Scientific data set naming specification checking method, system and equipment and storage medium

Abstract

The invention relates to the technical field of scientific data management, in particular to a scientific data set naming specification checking method, system and device and a storage medium. The method comprises the following steps: acquiring a matched information carrier of a scientific data set to be checked; establishing a naming rule according to the matched information carrier; obtaining naming semantic information according to the matched information carrier, constructing a naming feature network graph, and obtaining a reference naming structure from the graph based on a naming rule; according to the actual naming information characteristics of the scientific data set to be checked, obtaining associated corpora in a matched information carrier, and generating description information of the scientific data set to be checked based on a preset cue word template; obtaining an actual naming structure based on the actual data structure and the description information of the scientific data set to be checked; and comparing the actual naming structure with the reference naming structure, generating a naming specification check report, and updating the naming with the normativity problem. According to the method, normative detection of scientific data set naming is realized, and the readability and understandability of scientific data set naming are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of scientific data management, and more specifically, to a method, system, device and storage medium for checking naming standards of scientific data sets. Background Art

[0002] Scientific data mainly involves the fields of natural sciences, engineering and technology sciences, etc. It is the data generated through basic research, applied research, experimental development, etc., as well as the original data and its derivative data obtained and used for scientific research activities through observation, monitoring, investigation, inspection and testing. Scientific data sets are scientific data sets organized and managed based on certain rules and standards. Scientific data sets generally have problems such as irregular naming, redundant naming and incomplete information. These problems lead to difficulties in scientific data retrieval, reduced scientific data quality, and low reuse rate of scientific data. In particular, as the main form of scientific and technological achievements, the basis for acceptance of scientific research projects, and an important carrier for achievement transformation, scientific data has diverse and complex naming requirements and purposes. Irregular naming greatly reduces the acceptance efficiency of data review and the level of data sharing. It is difficult for relevant personnel who accept data, review data quality, and use data to find target data based on data folders and data file names. In order to ensure the readability, traceability and consistency of data, there is an urgent need for a scientific data set naming standard inspection method to assist the above-mentioned relevant personnel, which can automatically generate and update naming rules according to the characteristics and needs of different disciplines, ensure the naming standardization and normalization of scientific data sets, and improve the efficiency and quality of data management. Summary of the invention

[0003] The present invention aims to overcome at least one defect (shortcoming) of the above-mentioned prior art and provide a method, system, device and storage medium for checking the naming standard of scientific data sets to realize the standardization detection of the naming of scientific data sets.

[0004] According to a first aspect of the present application, a method for checking naming conventions of scientific datasets is provided, comprising:

[0005] Obtaining a reference supporting information carrier of the scientific data set to be checked, obtaining a reference naming requirement according to the reference supporting information carrier, and obtaining a reference naming rule according to the reference naming requirement;

[0006] Acquire naming semantic information according to the reference supporting information carrier, construct a naming feature network diagram based on the naming semantic information, and acquire a reference naming structure from the naming feature network diagram based on the reference naming rule;

[0007] Acquire an actual supporting information carrier corresponding to the scientific data set to be checked, acquire an actual naming requirement according to the actual supporting information carrier, and acquire an actual naming rule according to the actual naming requirement;

[0008] Acquire a prompt word template according to the actual supporting information carrier, the actual naming requirement and the actual naming rule, and generate description information of the scientific data set to be checked based on the prompt word template;

[0009] Extracting actual naming information features from the scientific data set to be checked, and acquiring an actual naming structure based on the actual naming information features and the description information;

[0010] The actual naming structure and the reference naming structure are compared to obtain difference data, the naming of the scientific data set to be checked is updated according to the difference data, and a naming update log is recorded.

[0011] It is understandable that by comparing the generated reference naming structure with the actual naming structure, it is possible to locate irregularities in the actual naming. The user can update the irregular naming according to the naming standard check report to ensure the standardization of the naming of the scientific dataset.

