on the lag time of each historical social event and the maximum lag time of synchronous social events between two geographic grids, identify whether historical social events occurring in different geographic grids are synchronous events.

[0021] Optionally, for each geographic grid: Historical social events occurring within that grid are sorted according to their chronological order, and it is determined whether the end time of each historical social event is greater than the start time of the next historical social event. If so, the two events are merged into a single event. This process is repeated until all events have been traversed, ultimately yielding a set of non-overlapping social events across all geographic grids. For example, for any two geographic grids i and j, the final occurrence times of the historical social events occurring within grid i are as follows: ( l =1,2,…… si The times of occurrence of the various historical and social events in grid j are as follows: ( m =1,2,…… sj ),in, These represent the total number of such events occurring in grid i and grid j, respectively.

[0022] Then, for any two historical social events occurring in any two geographic grids, the lag time between each historical social event and preceding and following historical social events occurring in the same geographic grid is determined; and based on the minimum lag time and the maximum lag time of synchronous social events occurring between two geographic grids, a lag time threshold for judging synchronous events is determined. Combined with... Figure 2 In the event synchronization method, for any two grid positions i and j, events occurring at any two times... and Two historical and social events can construct a local time scale. (i.e., variable lag time), used as a lag time threshold for determining whether the two historical social events are synchronous social events: (1) In addition, this embodiment also limits the maximum lag time for synchronous social events occurring between two geographic grids. That is, only when the time difference between two events does not exceed Only then are they considered the same social event (or synchronous event). Therefore, when > hour, Pick .

[0023] Based on the above threshold, we can use This represents the total probability that the same event occurs at position i after it occurs at position j (this probability is used to characterize the degree and its value is not limited to [0,1]): (2) in, As an intermediate variable, (3) The aforementioned total probability allows for dynamic delays between synchronous events at different locations, thus enabling the construction of the following symmetric and asymmetric combinations to measure the synchronicity and delay behavior of social events, respectively: , (4) in, It is an indicator that measures the intensity of synchronization of social events between geographic grids i and j, while It is an indicator that measures the delay deviation in the occurrence of social events between geographic grids i and j. The value ranges from [0,1], and is used when events are fully synchronized between grid positions i and j. . Between [-1, 1], when the event always occurs before the event at grid i. When the two grids are fully synchronized ,otherwise .

[0024] After the calculation is complete, two dimensions will be obtained. The matrix is ​​a square matrix, where N is the total number of grids in the region to be predicted where social events occur. One matrix provides the synchronization magnitude of such events between any two grid points (synchronization matrix), and the other matrix provides the delay deviation of such events between any two grids (delay matrix), to determine the order in which events occur between grids.

[0025] Step 2: Based on the identification results and complex network theory, an event synchronization network is constructed to characterize the synchronicity of social events among geographic grids. This social event synchronization network consists of all nodes (geographic grids) and a set of pairwise relationships between them. Each connection in the network represents the synchronization characteristics of events between two grids. In the event synchronization method, both the synchronization matrix and the delay matrix are symmetric, containing the synchronization values ​​and delay deviations of all possible grid location pairs. Therefore, these two matrices can be used to determine the connections and directions of the social event synchronization network.

[0026] Specifically, for any two geographic grids, the synchronization strength between the sets of social events occurring between the two geographic grids can be used as a measure. and delay deviation This determines whether two geographic grids are connected. Optionally, they are only connected if the time synchronization strength between the two geographic grids is above the 95th percentile, i.e., the synchronization strength significance level is 0.05, thus allowing for the construction of an appropriate connection density for the event synchronization network.

[0027] Then, the binary adjacency matrix A of the event synchronization network is constructed based on the synchronization matrix and the delay matrix. When the elements of the adjacency matrix... When the value is 1, it indicates that geographic grid i and geographic grid j have spatial synchronization characteristics, and edges can be connected between the two nodes. This process is repeated to construct an event synchronization network, and its formula is expressed as follows: (5) in, The 95th percentile of all non-zero synchronization values.

[0028] S130. Based on the event synchronization network, identify the spatial and temporal scales of the teleconnections of social events.

