Yangtze fish information database construction method, application and system

Abstract

The application discloses a Yangtze River fish information database construction method and application and system, and belongs to the technical field of Yangtze River fish resource protection. The application collects multi-dimensional fish data and carries out standardized management, establishes a four-dimensional metadata specification and fusion degree constraint model, realizes multi-library association fusion and hierarchical incremental synchronization by adopting a hybrid storage architecture, deploys multifunctional modules and constructs a multi-dimensional data quality evaluation system, and can be dynamically iteratively updated. The application can effectively solve problems of scattered Yangtze River fish data, low fusion precision, single application scene and the like, provides professional services for teaching, scientific research, engineering and management, simultaneously realizes intelligent early warning of rare fish, and provides efficient and reliable data support for fish protection, ecological restoration and supervision decision of the Yangtze River basin.

Description

Technical Field

[0001] This invention belongs to the technical field of Yangtze River fish resource protection, specifically relating to a method and application system for constructing a Yangtze River fish information database. Background Technology

[0002] The Yangtze River is a core habitat for freshwater fish resources in my country, home to over 400 species, including more than 100 endemic species, making it a crucial area for the conservation of aquatic biodiversity in the country. In recent years, affected by human activities such as water conservancy projects, water pollution, and overfishing, the population of fish in the Yangtze River has declined sharply, with some species facing extinction. The number of seedlings of the "four major freshwater fish" (Chinese carp, freshwater fish, and freshwater fish) has decreased by more than 90% compared to the 1950s, and the annual yield of Yangtze River knife fish has been significantly reduced from its historical high.

[0003] To implement the Yangtze River protection strategy, there is an urgent need to establish a comprehensive, accurate, and dynamic database of Yangtze River fish information to provide data support for resource assessment, ecological restoration, law enforcement supervision, and scientific research. Currently, existing fish databases have several shortcomings: First, data completeness is insufficient, failing to comprehensively cover information on Yangtze River endemic, rare, and endangered species; second, data sources are disorganized, with traditional survey data, molecular monitoring data, and environmental monitoring data not effectively integrated, resulting in inconsistent data quality; third, a dynamic update mechanism is lacking, leading to poor data timeliness and failing to meet the routine monitoring needs of Yangtze River protection; and fourth, an integrated support system for teaching, research, and application has not been established, making it difficult to serve university teaching, scientific research, engineering applications, and industry training, hindering deep integration of industry, academia, and research, and failing to fully realize the value of the data.

[0004] Therefore, developing a method and system for constructing a Yangtze River fish information database that integrates multi-source data, dynamically iterates and updates data, and has the ability to serve teaching, research and application has become an urgent technical problem to be solved in the field of Yangtze River protection. Summary of the Invention

[0005] To address the technical problems existing in the background art, the present invention provides a method, application, and system for constructing a Yangtze River fish information database.

[0006] This invention is achieved through the following technical solution: a method for constructing a Yangtze River fish information database, comprising the following steps: A comprehensive dataset of fish species in the Yangtze River Basin was collected, which includes basic data, monitoring data, environmental correlation data, and application scenario data. The full-dimensional dataset undergoes unified format conversion, missing value imputation, outlier removal, and ambiguity resolution to establish a unified data dictionary and metadata specifications, thus forming a standardized dataset. ; A hierarchical storage architecture is adopted to store standardized datasets. The database is divided into a basic information database, a monitoring and analysis database, an environmental correlation database, and an application service database. ETL tools are used to achieve the correlation, fusion, and incremental synchronization of data from multiple databases, thus constructing the ontology of the Yangtze River fish information database. Based on the Yangtze River fish information database, a data retrieval module, an intelligent early warning module, and a teaching and research support module are deployed, providing multi-terminal access interfaces and data services; Establish a data quality evaluation system The structure, content, and functional modules of the Yangtze River fish information database ontology are iteratively updated based on user feedback and monitoring needs.

[0007] In a further embodiment, the unified data dictionary includes fish entities, monitoring locations, environmental factors, application scenarios, and teaching and research cases; correspondingly, the metadata specification adopts a standardized four-dimensional structure and incorporates customized coding and verification rules for the Yangtze River Basin, expressed in the following form: ; in, The object is encoded using a 14-bit unique encoding structure, specifically in the format "object type code - watershed partition code - serial number". Required properties customized according to object type. Required attribute A unified validation standard for all required attribute settings in the document. This refers to cross-class linkage logic between objects, which includes: spatial association, ecological association, scene association, and teaching and research association.

