Maritime leisure fishery management system and method based on multi-source data

Abstract

The invention relates to the technical field of marine fishery, in particular to an offshore leisure fishery management system and method based on multi-source data. The method comprises the following steps: constructing a mapping layout of an overview management area based on a multi-level fishery restriction dimension, mapping multi-dimensional supervision information of offshore leisure fishery to the mapping layout through significance comparison, and then displaying a normal management area and a violation management area according to positive-negative invasion and occupation analysis; the method comprises the following steps of: performing distribution splicing of space-time trend change on corresponding past multi-source fishery data by taking identification of potential abnormal trend behaviors of trace strongholds experiencing seaborne leisure fishery activities on a violation management area as a mode premise, and tracing and positioning fishery points of this type and different types by analyzing a fishery space-time trend star tip map. According to the invention, different levels of management measures can be taken for offshore recreational fishery activities according to multi-source fishery data, the reliability and scientificity of offshore recreational fishery management are improved, the offshore recreational fishery industry is standardized, and sustainable development of fishery resources is promoted.

Description

Technical Field

[0001] This invention relates to the field of marine fisheries technology, and in particular to a marine recreational fisheries management system and method based on multi-source data. Background Technology

[0002] Recreational fishing is a new type of marine economic activity that has emerged in recent years, integrating sightseeing, fishing, and entertainment, and has good economic and social benefits. With the public's increasing pursuit of a higher quality of leisure life, recreational fishing has developed rapidly, with a continuous rise in the number of participants, and related fishing vessels, gear, and auxiliary services have become increasingly diversified. However, the development of this industry still faces problems such as increased pressure on fishery resource protection, the strong concealment of illegal fishing activities, and increased safety risks. Traditional management methods for recreational fishing mainly rely on on-site monitoring, user reporting, or regular patrols by fishery administration personnel, lacking multi-source, multi-dimensional, and multi-temporal data integration and analysis, which increases the difficulty of managing recreational fishing. At the same time, traditional management methods cannot detect in real time the occurrence of violations in localized areas, and lack effective tracking and compliance determination of the activities of anglers and boats, which may lead to management loopholes and irrationalities, reducing the standardization of recreational fishing and failing to meet the current needs for scientific and sustainable development management. To address these issues, it is necessary to develop a marine recreational fisheries management system and method that can effectively improve regulatory efficiency and promote the sustainable use of resources based on multi-source data analysis. Summary of the Invention

[0003] This invention overcomes the shortcomings of the prior art and provides a marine recreational fishery management system and method based on multi-source data.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] The first aspect of this invention provides a method for managing marine recreational fisheries based on multi-source data, comprising the following steps:

[0006] S102: Obtain multi-dimensional regulatory information on marine recreational fisheries, construct an overview map of the management area based on the multi-level fisheries constraints that meet management objectives, map the multi-dimensional regulatory information to the map, and then display the normal management area and the non-compliant management area based on positive-negative encroachment analysis;

[0007] S104: Obtain traces of marine recreational fishing activities in the overall management area and corresponding past multi-source fishery data. Identify potential abnormal trend behaviors of traces of fishery activities in violation management areas as a model premise. Then, splice the distribution of spatiotemporal trend changes of past multi-source fishery data to obtain a fishery spatiotemporal trend star map. By analyzing the fishery spatiotemporal trend star map, trace and locate the fishery points of this type and the fishery points of different types.

[0008] S106: For anomalous fishing sites within the non-compliant management area, use the causal relationship of pending multi-source fishery data samples to trace the fishery-induced accountability chain set that triggers the regulatory imbalance gap in the non-compliant management area. Based on the number of fishery-induced accountability chain sets, identify and take different management measures for the main users of recreational fishery.

[0009] S108: If unusual fishing spots frequently appear in areas under non-compliant management, the change pattern of the level of fishing activities within adjacent time steps will be calculated based on multi-source data. The change pattern will be analyzed according to the management rules and regulations system for marine recreational fisheries to adjust the supervision intensity in real time and generate management optimization plans.

[0010] More specifically, step S102 includes the following steps:

[0011] Obtain the overall management area and management objectives for marine recreational fisheries, and extract the multi-level fisheries constraint dimensions and the compliance scope of each multi-level fisheries constraint dimension from the management objectives.

[0012] Based on the multi-layered fisheries constraints, a mapping map of the overall management area with different regulatory channels is constructed, and multi-dimensional regulatory information on marine recreational fishery types and the overall management area when each type of marine recreational fishery is engaged in within a preset time sequence segment is simultaneously obtained.

[0013] The K-Medians algorithm is introduced to divide the mapping map into several sub-mapping maps based on the type of marine recreational fisheries. Using the sub-mapping maps as salient centers, each multidimensional regulatory information is assigned to the corresponding salient center cluster in the K-Medians algorithm with a Manhattan distance less than the preset Manhattan distance. The median of all samples in each cluster is calculated in each multi-level fisheries constraint dimension. The salient centers are updated and reorganized based on the median, and the salient aggregation results are output.

[0014] Based on the significant aggregation results, we characterize the significant local topography that attracts the attention of each sub-mapping map for each multidimensional regulatory information. We then establish a fusion mask according to the significant amplitude texture indicated by the significant local topography and generate a significant fusion weight model diagram for each multidimensional regulatory information.

[0015] Construct a comparative mapping domain, extract the current regulatory record values ​​of multi-dimensional regulatory information located in the multi-level fishery constraint dimensions, formulate the regulatory critical threshold for each regulatory channel based on compliance scope constraints, and map the multi-dimensional regulatory information to the mapping map in the comparative mapping domain based on the saliency fusion weight model graph. During the mapping process, determine whether the current regulatory record value exceeds the regulatory critical threshold and generate the positive-negative comparative mapping encroachment rate of each sub-mapping map.

[0016] If the positive-negative contrast mapping encroachment rate is less than the preset encroachment rate threshold, the marine recreational fishing area corresponding to the sub-mapping map is marked as a normal management area; if the positive-negative contrast mapping encroachment rate is greater than the preset encroachment rate threshold, the marine recreational fishing area corresponding to the sub-mapping map is marked as an illegal management area.

[0017] More specifically, step S104 includes the following steps:

[0018] The system acquires traces of marine recreational fishing activities within a preset time-series segment of the overall management area, and also acquires several past multi-source fisheries data for each trace.

[0019] Each trace point is defined as a high-dimensional focus of interest. The Dijkstra algorithm is introduced to calculate the shortest spatial distance that affects the overview management area when each high-dimensional focus of interest generates each past multi-source fishery data during marine recreational fishing activities. Based on the shortest spatial distance, a focus deformation function is assigned to each high-dimensional focus of interest. This function is used to scale or stretch the dependent position of the past multi-source fishery data around the focus, thereby obtaining the high-dimensional response layout structure of the trace points located in the overview management area where past multi-source fishery data occurs.

