Intelligent Accounting Platform and Method for Marine Industry Data

Abstract

The present invention relates to the technical field of marine industry data accounting, specifically to an intelligent accounting platform and method for marine industry data, comprising the following steps: multi-source data collection and preprocessing of marine industry data; nested analysis of multi-dimensional features in marine industry data based on a deep learning algorithm, and construction of characteristic graphs of marine resource utilization rate, ecological environment index and economic benefit ratio; based on the characteristic graph, a comprehensive benefit score function is constructed, and at the same time, the temporal and spatial variation laws of the data are integrated into the constructed function to achieve comprehensive accounting of marine industry data; the present invention introduces a graph neural network model for feature nested analysis, which can deeply explore the multi-dimensional features in marine industry data; based on an adaptive weight coefficient, a comprehensive benefit score function is constructed, and the temporal and spatial variation laws of the data are integrated into the function, thereby achieving intelligent accounting of the benefit changes of each marine region in different time periods and improving the timeliness and regional adaptability of the accounting.

Description

Technical Field

[0001] The present invention relates to the technical field of marine industry data accounting, and specifically to an intelligent accounting platform and method for marine industry data. Background Art

[0002] With the increasing development of technology, the economy brought by marine industry data has increasingly become an important part of economic development. However, at present, the following problems are faced in the accounting of marine industry data:

[0003] In the process of accounting for marine industry data, there are problems such as difficulty in aligning multi-source heterogeneous data, low data accounting accuracy, and lack of intelligent accounting means. Traditional accounting methods often rely on manual operations and empirical judgments, and cannot meet the complex and changing needs of marine industry data;

[0004] Traditional accounting methods for marine industry data usually adopt static weights and fixed algorithms for comprehensive evaluation, but fail to fully consider spatio-temporal changes and regional differences. For example, the impacts of seasonal factors, climate change, policy adjustments, etc. on resource utilization, ecological environment, and economic benefits have not been dynamically adapted, resulting in limitations in accounting results. In addition, traditional models often lack a flexible real-time feedback mechanism and are difficult to cope with the rapidly changing marine environment and industrial demands. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an intelligent accounting method for marine industry data, including the following steps,

[0007] Multi-source data collection and preprocessing of marine industry data, specifically:

[0008] Collect marine industry data in different formats through a unified standard, and preprocess the collected data based on an adaptive weighted dynamic fusion algorithm;

[0009] Based on a deep learning algorithm, perform nested analysis on multi-dimensional features in marine industry data, and construct feature maps of marine resource utilization rate, ecological environment index, and economic benefit ratio, specifically:

[0010] Take the fused marine industry data as the input to the node set in the graph neural network model, and construct a feature matrix of the nodes. At the same time, perform feature nested analysis based on the graph neural network, and introduce an attention mechanism to eliminate the influence of neighbor nodes in the data propagation process during the layer-by-layer analysis of graph convolution, and construct a feature map according to the calculated importance weights;

[0011] Based on the feature map, a comprehensive benefit scoring function is constructed. At the same time, the spatio-temporal variation law of the data is incorporated into the constructed function to achieve a comprehensive accounting of the marine industry data.

[0012] As a preferred embodiment of the intelligent accounting method for marine industry data according to the present invention, wherein: the collection of marine industry data in different formats through a unified standard is specifically as follows:

[0013] Set the source of marine industry data as , , wherein, represents the th marine industry data source, represents the th data in the th marine industry data source;

[0014] And mark the feature dimensions for each data source, including time feature, space feature, and attribute feature, specifically:

[0015]

[0016]

[0017]

[0018] Wherein, represents the time feature corresponding to the th marine industry data source, indicating the time stamp of the data record, represents the time stamp corresponding to the th data in the th marine industry data source, represents the space feature corresponding to the th marine industry data source, indicating the area label of the data record, represents the area label corresponding to the th data in the th marine industry data source, represents the attribute feature corresponding to the th marine industry data source, indicating the data field of the data record, represents the data field corresponding to the th data in the th marine industry data source.

[0019] As a preferred embodiment of the intelligent accounting method for marine industry data according to the present invention, wherein: the adaptive weighted dynamic fusion algorithm is specifically as follows:

[0020] For each time point Based on the attribute characteristics of corresponding different data sources, data fusion is performed based on adaptive weights, specifically as follows:

[0021]

[0022] Among them, represents the attribute characteristics of the th data source at , represents the total number of data sources, represents the data fused at , represents the th data source at The weight coefficient at is specifically:

[0023]

[0024] Among them, represents the th data source at , represents the average value of the attribute characteristic values of all data sources at , represents the total number of data sources, represents the th data source at The weight coefficient at, represents the th data source's noise parameter, used to reflect the overall noise level of the current data source.