[0008] In a further embodiment, the tiered storage architecture is a hybrid storage architecture, specifically implemented as follows: Standardize the dataset The database is divided into a basic information database, a monitoring and analysis database, an environmental correlation database, and an application service database. The basic information database is stored using a relational database: for structured data, it is categorized by required attributes. Define and build standardized data tables, and encode them as objects. As a global primary key, combined with a unified verification standard Design the field validation logic; The monitoring and analysis library uses a distributed file system for storage: for unstructured monitoring data, it is based on object encoding. The watershed partition code features are used to divide and store data in fragments, centrally storing monitoring data from the same watershed partition, while also using mandatory attributes. Establish a file index for the monitoring point attributes; The environmental association library and application service library are stored using a graph database: fish entities, environmental factors, application scenarios, and teaching and research cases are mapped to knowledge graph nodes, with cross-class linkage logic. Based on the edge, through object encoding Implement unique node identification and cross-database association.

[0009] In a further embodiment, the association fusion satisfies the following fusion degree constraint: ; in, This is the integration score. For spatial matching degree, For time consistency, For species-environment association, , and All are weighting coefficients. ; when At that time, the corresponding standardized dataset The ETL tool can then be used to complete the multi-database association and fusion and write the data into the Yangtze River fish information database ontology. The preset threshold for fish data fusion in the Yangtze River Basin.

[0010] In a further embodiment, the incremental synchronization adopts a hierarchical incremental synchronization mechanism based on metadata specifications, and the specific implementation method is as follows: Combining timestamps and integration scores Construct incremental triggering rules and extract standardized datasets. New and updated Valid data is automatically filtered to remove invalid and duplicate data. Partition synchronization involves the following steps: Basic information base: using the transaction incremental mechanism of relational databases, encoded in object form. Incremental writing of structured attributes is performed for unique primary keys, and the synchronization process follows a unified verification standard. Constrain and validate field compliance; Monitoring and Analysis Library: Based on Object Coding Based on the watershed partition code characteristics, the newly added unstructured monitoring data of the same watershed is batch synchronized to the corresponding storage nodes through the sharding incremental protocol of the distributed file system; Environment-related libraries and application service libraries: based on cross-class linkage logic Based on this, the newly added species-environment-scene relationships are synchronized through the node / edge incremental update interface of the graph database, and the synchronization process is performed using object encoding. Verify the uniqueness of the node.

[0011] In a further embodiment, the intelligent early warning is implemented based on a threshold discrimination model specifically designed for the protection of rare fish species in the Yangtze River Basin. The specific model expression is as follows: ; in, For the early warning judgment results, This indicates that an alert has been triggered. This indicates that there was no prior warning. For the first Population abundance of rare fish species The threshold for population abundance warning. For the first Measured values ​​of key environmental factors in the habitat of rare fish species For the first The appropriate range of values ​​for key environmental factors corresponding to rare fish species.

[0012] In a further embodiment, the teaching and research support module constructs four types of differentiated thematic data services according to user objects and application scenarios, specifically including: To cater to teaching scenarios, a standardized species case library and knowledge point dataset based on a unified data dictionary are constructed, and structured teaching materials are formed according to fish species, habitat characteristics, and protection levels for teaching demonstrations and popular science use; For scientific research scenarios, it provides open standardized API interfaces and batch data export interfaces, supporting data retrieval, model calling and offline analysis by species, watershed and time dimensions; For engineering application scenarios, it provides a thematic dataset on ecological restoration, including data on suitable fish habitats, environmental factor thresholds, habitat distribution and restoration potential evaluation, to provide data support for watershed ecological governance projects; For management decision-making scenarios, it provides specialized data services for regulatory decision-making, integrating species distribution, monitoring points, environmental anomalies and early warning information to form visualized regulatory data support.

[0013] In a further embodiment, the data quality evaluation system A customized multi-dimensional quantitative model for the Yangtze River Basin has the following iterative update mechanism: A quality dimension quantification model is created, decomposing data quality into five dimensions: completeness, accuracy, consistency, timeliness, and applicability, and calculating the quantification scores for each dimension as follows: , , , and ; Constructing a comprehensive quality evaluation function : ;in, , , , and These are the weighting coefficients; Quality evaluation is conducted according to the predetermined evaluation cycle, taking into account user feedback questionnaires and requests for changes to monitoring requirements. If any single dimension fails to meet the standard, the update process will be automatically triggered. This is a preset evaluation threshold.

[0014] An application for building a Yangtze River fish information database is proposed, which can be used for fish protection, teaching and research, ecological engineering construction, and basin management decision-making in the Yangtze River Basin.