[0020] To identify regulatory imbalance gaps in areas of illegal management, a spatiotemporal journey tree of past multi-source fishery data is established based on a high-dimensional response layout structure. The spatiotemporal journey tree is used to search and identify potential abnormal trend behaviors that form regulatory imbalance gaps in trace points, and to obtain multi-source radiation pattern access conclusions. The spatiotemporal distribution of past multi-source fishery data is spliced ​​with reference to the multi-source radiation pattern access conclusions to generate a fishery spatiotemporal trend star map for each trace point.

[0021] If there is no star-peak trend arm on the fishery spatiotemporal trend star-peak map with a star-peak area value greater than the star-peak area threshold, then mark the trace point as a fishery point of this type.

[0022] If there is at least one star-peak trend arm on the fishery spatiotemporal trend star-peak map with a star-peak area value greater than the star-peak area threshold, then the trace point is marked as an outlier fishery point.

[0023] More specifically, the process of identifying regulatory imbalance gaps in areas of non-compliance involves establishing a spatiotemporal journey tree based on a high-dimensional response layout structure to capture trace points from multiple sources of fisheries data. This spatiotemporal journey tree is then used to identify potential abnormal trend behaviors that could lead to regulatory imbalance gaps at these trace points, resulting in multi-source radiation pattern access conclusions. These conclusions are then used to spatiotemporally stitch together the past multi-source fisheries data to generate a spatiotemporal trend star map for each trace point. The specific steps include:

[0024] The high-dimensional response layout structure is analyzed and a hierarchical segmentation dimension is set. On the hierarchical segmentation dimension, several past multi-source fishery data of each trace point are divided into spatiotemporal order. The central balance fulcrum of different multi-source spatiotemporal levels is obtained. The left subtree and right subtree are constructed by recursively using past multi-source fishery data from the central balance fulcrum, forming a spatiotemporal journey tree of past multi-source fishery data of trace points.

[0025] Based on the preset star tip area threshold of the regulatory imbalance gap, starting from the root node of the spatiotemporal journey tree, the left subtree is recursively searched according to the current dimension of the regulatory imbalance gap and traversed down to the leaf node. If the distance between the hyperplane centered on the current leaf node and the query point is less than the preset minimum distance, the leaf node of the right subtree is visited, and the multi-source radiation pattern access conclusion of each trace point leading to the formation of the regulatory imbalance gap is output.

[0026] By accessing the conclusions of the multi-source radiation mode, we obtain the radiation concentric rings of each multi-source spatiotemporal level. Using the high-dimensional focus of attention as the concentric point, we adaptively distribute and stitch together the past multi-source fishery data of each multi-source spatiotemporal level on the radiation concentric ring according to the node degree followed by the spatiotemporal order, and finally generate the spatiotemporal trend star map of fishery for each trace point.

[0027] More specifically, step S106 includes the following steps:

[0028] Multi-source fishery data corresponding to non-compliant fishery points in the management area are independently extracted from the management log and defined as pending multi-source fishery data samples. Based on big data, the established logic is obtained when each pending multi-source fishery data sample triggers a regulatory imbalance event.

[0029] Construct a subject matrix of undetermined multi-source fishery data samples, and use non-circular constraints that follow a predetermined logical structure to solve the underlying objective function of the subject matrix causally to obtain a subject causal relationship diagram. Combine the regulatory imbalance gap with the subject causal relationship diagram to identify the causal relationship between subjects and the potential responsibility contribution intensity that induces key target events.

[0030] A causal reasoning algorithm is introduced, and the influence path of the causal relationship diagram between key target events and subjects is back-tracked and identified based on the intensity of potential responsibility contribution in the causal reasoning algorithm, so as to obtain the fishery-induced accountability chain set of anomalous fishery points for regulatory imbalance gaps.

[0031] If the number of multi-source fishery accountability chains in the fishery-induced accountability chain set exceeds the preset threshold, then warning measures will be taken against the main users of the recreational fishery at the aberrant fishery site, and the monitoring facilities along the way will be controlled to conduct observation and control management of the aberrant fishery site.

[0032] If the number of multi-source fishery accountability chains in the fishery-induced accountability chain set is equal to the preset number threshold, then restrictive and control measures will be taken against the main users of recreational fishery at the aberration fishery site, and the monitoring facilities along the way will be controlled to carry out follow-up intervention management of this type of fishery site.

[0033] If the number of multi-source fishery accountability chains in the fishery-induced accountability chain set exceeds a preset threshold, a regulatory warning will be issued to the main users of the recreational fishery at the aberration fishing site, and the monitoring facilities along the route will be controlled to implement lock-in mandatory management of this type of fishery site.

[0034] More specifically, the construction of the subject matrix of the undetermined multi-source fisheries data samples, the causal solution of the underlying objective function of the subject matrix using non-circular constraints following a predetermined logical order, the obtaining of the subject causal relationship graph, and the identification of the potential responsibility contribution intensity of the causal links between subjects leading to the induction of key target events by combining the regulatory imbalance gap with the subject causal relationship graph, specifically includes the following steps:

[0035] A subject matrix is ​​constructed using undetermined multi-source fishery data samples. Each data sample in the subject matrix has a causal objective function deployed on a local server. At the same time, a non-cyclic constraint is constructed based on the rule dependencies recorded by the established logical structure.

[0036] By combining the underlying objective function and non-cyclic constraints, the main causal constraint equation is obtained. The augmented Lagrange method is then introduced to solve the main causal constraint equation by alternating updates to approximate the optimal solution until the non-cyclic constraints converge to 0, thus obtaining the main causal relationship graph of the undetermined multi-source fishery data samples. Among them, the non-zero terms in the event causal relationship graph represent the directional causal relationship between the undetermined multi-source fishery data samples.

[0037] Define the regulatory imbalance gap as the key target event, obtain the loss drift parameter of the regulatory imbalance gap, calculate the implicit correlation degree between the loss drift parameter and each directed causal edge in the subject causal relationship graph, if the implicit correlation degree is greater than the preset implicit correlation degree, then assign a potential responsibility contribution intensity to the directed causal edge based on the implicit correlation degree.

[0038] More specifically, step S108 includes the following steps:

[0039] If the frequency of occurrence of non-standard fishing sites in the non-compliant management area is greater than the preset frequency, then the management rules and regulations currently followed by marine recreational fishing will be obtained.