[0029] Due to geographical and temporal limitations, the synchronicity of social events should exhibit a certain degree of clustering within a relatively small area, such as higher synchronicity within closely spaced clusters. However, on a large scale, due to the existence of some nonlinear evolutionary patterns, it also exhibits teleconnection characteristics between clusters. To align with this natural law, this step first defines specific spatial and temporal scales for the clusters in the teleconnection analysis based on the event synchronization network (hereinafter referred to as the spatial and temporal scales of social event teleconnection) to improve the rationality of subsequent analyses.

[0030] First, the time scale of the teleconnection of social events is identified. In one specific implementation, this process may include the following steps: Step 1: Adjust the maximum hysteresis time multiple times. And based on the adjusted maximum lag time, the event synchronization network is reconstructed using the complex network method. In this embodiment, the event synchronization similarity metric allows for a range of [0, ...]. [Considering dynamic delay] Therefore, when identifying time scales, this feature can be used to set the maximum lag time for multiple time resolutions. According to each Repeat the operation of S120 to obtain each The corresponding event synchronization network.

[0031] Step 2: Based on the aggregation characteristics of node degrees in each event synchronization network, determine the optimal value of the maximum lag time as the time scale for the teleconnection of social events. Optionally, for each The corresponding event synchronization network calculates the degree centrality (DC) of each node (i.e., the connected geographic grid). DC represents the number of social event grid points that are synchronized with that node. Degree can be defined as the number of connections between vertices; the degree of node i can be expressed as: (8) in, N1 is a node The total number of connected nodes. These are elements in the binary adjacency matrix of the event synchronization network.

[0032] Optional, for each Draw a DC distribution map, where color intensity represents the central characteristics of the DCs, with darker colors indicating stronger central characteristics. Observe each... The corresponding DC distribution plot shows that, with As the density of events increases, the degree of clustering (centrality) in the DC diagram gradually weakens, indicating that social events are beginning to spread and proliferate over long distances. However, according to the natural laws of material resource flow and information dissemination, the spread of social events should exhibit certain clustering characteristics within small-scale areas, and the corresponding DC diagram should also show obvious spatial clustering characteristics. Therefore, we select the DC diagram with a clear centrality and stable spatial pattern, and use that DC diagram as the basis for our analysis. The optimal value was determined and used as the time scale for the event teleconnection.

[0033] Optionally, the clustering of node degrees in the event synchronization network can also be calculated as follows: First, calculate the degree of each node in each event synchronization network within a given neighborhood radius. Then, for each event synchronization network, perform the following operations: determine the geographical neighborhood density of each node based on its degree within the given neighborhood radius; calculate the geographical clustering degree of the current event synchronization network based on its geographical neighborhood density. Finally, select the maximum lag time corresponding to the event synchronization network with the highest geographical clustering degree as the optimal value. For example, given a neighborhood radius r, for each node i in the current network, if other nodes j are in a neighborhood centered on node i with a radius of r (area r... Inside, and If node j is a neighbor of node i in its geographical neighborhood, then node j is considered a neighbor of node i; the number of nodes j That is, the degree of node i within its neighborhood radius; then the geographic neighborhood density of node i is... High density indicates high clustering; thus, the average neighborhood density of the entire network can be calculated. ,in, This represents the number of nodes in the current event synchronization network; therefore, the geographical clustering of the entire network is... , The larger the value, the stronger the network aggregation. (This is followed by a seemingly unrelated sentence about obtaining each...) corresponding Then, select The largest The optimal value is used as the event scale for event teleconnection.

[0034] Step 3: From the optimal event synchronization network corresponding to the optimal value, determine the geographical grid pairs with social event synchronization; and analyze the spatial scale of social event teleconnection based on the distance between each geographical grid pair. Once the optimal value is determined, the event synchronization network corresponding to that optimal value is the optimal event synchronization network.

[0035] Optionally, for any two nodes (i.e., geographic grids) i and j that are connected in the optimal event synchronization network, the Haversine geographic distance is calculated as follows: (6) in, For the Earth's radius, and Let be the latitude of the two nodes. and The difference in latitude and longitude between the two nodes.

[0036] Then, the spatial distance of all node pairs that have established connections will be calculated. To fully capture the distribution characteristics from short to long distances, the sample set D is constructed using a logarithmic interval method, dividing it into several distance intervals. ( , Let represent the two endpoints of the l-th interval, and construct the probability density function using the normalized histogram method as follows: (7) in, This represents the number of node pairs (i.e., geographic grid pairs) that have established connections in the optimal event synchronization network. Geographical distance In the interval The number of node pairs, For indicator functions, when hour ,otherwise , This represents the probability density corresponding to each interval.