[0025] As a preferred solution of the intelligent accounting method for marine industry data described in the present invention, among them: The graph neural network model is specifically as follows:

[0026] According to the constructed graph neural network model, the marine industry data is used as the node set input into the graph neural network model, including different dimensional characteristics of the marine industry data as the node set , and the time series correlation between data as the edge set , among which, represents nodes, represents the edge relationship determined between node and node ;

[0027] Construct the feature matrix of the nodes, then there is, , among which, represents the constructed node feature matrix, represents the number of nodes, Denote the feature dimensions, including resource utilization rate, ecological environment index, economic benefit ratio, ocean temperature, and pollution index. Meanwhile, construct the node adjacency matrix , and Denote the node and the node can determine the edge relationship, including the mutual influence of data between adjacent sea areas, the industrial connection between the fishery production center and the port, and the impact of pollution in a certain area on the surrounding areas. Otherwise, it is 0.

[0028] As a preferred scheme of the intelligent accounting method for ocean industry data described in the present invention, wherein: the feature nesting analysis based on the graph neural network is specifically as follows:

[0029] Initialize the node features ;

[0030] Based on the graph convolution layer to achieve the layer-by-layer propagation of data, then there is

[0031]

[0032] wherein Denote the node feature matrix of the th layer Denote the weight matrix of the th layer Denote the adjacency matrix Denote the node feature matrix of the th layer;

[0033]

[0034] wherein Denote the attention vector Denote the transpose matrix Denote the introduced transformation matrix 、 Denote the feature vectors of the node and the node respectively Denote the calculated importance weight, which is the influence weight of neighbor nodes.

[0035] As a preferred scheme of the intelligent accounting method for ocean industry data described in the present invention, wherein: the construction of the feature graph according to the calculated importance weight is specifically as follows:

[0036] The ocean resource utilization rate represents the development degree of ocean resources in each region, including fishery resources and mineral development. Specifically

[0037]

[0038] Among them, represents the importance weight of calculation, represents the catch, represents the total amount of exploitable resources in the current area, represents the utilization rate of marine resources in the current area;

[0039] The ecological environment index represents the environmental health of each area, including the pollution index and water quality data, specifically:

[0040]

[0041] Among them, represents the importance weight of calculation, represents the pollutant concentration, represents the environmental carrying capacity of the current area, represents the ecological environment index of the current area;

[0042] The economic benefit ratio represents the ratio of industrial input to income, specifically:

[0043]

[0044] Among them, represents the importance weight of calculation, represents the total industrial income of the current area, including fishery, tourism and port economy, represents the total industrial cost of the current area, including labor cost, resource consumption and environmental governance, represents the economic benefit ratio of the current area.

[0045] As a preferred solution of the intelligent accounting method for marine industry data described in the present invention, wherein: the comprehensive benefit score function is specifically as follows:

[0046] Based on the adaptive weight coefficient, a comprehensive benefit score function is constructed, then

[0047]

[0048] Among them, , , are respectively the adaptive weight coefficients, represents the resource utilization rate feature map, represents the ecological environment index feature map, represents the economic benefit feature map, represents the comprehensive benefit score corresponding to the current area;

[0049] Based on the constructed comprehensive benefit score function, the accounting of marine industry data is realized, specifically:

[0050]

[0051] Among them, represents the importance weight of calculation, represents the comprehensive benefit score corresponding to the current area, represents the total number of ocean areas, represents the accounting results of all ocean areas, which are the comprehensive accounting results of ocean industry data.

[0052] As a preferred solution of the intelligent accounting method for ocean industry data described in the present invention, wherein: the specific form of integrating the spatio-temporal variation law of data into the constructed function is as follows:

[0053] Adaptive weight coefficient , and are dynamically regulated by the time characteristics and spatial characteristics in the ocean industry data, specifically:

[0054] Regulation of time characteristics ,

[0055]

[0056] Among them, represents the seasonal change factor, which is the seasonal change time node in the time characteristics. When the time distance from the seasonal change is shorter, the seasonal change factor is smaller, represents the basic adaptive weight coefficient, represents the adjusted adaptive weight coefficient;

[0057] Regulation of spatial characteristics ,

[0058]

[0059]

[0060]

[0061] Among them, , , represent the adaptive weight coefficients before adjustment, , , represent the adjusted adaptive weight coefficients, represents the resource density of the current area, represents the pollution concentration of the current area, Indicates the intensity of economic activity in the current region.