[0040] By extracting the assessment coefficients of regulatory risks in response to changes in the scale of fishery activities at different levels from the management rules and regulations system, and the scale of management mobilization that can be implemented to address various regulatory risks, these coefficients are set as the first time step, and the previous time step is set as the second time step.

[0041] A multi-source fishery dataset of outlier fishery points that appear in the illegal management area within a first time step is defined as the first multi-source data queue, and a multi-source fishery dataset of outlier fishery points that appear in the illegal management area within a second time step is defined as the second multi-source data queue.

[0042] A state incremental learning model is constructed, and the first multi-source data queue and the second multi-source data queue are imported into the state incremental learning model to perform incremental learning calculations, so as to obtain the online change law of the fishery activity level within the first time step and the historical change law of the second time step.

[0043] A Bayesian inference network is introduced to perform Bayesian inference on the shift of fishery activity magnitude from the previous time step to the current time step based on the online and historical change patterns. This yields the real-time joint probability distribution of fishery activity magnitude changes. Based on the real-time joint probability distribution, a Lagrange mutual information curve for assessing regulatory risks during interpolation magnitude changes is fitted.

[0044] An algebraic equation solver is created to map the regulatory risk assessment coefficient to the management mobilization scale. The algebraic equation of the Lagrange interpolation curve is solved by the curve equation solver to obtain a series of algebraic equation solutions for the management mobilization scale.

[0045] The management mobilization scale is determined by solving a series of algebraic equations, and the regulatory intensity of the non-compliant management areas is adjusted in real time based on the management mobilization scale, thus outputting a management optimization plan.

[0046] A second aspect of the present invention provides a marine recreational fishery management system based on multi-source data. The marine recreational fishery management system includes a memory and a processor. The memory stores a marine recreational fishery management method program based on multi-source data. When the marine recreational fishery management method program is executed by the processor, it implements any of the steps of the marine recreational fishery management method described in the present invention.

[0047] This invention addresses the technical deficiencies in the prior art, and its beneficial technical effects are as follows:

[0048] Multidimensional regulatory information on marine recreational fisheries is acquired. Based on the multi-layered fisheries constraints that meet management objectives, a mapping map of the overall management area is constructed. The saliency of the multidimensional regulatory information is compared and mapped onto the mapping map. Subsequently, positive-negative encroachment analysis is used to display normal management areas and non-management areas. Traces of marine recreational fisheries activities in the overall management area and corresponding past multi-source fisheries data are acquired. Using the identification of potential abnormal trend behaviors of traces of these traces in non-management areas as a model premise, the distribution of past multi-source fisheries data is spliced ​​to obtain a fisheries spatiotemporal trend star map. By analyzing the spatiotemporal trends of fisheries... The invention uses a star-shaped data map to trace and locate fishing sites of the same and different types. For different fishing sites within areas of illegal management, it uses causal relationships from undetermined multi-source fisheries data samples to trace the fisheries-induced accountability chains that trigger regulatory imbalances in the illegal management area. Based on the number of these chains, different management measures are taken for the main users of recreational fishing. If different fishing sites frequently appear in areas of illegal management, the invention calculates the changing patterns of fisheries activity levels within adjacent time steps based on multi-source data. These patterns are then analyzed according to the management regulations for marine recreational fishing to adjust regulatory intensity in real time and generate optimized management plans. This invention can capture and display illegal areas of marine recreational fishing activities based on multi-source fisheries data, accurately trace and locate fishing sites causing abnormal violations, and take different levels of management measures for the main users of these violations. This improves the reliability and scientific nature of marine recreational fishing management, optimizes regulatory efficiency and quality, improves the fisheries ecology, standardizes the marine recreational fishing industry, and promotes the sustainable development of fisheries resources. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.

[0050] Figure 1 A flowchart of the first method for marine recreational fisheries management based on multi-source data is shown;

[0051] Figure 2 A flowchart of a second method for marine recreational fisheries management based on multi-source data is shown.

[0052] Figure 3 A system framework diagram of a marine recreational fisheries management system based on multi-source data is shown. Detailed Implementation

[0053] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0054] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0055] The first aspect of this invention provides a method for managing marine recreational fisheries based on multi-source data, such as... Figure 1 As shown, it includes the following steps:

[0056] S102: Obtain multi-dimensional regulatory information on marine recreational fisheries, construct an overview map of the management area based on the multi-level fisheries constraints that meet management objectives, map the multi-dimensional regulatory information to the map, and then display the normal management area and the non-compliant management area based on positive-negative encroachment analysis;

[0057] S104: Obtain traces of marine recreational fishing activities in the overall management area and corresponding past multi-source fishery data. Identify potential abnormal trend behaviors of traces of fishery activities in violation management areas as a model premise. Then, splice the distribution of spatiotemporal trend changes of past multi-source fishery data to obtain a fishery spatiotemporal trend star map. By analyzing the fishery spatiotemporal trend star map, trace and locate the fishery points of this type and the fishery points of different types.

[0058] S106: For anomalous fishing sites within the non-compliant management area, use the causal relationship of pending multi-source fishery data samples to trace the fishery-induced accountability chain set that triggers the regulatory imbalance gap in the non-compliant management area. Based on the number of fishery-induced accountability chain sets, identify and take different management measures for the main users of recreational fishery.

[0059] S108: If unusual fishing spots frequently appear in areas under non-compliant management, the change pattern of the level of fishing activities within adjacent time steps will be calculated based on multi-source data. The change pattern will be analyzed according to the management rules and regulations system for marine recreational fisheries to adjust the supervision intensity in real time and generate management optimization plans.

[0060] More specifically, step S102 includes the following steps:

[0061] Obtain the overall management area and management objectives for marine recreational fisheries, and extract the multi-level fisheries constraint dimensions and the compliance scope of each multi-level fisheries constraint dimension from the management objectives.

[0062] Based on the multi-layered fisheries constraints, a mapping map of the overall management area with different regulatory channels is constructed, and multi-dimensional regulatory information on marine recreational fishery types and the overall management area when each type of marine recreational fishery is engaged in within a preset time sequence segment is simultaneously obtained.

[0063] The K-Medians algorithm is introduced to divide the mapping map into several sub-mapping maps based on the type of marine recreational fisheries. Using the sub-mapping maps as salient centers, each multidimensional regulatory information is assigned to the corresponding salient center cluster in the K-Medians algorithm with a Manhattan distance less than the preset Manhattan distance. The median of all samples in each cluster is calculated in each multi-level fisheries constraint dimension. The salient centers are updated and reorganized based on the median, and the salient aggregation results are output.

[0064] Based on the significant aggregation results, we characterize the significant local topography that attracts the attention of each sub-mapping map for each multidimensional regulatory information. We then establish a fusion mask according to the significant amplitude texture indicated by the significant local topography and generate a significant fusion weight model diagram for each multidimensional regulatory information.