[0015] A system for constructing a Yangtze River fish information database, used to implement the construction method described above, includes: The first module is set to collect a full-dimensional dataset of fish species in the Yangtze River Basin. The full-dimensional dataset includes basic data, monitoring data, environmental correlation data, and application scenario data. The second module is configured to perform unified format conversion, missing value imputation, outlier removal, and ambiguity resolution on the full-dimensional dataset, establish a unified data dictionary and metadata specifications, and form a standardized dataset. ; The third module is configured to use a hierarchical storage architecture to store standardized datasets. The database is divided into a basic information database, a monitoring and analysis database, an environmental correlation database, and an application service database. ETL tools are used to achieve the correlation, fusion, and incremental synchronization of data from multiple databases, thus constructing the ontology of the Yangtze River fish information database. The fourth module is configured to deploy a data retrieval module, an intelligent early warning module, and a teaching and research support module based on the Yangtze River fish information database, providing multi-terminal access interfaces and data services; The fifth module is set up to establish a data quality evaluation system. The structure, content, and functional modules of the Yangtze River fish information database ontology are iteratively updated based on user feedback and monitoring needs.

[0016] The beneficial effects of this invention are: This invention unifies the format, verifies constraints, and defines associations for basic data, monitoring data, and environmental data in the Yangtze River Basin by using a unified data dictionary and four-dimensional metadata specifications. This solves the problems of traditional data being scattered, heterogeneous, and highly ambiguous, and significantly improves data consistency, standardization, and reusability.

[0017] This invention introduces a fusion constraint model to quantitatively determine the quality of data fusion from three dimensions: spatial matching, temporal consistency, and species-environment association. Only data that meets the standards can be included in the database, thereby reducing erroneous associations, spatiotemporal misalignments, and contradictory ecological logic data from the source and significantly improving the overall accuracy of the database.

[0018] For structured, unstructured, and relational data, relational databases, distributed file systems, and graph databases are used for collaborative storage, balancing query efficiency, large-capacity storage, and complex ecosystem relationship mining, to achieve efficient cross-database association, fast retrieval, and parallel read and write.

[0019] This invention uses quantitative scoring based on completeness, accuracy, consistency, timeliness, and applicability to form a quality closed loop that can be automatically judged, traced, and iteratively updated, ensuring that the database is reliable, usable, and up-to-date in the long term.

[0020] This invention provides a case library, open API, ecological restoration dataset, and regulatory decision-making services for teaching, scientific research, engineering applications, and management decision-making, enabling multiple uses of a single database and on-demand supply, significantly enhancing the application value of Yangtze River fish data in protection, teaching, scientific research, and governance. Attached Figure Description

[0021] Figure 1 This is a flowchart of the method for constructing the Yangtze River fish information database in Example 1. Detailed Implementation

[0022] The present invention will now be further described with reference to the accompanying drawings and embodiments.

[0023] like Figure 1 As shown in the figure, this embodiment discloses a method for constructing a Yangtze River fish information database, including the following steps: A comprehensive dataset of fish species in the Yangtze River Basin was collected, which includes basic data, monitoring data, environmental correlation data, and application scenario data. The full-dimensional dataset undergoes unified format conversion, missing value imputation, outlier removal, and ambiguity resolution to establish a unified data dictionary and metadata specifications, thus forming a standardized dataset. ; A hierarchical storage architecture is adopted to store standardized datasets. The database is divided into a basic information database, a monitoring and analysis database, an environmental correlation database, and an application service database. ETL tools are used to achieve the correlation, fusion, and incremental synchronization of data from multiple databases, thus constructing the ontology of the Yangtze River fish information database. Based on the Yangtze River fish information database, a data retrieval module, an intelligent early warning module, and a teaching and research support module are deployed, providing multi-terminal access interfaces and data services; Establish a data quality evaluation system The structure, content, and functional modules of the Yangtze River fish information database ontology are iteratively updated based on user feedback and monitoring needs.

[0024] In this embodiment, the basic data includes a list of fish species, taxonomic information, morphological characteristics, distribution areas, living habits, reproductive characteristics, resource quantity, and protection level.

[0025] The monitoring data includes data on species spatiotemporal distribution, population dynamics, and habitat factors obtained based on traditional surveys, environmental DNA (eDNA) technology, acoustic monitoring, and remote sensing GIS technology.

[0026] The environmental data includes water temperature, pH value, dissolved oxygen, flow velocity, sediment type, hydrological rhythm, and data on the impact of water conservancy projects.

