Visualization method and system of biological invasion in shipping data based on high-order network

Abstract

This invention discloses a method and system for visualizing biological invasions using shipping data based on a high-order network. The method comprises: obtaining shipping data and generating biological invasion probabilities between ports based on the shipping data; constructing a high-order biological invasion network based on the biological invasion probabilities between ports and the shipping data; and generating port invasion risks based on the high-order biological invasion network and visually displaying them. The system comprises a probability calculation module, a network construction module, and a display module. Using this method, a high-order network can be constructed using shipping data and various data from the high-order network can be effectively displayed, thereby enabling better visualization and analysis of biological invasions. As a method and system for visualizing biological invasions using shipping data based on a high-order network, this invention can be widely applied in the field of biological invasion prediction.

Description

Technical Field

[0001] The present invention relates to the field of biological invasion prediction, and in particular to a method and system for visualizing biological invasion of shipping data based on a high-order network. Background Art

[0002] In the global ocean shipping network, some organisms can enter other ecosystems through ship ballast or, unknowingly, through the hulls themselves, "hitchhiking" and thus causing bioinvasions. Traditional maritime data research rarely uses visualization systems, instead relying on process-oriented mathematical models to gradually explore key or interesting information. This approach is inflexible and does not allow for direct observation of relationships between data. Summary of the Invention

[0003] In order to solve the above technical problems, the purpose of the present invention is to provide a method and system for visualizing biological invasions based on shipping data using a high-order network, which can use shipping data to construct a high-order network and effectively display various data of the high-order network, thereby better performing biological invasion visualization analysis.

[0004] The first technical solution adopted by the present invention is: a method for visualizing biological invasions of shipping data based on a high-order network, comprising the following steps:

[0005] Obtain shipping data and generate the probability of biological invasion between ports based on shipping data;

[0006] Based on the probability of biological invasion between ports, a high-order biological invasion network was constructed in combination with shipping data.

[0007] Generate port invasion risk based on biological invasion high-order network and perform visualization.

[0008] Furthermore, the step of obtaining shipping data and generating the probability of biological invasion between ports based on the shipping data is as follows:

[0009] Acquire shipping data, including time series data of different ships arriving at different ports and ecological data of the ports;

[0010] Based on the time series data of different ships arriving at different ports and the ecological data of the ports, the invasion risk brought by ships between two ports is defined;

[0011] Based on the invasion risk brought by ships between two ports, the ships in the shipping data are integrated to obtain the probability of invasion between the two ports.

[0012] Furthermore, the ecological data of the port includes average temperature, salinity, sea area, ecosystem and adjacent ecological areas.

[0013] Furthermore, the probability formula of invasion between the two ports is expressed as follows:

[0014]

[0015] In the above formula, the p ij represents the probability of invasion between port i and port j, the pr ij Indicates whether port i and port j are in different ecological areas. ij represents the similarity of average temperature and salinity between port i and port j, represents the invasion risk brought by a ship between port i and port j, where i and j represent different ports respectively, and v represents a ship.

[0016] Furthermore, the step of constructing a biological invasion high-order network based on the biological invasion probability between ports and combined with shipping data specifically includes:

[0017] Based on the biological invasion probability between ports and the historical trajectory of ships in shipping data, the invasion probability of the trajectory is generated;

[0018] According to the shipping data, the invasion probability of the trajectory is calculated in sequence and cut to obtain high-order nodes;

[0019] Taking trajectories as edges, a high-order network is constructed according to high-order nodes to obtain a high-order network of biological invasion.

[0020] Furthermore, the step of generating port invasion risk based on the biological invasion high-order network and visually displaying it specifically includes:

[0021] The invasion probability of the port is calculated based on the invasion probability of the edges and trajectories in the biological invasion high-order network, and the port invasion risk is obtained;

[0022] Integrate the port intrusion risk in the high-order network and create dependency view, subgraph view and aggregation view based on the geographical view for visualization.

[0023] Furthermore, the creation of dependency views, subgraph views, and aggregate views based on the geographic view for visual display specifically includes:

[0024] Displays the probability of invasion of ports based on a geographical view;

[0025] Display the high-order nodes corresponding to the ports in the dependency view;

[0026] Display the entire high-level network through subgraph view;

[0027] The weight statistics of high-order network communities are displayed through aggregate views.