[0012] Optionally, the obtaining of a reference supporting information carrier of the scientific data set to be checked, obtaining a reference naming requirement according to the reference supporting information carrier, and obtaining a reference naming rule according to the reference naming requirement specifically includes:

[0013] Acquire a reference naming requirement according to the reference matching information carrier, and extract the reference naming requirement related corpus from the reference naming requirement;

[0014] Acquire text content related to the naming of the scientific data set to be checked from the reference naming requirement related corpus to form a rule description text set;

[0015] Extracting keywords and topics from the rule description text set, obtaining associated semantic features of the keywords and topics through vector clustering analysis, and optimizing the rule description text set;

[0016] The optimized rule description text set is identified to obtain reference naming rule features, the structure of the reference naming rule is defined, and then the reference naming rule features are formatted into reference naming rules according to the structure of the reference naming rules, wherein the reference naming rules include naming fields and rules that the naming fields need to follow.

[0017] It is understandable that the content extracted from the semantic features of the reference naming requirements is converted into specific and operational small reference naming rules to ensure the systematicity, executability and applicability of the reference naming rules.

[0018] Optionally, the acquiring naming semantic information according to the reference matching information carrier specifically includes:

[0019] Extracting named entities and contextual semantic features associated with the named entities from the reference supporting information carrier to form a naming information reference set;

[0020] Match the naming field of the reference naming rule to the naming entity of the naming information reference set, and obtain key naming information of the naming field of the reference naming rule from the naming information reference set;

[0021] Constructing a semantic network, performing concept clustering on the key naming information based on the semantic network, and generating a reference concept cluster, wherein the reference concept cluster includes a concept-named entity feature set and a concept-semantic feature set;

[0022] The naming feature distribution is acquired according to the reference concept clustering, and the naming semantic information is generated according to the naming feature distribution.

[0023] It is understandable that by obtaining the naming semantic information of the supporting information carrier of the scientific dataset to be checked, the basis for naming can be obtained from the naming semantic information, so that the naming of the scientific dataset to be checked is more standardized and corresponds one-to-one with the content in the relevant supporting information carrier, which is convenient for data management personnel to identify and manage.

[0024] Optionally, constructing a naming feature network diagram based on the naming semantic information, and acquiring a reference naming structure from the naming feature network diagram based on the reference naming rule specifically includes:

[0025] Initialize the named feature network graph;

[0026] Taking the concept words in the concept-named entity feature set and the concept-semantic feature set in the reference concept cluster as nodes of the naming feature network diagram, and defining a semantic label and a feature description for each of the nodes;

[0027] According to the naming feature distribution corresponding to the reference concept cluster, establishing the connection edge of the naming feature network diagram, and logically or semantically associating the concept words;

[0028] Mapping the reference naming rule to the naming feature network diagram to obtain a reference naming structure that complies with the naming rule;

[0029] The reference naming structure includes the reference naming format of the scientific dataset to be checked, the acceptable content of the reference naming field, the reference hierarchical relationship of the folder where it is located, and the reference supporting information carrier.

[0030] It is understandable that establishing a naming feature network diagram can clearly clarify the relationship between various concepts and naming subjects, so as to improve the efficiency of naming information processing and the efficiency of establishing a reference naming structure.

[0031] Optionally, the acquiring the prompt word template according to the actual matching information carrier, the actual naming requirement and the actual naming rule specifically includes:

[0032] Defining an actual naming requirement input framework according to the actual matching information carrier;

[0033] Define the actual concept clustering input framework according to the actual naming rules;

[0034] Determine the information extraction task according to the actual supporting information carrier;

[0035] According to the actual naming requirement, the association between the actual concept clusters and the actual supporting information carriers, the description information generation task of the scientific data set to be checked is determined.

[0036] It is understandable that the setting of the prompt word template can generate auxiliary information for actual naming, and the defined input framework can simplify the input and operation of users during the naming process. At the same time, defining the input framework according to the actual supporting information carrier and the actual naming rules can ensure that the standardization of the actual naming meets the requirements to a certain extent, thereby guiding the creation of the actual naming structure.

[0037] Optionally, extracting actual naming information features from the scientific data set to be checked, and acquiring an actual naming structure based on the actual naming information features and the description information, specifically includes:

[0038] Extracting the actual naming information features according to the folder where the scientific data set to be checked is located, other file name keywords in the folder, and key fields in the description information;

[0039] Acquire the actual naming structure according to the hierarchy of the folder, the structure of the other file names and the key field;

[0040] The actual naming structure includes the actual naming format of the scientific data set to be checked, the actual content of the actual naming field, the actual hierarchical relationship of the folder where it is located, and the actual supporting information carrier.