[0037] Based on the above calculation results, a distance distribution map can be drawn, such as... Figure 3 As shown in the figure, the Y-axis represents the probability density function value of geographic distance, and the X-axis represents geographic distance. The dual-axis data is represented using log-log coordinates. In the distance distribution map, the spatial scale is identified according to the following steps: based on the PDF curve of geographic distance between nodes, the characteristics of the curve's variation with distance are identified, and the locations of significant inflection points or local extrema appearing in the curve are determined; the first inflection point or extrema location is used as the regional scale threshold for short-connection distances; the second inflection point or extrema location is used as the teleconnection distance threshold from local to large regional scales. The first inflection point or extrema location aligns with the natural law that the flow of material resources and the dissemination of information are constrained by geographic distance, and can be used as the spatial scale for subsequent analysis of teleconnections of social events.

[0038] S140. Using the spatial scale as the neighborhood radius of the sample points in the clustering algorithm, spatial clustering is performed on the geographic grids with social event synchronization in the optimal event synchronization network.

[0039] This step identifies nodes with significant correlations in the optimal event synchronization network and performs spatial clustering on these nodes based on the aforementioned spatial scale to identify the structural features (including hubs and directions) of social events that are remotely related.

[0040] In one specific implementation, the first step is to identify the remotely related hubs. The degree distribution of complex networks in many real-world systems does not follow a uniform or normal distribution, but often exhibits a power-law distribution. This means that most nodes have a small number of connections, while a few nodes have a large number of connections. Therefore, by deriving the degree of the event synchronization network and utilizing the DBSCAN algorithm, the remotely related hubs of social events can be identified and analyzed. Optionally, this process may include the following steps: First, when constructing the optimal event synchronization network, the degree of each node in the network has been obtained through the calculation using formula (8). Nodes with extraction scores significantly higher than the average level are identified and marked on the map. For ease of distinction, these marked nodes on the map will be referred to as labeled nodes.

[0041] Then, the spatial scale of the teleconnections of social events identified in S130 is used as the neighborhood radius ε of the sample points. The DBSCAN algorithm is then executed to spatially cluster all labeled nodes, accurately extracting the hubs (clusters) of teleconnections. Specifically, DBSCAN uses several core definitions to describe and implement the clustering process. Assume D represents the height of the optimal event in different networks. The locations where social events occur (i.e., geographic grids) The set of: (9) For any sample ε neighborhood It can be defined as: (10) The shape corresponding to the ε-neighborhood consists of two points. and Distance function between Decision. Optional. Use Haversine as the distance function. ε is used. Figure 3 The first distance threshold in the spatial distance distribution (i.e., the spatial scale determined in S130).

[0042] Cluster C is a non-empty subset of D. For any geographic grid and ,if Belongs to C, and Depend on If the density is achievable, then It also belongs to C. Correspondingly... and When both points belong to the same cluster, the density between them is achievable. For each point in cluster C... A given radius ε must contain at least a certain number of points in its neighborhood, i.e. Then Defined as the core point. From this, different clusters are obtained, namely teleconnection hubs (clusters).

[0043] After identifying the teleconnection hubs, the next step is to determine the teleconnection directions between each hub. In an event synchronization network, the in-degree reflects the degree and frequency of how social events in one region are affected by events in other regions. The out-degree, on the other hand, reflects the degree and frequency of how social events in one region directly affect events in other regions. Therefore, based on the optimal event synchronization network, the out-degree and in-degree network characteristics can be used to identify the directions of teleconnections, i.e., the source centers and affected sink centers of events. Optionally, this process may include the following steps: First, based on the adjacency matrix of the optimal event synchronization network, calculate the out-degree and in-degree of each node that has established a connection: (11) (12) in, This represents the out-degree of node i. This indicates that there exists an edge from node i to node j; Indicates the in-degree of node i. This indicates that there exists an edge from node j to node i.

[0044] Then, the out-degree and in-degree are respectively used as the values ​​in the above steps. The DBSCAN algorithm is used again to identify nodes or regions with significantly high out-degree as event source centers (source nodes) and nodes or regions with significantly high in-degree as event sink centers (affected nodes).