[0027] The application scenario data includes data on stock enhancement and release, fishery resource assessment, ecological restoration, law enforcement supervision, and collaborative teaching and research.

[0028] Based on this, the unified data dictionary includes fish entities, monitoring locations, environmental factors, application scenarios, and teaching and research cases; correspondingly, the metadata specification adopts a standardized four-dimensional structure and incorporates customized coding and verification rules for the Yangtze River Basin, expressed in the following form: ; in, For object coding, a 14-bit unique coding structure is adopted, specifically in the format of "object type code - watershed zoning code - serial number". In this embodiment, the object type code is a 2-digit letter code, with fixed values ​​assigned as: fish entity (FI), monitoring point (SI), environmental factor (EV), application scenario (SC), and teaching and research case (TE); the watershed zoning code is a 2-digit numeric code, divided according to the administrative and ecological zoning rules of the Yangtze River Basin (upper reaches of the main stream 01, middle reaches of the main stream 02, lower reaches of the main stream 03, tributaries 04, lakes 05); the serial number is an 8-digit numeric code, which increments sequentially from 00000001 under the same object type and the same watershed zoning to ensure global uniqueness of the code.

[0029] Required attributes are customized according to object type. For example, the required attributes for fish entities (FI) include: Chinese name, scientific name, taxonomic rank, protection level, distribution section, suitable habitat, and reproductive characteristics. Required attributes for monitoring sites (SI) include site name, latitude and longitude coordinates, water area type, monitoring frequency, and deployment time. Required attributes for environmental factors (EV) include factor name, monitoring value, monitoring time, monitoring unit, and data source. Required attributes for application scenarios (SC) include scenario name, scenario type, implementation area, implementation time, and related requirements. Required attributes for teaching and research cases (TE) include case name, applicable objects, related species codes, core content, and application scenario.

[0030] Is a mandatory attribute The unified verification standard for the setting of each mandatory attribute in it includes three types of constraints: data type, unit, and value range; the data type is fixed as string type, numerical type, enumeration type, date-time type, and geographic coordinate type; the unit uniformly follows international standards and industry specifications; the value range adapts to the ecological characteristics of the Yangtze River Basin, such as the water temperature is limited to 0~32°C, and the protection level is limited to national first-class / national second-class / vulnerable / endangered / no protection.

[0031] Is the cross-class linkage logic between objects. It includes spatial association (fish and monitoring points), ecological association (fish and environmental factors), scenario association (application scenario and monitoring points), and teaching-research-application association (teaching cases and fish entities), realizing the integrated association of multi-source data.

[0032] Based on the above description, this embodiment is further illustrated by taking the fish entity (Myxocyprinus asiaticus) as an example. , where Is: Chinese name: Myxocyprinus asiaticus, scientific name: Myxocyprinusasiaticus , taxonomic rank: Cypriniformes-Catostomidae, protection level: national second-class, distribution section: from Jinsha River to Yibin section in the upper reaches of the Yangtze River, suitable habitat: rapid gravel area, breeding characteristics: March and April.

[0033] In a further embodiment, the hierarchical storage architecture is a hybrid storage architecture, and the specific implementation method is as follows: Divide the standardized data set Into a basic information library, a monitoring and analysis library, an environmental association library, and an application service library; Among them, the basic information library is stored using a relational database: for structured data, define and build standardized data tables according to the mandatory attributes , and use the object code As the global primary key, design the field verification logic in combination with the unified verification standard Ensure the storage standardization and query accuracy of structured attributes such as protection level and distribution section, and adapt to high-concurrency transaction processing scenarios; the structured data is fish entities, monitoring points, etc.

[0034] The monitoring and analysis library is stored using a distributed file system: for unstructured monitoring data, divide and store shards based on the characteristics of the basin partition code of the object code , and centrally store the monitoring data in the same basin partition. At the same time, through the mandatory attribute A file index is established based on the monitoring point attributes. This enables rapid association between unstructured data and structured attributes, improving the efficiency of parallel reading and writing of large amounts of data. The unstructured monitoring data described in this embodiment includes eDNA sequencing sequences, remote sensing images, and acoustic detection audio, etc.