[0062] A computer device includes a memory and a processor, wherein the memory stores a computer program and the processor implements the steps of an intelligent accounting method for marine industry data when executing the computer program.

[0063] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of an intelligent accounting method for marine industry data.

[0064] Beneficial effects of the present invention:

[0065] By collecting marine industry data in different formats through unified standards and pre-processing the data based on an adaptive weighted dynamic fusion algorithm, the system achieves alignment and fusion of multi-source heterogeneous data, improving data accuracy and reliability.

[0066] The introduction of a graph neural network model for feature nesting analysis can deeply explore the multi-dimensional features in marine industry data, construct characteristic maps of marine resource utilization rate, ecological environment index, and economic benefit ratio, and provide more comprehensive data support for accounting;

[0067] A comprehensive benefit scoring function is constructed based on adaptive weight coefficients, and the temporal and spatial variation patterns of the data are integrated into the function, which realizes the intelligent calculation of the benefit changes of various ocean regions in different time periods and improves the timeliness and regional adaptability of the calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0069] Figure 1 This is a schematic diagram of the overall method steps of the intelligent accounting method for marine industry data of the present invention. DETAILED DESCRIPTION

[0070] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0071] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways than those specifically described herein, and those skilled in the art can make similar extensions without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0072] Secondly, as used herein, "one embodiment" or "an embodiment" refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other.

[0073] The present invention is described in detail in conjunction with schematic diagrams. When describing the embodiments of the present invention in detail, for the sake of convenience of explanation, the cross-sectional views showing the device structure will be enlarged locally in a non-general proportion, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0074] At the same time, in the description of the present invention, it should be noted that the terms "first, second, or third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0075] Unless otherwise clearly defined and limited in the present invention, the terms "mounted, connected, and coupled" shall be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can also be a mechanical connection, an electrical connection, or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0076] Embodiment 1

[0077] Referring to Figure 1 , for an embodiment of the present invention, an intelligent accounting method for marine industry data is provided, including the following steps:

[0078] S1: Multi-source data collection and preprocessing of marine industry data.

[0079] Specifically, the multi-source data collection and preprocessing of the marine industry data is to collect marine industry data from different sources and in different formats through unified standardization, and preprocess the collected data based on an adaptive weighted dynamic fusion algorithm and a time series algorithm. The specific implementation is as follows:

[0080] Set the source of marine industry data as , , where represents the One marine industry data source, indicating the th data in the th marine industry data source;

[0081] And mark the feature dimensions for each data source, including temporal features, spatial features, and attribute features, specifically:

[0082]

[0083]

[0084]

[0085] Among them, indicating the temporal feature corresponding to the th marine industry data source, representing the timestamp of the data record, indicating the th data in the th marine industry data source corresponding to the timestamp, indicating the spatial feature corresponding to the th marine industry data source, representing the area label of the data record, indicating the th data in the th marine industry data source corresponding to the area label, indicating the attribute feature corresponding to the th marine industry data source, representing the data field of the data record, indicating the th data in the th marine industry data source corresponding to the data field;

[0086] Perform data conversion on the data with different feature dimensions, specifically:

[0087] Set a global reference time axis , and map the temporal feature of each data source to the global reference time axis, thereby completing the unification of data in different time dimensions in terms of time;

[0088] Meanwhile, map the spatial feature of each data to a unified spatial network to complete the unification of data in different spatial dimensions in terms of space;

[0089] And perform normalization on the attribute features of all data to complete the attribute standardization of data in different attribute dimensions;

[0090] For each time point Based on the adaptive weights, the attribute features of different data sources corresponding to each time point are fused, specifically as follows:

[0091]

[0092] Among them, represents the attribute feature of the th data source at , represents the total number of data sources, represents the fused data at , represents the weight coefficient of the th data source at , specifically:

[0093]

[0094] Among them, represents the attribute feature of the th data source at , represents the average value of the attribute feature values of all data sources at , represents the total number of data sources, represents the weight coefficient of the th data source at , represents the noise parameter of the th data source, which is used to reflect the overall noise level of the current data source and is determined by the characteristics of the data source itself.