[0065] Construct a comparative mapping domain, extract the current regulatory record values ​​of multi-dimensional regulatory information located in the multi-level fishery constraint dimensions, formulate the regulatory critical threshold for each regulatory channel based on compliance scope constraints, and map the multi-dimensional regulatory information to the mapping map in the comparative mapping domain based on the saliency fusion weight model graph. During the mapping process, determine whether the current regulatory record value exceeds the regulatory critical threshold and generate the positive-negative comparative mapping encroachment rate of each sub-mapping map.

[0066] If the positive-negative contrast mapping encroachment rate is less than the preset encroachment rate threshold, the marine recreational fishing area corresponding to the sub-mapping map is marked as a normal management area; if the positive-negative contrast mapping encroachment rate is greater than the preset encroachment rate threshold, the marine recreational fishing area corresponding to the sub-mapping map is marked as an illegal management area.

[0067] It should be noted that the multi-layered constraints on fisheries include ecological protection, regional legitimacy, activity behavior, and user personnel. Types of marine recreational fisheries include recreational fishing, sightseeing and experiential fishing, participatory operations, comprehensive entertainment, and education and science popularization. Multi-dimensional regulatory information includes ecologically sensitive regulatory information, vessel and equipment use regulatory information, spatial geographic service regulatory information, and activity behavior regulatory information.

[0068] It should be noted that marine recreational fishing activities are characterized by their diversity, wide scope, and dispersed distribution, making regional management of these activities significantly challenging. It is difficult to detect in real time whether user behavior violates reasonable regulations across different local areas within the management region, thus hindering the identification of areas with illegal fishing activities and greatly reducing the effectiveness of management guidance for anomalies in marine recreational fishing. To address this, this method first constructs a mapping map of the overall management region, referencing multi-layered fisheries constraints, with different regulatory channels. This mapping map can display corresponding regulatory information across different fisheries constraint dimensions; for example, in the ecological protection dimension, it can show whether the ecological sensitivity structure of a certain area has been damaged by fishing users. By precisely aggregating the multi-dimensional regulatory information collected uniformly within the management region according to the affiliation of different local areas, the visualization of regulatory information across different regulatory channels for different local areas becomes more detailed and complete, providing a reliable basis for subsequent regional violation management. Then, based on the local attribution definition of each multidimensional regulatory information in the significant aggregation results, it is compared and mapped onto the mapping map. This allows for a multidimensional presentation of the overall regulatory standardization of marine recreational fishing activities within the management area in the form of a planar macroscopic map. Each regulatory standard channel on the mapping map is constrained by a standardization judgment value, i.e., a standardization threshold, based on the compliance range of each multi-level fishery constraint dimension. This enables refined violation identification, achieving abnormal guidance and capture of violations of management regulations at different fishing locations within the management area. This significantly improves the accuracy and perception guidance capabilities of marine recreational fishing violation management, effectively solving the problem of high management difficulty that traditional methods rely on on-site monitoring, user reporting, or regular patrols by fishery administration personnel.

[0069] It should be noted that if the Manhattan distance is less than the preset Manhattan distance, it indicates a high degree of affiliation between the multidimensional regulatory information and the regulatory category classification of a certain sub-mapping area. Therefore, it is assigned to the cluster containing the corresponding saliency center of that sub-mapping area. This step effectively ensures that multidimensional information samples are classified into the most similar regional clusters, improving the detail, fidelity, and accuracy of the visualization of regulatory information on different regulatory channels in different local areas. Updating and reorganizing saliency centers based on the median allows the expression of multidimensional regulatory information to more closely resemble the true distribution of the fishery constraints specified within the area, ensuring the visualization accuracy and credibility of global information and avoiding omissions or misjudgments of illegal capture. The saliency local topology depicts the regions with the most multidimensional regulatory information in the sub-mapping area, used to guide fusion. The fusion mask assigns different fusion weights to each sub-mapping area according to the saliency of the multidimensional regulatory information, ultimately generating a saliency fusion weight model map for different regional locations. During the mapping process, if the current regulatory record value does not exceed the regulatory threshold, it means that the multidimensional regulatory information is compliant in this area. Therefore, a positive contrast mapping value is added to the area where the multidimensional regulatory information is located on the sub-mapping map. If the current regulatory record value does not exceed the regulatory threshold, it means that the multidimensional regulatory information has compliant behavior in this area. Therefore, a negative contrast mapping value is added to the area where the multidimensional regulatory information is located on the sub-mapping map. The ratio of the positive contrast mapping value to the negative contrast mapping value is the positive-negative contrast mapping encroachment rate of each sub-mapping map. This positive-negative contrast mapping encroachment rate represents the degree of regulatory compliance of different sub-mapping map areas after experiencing marine recreational fishing activities.

[0070] More specifically, step S104 includes the following steps:

[0071] The system acquires traces of marine recreational fishing activities within a preset time-series segment of the overall management area, and also acquires several past multi-source fisheries data for each trace.

[0072] Each trace point is defined as a high-dimensional focus of interest. The Dijkstra algorithm is introduced to calculate the shortest spatial distance that affects the overview management area when each high-dimensional focus of interest generates each past multi-source fishery data during marine recreational fishing activities. Based on the shortest spatial distance, a focus deformation function is assigned to each high-dimensional focus of interest. This function is used to scale or stretch the dependent position of the past multi-source fishery data around the focus, thereby obtaining the high-dimensional response layout structure of the trace points located in the overview management area where past multi-source fishery data occurs.

[0073] To identify regulatory imbalance gaps in areas of illegal management, a spatiotemporal journey tree of past multi-source fishery data is established based on a high-dimensional response layout structure. The spatiotemporal journey tree is used to search and identify potential abnormal trend behaviors that form regulatory imbalance gaps in trace points, and to obtain multi-source radiation pattern access conclusions. The spatiotemporal distribution of past multi-source fishery data is spliced ​​with reference to the multi-source radiation pattern access conclusions to generate a fishery spatiotemporal trend star map for each trace point.

[0074] If there is no star-peak trend arm on the fishery spatiotemporal trend star-peak map with a star-peak area value greater than the star-peak area threshold, then mark the trace point as a fishery point of this type.

[0075] If there is at least one star-peak trend arm on the fishery spatiotemporal trend star-peak map with a star-peak area value greater than the star-peak area threshold, then the trace point is marked as an outlier fishery point.