[0035] The environmental association library and application service library are stored using a graph database: fish entities, environmental factors, application scenarios, and teaching and research cases are mapped to knowledge graph nodes, with cross-class linkage logic. Based on the edge, through object encoding It enables unique node identification and cross-database association, supporting full-link association analysis and complex relationship mining of "species-habitat-scenario-case"; Based on object encoding Global uniqueness and unified verification standard Based on the standardized features, a unified data synchronization interface is constructed to achieve real-time data synchronization between relational databases, distributed file systems, and graph databases. During the synchronization process, data consistency is verified through metadata specifications to ensure the logical closed loop of "attribute-constraint association" when querying across databases. Data access latency is ≤50ms, and the accuracy of association queries is ≥99%.

[0036] When constructing the Yangtze River fish information database ontology, the association fusion satisfies the following fusion degree constraints: ; in, For the integration score function, For spatial matching degree, For time consistency, For species-environment association, , and All are weighting coefficients. ; , and It can be adaptively adjusted according to the business scenario: such as in a scientific research scenario. , , In law enforcement scenarios , , This is to ensure that the integration constraints are adapted to different business needs.

[0037] when At that time, the corresponding standardized dataset The ETL tool can then be used to complete the multi-database association and fusion and write the data into the Yangtze River fish information database ontology. This is a preset threshold for data fusion of fish species in the Yangtze River basin. Generally, This threshold is the empirically optimal threshold for multi-source data fusion of fish in the Yangtze River Basin. Data below this threshold may have problems such as spatial misalignment, temporal asynchrony, or ecological conflicts, and its inclusion in the database may affect the reliability and authority of the database.

[0038] In this embodiment, spatial matching degree The calculation formula is: , To monitor the area of ​​fish distribution covered by the monitoring points, This represents the total distribution area of ​​this fish species. The fish distribution area described in this embodiment. and total distribution area All data were obtained by combining historical distribution data of fish in the Yangtze River Basin, remote sensing image data, and coordinates of on-site monitoring points with GIS spatial analysis tools. The total distribution area vector layer was constructed based on the historical distribution sections and habitat ranges of fish entities. A preset monitoring buffer zone (e.g., a radius of 5km) was delineated with the latitude and longitude coordinates of the monitoring points as the center. The intersection area between the buffer zone and the total distribution area was calculated through spatial overlay analysis, which is the overlap. The area of ​​the total distribution area vector layer is the fish distribution, thereby realizing the automated and quantifiable calculation of spatial matching degree.

[0039] Time Consistency The calculation formula is: ,in, This represents the difference between the fish monitoring time and the environmental monitoring time. The standard time window for fish monitoring in the Yangtze River Basin (e.g., pre-set to 30 days). .

[0040] Species-environment association Based on the monitoring values ​​of suitable habitat attributes (such as water temperature, dissolved oxygen, and flow rate) and environmental factors (EV) of fish entities (FI), the degree of matching between environmental parameters and fish habitat requirements is calculated. The calculation formula is as follows: ;in, The number of key environmental factors (such as water temperature, dissolved oxygen, and flow rate, a total of 3). For the first Individual matching degree of environmental factors: when the monitored value is within the suitable range for fish. When the value exceeds the range, the decay rate decreases linearly according to the degree of deviation. If the water temperature monitoring value of the red-lipped barb is 8°C below the suitable lower limit, the temperature should be gradually reduced as the deviation increases. If the deviation is too large... .

[0041] Correspondingly, incremental synchronization adopts a hierarchical incremental synchronization mechanism based on metadata specifications, and the specific implementation method is as follows: Combining timestamps and integration scores Construct incremental triggering rules and extract standardized datasets. New and updated Valid data is automatically filtered to remove invalid and duplicate data. Partition synchronization involves the following steps: Basic information base: using the transaction incremental mechanism of relational databases, encoded in object form. Incremental writing of structured attributes is performed for unique primary keys, and the synchronization process follows a unified verification standard. Constrain and validate field compliance; Monitoring and Analysis Library: Based on Object Coding Based on the watershed partition code characteristics, the newly added unstructured monitoring data of the same watershed is batch synchronized to the corresponding storage nodes through the sharding incremental protocol of the distributed file system; Environment-related libraries and application service libraries: based on cross-class linkage logic Based on this, the newly added species-environment-scene relationships are synchronized through the node / edge incremental update interface of the graph database, and the synchronization process is performed using object encoding. Verify the uniqueness of the node.