[0045] S150. Based on the characteristics of each cluster and the optimal event synchronization network, identify the spatial and temporal patterns of telecorrelation of social events within the area to be warned.

[0046] This step, based on the extracted teleconnection clusters, analyzes the mutual influence patterns among these clusters, revealing the influence relationships and interaction patterns between them. Spatial patterns can be characterized by the spatial location of each cluster, as well as the direction and intensity of their mutual influence, while temporal patterns can be characterized by the changes in teleconnection characteristics over time.

[0047] In one specific implementation, the network characteristics of the event synchronization network represent the degree of network nodes or clusters. Therefore, the above pattern recognition process may include the following steps: Step 1: Determine the out-degrees between clusters based on the out-degrees of each geographic grid within each cluster in the optimal event synchronization network. Assume there are K clusters in total, including... If there are n nodes, then the set of nodes that make up the entire cluster is . The cluster set is The adjacency matrix of the network composed of these nodes is ,in: (13) This allows us to calculate the cluster. Out-degree of a single node to other clusters: For a node For clusters out of degree It can be represented as: (14) This means that node i points to the cluster. The sum of the number of edges or weights of all nodes in the array.

[0048] This allows for the creation of clusters. For cluster out of degree : (15) if This gives the out-degree within the cluster; if This gives the out-degree between clusters.

[0049] Step 2: Based on the out-degree of each cluster, determine the telecorrelation strength of social events among the clusters, which serves as the spatial pattern of social event telecorrelation within the area to be warned. Specifically, based on... It can compute clusters Total out-degree (Number of edges pointing to all clusters): (16) Thus, the cluster is calculated. For cluster Out-of-degree ratio : (17) This ratio represents the cluster. What percentage of the total impact points to the cluster? ,satisfy: (18) Following the steps above, the total out-degree and out-degree percentage of each cluster to other clusters can be calculated, thus obtaining the teleconnection strength matrix between clusters. Each row and column of this matrix corresponds to a cluster, where the matrix element in the m-th row and n-th column is... This represents a cluster. For cluster The teleconnection strength.

[0050] Step 3: Analyze the change in the total out-degree of each cluster to other clusters over time, as a temporal pattern of the teleconnection of social events within the area to be warned. Optionally, the interannual variation of the teleconnection intensity can be analyzed to reveal the temporal evolution of the teleconnection of social events. The specific steps are as follows: First, for each cluster Calculate the total number of network connections for the teleconnected clusters of the directed synchronization network (i.e., the optimal event synchronization network) on an annual basis: (19) in, Represents a cluster The total number of connections in year y, N2 is the number of connections in the cluster. The total number of internal nodes.

[0051] For each cluster After performing the above calculations, we can continue to calculate each cluster. Standardized teleconnection index : (20) in, and Representing clusters All years The mean and standard deviation of .

[0052] Based on the above calculation results, an interannual variation map of the teleconnection index can be plotted, such as... Figure 4 As shown. By calculating the slope of the curve, the temporal trend of the teleconnection index can be identified. By comparing different clusters, the differences in the changes of the teleconnection index among the clusters can also be identified.

[0053] S160. Based on the spatial and temporal patterns, provide early warnings and / or resource allocation suggestions for future social events.

[0054] The aforementioned spatial and temporal patterns reflect the propagation patterns of social events from one cluster to other clusters. Based on these patterns, early warnings and / or resource allocation suggestions can be made for potential future social events.

[0055] Optionally, in response to a new social event occurring within a certain cluster, social event warnings and / or resource allocation suggestions can be made for other clusters that have a significant social event telecorrelation strength with the aforementioned cluster, based on the telecorrelation matrix, and the corresponding telecorrelation strength can be used as the confidence level of the warning and / or suggestion. For example, other clusters with telecorrelation strengths higher than a set threshold can be designated as warning targets, and warnings and / or resource allocation suggestions can be made in advance. For instance, in response to a peak in disease outbreaks, disease prevention warnings or suggestions for allocating protective materials can be made in advance; in response to cyberattacks on oil and gas pipelines, cybersecurity warnings or suggestions for deploying cybersecurity equipment can be made in advance; and in response to conflicts caused by food shortages, food warnings or suggestions for food allocation can be made in advance, etc.