[0076] It should be noted that due to the high density of users in marine recreational fishing during certain periods, traditional management methods cannot quickly and accurately track and locate the offending users after regional violations occur. This results in blindness and lag in the source tracing management of marine recreational fishing, indicating a weak link in violation management. To address this, this method collects the locations of marine recreational fishing projects that have occurred within the overall management area (trace points) and historical multi-source data generated at each location (past multi-source fishery data). Then, it constructs a spatiotemporal journey tree for different trace points and different past multi-source fishery data. This spatiotemporal journey tree traces the mileage of marine recreational fishing activities at trace points in time and space as they experienced past multi-source fishery data, providing a concrete expression of the trend behavior patterns of trace points. Through this spatiotemporal journey tree, it is possible to retrieve and identify radiation patterns from different data sources, thereby exploring the potential abnormal trend behaviors of the main users at each trace point when they experienced multi-source fishery data, and obtaining multi-source radiation pattern access conclusions. These multi-source radiation pattern access conclusions provide a reliable framework for tracking trend changes in multi-source fishery data across different behavioral indicators. Subsequently, referencing the conclusions of the multi-source radiation model access, the past multi-source fisheries data were spatiotemporally distributed and stitched together to generate a spatiotemporal trend star map of fisheries for each trace site. This spatiotemporal trend star map of fisheries expresses the trend changes of various fisheries project indicators with different regions and time periods. It is a high-dimensional dynamic display structure. The star-point trend arm from the center to the side length represents the trend vector of different indicators. Therefore, the star-point area reflects the standardization and rationality of the activities of the subject users in carrying out fisheries projects within the trace site. If there are no star-peak trend arms with a star-peak area value greater than the star-peak area threshold on the fisheries spatiotemporal trend star-peak map, it indicates that the marine recreational fishing activities of the fishermen within the trace point are not the main trend cause of the regulatory imbalance gap in the non-compliant management area, representing that the behavior of the main user is compliant. Therefore, the trace point is marked as this type of fishery point. If there are at least one star-peak trend arm with a star-peak area value greater than the star-peak area threshold on the fisheries spatiotemporal trend star-peak map, it indicates that the marine recreational fishing activities of the fishermen have caused a considerable regulatory imbalance gap in the non-compliant management area. Tracing this as the main trend cause, the trace point is marked as an outlier fishery point.

[0077] It should be noted that regulatory imbalance gaps refer to non-compliance deficiencies in a specific fisheries constraint dimension within the non-compliant management area, such as environmental damage in ecologically sensitive areas within the ecological protection dimension. The Dijkstra algorithm is used to calculate the shortest spatial distance between each past multi-source fisheries data point generated by each high-dimensional focus of attention engaging in recreational fishing activities and the overall management area. This shortest spatial distance quantifies the degree of correlation between each past multi-source fisheries data point's impact on the overall management area and each high-dimensional focus of attention, avoiding bias from a single-point perspective and achieving a tight semantic organization of multiple trend sources. The focus deformation function assigned to each piece of past multi-source fisheries data influences the position of surrounding past multi-source fisheries data by the position of each high-dimensional focus, highlighting each high-dimensional focus and its neighborhood. This scales or stretches the dependent positions of the surrounding past multi-source fisheries data, compressing the trend expression of the peripheral multi-source data, avoiding trend information obscuring, and achieving a graph layout method that simultaneously presents multiple traceability focus points (trace points) and their surrounding contextual trend information (trend behaviors contained in multi-source information). This high-dimensional response layout structure makes the subsequent spatiotemporal journey tree architecture construction and potential abnormal trend behavior identification more stable and accurate. This method can quickly and efficiently track and locate anomalies that induce regulatory imbalances within the management area, achieving a pinpoint source tracing effect for violations. This makes the management of violations in marine recreational fisheries more accurate and targeted, reaching the management and crackdown level for eliminating fisheries violations.

[0078] More specifically, the process of identifying regulatory imbalance gaps in areas of non-compliance involves establishing a spatiotemporal journey tree based on a high-dimensional response layout structure to capture trace points from multiple sources of fisheries data. This spatiotemporal journey tree is then used to identify potential abnormal trend behaviors that could lead to regulatory imbalance gaps at these trace points, resulting in multi-source radiation pattern access conclusions. These conclusions are then used to spatiotemporally stitch together the past multi-source fisheries data to generate a spatiotemporal trend star map for each trace point. The specific steps include:

[0079] The high-dimensional response layout structure is analyzed and a hierarchical segmentation dimension is set. On the hierarchical segmentation dimension, several past multi-source fishery data of each trace point are divided into spatiotemporal order. The central balance fulcrum of different multi-source spatiotemporal levels is obtained. The left subtree and right subtree are constructed by recursively using past multi-source fishery data from the central balance fulcrum, forming a spatiotemporal journey tree of past multi-source fishery data of trace points.

[0080] Based on the preset star tip area threshold of the regulatory imbalance gap, starting from the root node of the spatiotemporal journey tree, the left subtree is recursively searched according to the current dimension of the regulatory imbalance gap and traversed down to the leaf node. If the distance between the hyperplane centered on the current leaf node and the query point is less than the preset minimum distance, the leaf node of the right subtree is visited, and the multi-source radiation pattern access conclusion of each trace point leading to the formation of the regulatory imbalance gap is output.

[0081] By accessing the conclusions of the multi-source radiation mode, we obtain the radiation concentric rings of each multi-source spatiotemporal level. Using the high-dimensional focus of attention as the concentric point, we adaptively distribute and stitch together the past multi-source fishery data of each multi-source spatiotemporal level on the radiation concentric ring according to the node degree followed by the spatiotemporal order, and finally generate the spatiotemporal trend star map of fishery for each trace point.

[0082] It should be noted that, regarding the construction steps of the spatiotemporal journey tree, this method divides several past multi-source fisheries data of each trace point into spatiotemporal order according to hierarchical segmentation dimensions based on a high-dimensional response layout structure. This effectively avoids bias in trend behavior identification towards a single dimension, such as only identifying abnormal behavior in fishing vessel navigation trajectories, ensuring the balanced identification of the tree structure and reasonable multi-dimensional allocation. Recursively constructing the left and right subtrees from the central balancing pivot point based on past multi-source fisheries data maximizes the balance of data traversal across different dimensional paths, significantly improving retrieval and identification accuracy. The multi-source radiation pattern access conclusions reveal the magnitude, angle, and direction of the trend radiation induced by trace points to form regulatory imbalance gaps, reflecting the severity of changes in potential abnormal trend behaviors, and providing a unique characterization of abnormal fisheries trend behaviors at trace points.

[0083] More specifically, step S106, as follows: Figure 2 As shown, the specific steps include:

[0084] S202: Independently extract multi-source fishery data corresponding to non-compliant fishery points in the non-compliant management area through management logs, define them as pending multi-source fishery data samples, and obtain the established logic when regulatory imbalance events are triggered by each pending multi-source fishery data sample based on big data.