[0095] It should be noted that by combining the data source noise level and time to dynamically adjust the weights, the robustness and accuracy of the fusion structure can be improved. Moreover, for the unification of different feature dimensions of the data, the alignment problem of multi-source heterogeneous data can be solved.

[0096] S2: Based on the deep learning algorithm, nested analysis is performed on the multi-dimensional features in the marine industry data to construct the feature maps of marine resource utilization rate, ecological environment index, and economic benefit ratio.

[0097] Specifically, the nested analysis of the multi-dimensional features in the marine industry data based on the deep learning algorithm is carried out for the fused marine industry data. By building a graph neural network model and using the marine industry data as the input of the model, the feature maps of marine resource utilization rate, ecological environment index, and economic benefit ratio are constructed according to the output results of the model. The specific implementation is as follows:

[0098] According to the constructed graph neural network model, the marine industry data is used as the input to the node set in the graph neural network model, including the different dimensional features of the marine industry data as the node set , and the time series correlation between the data is used as the edge set , where represents the th node, represents the node and the node determines the edge relationship between them;

[0099] Construct the feature matrix of the node, then there is , where represents the constructed node feature matrix, represents the number of nodes, represents the feature dimension, including resource utilization rate, ecological environment index, economic benefit ratio, ocean temperature, and pollution index. At the same time, construct the node adjacency matrix , and represents the node and the node can determine the edge relationship between them, including the mutual influence of data between adjacent sea areas, the industrial connection between the fishery production center and the port, and the impact of pollution in a certain area on the surrounding areas. Otherwise, it is 0;

[0100] Based on the graph neural network, perform feature nesting analysis, specifically:

[0101] Initialize the node features ;

[0102] Based on the graph convolution layer, realize the layer-by-layer propagation of data, then there is

[0103]

[0104] where represents the node feature matrix of the th layer, represents the weight matrix of the th layer, represents the adjacency matrix, represents the node feature matrix of the th layer;

[0105] When the data is propagated layer by layer through the graph convolution layer, the influence brought by the neighbor nodes in the data propagation process is eliminated by introducing the attention mechanism. Specifically, it is by calculating the importance weight of each node and the neighbor node , specifically:

[0106]

[0107] Among them, represents the attention vector, represents the transpose matrix, represents the introduced transformation matrix, , respectively represent the feature vectors of nodes and node ; represents the calculated importance weight, which is the influence weight of neighbor nodes;

[0108] According to the calculated importance weights, feature maps of the marine resource utilization rate, ecological environment index, and economic benefit ratio are constructed respectively, specifically as follows:

[0109] The marine resource utilization rate represents the development degree of marine resources in each region, including fishery resources and mineral development, specifically:

[0110]

[0111] Among them, represents the calculated importance weight, represents the fish catch, represents the total amount of resources available for development in the current region, represents the marine resource utilization rate of the current region;

[0112] The ecological environment index represents the environmental health degree of each region, including the pollution index and water quality data, specifically:

[0113]

[0114] Among them, represents the calculated importance weight, represents the pollutant concentration, represents the environmental carrying capacity of the current region, represents the ecological environment index of the current region;

[0115] The economic benefit ratio represents the ratio of industrial input to output, specifically:

[0116]

[0117] Among them, represents the calculated importance weight, represents the total industrial income of the current region, including fishery, tourism, and port economy, represents the total industrial cost of the current region, including labor cost, resource consumption, and environmental governance, represents the economic benefit ratio of the current region.

[0118] It should be noted that for the calculated ocean resource utilization rate, ecological environment index, and economic benefit ratio, corresponding characteristic maps are constructed to provide a data basis for subsequent intelligent accounting.

[0119] S3: Based on the constructed characteristic maps, a multi-objective optimization model is introduced to comprehensively evaluate the relationships among resource utilization rate, ecological environment impact, and economic benefits, and realize the intelligent accounting of ocean industry data.

[0120] Specifically, the multi-objective optimization model constructs a comprehensive benefit score function through adaptive weight coefficients and based on the characteristic maps. At the same time, the spatio-temporal variation law of the data is incorporated into the constructed function to analyze the benefit changes in different ocean regions at different time periods and achieve a comprehensive accounting of ocean industry data. The specific implementation is as follows:

[0121] Based on the adaptive weight coefficients, a comprehensive benefit score function is constructed, so there is

[0122]

[0123] Among them, 、 、 are the adaptive weight coefficients respectively, represents the characteristic map of resource utilization rate, represents the characteristic map of ecological environment index, represents the characteristic map of economic benefits, represents the comprehensive benefit score corresponding to the current region;

[0124] Based on the constructed comprehensive benefit score function, the accounting of ocean industry data is realized, specifically as follows:

[0125]

[0126] Among them, represents the calculated importance weight, represents the comprehensive benefit score corresponding to the current region, represents the total number of ocean regions, represents the accounting results of all ocean regions, which is the comprehensive accounting result of ocean industry data.