[0028] The second technical solution adopted by the present invention is: a marine data biological invasion visualization system based on a high-order network, comprising:

[0029] A probability calculation module is used to obtain shipping data and generate the probability of biological invasion between ports based on the shipping data;

[0030] The network construction module constructs a high-order biological invasion network based on the probability of biological invasion between ports and combined with shipping data;

[0031] The display module generates port invasion risks based on the biological invasion high-order network and displays them visually.

[0032] The beneficial effects of the method and system of the present invention are as follows: the present invention uses maritime transport data to construct an invasion probability mathematical model, and further proposes a method for dividing high-order nodes and high-order network edges, thereby constructing a high-order network that conforms to reality and has physical significance for biological invasions, effectively displaying various data of the constructed high-order network, and being able to intuitively find data of interest, thereby better conducting exploration and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flowchart of the steps of a method for visualizing biological invasions of shipping data based on a high-order network according to the present invention;

[0034] Figure 2 This is a structural block diagram of a marine data biological invasion visualization system based on a high-order network according to the present invention;

[0035] Figure 3 is a schematic diagram of a geographical view of a specific embodiment of the present invention;

[0036] Figure 4 is a schematic diagram of a dependency view according to a specific embodiment of the present invention;

[0037] Figure 5 It is a practical schematic diagram of the subgraph view and the aggregate graph according to a specific embodiment of the present invention. DETAILED DESCRIPTION

[0038] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The step numbers in the following embodiments are provided for ease of description only and do not limit the order of the steps. The order of execution of the steps in the embodiments can be adaptively adjusted based on the understanding of those skilled in the art.

[0039] like Figure 1 As shown, the present invention provides a method for visualizing biological invasions of shipping data based on a high-order network, the method comprising the following steps:

[0040] S1. Obtain shipping data and generate the probability of biological invasion between ports based on the shipping data;

[0041] S1.1. Acquire shipping data, including time series data of different ships arriving at different ports and ecological data of the ports;

[0042] S1.2. Define the invasion risk posed by ships between two ports based on time series data of different ships arriving at different ports and the ecological data of the ports;

[0043] Specifically, based on the time series data of different ships arriving at different ports, the historical trajectory of each ship and the interaction information between each port and the ship can be obtained. The ecological data of the port includes average temperature, salinity, the sea area where it is located, the ecosystem where it is located, and the adjacent ecological areas.

[0044] If we want to quantify biological invasions, we can think of them as the probability of a biological invasion. For a ship, every time it passes between two different ecological zones, a biological invasion is possible. The ship may carry organisms from the previous port, or even several ports, contaminating the ecology of the current port. Now, let's define the invasion risk posed by a ship between two ports as:

[0045]

[0046] Among them, i and j represent port i and port j, pr ij Indicates whether ports i and j are in the same ecological region, pe ij Indicates the similarity between port environments i and j, represents the invasion risk of ship v from port i to port j. The following defines pr ij 、pe ij and risk of intrusion.

[0047] pr ij is defined as:

[0048]

[0049] The risk of invasion only arises when the two ecological zones are different.

[0050] pe ij Defined as:

[0051]

[0052] Where T represents the average temperature of the port, S represents the salinity of the port, Δ represents the difference between the values ​​for ports i and j, and δT and δS represent the standard deviation of the average temperature and salinity between the ports, respectively. Assume δT = 2°C, δS = 10 ppt, and α = 0.00015. The value of this formula represents the similarity of the average temperature and salinity between the two ports. If two ports are in different ecological zones but have similar average temperatures and salinities, then they are more habitable for invasive organisms, and the probability of invasion is greater.

[0053] is defined as:

[0054]

[0055] Where v represents the ship, and v is the same for the same ship. w represents the proportion of ships that do not operate anti-pollution operations in this category. μ is the mortality rate of invasive organisms in the ship. Indicates the time the ship passes The proportion of organisms that die after an invasion. The value of w is estimated based on actual surveys and is 0.19 for container ships, 0.20 for car carriers, 0.30 for oil tankers, 0.31 for passenger ships, 0.42 for bulk carriers, 0.53 for general vessels, and 0.60 for other merchant ships. μ is a constant of 0.002. It can be found in shipping data. ij It represents the maritime policy control of all ships in the port. Adjusting the weight will reduce the invasion risk from port i to port j.

[0056] S1.3. Based on the invasion risk brought by ships between two ports, the ships in the shipping data are integrated to obtain the probability of invasion between the two ports.