[0041] It is understandable that this process obtains the actual naming structure of the scientific data set, clarifies the logic and source of the actual naming structure, and facilitates subsequent inspections to accurately locate the standardization issues of the actual naming structure.

[0042] Optionally, comparing the actual naming structure with the reference naming structure to obtain difference data, updating the naming of the scientific data set to be checked according to the difference data, and recording a naming update log specifically includes:

[0043] Comparing the actual naming structure with the reference naming structure in terms of naming format, possible content of naming fields, hierarchical relationship of folders, and supporting information carriers, to obtain difference data;

[0044] The name of the scientific data set to be checked is updated according to the difference data, and a name update log is recorded.

[0045] It is understandable that the naming of the scientific data set with irregular naming is changed according to the naming standard check report to ensure the standardization of the scientific data set.

[0046] According to a second aspect of the present application, a scientific dataset naming specification checking system is provided, comprising:

[0047] A reference naming structure module is constructed, which is used to obtain a reference supporting information carrier of a scientific data set to be checked, obtain a reference naming requirement according to the reference supporting information carrier, and obtain a reference naming rule according to the reference naming requirement; obtain naming semantic information according to the reference supporting information carrier, construct a naming feature network diagram based on the naming semantic information, and obtain a reference naming structure from the naming feature network diagram based on the reference naming rule;

[0048] Constructing an actual naming structure module, used to obtain the actual supporting information carrier corresponding to the scientific data set to be checked, obtain the actual naming requirement according to the actual supporting information carrier, and obtain the actual naming rule according to the actual naming requirement; obtain a prompt word template according to the actual supporting information carrier, the actual naming requirement and the actual naming rule, and generate description information of the scientific data set to be checked based on the prompt word template; extracting actual naming information features from the scientific data set to be checked, and obtaining an actual naming structure based on the actual naming information features and the description information;

[0049] The comparison module compares the actual naming structure with the reference naming structure to obtain difference data, updates the naming of the scientific data set to be checked according to the difference data, and records a naming update log.

[0050] According to a third aspect of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement a scientific data set naming specification checking method described in the first aspect above.

[0051] According to a fourth aspect of the present application, a computer storage medium is provided, on which a computer program is stored. When the computer program is executed, a method for checking the naming convention of a scientific dataset as described in the first aspect above is implemented.

[0052] Based on any of the above aspects, the embodiments of the present application provide a scientific data set naming standard checking method, system, device and storage medium, which can obtain the naming rules according to the supporting information carrier of the scientific data set to be checked, thereby obtaining the reference naming structure, and at the same time obtain the actual naming structure according to the actual naming information of the scientific data set to be checked, and by comparing the reference naming structure with the actual naming structure, find out the standard problems of the actual naming, and put forward the naming standard suggestions. Compared with the prior art, the beneficial effects of the present invention are:

[0053] (1) This method improves the efficiency of data management, reduces the time and human errors of manual inspection, and ensures the accuracy and consistency of data. The automated naming check and update process improves the efficiency and quality of scientific data management and reduces costs.

[0054] (2) Enhanced readability, understandability, and traceability of scientific data sets. The standardized naming structure enables researchers to easily trace the source and usage history of data, promotes data sharing and collaboration, and improves data reusability. The naming check method based on semantic association ensures the accuracy and adaptability of the naming specification and meets the dynamically changing naming needs in the field of scientific research. This method realizes the standardization detection of scientific data set naming, improves the efficiency of scientific data submission and review, and enhances the readability and understandability of scientific data set naming. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0056] Figure 1 A flow chart of a method for checking scientific dataset naming standards provided in this embodiment.

[0057] Figure 2 A flowchart for establishing a naming rule is provided in this embodiment.

[0058] Figure 3 A flow chart for obtaining the naming semantic information is provided in this embodiment.

[0059] Figure 4 A flow chart of obtaining a reference naming structure provided in this embodiment.

[0060] Figure 5 A flow chart of obtaining the prompt word template provided in this embodiment.

[0061] Figure 6 A flow chart for obtaining an actual naming structure is provided in this embodiment.