[0056] Simultaneously, based on the time-varying pattern of the total out-degree of a certain cluster to other clusters, for other clusters that have a significant telecorrelation strength with the aforementioned cluster, detailed annual social event warnings and / or resource allocation suggestions are provided. For example, years with higher telecorrelation indices are designated as key years for warnings or resource allocation, and combined with the warning regions and confidence levels in spatial model warnings, more refined warning or suggestion information is constructed.

[0057] The entire process described above can also be combined with Figure 5 This is understood. It should be noted that all information and data involved in this application are authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users or parties to choose to authorize or refuse.

[0058] In summary, this embodiment provides a social event early warning method based on complex networks, utilizing complex network theory to reveal the teleconnectivity between social events. Based on this, it identifies the structural features of teleconnections through network characteristics and extracts teleconnection clusters using the DBSCAN clustering algorithm, identifying their interaction patterns and temporal evolution characteristics, thus providing decision support for early warning of social events and resource allocation.

[0059] Specifically, this method first quantifies the synchronization intensity of social events using event synchronization methods and constructs a directed synchronization network using complex network methods. Then, considering the natural laws governing the flow of material resources and the dissemination of information, which should exhibit characteristics of aggregation in small-scale regions and teleconnection in large-scale regions, this method quantifies the spatiotemporal thresholds of teleconnection based on the synchronization network, identifying the spatiotemporal scales for teleconnection analysis. Finally, by deriving network characteristics and using DBSCAN clustering, the method identifies the structural characteristics (hubs and directions) and interaction patterns of teleconnection, thereby revealing the nonlinear teleconnection laws between social events. This law eliminates the interference of uncertain factors in the data and retains the universally stable parts under the influence of natural laws, providing more information for early warning and resource allocation of social events.

[0060] Figure 6 This is a schematic diagram of the structure of a social event early warning system based on a complex network, provided in an embodiment of the present invention. Figure 6 As shown, the system includes: The acquisition module is used to acquire historical social events that occurred in each geographic grid within the area to be warned. The network construction module is used to construct an event synchronization network to characterize the synchronicity of social events between geographic grids based on the lag time between various historical social events and using complex network methods. The scale identification module is used to identify the spatial and temporal scales of social events based on the event synchronization network. The clustering module is used to select the event synchronization network at the time scale as the optimal event synchronization network, and the spatial scale as the neighborhood radius of the sample points in the clustering algorithm, and to perform spatial clustering on the geographic grids with social event synchronization in the optimal event synchronization network. The pattern recognition module is used to identify the spatial and temporal patterns of telecorrelation of social events within the area to be warned, based on the network characteristics of each cluster and the optimal event synchronization network. The early warning module is used to provide early warnings and / or resource allocation suggestions for future social events based on the spatial and temporal patterns.

[0061] Optionally, the network construction module constructs an event synchronization network to characterize the synchronicity of social events among geographic grids by using complex network methods based on the lag time between each historical social event: Based on the lag time of each historical social event and the maximum lag time of synchronous social events between two geographic grids, we can identify whether historical social events occurring in different geographic grids are synchronous events. Based on the identification results and complex network theory, an event synchronization network is constructed to characterize the synchronicity of social events among geographic grids.

[0062] Optionally, the network construction module identifies whether historical social events occurring in different geographic grids are synchronous events based on the lag time of each historical social event and the maximum lag time of synchronous social events between two geographic grids: For any two historical social events that occur in any two geographic grids, determine the lag time between each historical social event and the preceding and following historical social events that occur in the same geographic grid. Based on the minimum value among the various lag durations and the maximum lag duration of synchronous social events occurring between two geographic grids, a lag duration threshold for judging synchronous events is determined. Based on the lag time threshold and the lag time between any two historical social events, determine whether any two historical social events are synchronous events.

[0063] Optionally, the scale recognition module identifies the spatial and temporal scales of social events based on the event synchronization network in the following ways: The maximum lag time is adjusted multiple times, and the event synchronization network is reconstructed using the complex network method based on each adjusted maximum lag time. Based on the aggregation characteristics of node degree in the synchronization network of each event, the optimal value of the maximum lag time is determined as the time scale of the teleconnection of social events. From the optimal event synchronization network corresponding to the optimal value, determine the geographical grid pairs with social event synchronization. Based on the distances between each geographic grid pair, the spatial scale of the teleconnection of social events is analyzed.