[0057] Specifically, the above calculation only describes the invasion risk posed by a single ship between two ports. However, if a port is visited by many ships from around the world, the invasion risk will be much higher than if only a few ships pass through. If we consider each ship passing through the shipping data as a random event, then, by integrating all ships in the data and assuming that there are n ships passing through this port, we can obtain the probability of an invasion between two ports as:

[0058]

[0059] The probability in the product symbol represents the probability that no invasion occurred on vessel v between port i and port j. The entire product represents the probability that no invasion occurred on any vessel. Subtracting this value from 1 equals the probability that at least one invasion occurred. This determines the probability of an invasion under first-order dependence in the shipping data.

[0060] S2. Based on the probability of biological invasion between ports, a high-order biological invasion network is constructed in combination with shipping data;

[0061] S2.1. Based on the biological invasion probability between ports and the historical trajectory of ships in shipping data, the invasion probability of the trajectory is generated;

[0062] Specifically, formula (5) calculates the invasion probability between two ports. However, in reality, new invasions may occur between the two ports due to the ship's historical trajectory. For example, consider trajectory A->B->C, where ports B and C are in the same ecological zone, while ports A and B are in different ecological zones. If we only consider port B->C, there is no invasion risk. However, if we also consider port A, the ship may bring alien organisms from port A to port C. Therefore, we need to further consider the changes in the invasion probability along the trajectory.

[0063] Let's define the invasion probability for a given trajectory. Consider a ship with a trajectory A->B->C. After arriving at Port B from Port A, the number of organisms carried may decrease due to natural mortality or the ship's water changes. At this point, the ship will contain organisms from both Ports A and B, and will continue to carry these organisms to Port C. During this process, we need to consider how the organisms on the ship change with each port and how these organisms affect the invasion risk. For a long path AB...MN, the invasion probability can be defined as:

[0064]

[0065] Among them is the formula:

[0066]

[0067] cw represents the coefficient of water exchange, cw∈[0,1] is a constant, k represents the length of the trajectory, and represents the invasion probability of port N under the historical trajectory. After each water change, it will affect the organisms in the historical trajectory ((1-cw v ) k-1 p(M|AB...L)), and at each node passed, there will be natural death of organisms (ps), or the invasion probability will be reduced due to environmental influences (pr), or there will be an impact due to different biological regions (pe). Finally, when the path is short enough and there are only two ports, the invasion risk calculation of the first-order dependence in (5) is performed.

[0068] So we can already know how to calculate the invasion probability under a trajectory. Similarly, if there are multiple ships passing through such a trajectory, in order to integrate this data, each ship is considered as a random event, then:

[0069]

[0070] S2.2. Based on the shipping data, calculate the invasion probability of the trajectory and cut it to obtain high-order nodes;

[0071] Specifically, let's consider how to slice the shipping data sequence to obtain higher-order nodes. Simply calculate the intrusion probability of each trajectory until the intrusion probability remains constant. Each path in the trajectory can then be sliced ​​and treated as multiple higher-order nodes. For example, if there is a trajectory A->B->C->D->E->F, where the intrusion probabilities for A->B->C and A->B->C->D are the same, we can obtain the higher-order node {C|AB,B|A,C|B}. We can then continue calculating subsequent higher-order nodes starting from D.

[0072] S2.3. Using trajectories as edges, a high-order network is constructed based on high-order nodes to obtain a high-order network of biological invasion.

[0073] We can then use shipping data to calculate the intrusion probability of each trajectory and cut out all high-order nodes. We can then use the high-order nodes to form a high-order network to complete the graph network organization.

[0074] It is worth noting that the definition There is a "po ij The parameter of "Maritime Policy Control of Ships" will have an impact on the generation of high-order networks. When this parameter changes, the calculated invasion risk between certain ports will change, and the high-order nodes cut out will also change, causing the entire high-order network to change, that is, a different high-order network is obtained. Using this property, after constructing different maritime policies, different high-order networks are obtained. These different networks are used for comparison to obtain the similarities and differences between different maritime policies.

[0075] S3. Generate port invasion risk based on biological invasion high-order network and perform visual display.

[0076] S3.1. Calculate the probability of port invasion based on the invasion probabilities of edges and trajectories in the biological invasion high-order network to obtain the port invasion risk.

[0077] Now that we've segmented the shipping data into high-level nodes and organized the high-level network, we need to calculate the probability of a breach for each port to achieve better visualization, allowing for better interaction and selection of ports of interest.