[0062] Figure 7 A flow chart of obtaining difference data by comparing an actual naming structure with a reference naming structure is provided in this embodiment.

[0063] Figure 8 A module diagram of a scientific data set naming standard checking system provided in this embodiment.

[0064] Fig. 9 A device structure diagram of an electronic device provided in this embodiment. DETAILED DESCRIPTION

[0065] The drawings of this application are only used for illustrative purposes and should not be construed as limiting the present application. In order to better illustrate the following embodiments, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product; it is understandable to those skilled in the art that some well-known structures and their descriptions in the drawings may be omitted.

[0066] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0067] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0068] This embodiment provides a technical solution that can solve the above-mentioned problem. The specific implementation methods of this application are described in detail below in conjunction with the accompanying drawings.

[0069] For example, a flowchart of a method for checking the naming standard of a scientific data set provided in an embodiment of the present application is provided. Figure 1 As shown, the method comprises the following steps:

[0070] S110, obtaining a supporting information carrier of a scientific data set to be checked, obtaining a reference naming requirement according to the reference supporting information carrier, and obtaining a reference naming rule according to the reference naming requirement;

[0071] It is understandable that the supporting information carrier of the scientific data set to be checked is specifically designated by business specialists such as scientific data, project management, and achievement management. Reference standards, review guidelines, etc. can be used as supporting information carriers of the scientific data set to be checked to clarify that the naming of the scientific data set to be checked must be based on the content of the corresponding supporting information carrier;

[0072] It is understandable that the supporting information carrier includes a structured supporting information carrier and an unstructured information carrier;

[0073] The structured supporting information carrier refers to the explanatory information related to data collection, submission, preservation, sharing, and utilization, which is filled in and submitted according to the prescribed template during the scientific data management process. It is usually stored in a database, spreadsheet or other file with a clear structure, such as metadata tables, data resource lists, data management plans, data submission plans, data quality reports, etc.

[0074] The unstructured supporting information carrier refers to the descriptive information related to data collection, submission, storage, sharing, and utilization that has not been strictly formatted, and is stored in a file with a non-standard structure but with text structure characteristics. The file with text structure characteristics can be customized by the user, such as an explanation of data sharing methods, data classification, data application case description, instructions for using data auxiliary tool software, etc.

[0075] like Figure 2 As shown, step S110 can be subdivided into the following steps: Figure 2 A flowchart for establishing a naming rule provided in this embodiment includes:

[0076] S111, obtaining a reference naming requirement according to the reference matching information carrier, and extracting the reference naming requirement related corpus from the reference naming requirement;

[0077] The reference naming requirement is obtained by self-definition according to the reference supporting information carrier. For example, it can obtain the purpose, application scenario, sharing and dissemination requirements of the scientific dataset to be checked, and the management and use specifications of the scientific dataset according to the reference supporting information carrier, and then propose the reference naming requirement of the scientific dataset to be checked. In specific implementation, it can be obtained by the institution, organization or individual who reviews or manages the scientific dataset, or it can be executed by a computer based on preset execution rules. Among them, the naming requirements include structured naming requirements and structured naming requirements;

[0078] For example, in the submission of scientific data for scientific and technological projects, the structured naming requirements refer to the scientific data set naming requirements proposed by the organization that reviews or accepts the scientific data, based on the acceptance requirements of the results or projects associated with the scientific data set, or based on the requirements of data sharing and data dissemination;

[0079] For example, the structured naming requirements are shown in Table 1:

[0080] Table 1 Example of structured naming requirements

[0081]

[0082]

[0083] Exemplarily, the unstructured naming requirement means that the naming of the scientific data set submitted by the project should be able to be found in the supporting information carrier, such as the supporting information carrier can be a project task book or a scientific report. In the folder hierarchy of the scientific data set submitted by the project, it should comply with the list of scientific data resources, or reflect the source of scientific data corresponding to the project task or research. Scientific data sets at different stages need to add corresponding identifiers in the file name. The naming of scientific data sets should be kept simple and unified, and no confusion or duplication of names should occur.