[0064] Optionally, the scale identification module determines the optimal value of the maximum lag time based on the aggregation characteristics of node degrees in each event synchronization network in the following ways: Optionally, the scale identification module analyzes the spatial scale of the teleconnection of social events based on the distance between each pair of geographic grids in the following way: Divide the distance between geographic grid pairs into multiple intervals; Based on the number of geographic grid pairs within each interval, a probability density function is constructed to characterize the intensity of synchronization of social events. Based on the probability density function, the maximum distance at which social events show significant synchronicity is determined, which serves as the spatial scale for the teleconnection of social events.

[0065] Optionally, the pattern recognition module identifies the spatial and temporal patterns of telecorrelation of social events within the area to be warned, based on the network characteristics of each cluster and the optimal event synchronization network, in the following manner: The out-degree between clusters is determined based on the out-degree of each geographic grid in each cluster in the optimal event synchronization network; Based on the out-degree between each cluster, the telecorrelation strength of social events between each cluster is determined, which serves as the spatial pattern of the telecorrelation of social events within the area to be warned. The variation of the total out-degree of each cluster to other clusters over time is analyzed to serve as the time pattern of the telecorrelation of social events in the area to be warned.

[0066] Optionally, the early warning module can provide early warnings and / or resource allocation suggestions for future social events based on the spatial and temporal patterns in the following ways: In response to a new social event occurring within a certain cluster, social event warnings and / or resource allocation suggestions are provided to other clusters that have a significant telecorrelation strength with the aforementioned cluster. Based on the time-varying pattern of the total out-degree of a certain cluster to other clusters, for other clusters that have a significant telecorrelation strength with social events related to the certain cluster, detailed annual social event warnings and / or resource allocation suggestions are provided.

[0067] It is worth mentioning that this embodiment is based on the same inventive concept as the above method embodiments, and any limitation in the above method embodiments is applicable to this embodiment, and this embodiment can achieve the same technical effect as any of the above method embodiments.

[0068] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, such as... Figure 7 As shown, the device includes a processor 60, a memory 61, an input device 62, and an output device 63; the number of processors 60 in the device can be one or more. Figure 7 Taking a processor 60 as an example; the processor 60, memory 61, input device 62, and output device 63 in the device can be connected via a bus or other means. Figure 7 Taking the example of a connection between China and Israel via a bus.

[0069] The memory 61, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the social event early warning method based on complex networks in this embodiment of the invention. The processor 60 executes various functional applications and data processing of the device by running the software programs, instructions, and modules stored in the memory 61, thereby realizing the aforementioned social event early warning method based on complex networks.

[0070] The memory 61 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on terminal usage. Furthermore, the memory 61 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, the memory 61 may further include memory remotely located relative to the processor 60, which can be connected to the device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0071] Input device 62 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the device. Output device 63 may include display devices such as a display screen.

[0072] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements any of the complex network-based social event early warning methods of any embodiment.

[0073] The computer storage medium of this invention can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0074] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0075] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0076] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—as well as conventional procedural programming languages—such as C or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A social event early warning method based on complex networks, characterized in that, include: Obtain historical social events occurring in each geographic grid within the area to be warned; Based on the lag time of each historical social event and the maximum lag time of synchronous social events between two geographic grids, we can identify whether historical social events occurring in different geographic grids are synchronous events. Based on the identification results and complex network theory, an event synchronization network is constructed to characterize the synchronicity of social events among geographic grids. The maximum lag time is adjusted multiple times, and the event synchronization network is reconstructed using the complex network method based on each adjusted maximum lag time. Based on the aggregation characteristics of node degree in the synchronization network of each event, the optimal value of the maximum lag time is determined as the time scale of the teleconnection of social events. Specifically, the optimal value is the maximum lag time that maximizes the geographical clustering of the entire network. From the optimal event synchronization network corresponding to the optimal value, determine the geographical grid pairs with social event synchronization; divide the distance between the geographical grid pairs into multiple intervals; construct a probability density function to characterize the intensity of social event synchronization based on the number of geographical grid pairs in each interval; The first inflection point or extreme value in the probability density function curve is used as the spatial scale of the teleconnection of social events. The event synchronization network at the time scale is taken as the optimal event synchronization network, and the spatial scale is taken as the neighborhood radius of the sample points in the clustering algorithm. Spatial clustering is performed on the geographic grids with social event synchronization in the optimal event synchronization network. Based on the characteristics of each cluster and the optimal event synchronization network, spatial and temporal patterns of telecorrelation of social events within the area to be warned are identified. Based on the spatial and temporal patterns, early warnings and / or resource allocation suggestions for future social events are made.