[0042] It is worth mentioning that the intelligent early warning described in this embodiment is based on a threshold discrimination model specifically for the protection of rare fish species in the Yangtze River Basin. The specific model expression is as follows: ; in, For the early warning judgment results, This indicates that an alert has been triggered. This indicates that there was no prior warning. For the first The population abundance of rare fish species (unit: fish / square kilometer) can be determined based on monitoring data of fish individuals and required attributes. The distribution of the river sections was calculated. The population abundance warning threshold is set based on historical population data and protection levels of rare fish species in the Yangtze River Basin (such as the Chinese paddlefish and the Chinese sturgeon), and includes national first-class protected fish species. The target is 5 fish per square kilometer; this is a Class II protected fish species in China. The target is 10 tails per square kilometer; For the first Measured values ​​of key environmental factors in the habitat of rare fish species (including water temperature, dissolved oxygen, flow velocity, and mandatory attributes of environmental factors). ).

[0043] For the first The appropriate range of values ​​for key environmental factors corresponding to rare fish species.

[0044] In other words, At that time, an emergency warning for endangered rare fish species will be automatically triggered. or habitat degradation early warning .

[0045] It pushes early warning information containing warning type, fish code, abnormal area, and emergency response suggestions through multi-terminal access interface, with warning response delay ≤5 minutes and warning accuracy ≥95%.

[0046] To meet the needs of different application scenarios in the future, the teaching and research support module constructs four types of differentiated thematic data services based on user groups and application scenarios, specifically including: For teaching scenarios, a standardized species case library and knowledge point dataset based on a unified data dictionary are constructed. Structured teaching materials are formed according to fish species, habitat characteristics, and protection levels for teaching demonstrations and popular science use. For scientific research scenarios, it provides open standardized API interfaces and batch data export interfaces, supporting data retrieval, model calling and offline analysis by species, watershed and time dimensions; For engineering application scenarios, it provides a thematic dataset on ecological restoration, including data on suitable fish habitats, environmental factor thresholds, habitat distribution and restoration potential evaluation, to provide data support for watershed ecological governance projects; For management decision-making scenarios, it provides specialized data services for regulatory decision-making, integrating species distribution, monitoring points, environmental anomalies and early warning information to form visualized regulatory data support.

[0047] To meet compatibility requirements and improve customer experience, the data quality evaluation system... A customized multi-dimensional quantitative model for the Yangtze River Basin has the following iterative update mechanism: A quality dimension quantification model is created, decomposing data quality into five dimensions: completeness, accuracy, consistency, timeliness, and applicability, and calculating the quantification scores for each dimension as follows: , , , and ; Furthermore, completeness quantification score Required attribute The fill rate, such as If the result is , it indicates that the integrity standard has been met.

[0048] Accuracy Quantification Score for percentage, such as If the percentage is greater than the preset percentage, it indicates that the accuracy meets the standard.

[0049] Consistency Quantification Score To achieve a cross-database data consistency accuracy of ≥99.5% through Code encoding uniqueness verification, Const constraint attribute format and value range verification, and Link association logic verification.

[0050] Timeliness Quantification Score Based on the above-mentioned hierarchical incremental synchronization results, according to the standard that the synchronization delay of structured data is ≤10 minutes and the synchronization delay of unstructured data is ≤30 minutes, the proportion of data that meets the standard is counted. When the proportion is ≥98%, the timeliness meets the standard. Applicability Quantification Score The applicability of data services is assessed based on the satisfaction of user needs across four categories of thematic data services (teaching, research, engineering applications, and management decision-making). When the satisfaction rate is ≥90%, the applicability is considered satisfactory.

[0051] Based on the above individual items, a comprehensive quality evaluation function is constructed simultaneously. : ;in, , , , and These are the weighting coefficients; Quality evaluation will be conducted according to the predetermined evaluation cycle, taking into account user feedback questionnaires and requests for changes to monitoring requirements; when If any single dimension fails to meet the standard, the update process will be automatically triggered. This is a preset evaluation threshold.

[0052] The update process in this embodiment includes structural updates, content updates, and functional updates, which can be implemented in the following ways: Structural Update: Based on Cross-Class Linkage Logic Optimize database table structure and knowledge graph node / edge design to adapt to new monitoring indicators or application scenarios; Content Update: Corrected substandard data (labeled problematic data), supplemented missing fish entities, monitoring data, or thematic datasets, and updated unified verification standard constraints. The threshold and range of values; Feature Update: Optimized the interface response efficiency of the data service, added frequently requested functional modules, and simultaneously upgraded the threshold parameters of the intelligent early warning model; after the update, the quality evaluation was re-executed until the requirements were met. This ensures that the database continuously adapts to the dynamic needs of fish conservation, teaching, research, and management decisions in the Yangtze River Basin.

[0053] Example 2 Based on the construction method disclosed in Example 1, this example applies it to fish protection, teaching and research, ecological engineering construction, and basin management decision-making in the Yangtze River Basin.