[0084] S112, acquiring text content related to the naming of the scientific data set to be checked from the reference naming requirement related corpus to form a rule description text set;

[0085] Remove stop words irrelevant to the scientific data from the reference naming requirement-related corpus, extract text content related to the naming rule through vector similarity analysis, including keywords and context sentences related to the reference naming rule, obtain key semantic information of the reference naming requirement, and generate a rule description text set;

[0086] Exemplarily, the rule description text set is in the following format, including the rule entry number and the rule text description information:

[0087] "Rule 1: The naming of scientific datasets must include the project number, task name, data type, generation date, and version number."

[0088] "Rule 2: The project number field must be unique and conform to the format [AZ]{3}[0-9]{4}."

[0089] “Rule 3: The Project Number field should be associated with the Task Name field, with each Project Number corresponding to a specific Task Name.”

[0090] “Rule 4: The combination of project number and task name should be consistent with that recorded in the project management plan.”

[0091] “Rule 5: All fields involved in dataset naming must be derived from the content specified in the data management plan and project task book.”

[0092] S113, extracting keywords and topics from the rule description text set, obtaining associated semantic features of the keywords and topics through vector clustering analysis, and optimizing the rule description text set;

[0093] In the rule description text set generated in step S112, there may be several constraint rules for the same field, which must be merged; or there may be several naming rules that rely on the same supporting information carrier, and each corresponding rule must be optimized. Through keyword extraction and vector clustering, the rule text description information of similar topics or similar features can be identified, so that the rule text description information can be merged to optimize the rule text description set.

[0094] Exemplarily, through keyword extraction and vector clustering, it can be clarified that the above rule text description information all has the "project number" feature, and can thus be merged into one rule "the named field content includes the project number field, extracted from the project task book, and the project number must be unique", which is equivalent to decomposing Rule 1 and clarifying the specific rule description of the project number field.

[0095] S114. Identify the optimized rule description text set to obtain reference naming rule features, define the structure of the reference naming rule, and then format the reference naming rule features into reference naming rules according to the structure of the reference naming rules; wherein the reference naming rules include naming fields and rules that the naming fields need to follow.

[0096] It is understandable that the identification of the optimized rule description text set specifically includes: identifying the naming rule features in each reference naming rule in the optimized rule description text set, and the naming rule features include the key fields, field formats, field orders and the correlation between the fields of the optimized rule description text set. By analyzing the naming rule features, it can be determined which different types of scientific data set naming these naming rules are applicable to, such as specific data types, data collection methods or data application methods.

[0097] Exemplarily, the following naming rule features are extracted from the optimized rule description text set:

[0098] Feature 1: The project number field must be unique and conform to the format of [AZ]{3}[0-9]{4}.

[0099] Feature 2: The name of a scientific dataset must contain the generation date in the format of YYYYMMDD.

[0100] Field association information: The project number field must be consistent with the project number field in the project task book, and the generation date field must be consistent with the date field in the data collection log.

[0101] It is understandable that the structure of the reference naming rules refers to a set of general rules and structured templates that define the naming format of scientific datasets, which can be configured and extended, can be extracted from information sources of specific naming requirements of the relevant neighborhood based on natural language processing, and can also be defined by the above-mentioned business specialists.

[0102] The structure of the reference naming rule may include but is not limited to the following elements:

[0103] [Field name: the specific field that needs to be included in the naming of scientific datasets];

[0104] [Field order: the order in which fields are arranged in the naming of scientific datasets];

[0105] [Field format: specific format requirements for each field (such as date format, number format, etc.)];

[0106] [Specific field restrictions: Specific restrictions on certain field values ​​(such as device models can only contain specific values)].

[0107] For example, the structure of the reference naming rule defined in a scientific data file is:

[0108] [Task name]_[Data type]_[Collection time]_[Device model]_[Phase ID].[File extension];

[0109] Based on the extracted naming rule features, a standardized template corresponding to the structure of the reference naming rule is constructed. This template should contain all key elements of the structure of the reference naming rule, such as field name, field format, field order, correlation between fields, and specific restrictions for each field.