[0127] It should be noted that for the adaptive weight coefficients in the process of constructing the comprehensive benefit score function, they are regulated based on the spatio-temporal variation law of ocean data. The specific regulation process is as follows:

[0128] The adaptive weight coefficients 、 and are affected by the time characteristics and spatial characteristics Dynamic regulation, specifically:

[0129] Regulation of time characteristics ,

[0130]

[0131] Among them, represents the seasonal change factor, which is the seasonal change time node in the time characteristics. The shorter the time distance from the seasonal change, the smaller the seasonal change factor. The specific value is set by the implementer according to the actual application scenario. represents the basic adaptive weight coefficient, represents the adjusted adaptive weight coefficient;

[0132] Regulation of spatial characteristics ,

[0133]

[0134]

[0135]

[0136] Among them, , , represent the adaptive weight coefficients before adjustment, , , represent the adjusted adaptive weight coefficients, represents the resource density of the current area, including fishery resources and mineral resources, represents the pollution concentration of the current area, represents the intensity of economic activities in the current area, including tourism.

[0137] It should be noted that by introducing the adaptive weight coefficient regulation mechanism and integrating time characteristics and spatial characteristics, not only the accuracy of marine industry data accounting is improved, but also the weight coefficients can be dynamically adjusted for different time periods and regions to ensure the timeliness and regional adaptability of the accounting results.

[0138] Furthermore, if the above-mentioned functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0139] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definable sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in conjunction with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0140] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or otherwise processing it as appropriate, and then storing it in a computer memory.

[0141] In addition, to provide a concise description of the exemplary embodiments, not all features of the actual embodiments may be described (i.e., those features that are not relevant to the currently contemplated best mode of implementing the present invention, or those features that are not relevant to the implementation of the present invention).

[0142] It should be understood that, during the development of any actual implementation, such as in any engineering or design project, a large number of specific implementation decisions can be made. Such development efforts may be complex and time-consuming, but for those of ordinary skill in the art who benefit from this disclosure, without undue experimentation, such development efforts will be routine work of design, manufacturing, and production.

[0143] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An intelligent accounting method for marine industry data, characterized in that: It includes the following steps: Multi-source data collection and preprocessing of marine industry data, specifically: Collect marine industry data in different formats through a unified standard, and preprocess the collected data based on an adaptive weighted dynamic fusion algorithm; Perform nested analysis on multi-dimensional features in marine industry data based on deep learning algorithms, and construct feature maps of marine resource utilization rate, ecological environment index, and economic benefit ratio, specifically: Take the fused marine industry data as the input to the node set in the graph neural network model, and construct the feature matrix of the nodes. At the same time, perform feature nesting analysis based on the graph neural network, and introduce an attention mechanism to eliminate the influence of neighbor nodes during data propagation during the layer-by-layer analysis process of graph convolution, and construct a feature map according to the calculated importance weights; Based on the feature map, construct a comprehensive benefit scoring function, and at the same time, incorporate the spatio-temporal variation law of the data into the constructed function to achieve a comprehensive accounting of marine industry data; The specific form of the comprehensive benefit scoring function is as follows: Construct a comprehensive benefit scoring function based on an adaptive weight coefficient, then there is, ; Among them, , , are adaptive weight coefficients respectively, represents the resource utilization feature map, represents the ecological environment index feature map, represents the economic benefit feature map, represents the comprehensive benefit score corresponding to the current area; Based on the constructed comprehensive benefit scoring function, achieve the accounting of marine industry data, specifically: ; Among them, represents the importance weight of the calculation, represents the comprehensive benefit score corresponding to the current area, represents the total number of ocean areas, represents the accounting results of all ocean areas, which are the comprehensive accounting results of ocean industry data; The specific form of incorporating the spatio-temporal variation law of the data into the constructed function is as follows: Adaptive weight coefficient , and , are dynamically regulated by the time characteristics and spatial characteristics in the marine industry data, specifically as follows: Temporal feature regulation ; Among them, represents the seasonal change factor, which is the seasonal change time node in the time characteristics. The shorter the time distance from the seasonal change, the smaller the seasonal change factor. represents the basic adaptive weight coefficient, represents the adjusted adaptive weight coefficient; Spatial features regulation ; ; ; Among them, , , represent the adaptive weight coefficients before adjustment, , , represent the adaptive weight coefficients after adjustment, represents the resource density of the current area, represents the pollution concentration of the current area, represents the intensity of economic activities in the current area.