[0054] A further disclosed system for constructing a Yangtze River fish information database, used to implement the construction method described in Example 1, includes: The first module is set to collect a full-dimensional dataset of fish species in the Yangtze River Basin. The full-dimensional dataset includes basic data, monitoring data, environmental correlation data, and application scenario data. The second module is configured to perform unified format conversion, missing value imputation, outlier removal, and ambiguity resolution on the full-dimensional dataset, establish a unified data dictionary and metadata specifications, and form a standardized dataset. ; The third module is configured to use a hierarchical storage architecture to store standardized datasets. The database is divided into a basic information database, a monitoring and analysis database, an environmental correlation database, and an application service database. ETL tools are used to achieve the correlation, fusion, and incremental synchronization of data from multiple databases, thus constructing the ontology of the Yangtze River fish information database. The fourth module is configured to deploy a data retrieval module, an intelligent early warning module, and a teaching and research support module based on the Yangtze River fish information database, providing multi-terminal access interfaces and data services; The fifth module is set up to establish a data quality evaluation system. The structure, content, and functional modules of the Yangtze River fish information database ontology are iteratively updated based on user feedback and monitoring needs.

Claims

1. A method for constructing a Yangtze River fish information database, characterized in that, Includes the following steps: A comprehensive dataset of fish species in the Yangtze River Basin was collected, which includes basic data, monitoring data, environmental correlation data, and application scenario data. The full-dimensional dataset undergoes unified format conversion, missing value imputation, outlier removal, and ambiguity resolution to establish a unified data dictionary and metadata specifications, thus forming a standardized dataset. ; A hierarchical storage architecture is adopted to store standardized datasets. The database is divided into a basic information database, a monitoring and analysis database, an environmental correlation database, and an application service database. ETL tools are used to achieve the correlation, fusion, and incremental synchronization of data from multiple databases, thus constructing the ontology of the Yangtze River fish information database. Based on the Yangtze River fish information database, a data retrieval module, an intelligent early warning module, and a teaching and research support module are deployed, providing multi-terminal access interfaces and data services; Establish a data quality evaluation system The structure, content, and functional modules of the Yangtze River fish information database ontology are iteratively updated based on user feedback and monitoring needs.

2. The method for constructing a Yangtze River fish information database according to claim 1, characterized in that, The unified data dictionary includes fish entities, monitoring locations, environmental factors, application scenarios, and teaching and research cases; correspondingly, the metadata specification adopts a standardized four-dimensional structure and incorporates customized coding and verification rules for the Yangtze River Basin, expressed in the following form: ； in, The object is encoded using a 14-bit unique encoding structure, specifically in the format "object type code - watershed partition code - serial number". Required properties customized according to object type. Required attribute A unified validation standard for all required attribute settings in the document. This refers to cross-class linkage logic between objects, which includes: spatial association, ecological association, scene association, and teaching and research association.

3. The method for constructing a Yangtze River fish information database according to claim 1, characterized in that, The tiered storage architecture is a hybrid storage architecture, and its specific implementation is as follows: Standardize the dataset The database is divided into a basic information database, a monitoring and analysis database, an environmental correlation database, and an application service database. The basic information database is stored using a relational database: for structured data, it is categorized by required attributes. Define and build standardized data tables, and encode them as objects. As a global primary key, combined with a unified verification standard Design the field validation logic; The monitoring and analysis library uses a distributed file system for storage: for unstructured monitoring data, it is based on object encoding. The watershed partition code features are used to divide and store data in fragments, centrally storing monitoring data from the same watershed partition, while also using mandatory attributes. Establish a file index for the monitoring point attributes; The environmental association library and application service library are stored using a graph database: fish entities, environmental factors, application scenarios, and teaching and research cases are mapped to knowledge graph nodes, with cross-class linkage logic. Based on the edge, through object encoding Implement unique node identification and cross-database association.

4. The method for constructing a Yangtze River fish information database according to claim 1, characterized in that, The association fusion satisfies the following fusion degree constraints: ； in, This is the integration score. For spatial matching degree, For time consistency, For species-environment association, , and All are weighting coefficients. ; when At that time, the corresponding standardized dataset The ETL tool can then be used to complete the multi-database association and fusion and write the data into the Yangtze River fish information database ontology. The preset threshold for fish data fusion in the Yangtze River Basin.