[0078] Now for a high-order network, there are edges such as E(C|AB,D|BC), where this edge points to port D as the end point. If we extract all edges that point to port D as the end point, we can continue to treat each trajectory as a random event and calculate the port invasion probability. The formula is:

[0079]

[0080] In the actual process, it is found that if this formula is used directly for calculation, the invasion risk of all ports will be too high, which is not conducive to visual observation. Therefore, before calculating P, it is possible to consider reducing p first. ij , then we can sum and normalize the intrusion probabilities of all intrusions into port D to get p' ij , and then recalculate the formula to get the weight value of port D being invaded.

[0081] S3.2. Integrate the port intrusion risk in the high-order network and create dependency view, subgraph view and aggregation view based on the geographic view for visualization.

[0082] As a further preferred embodiment of the method, the step of creating a dependency view, a subgraph view, and an aggregate view based on the geographic view for visual display specifically includes:

[0083] The original data is the time series data of different ships passing through various ports at different times. It can be considered that each piece of original data is the time series data of a ship passing through different ports.

[0084] High-order nodes use different cutting methods to cut each time series data into different small pieces, or small sequences. Each small piece is a high-order node. For example, A->B->C->B is a small sequence cut out, which is also a high-order node.

[0085] A high-order network is the connection of these small sequences. There's an edge connecting the two high-order nodes before and after the cut. The network composed of all edges and high-order nodes is a high-order network.

[0086] Therefore, the dependency view displays the data for the aforementioned high-order nodes. The selected high-order nodes are displayed in the view. Next, all high-order nodes are mapped onto a two-dimensional space using the ForceAtlas2 method for visualization. The resulting image is the subgraph view. A high-order network can be considered a special graph network, so its properties can be leveraged to calculate the network's distinct communities. By calculating the weights within the communities, a view consisting of a rectangular map and a heat map is generated. This view is part of the subgraph view.

[0087] Displays the probability of invasion of ports based on a geographical view;

[0088] Specifically, refer to Figure 3 The geographic view shows the weighted distribution of all ports. The map can be interacted with using the mouse: drag left or right to pan the map, creating a circular display; drag up or down to pan the map; scroll to zoom in or out, especially zooming in to reveal ports that are hidden in densely populated areas; and click to obtain detailed information about a port.

[0089] Display the high-order nodes corresponding to the ports in the dependency view;

[0090] Specifically, refer to Figure 4 The circular node on the left represents a port. The color of the circular node corresponds to the color in the geographic view, indicating the port's biological invasion risk. Each edge represents a higher-order node, representing the biological invasion risk along that path. The thicker the edge and the darker the color, the higher the probability of biological invasion. The small rectangle in the middle represents the transition probability of the higher-order node. The darker the color, the higher the transition probability, meaning the higher the probability that the high-order node will transition to the next destination. A high-order node can have different destinations, so a high-order node can correspond to multiple biological invasion risks and different transition probabilities.

[0091] Display the entire high-level network through subgraph view;

[0092] Specifically, refer to Figure 5 , when a port is selected in the geographic view, the corresponding high-order nodes of the port will be visualized in the dependency view. There are two types of visualization results. The first is to visualize the high-order nodes starting from the port, and the second is to visualize the high-order nodes ending at the port. The choice of these two modes can be obtained by checking the "Trace Back" button in the dependency view in the e-interaction toolbar. At the same time, after selecting one of the ports, the high-order node sequence of this port will be highlighted and displayed in the geographic view and subgraph view. Figure 5In the geographic view, the highlighted ports are those in the Mediterranean region. Secondly, the darker the color of the high-order node in the subgraph view, the greater the risk of biological invasion of the corresponding high-order node.

[0093] The weight statistics of high-order network communities are displayed through aggregate views.

[0094] Specifically, if Figure 5 As shown in the lower right corner, the Aggregate View consists of a heat map and a bar chart. Each column represents a category, and each row represents a high-order node. If the number of high-order nodes is too large, the scroll bar on the right will display additional high-order nodes. To interact with this view, select a few high-order nodes in the Geographic View or Dependency View, and these high-order nodes will be highlighted in the Subgraph View. Click the Trace button in the interactive toolbar to perform a propagation exploration of the selected high-order nodes in the Subgraph View. This propagation can be forward or backward. For a node in the high-order network with directed edges, backward propagation finds the out-degree of the high-order node, while forward propagation finds the in-degree of the node. By searching for the forward or backward high-order nodes of the selected high-order nodes, both the Subgraph View and the Geographic View will be updated to display the nodes discovered through propagation. The Aggregate View calculates the impact of the high-order nodes initially used for propagation on each community. These communities are identified using the Louwan method. When calculating community weights, higher columns represent a greater contribution rate to the corresponding community, indicating a greater risk of biological invasion. The darker the squares in the statistical chart, the greater the contribution rate of the corresponding high-order node in the current community.