[0085] S204: Construct the subject matrix of the undetermined multi-source fishery data samples, use the non-circular constraint following the established logical order to solve the underlying objective function of the subject matrix causally, obtain the subject causal relationship diagram, and combine the regulatory imbalance gap with the subject causal relationship diagram to identify the causal relationship between subjects that leads to the potential responsibility contribution intensity of key target events.

[0086] S206: Introduce a causal reasoning algorithm. Based on the intensity of potential responsibility contribution, backtrack and identify the influence path of the causal relationship diagram between key target events and the subject in the causal reasoning algorithm to obtain the fishery-induced accountability chain set of anomalous fishery points for regulatory imbalance gaps.

[0087] S208: If the number of multi-source fishery accountability chains in the fishery-induced accountability chain set exceeds the preset threshold, then warning measures will be taken for the main users of recreational fishery at the aberrant fishery site, and the monitoring facilities along the way will be controlled to conduct observation and control management of the aberrant fishery site.

[0088] S210: If the number of multi-source fishery accountability chains in the fishery-induced accountability chain set is equal to the preset number threshold, then restrictive and control measures shall be taken for the main users of recreational fishery at the aberrant fishery site, and the monitoring facilities along the way shall be controlled to carry out follow-up intervention management of this type of fishery site.

[0089] S212: If the number of multi-source fishery accountability chains in the fishery-induced accountability chain set exceeds the preset threshold, a regulatory warning will be issued to the main users of the recreational fishery at the aberration fishery site, and the monitoring facilities along the route will be controlled to implement lock-in mandatory management of this type of fishery site.

[0090] It should be noted that the regulatory imbalance gaps generated by different fishery users vary. For example, excessive fixed-point fishing by a certain fishery user may lead to a significant ecological imbalance in the non-managed area. Traditional methods may have errors in the responsibility management decisions for different fishery users, potentially resulting in an imbalance in the management of marine recreational fisheries. To address this, this method first uses a predetermined logical framework as a non-circular constraint to explore the causal relationships when uncertain multi-source fishery data samples trigger regulatory imbalance events. This allows for the creation of a causal relationship diagram showing the main actors causing violations in abnormal management areas due to outlier fishery points located on different fishery constraint dimensions. Subsequently, based on the implicit correlation between this causal relationship diagram and the regulatory imbalance gap, the responsibility intensity (potential responsibility contribution intensity) of outlier fishery points on different fishery constraint dimensions is clarified. This further allows for the backtracking and identification of the impact paths between key target events and the main actor causal relationship diagram, resulting in a fishery-induced accountability chain set for outlier fishery points in relation to the regulatory imbalance gap. This fishery-induced accountability chain set is weighed with parameters of the regulatory imbalance gap, thus reflecting the corresponding management effort that fishery users should bear when they cause regulatory imbalance gaps in non-managed areas. This method enables the logical exploration of positive causal relationships between regulatory gaps based on multi-source data from fishing activity users, leading to the decision-making of reasonable management measures. This effectively improves the management fit of marine recreational fisheries, achieves targeted management of user violations through the accountability system for marine recreational fisheries, and optimizes management efficiency.

[0091] More specifically, the construction of the subject matrix of the undetermined multi-source fisheries data samples, the causal solution of the underlying objective function of the subject matrix using non-circular constraints following a predetermined logical order, the obtaining of the subject causal relationship graph, and the identification of the potential responsibility contribution intensity of the causal links between subjects leading to the induction of key target events by combining the regulatory imbalance gap with the subject causal relationship graph, specifically includes the following steps:

[0092] A subject matrix is ​​constructed using undetermined multi-source fishery data samples. A low-level objective function with causal relationships is deployed for each data sample in the subject matrix. At the same time, a non-cyclic constraint is constructed based on the rule dependencies recorded by the established logical structure.

[0093] By combining the underlying objective function and non-cyclic constraints, the main causal constraint equation is obtained. The augmented Lagrange method is then introduced to solve the main causal constraint equation by alternating updates to approximate the optimal solution until the non-cyclic constraints converge to 0, thus obtaining the main causal relationship graph of the undetermined multi-source fishery data samples. Among them, the non-zero terms in the event causal relationship graph represent the directional causal relationship between the undetermined multi-source fishery data samples.

[0094] Define the regulatory imbalance gap as the key target event, obtain the loss drift parameter of the regulatory imbalance gap, calculate the implicit correlation degree between the loss drift parameter and each directed causal edge in the subject causal relationship graph, if the implicit correlation degree is greater than the preset implicit correlation degree, then assign a potential responsibility contribution intensity to the directed causal edge based on the implicit correlation degree.

[0095] It should be noted that, in the calculation steps for the intensity of potential liability contribution, this method employs a regression model to deploy a low-level objective function that establishes a causal relationship between each data sample in the main matrix. This low-level objective function is the dependency objective fitted to the data samples, ensuring that each data sample variable can be approximated by a linear combination of other data sample variables. This allows the main causal relationship diagram to more clearly and accurately express the causal relationships between the data samples. Non-cyclic constraints, through matrix exponential function constraints dominated by rule dependencies recorded in a predetermined logical framework, transform acyclic combinations into continuous, traceable logical threads. For example, the complete logical chain triggered by the regulatory imbalance event, "the fishing boat's excessive speed led to entry into an ecologically sensitive area—leading to environmental damage and imbalance in the ecologically sensitive area—forming a regulatory imbalance gap," ensures the accuracy and reliability of the subsequent calculation of the intensity of potential liability contribution. The loss drift parameter of the regulatory imbalance gap is a loss-measured value, such as the deviation of the actual pollution parameters of the ecological environment in the illegally managed area from the compliance benchmark value. If the implicit correlation is greater than the preset implicit correlation, it indicates that the regulatory imbalance gap is highly likely to be caused by the undetermined multi-source fisheries data samples connected by the directed causal edge, and therefore a potential responsibility contribution intensity is assigned.

[0096] More specifically, step S108 includes the following steps:

[0097] If the frequency of occurrence of non-standard fishing sites in the non-compliant management area is greater than the preset frequency, then the management rules and regulations currently followed by marine recreational fishing will be obtained.

[0098] By extracting the assessment coefficients of regulatory risks in response to changes in the scale of fishery activities at different levels from the management rules and regulations system, and the scale of management mobilization that can be implemented to address various regulatory risks, these coefficients are set as the first time step, and the previous time step is set as the second time step.