2. The intelligent accounting method for marine industry data according to claim 1, characterized in that: The specific form of collecting marine industry data in different formats through a unified standard is as follows: Set the data source of the marine industry as , , where represents the th marine industry data source, represents the th data source of the marine industry and the th data; Mark feature dimensions for each data source, including time features, spatial features, and attribute features, specifically: ; ; ; Among them, represents the time feature corresponding to the th marine industry data source, indicating the timestamp of the data record, represents the th marine industry data source, and the th data record's corresponding timestamp, represents the spatial feature corresponding to the th marine industry data source, indicating the area label of the data record, represents the th marine industry data source, and the th data record's corresponding area label, represents the attribute feature corresponding to the th marine industry data source, indicating the data field of the data record, represents the th marine industry data source, and the th data record's corresponding data field.

3. The intelligent accounting method for marine industry data according to claim 2, characterized in that: The specific form of the adaptive weighted dynamic fusion algorithm is as follows: For each time point Based on the adaptive weights, data fusion is performed on the attribute features of different data sources corresponding thereto, as follows: ; Among them, represents the attribute characteristics of the th data source at ; represents the total number of data sources; represents the data fused at ; represents the weight coefficient of the th data source at , specifically: ; Among them, represents the attribute characteristics of the th data source at ; represents the average value of the attribute characteristic values of all data sources at ; represents the total number of data sources, represents the weight coefficient of the th data source at ; represents the noise parameter of the th data source, which is used to reflect the overall noise level of the current data source.

4. The intelligent accounting method for marine industry data according to claim 3, characterized in that: The specific form of the graph neural network model is as follows: According to the constructed graph neural network model, the marine industry data is used as the node set input into the graph neural network model, including different dimensional features of the marine industry data as the node set , and the time series correlation between the data is used as the edge set , where represents nodes, represents the node and the node the determined edge relationship between them; Construct the feature matrix of the nodes, then there is, , where, represents the constructed node feature matrix, represents the number of nodes, represents the feature dimension, including resource utilization rate, ecological environment index, economic benefit ratio, ocean temperature, and pollution index. At the same time, construct the node adjacency matrix , and, represents node and node can determine the edge relationship between them, including the mutual influence of data between adjacent sea areas, the industrial connection between the fishery production center and the port, and the impact of pollution in a certain area on the surrounding areas. Otherwise, it is 0.

5. The intelligent accounting method for marine industry data according to claim 4, characterized in that: The specific form of performing feature nesting analysis based on the graph neural network is as follows: Initialize node features ; Based on the graph convolution layer, realize the layer-by-layer propagation of data, then there is, ; Among them, represents the node feature matrix of the th layer, represents the weight matrix of the th layer, represents the adjacency matrix, represents the node feature matrix of the th layer; ; Among them, represents the attention vector, represents the transpose matrix, represents the introduced transformation matrix, 、 respectively represent the and feature vectors of the nodes represents the calculated importance weight, which is the influence weight of neighbor nodes.

6. The intelligent accounting method for marine industry data according to claim 5, characterized in that: The specific form of constructing a feature map according to the calculated importance weights is as follows: The marine resource utilization rate represents the degree of development of marine resources in each region, including fishery resources and mineral development, specifically: ; Among them, represents the importance weight of the calculation, represents the catch, represents the total amount of resources available for development in the current area, represents the utilization rate of marine resources in the current area; The ecological environment index represents the environmental health of each region, including pollution index and water quality data, specifically: ; Among them, represents the importance weight of the calculation, represents the pollutant concentration, represents the environmental carrying capacity of the current area, represents the ecological environment index of the current area; The economic benefit ratio represents the ratio of industrial input to output, specifically: ; Among them, represents the importance weight of the calculation, represents the total industrial income of the current region, including fishery, tourism and port economy, represents the total industrial cost of the current region, including labor cost, resource consumption and environmental governance, represents the economic benefit ratio of the current region.

7. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method described in any one of claims 1 to 6.

Images