5. The method for constructing a Yangtze River fish information database according to claim 4, characterized in that, The incremental synchronization adopts a hierarchical incremental synchronization mechanism based on metadata specifications, and the specific implementation method is as follows: Combining timestamps and integration scores Construct incremental triggering rules and extract standardized datasets. New and updated Valid data is automatically filtered to remove invalid and duplicate data. Partition synchronization involves the following steps: Basic information base: using the transaction incremental mechanism of relational databases, encoded in object form. Incremental writing of structured attributes is performed for unique primary keys, and the synchronization process follows a unified verification standard. Constrain and validate field compliance; Monitoring and Analysis Library: Based on Object Coding Based on the watershed partition code characteristics, the newly added unstructured monitoring data of the same watershed is batch synchronized to the corresponding storage nodes through the sharding incremental protocol of the distributed file system; Environment-related libraries and application service libraries: based on cross-class linkage logic Based on this, the newly added species-environment-scene relationships are synchronized through the node / edge incremental update interface of the graph database, and the synchronization process is performed using object encoding. Verify the uniqueness of the node.

6. The method for constructing a Yangtze River fish information database according to claim 1, characterized in that, The intelligent early warning is implemented based on a threshold discrimination model specifically designed for the protection of rare fish species in the Yangtze River Basin. The specific model expression is as follows: ； in, For the early warning judgment results, This indicates that an alert has been triggered. This indicates that there was no warning. For the first Population abundance of rare fish species The threshold for population abundance warning. For the first Measured values ​​of key environmental factors in the habitat of rare fish species For the first The appropriate range of values ​​for key environmental factors corresponding to rare fish species.

7. The method for constructing a Yangtze River fish information database according to claim 1, characterized in that, The teaching and research support module constructs four types of differentiated thematic data services based on user groups and application scenarios, specifically including: To cater to teaching scenarios, a standardized species case library and knowledge point dataset based on a unified data dictionary are constructed, and structured teaching materials are formed according to fish species, habitat characteristics, and protection levels for teaching demonstrations and popular science use; For scientific research scenarios, it provides open standardized API interfaces and batch data export interfaces, supporting data retrieval, model calling and offline analysis by species, watershed and time dimensions; For engineering application scenarios, it provides a thematic dataset on ecological restoration, including data on suitable fish habitats, environmental factor thresholds, habitat distribution and restoration potential evaluation, to provide data support for watershed ecological governance projects; For management decision-making scenarios, it provides specialized data services for regulatory decision-making, integrating species distribution, monitoring points, environmental anomalies and early warning information to form visualized regulatory data support.

8. The method for constructing a Yangtze River fish information database according to claim 1, characterized in that, The data quality evaluation system A customized multi-dimensional quantitative model for the Yangtze River Basin has the following iterative update mechanism: A quality dimension quantification model is created, decomposing data quality into five dimensions: completeness, accuracy, consistency, timeliness, and applicability, and calculating the quantification scores for each dimension as follows: , , , and ; Construct a comprehensive quality evaluation function : ;in, , , , and These are the weighting coefficients; Quality evaluation will be conducted according to the predetermined evaluation cycle, taking into account user feedback questionnaires and requests for changes to monitoring requirements; when If any single dimension fails to meet the standard, the update process will be automatically triggered. This is a preset evaluation threshold.

9. An application for constructing a Yangtze River fish information database, characterized in that, It is applied to fish conservation, teaching and research, ecological engineering construction, and basin management decision-making in the Yangtze River Basin.

10. A system for constructing a Yangtze River fish information database, used to implement the construction method described in any one of claims 1 to 8, characterized in that, include: The first module is set to collect a full-dimensional dataset of fish species in the Yangtze River Basin. The full-dimensional dataset includes basic data, monitoring data, environmental correlation data, and application scenario data. The second module is configured to perform unified format conversion, missing value imputation, outlier removal, and ambiguity resolution on the full-dimensional dataset, establish a unified data dictionary and metadata specifications, and form a standardized dataset. ; The third module is configured to use a hierarchical storage architecture to store standardized datasets. The database is divided into a basic information database, a monitoring and analysis database, an environmental correlation database, and an application service database. ETL tools are used to achieve the correlation, fusion, and incremental synchronization of data from multiple databases, thus constructing the ontology of the Yangtze River fish information database. The fourth module is configured to deploy a data retrieval module, an intelligent early warning module, and a teaching and research support module based on the Yangtze River fish information database, providing multi-terminal access interfaces and data services; The fifth module is set up to establish a data quality evaluation system. The structure, content, and functional modules of the Yangtze River fish information database ontology are iteratively updated based on user feedback and monitoring needs.

Images