[0099] A multi-source fishery dataset of outlier fishery points that appear in the illegal management area within a first time step is defined as the first multi-source data queue, and a multi-source fishery dataset of outlier fishery points that appear in the illegal management area within a second time step is defined as the second multi-source data queue.

[0100] A state incremental learning model is constructed, and the first multi-source data queue and the second multi-source data queue are imported into the state incremental learning model to perform incremental learning calculations, so as to obtain the online change law of the fishery activity level within the first time step and the historical change law of the second time step.

[0101] A Bayesian inference network is introduced to perform Bayesian inference on the shift of fishery activity magnitude from the previous time step to the current time step based on the online and historical change patterns. This yields the real-time joint probability distribution of fishery activity magnitude changes. Based on the real-time joint probability distribution, a Lagrange mutual information curve for assessing regulatory risks during interpolation magnitude changes is fitted.

[0102] An algebraic equation solver is created to map the regulatory risk assessment coefficient to the management mobilization scale. The algebraic equation of the Lagrange interpolation curve is solved by the curve equation solver to obtain a series of algebraic equation solutions for the management mobilization scale.

[0103] The management mobilization scale is determined by solving a series of algebraic equations, and the regulatory intensity of the non-compliant management areas is adjusted in real time based on the management mobilization scale, thus outputting a management optimization plan.

[0104] It should be noted that if the frequency of unusual fishing spots appearing in the non-compliant management area exceeds the preset frequency, it indicates that the management intensity for marine recreational fishing projects in that area is unreasonable. Furthermore, user traffic is constantly changing; applying a uniform management intensity could lead to omissions, errors, and decreased regulatory efficiency in the detection of violations. To address this, this method utilizes multi-source data to calculate the change pattern of fishery activity levels from the first time series step to the second. Based on the real-time joint probability distribution inferred between online and historical change patterns, the change in regulatory risk assessment is fitted as a more detailed Lagrange mutual information curve, thus providing a smoother description of the changes in the required regulatory risk assessment coefficient for the current non-compliant management area. Finally, solving the algebraic equation of this Lagrange mutual information curve yields the regulatory intensity that should be adjusted for the required regulatory risk assessment. This method can achieve a reasonable allocation of management intensity in marine recreational fishing areas, reducing the incidence of regulatory risks, alleviating management pressure, improving management efficiency, and ensuring the sustainable development of marine recreational fishing areas.

[0105] A second aspect of this invention provides a marine recreational fisheries management system based on multi-source data, such as... Figure 3 As shown, the marine recreational fishery management system includes a memory 31 and a processor 32. The memory 31 stores a marine recreational fishery management method program based on multi-source data. When the marine recreational fishery management method program is executed by the processor 32, it implements any of the steps of the marine recreational fishery management method described above.

[0106] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for managing offshore recreational fisheries based on multi-source data, characterized in that, The method comprises the following steps: S102: obtaining multi-dimensional monitoring information of the offshore recreational fishery, constructing a mapping map of the overview management area based on multi-layer fishery restriction dimensions meeting the management target, and mapping the multi-dimensional monitoring information to the mapping map for significant comparison, and then displaying normal management areas and illegal management areas according to positive-negative encroachment analysis; S104: obtaining trace points of the overview management area experiencing offshore recreational fishery activities and corresponding past multi-source fishery data, identifying potential abnormal trend behaviors of the trace points to the illegal management area as a mode premise, and performing spatiotemporal trend change distribution splicing on the past multi-source fishery data to obtain a fishery spatiotemporal trend star point graph, and tracing and positioning the same type of fishery points and different types of fishery points by analyzing the fishery spatiotemporal trend star point graph; S106: for different types of fishery points in the illegal management area, a fishery-induced accountability chain set of the different types of fishery points triggering the regulatory imbalance gap of the illegal management area is traced back by using a pending multi-source fishery data sample causal correlation formula, and different measures are taken to manage the recreational fishery subject user according to the number of the fishery-induced accountability chain set; S108: if different types of fishery points frequently appear in the illegal management area, the change law of the magnitude of fishery activities in the adjacent time sequence step is calculated according to the multi-source data, the change law is analyzed according to the management rules and system of the offshore recreational fishery, the regulatory strength is adjusted in real time, and a management optimization scheme is generated.

2. The method for offshore recreational fishery management based on multi-source data according to claim 1, characterized in that, The S102 step specifically comprises the following steps: obtaining an overview management area of the offshore recreational fishery and a management target, extracting multi-layer fishery restriction dimensions of the overview management area and compliance ranges of each multi-layer fishery restriction dimension through the management target; constructing a mapping map of the overview management area having different regulatory specification channels based on the multi-layer fishery restriction dimensions, and synchronously obtaining multi-dimensional monitoring information of the offshore recreational fishery type and the overview management area engaged in each offshore recreational fishery type in a preset time sequence segment; introducing a K-Medians algorithm, dividing the mapping map into a plurality of sub-mapping maps based on the offshore recreational fishery type, assigning each multi-dimensional monitoring information to an aggregated cluster in which the Manhattan distance of the corresponding significant center is less than a preset Manhattan distance in the K-Medians algorithm, calculating the median of all samples in each aggregated cluster on each multi-layer fishery restriction dimension, updating and reorganizing the significant center based on the median, and outputting a significant aggregation result; according to the significant aggregation result, describing a significant local texture of each multi-dimensional monitoring information attracting the attention of each sub-mapping map, establishing a fusion mask according to the significant amplitude texture prompted by the significant local texture, and generating a significant fusion weight model graph of each multi-dimensional monitoring information; The multi-dimensional regulatory information is mapped into a mapping area in the contrast mapping field based on a saliency fusion weight model diagram. In the mapping process, it is judged whether the present specification record value exceeds the specification critical threshold value, and a positive-negative contrast mapping occupation rate of each sub-mapping area is generated. If the positive-negative contrast mapping occupation rate is less than a preset occupation rate threshold value, the sea leisure fishery area corresponding to the sub-mapping area is marked as a normal management area; if the positive-negative contrast mapping occupation rate is greater than the preset occupation rate threshold value, the sea leisure fishery area corresponding to the sub-mapping area is marked as a rule violation management area.