[0095] like Figure 2 As shown in FIG, a biological invasion visualization system for shipping data based on a high-order network includes:

[0096] A probability calculation module is used to obtain shipping data and generate the probability of biological invasion between ports based on the shipping data;

[0097] The network construction module constructs a high-order biological invasion network based on the probability of biological invasion between ports and combined with shipping data;

[0098] The display module generates port invasion risks based on the biological invasion high-order network and displays them visually.

[0099] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0100] A high-order network-based visualization device for biological invasions in maritime data:

[0101] at least one processor;

[0102] at least one memory for storing at least one program;

[0103] When the at least one program is executed by the at least one processor, the at least one processor implements the above-mentioned method for visualizing biological invasions of shipping data based on a high-order network.

[0104] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0105] A storage medium storing processor-executable instructions is characterized in that the processor-executable instructions, when executed by the processor, are used to implement the above-mentioned method for visualizing biological invasions of shipping data based on a high-order network.

[0106] The contents of the above method embodiments are all applicable to the present storage medium embodiment. The functions specifically implemented by the present storage medium embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0107] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for visualizing biological invasions in marine data based on high-order networks, characterized by: The following steps are involved: Obtain shipping data and generate the probability of biological invasion between ports based on shipping data; Based on the probability of biological invasion between ports, a high-order biological invasion network was constructed in combination with shipping data. Generate port invasion risk based on biological invasion high-order networks and perform visual display; The step of constructing a biological invasion high-order network based on the biological invasion probability between ports and combined with shipping data specifically includes: Based on the biological invasion probability between ports and the historical trajectory of ships in shipping data, the invasion probability of the trajectory is generated; According to the shipping data, the invasion probability of the trajectory is calculated in sequence and cut to obtain high-order nodes; Using trajectories as edges, a high-order network is constructed based on high-order nodes to obtain a high-order network of biological invasions; The step of generating port invasion risk based on the biological invasion high-order network and visually displaying it specifically includes: The invasion probability of the port is calculated based on the invasion probability of the edges and trajectories in the biological invasion high-order network, and the port invasion risk is obtained; Integrate port intrusion risks in high-order networks and create dependency views, subgraph views, and aggregation views based on geographic views for visualization; The creation of dependency views, subgraph views, and aggregate views based on the geographic view for visual display specifically includes: Displays the probability of invasion of ports based on a geographical view; Display the high-order nodes corresponding to the ports in the dependency view; Display the entire high-level network through subgraph view; The weight statistics of high-order network communities are displayed through aggregate views.

2. The method for visualizing biological invasions of shipping data based on high-order networks according to claim 1 is characterized in that: The step of obtaining shipping data and generating the probability of biological invasion between ports based on the shipping data specifically includes: Acquire shipping data, including time series data of different ships arriving at different ports and ecological data of the ports; Based on the time series data of different ships arriving at different ports and the ecological data of the ports, the invasion risk brought by ships between two ports is defined; Based on the invasion risk brought by ships between two ports, the ships in the shipping data are integrated to obtain the probability of invasion between the two ports.

3. The method for visualizing biological invasions of shipping data based on high-order networks according to claim 2 is characterized in that: The ecological data of the port include average temperature, salinity, sea area, ecosphere and adjacent ecological areas.

4. The method for visualizing biological invasions of shipping data based on high-order networks according to claim 3 is characterized in that: The probability formula of invasion between the two ports is expressed as follows: In the above formula, the p ij represents the probability of invasion between port i and port j, the pr ij Indicates whether port i and port j are in different ecological areas. ij represents the similarity of average temperature and salinity between port i and port j, represents the invasion risk brought by a ship between port i and port j, where i and j represent different ports respectively, and v represents a ship.

5. A marine data biological invasion visualization system based on a high-order network, characterized by: The method for visualizing biological invasions of shipping data based on a high-order network as claimed in claim 1 comprises: A probability calculation module is used to obtain shipping data and generate the probability of biological invasion between ports based on the shipping data; The network construction module constructs a high-order biological invasion network based on the probability of biological invasion between ports and combined with shipping data; The display module generates port invasion risks based on the biological invasion high-order network and displays them visually.

Images