2. The method according to claim 1, characterized in that, The method of identifying whether historical social events occurring in different geographic grids are synchronous events based on the lag time of each historical social event and the maximum lag time of synchronous social events between two geographic grids includes: For any two historical social events that occur in any two geographic grids, determine the lag time between each historical social event and the preceding and following historical social events that occur in the same geographic grid. Based on the minimum value among the various lag durations and the maximum lag duration of synchronous social events occurring between two geographic grids, a lag duration threshold for judging synchronous events is determined. Based on the lag time threshold and the lag time between any two historical social events, determine whether any two historical social events are synchronous events.

3. The method according to claim 1, characterized in that, The step of determining the optimal value of the maximum lag time based on the aggregation characteristics of node degrees in each event synchronization network includes: Calculate the degree of each node in the event synchronization network within a given neighborhood radius; For each event synchronization network: determine the geographical neighborhood density of each node based on the degree of each node within a given neighborhood radius in the current event synchronization network; calculate the geographical clustering degree of the current event synchronization network based on the geographical neighborhood density of each node. The optimal value is the maximum latency corresponding to the event synchronization network with the highest geographical clustering.

4. The method according to claim 1, characterized in that, The step of identifying spatial and temporal patterns of telecorrelation of social events within the area to be warned, based on the network characteristics of each cluster and the optimal event synchronization network, includes: The out-degree between clusters is determined based on the out-degree of each geographic grid in each cluster in the optimal event synchronization network; Based on the out-degree between each cluster, the telecorrelation strength of social events between each cluster is determined, which serves as the spatial pattern of the telecorrelation of social events within the area to be warned. The variation of the total out-degree of each cluster to other clusters over time is analyzed to serve as the time pattern of the telecorrelation of social events in the area to be warned.

5. The method according to claim 4, characterized in that, The provision of early warnings and / or resource allocation suggestions for future social events based on the spatial and temporal patterns includes: In response to a new social event occurring within a certain cluster, social event warnings and / or resource allocation suggestions are provided to other clusters that have a significant telecorrelation strength with the aforementioned cluster. Based on the time-varying pattern of the total out-degree of a certain cluster to other clusters, for other clusters that have a significant telecorrelation strength with social events related to the certain cluster, detailed annual social event warnings and / or resource allocation suggestions are provided.

6. A social event early warning system based on complex networks, characterized in that, include: The acquisition module is used to acquire historical social events that occurred in each geographic grid within the area to be warned. The network construction module is used to identify whether historical social events occurring in different geographic grids are synchronous events based on the lag time of each historical social event and the maximum lag time of synchronous social events between two geographic grids. Based on the identification results and complex network theory, an event synchronization network is constructed to characterize the synchronicity of social events among geographic grids. The scale identification module is used to adjust the maximum lag time multiple times, and reconstruct the event synchronization network using complex network methods based on each adjusted maximum lag time. Based on the aggregation characteristics of node degree in the synchronization network of each event, the optimal value of the maximum lag time is determined as the time scale of the teleconnection of social events. Specifically, the maximum lag time with the highest geographical clustering in the entire network is selected as the optimal value; from the optimal event synchronization network corresponding to the optimal value, geographical grid pairs with social event synchronization are determined; the distance between geographical grid pairs is divided into multiple intervals; based on the number of geographical grid pairs in each interval, a probability density function is constructed to characterize the intensity of social event synchronization. The first inflection point or extreme value in the probability density function curve is used as the spatial scale of the teleconnection of social events. The clustering module is used to select the event synchronization network at the time scale as the optimal event synchronization network, and the spatial scale as the neighborhood radius of the sample points in the clustering algorithm, and to perform spatial clustering on the geographic grids with social event synchronization in the optimal event synchronization network. The pattern recognition module is used to identify the spatial and temporal patterns of telecorrelation of social events within the area to be warned, based on the network characteristics of each cluster and the optimal event synchronization network. The early warning module is used to provide early warnings and / or resource allocation suggestions for future social events based on the spatial and temporal patterns.

7. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the social event early warning method based on complex networks as described in any one of claims 1-5.

Images