3. The method of claim 1, wherein, The S104 step specifically includes the following steps: Obtain trace points of the total management area located in the preset time sequence segment that have experienced sea leisure fishery activities, and obtain a plurality of past multi-source fishery data of each trace point; Define each trace point as a high-dimensional focus point, introduce a Dijkstra algorithm to calculate the shortest spatial distance of each high-dimensional focus point affecting the total management area when engaging in sea leisure fishery activities to generate each past multi-source fishery data, assign a focus deformation function to each high-dimensional focus point based on the shortest spatial distance, and scale or stretch the dependent positions of the past multi-source fishery data around the focus point to obtain a high-dimensional response layout structure of the past multi-source fishery data of the trace points located in the total management area; Obtain a regulatory imbalance gap of the rule violation management area, establish a space-time journey tree of the past multi-source fishery data of the trace points based on the high-dimensional response layout structure, use the space-time journey tree to search and identify potential abnormal trend behaviors of the trace points forming the regulatory imbalance gap, obtain a multi-source radiation mode access conclusion, reference the multi-source radiation mode access conclusion to spatiotemporally distribute the past multi-source fishery data, and generate a fishery space-time trend star point graph of each trace point; If there is no star point trend arm with a star point area value greater than a star point area threshold value on the fishery space-time trend star point graph, the trace point is marked as a same-class fishery point; If there is at least one and more star point trend arm with a star point area value greater than a star point area threshold value on the fishery space-time trend star point graph, the trace point is marked as a different-class fishery point.

4. The method according to claim 3, wherein, The step of obtaining a regulatory imbalance gap of the rule violation management area, establishing a space-time journey tree of the past multi-source fishery data of the trace points based on the high-dimensional response layout structure, using the space-time journey tree to search and identify potential abnormal trend behaviors of the trace points forming the regulatory imbalance gap, obtaining a multi-source radiation mode access conclusion, referencing the multi-source radiation mode access conclusion to spatiotemporally distribute the past multi-source fishery data, and generating a fishery space-time trend star point graph of each trace point specifically includes the following steps: The high-dimensional response layout structure is analyzed to set a hierarchical dimension partition, time and space order of a plurality of past multi-source fishery data of each trace point is divided on the hierarchical dimension partition, a central balanced fulcrum of different multi-source time and space levels is obtained, a left sub-tree and a right sub-tree are constructed by recursively past multi-source fishery data from the central balanced fulcrum, and a time and space journey tree of past multi-source fishery data of the trace point is formed; A preset star point area threshold of the regulatory imbalance gap is based on, and a left sub-tree is recursively searched and traversed to a leaf node from a root node of the time and space journey tree according to a current dimension of the regulatory imbalance gap, if a distance between a hyperplane with the current leaf node as a center and a query point is less than a preset shortest distance, a leaf node of the right sub-tree is accessed, and a multi-source radiation mode access conclusion of each trace point leading to formation of the regulatory imbalance gap is output; Radiation concentric circles of each layer of multi-source time and space levels are obtained through the multi-source radiation mode access conclusion, past multi-source fishery data of each layer of multi-source time and space levels is distributed and spliced on the radiation concentric circles according to a node degree followed by time and space order, and finally a fishery time and space trend star point graph of each trace point is generated.

5. The method for offshore recreational fishery management based on multi-source data according to claim 1, characterized in that, The S106 step specifically includes the following steps: The S106 step specifically includes the following steps: The S106 step specifically includes the following steps: The S106 step specifically includes the following steps: The S106 step specifically includes the following steps: The S106 step specifically includes the following steps: The S106 step specifically includes the following steps:

6. The method for offshore recreational fisheries management based on multi-source data according to claim 5, characterized in that, The subject matrix of the to-be-determined multi-source fishery data sample is constructed, a non-cyclic constraint following a certain logical order is used to causally solve the bottom layer objective function of the subject matrix, a subject causal association graph is obtained, and the potential responsibility contribution intensity of the causal relationship between the subjects causing the key target event is found out by combining the regulatory imbalance gap and the subject causal association graph, specifically including the following steps: A subject matrix is constructed using the to-be-determined multi-source fishery data sample, a bottom layer objective function of each data sample in the subject matrix having a causal relationship is deployed on a local server, and a non-cyclic constraint is constructed based on the rule dependency recorded in the certain logical order; The bottom layer objective function and the non-cyclic constraint are combined to obtain a subject causal constraint equation, and an augmented Lagrange method is introduced to alternately update and solve the subject causal constraint equation to approach the optimal solution until the non-cyclic constraint converges to 0, thereby obtaining a subject causal association graph of the to-be-determined multi-source fishery data sample; wherein a non-zero term of the event causal association graph represents a directional causal relationship between the to-be-determined multi-source fishery data samples; The regulatory imbalance gap is defined as a key target event, the missing drift parameters of the regulatory imbalance gap are obtained, the implicit association degree between the missing drift parameters and each directional causal edge in the subject causal association graph is calculated, and if the implicit association degree is greater than a preset implicit association degree, a potential responsibility contribution intensity is assigned to the directional causal edge based on the implicit association degree.

7. The method for offshore recreational fisheries management based on multi-source data according to claim 1, characterized in that, The S108 step specifically includes the following steps: If the appearance frequency of the heterogeneous fishery point in the illegal management area is greater than the preset appearance frequency, the management rules and regulations system currently followed by the marine recreational fishery is obtained; The management rules and regulations system is used to extract the regulatory risk assessment coefficients when facing fishery activity level changes of different degrees and the management adjustment scale that can be dynamically implemented for each regulatory risk assessment coefficient, and a first time step is set, and a second time step located at a step before the first time step is set, A multi-source fishery data set of the heterogeneous fishery point appearing in the illegal management area in the first time step is obtained and defined as a first multi-source data queue, and a multi-source fishery data set of the heterogeneous fishery point appearing in the illegal management area in the second time step is obtained and defined as a second multi-source data queue; A state increment learning model is constructed, the first multi-source data queue and the second multi-source data queue are imported into the state increment learning model to perform increment learning calculation, and the online change law of the fishery activity level in the first time step and the historical change law of the second time step are obtained; A Bayesian inference network is introduced, the Bayesian inference network is used to perform Bayesian inference of the fishery activity level transferred from the previous time step to the current time step based on the online change law and the historical change law, a real-time joint probability distribution of the fishery activity level change is obtained, and a Lagrange mutual information curve of the regulatory risk assessment during the interpolation level change is fitted according to the real-time joint probability distribution. Create regulatory risk assessment coefficient-algebraic equation solver of management mobilization scale mapping, through the algebraic equation solver of Lagrange interpolation curve to solve the algebraic equation of management mobilization scale, obtain a series of algebraic equation solution of management mobilization scale; According to a series of algebraic equation solution of management mobilization scale to determine the management mobilization scale, based on the management mobilization scale to adjust the supervision intensity of the illegal management area in real time, output the management optimization scheme.

8. A multi-source data-based offshore recreational fishery management system, characterized by, The offshore recreational fishery management system comprises a memory and a processor, the memory stores a kind of offshore recreational fishery management method program based on multi-source data, when the offshore recreational fishery management method program is executed by the processor, the offshore recreational fishery management method steps of any one of claims 1-7 are realized.

Images