[0110] Exemplarily, the standardized template corresponding to the reference naming rule structure is as follows:

[0111] The naming rule structure corresponds to the structure of the standardized template: [Task Name]_[Data Type]_[Collection Time]_[Device Model]_[Phase Identifier].[File Extension]

[0112] Specific field rules:

[0113] Task name: cannot contain spaces or special characters

[0114] Data type: Possible values ​​are {remote sensing images, table data, text data, historical atlas, video, photograph}

[0115] Collection time: The format is YYYYMMDD

[0116] Device Model: Allow only specific device models

[0117] Phase identification: Possible values ​​are {theoretical discovery, model verification, product development, test verification}

[0118] The reference naming rule features are extracted from the optimized rule description text set and mapped to the structure of the naming rule, ensuring that each rule in the structure of the reference naming rule can accurately reflect the reference naming requirements, thereby obtaining the reference naming rule.

[0119] S120, acquiring naming semantic information according to the reference supporting information carrier, constructing a naming feature network diagram based on the naming semantic information, and acquiring a reference naming structure from the naming feature network diagram based on the reference naming rule;

[0120] like Figure 3 As shown, Figure 3 A flow chart for obtaining the naming semantic information provided in this embodiment includes:

[0121] S211, extracting named entities and contextual semantic features associated with the named entities from the reference matching information carrier to form a naming information reference set;

[0122] It is understandable that, because the structured supporting information carriers are all written in accordance with certain templates, the contextual semantic features associated with the named entities can be extracted according to the original templates. For example, corresponding fields can be directly extracted from tabular files such as metadata tables and data resource lists as named entities.

[0123] For templated documents such as data management plans, data submission plans, and data quality reports, these documents follow specific templates and formats, and can be processed in blocks based on the templates. The documents can be divided into sections based on information such as chapter titles and paragraph identifiers. For example, the "Dataset Description" section in the data management plan can be automatically located and extracted using specific keywords or paragraph identifiers.

[0124] Based on the block processing, the named entities in each block are located using the known writing requirements in the template. The named entities usually include dataset names, variable names, indicator names, etc. For example, in the "Quality Control Measures (Source, Collection, Processing, Treatment Method, etc.)" section of the data quality report, the named entities in the "Data Source" field can be directly located and extracted.

[0125] Using the template structure, extract the contextual semantic features of the block or paragraph where the named entity is located. The contextual semantic features may include the definition, attribute description, and usage description of the key named entity. These information usually have clear writing requirements in the template. For example, in the "Sample Description" section of the data quality report, after extracting the named entity of "Sample Name", you can extract the contextual semantic features of the sample source, sample type, sample quantity, sample quality, etc. in the paragraph.

[0126] Since the structured supporting information carrier has a fixed format, the template rules and semantic analysis technology can be combined to improve the extraction accuracy of named entities and contextual semantic features. The named subject and its contextual semantic features are extracted using the specific format and specific fields specified in the template. For example, in the "Data Retention Strategy" section of the data management plan, the named subject and its contextual semantic features such as "Retention Period" and "Backup Strategy" can be accurately extracted through the format specified by the template.

[0127] For unstructured supporting information carriers, we can first extract the chapter and paragraph structure, extract the paragraph content, and use the pre-trained key named entity recognition model and custom named entity library to identify named entities from the text.

[0128] Preferably, the extraction of named entities can be completed in combination with a custom named entity library. The custom named entity library contains Chinese professional names involved in data classification, data sharing agreements, etc., which can generally be obtained from relevant standards, specifications, work guidelines and other documents. Set a context semantic feature window to extract a certain number of words or sentences before and after the named entity as context semantic features. Preferably, 5 words or 1-2 sentences before and after the named entity can be extracted as its context semantic information. Use NLP techniques, such as dependency syntactic analysis and word vector representation, to analyze semantic features in the context.

[0129] The extracted named entities and their contextual semantic features are integrated into a structured naming information reference set. The naming information reference set can be directly used for subsequent naming standard inspection and analysis. For example, in the data submission scheme, all the extracted named entities and their contextual features are integrated into a naming information reference set for comparison and reference during the naming standard inspection process. The inspection items corresponding to the naming rules are compared with the information in the naming information reference set to obtain naming reference suggestions.

[0130] For example, in the scientific data management business, a scientific data submission and review work is in progress. The supporting information carriers submitted by the data producer include the project task book (task number, project leader, research phase division, data collection requirements, etc.), the scientific data management plan (covering data types, data formats, data life cycle management, data quality control measures, etc.), and the data sharing agreement (describing the rights of data users, the scope of data sharing, data security, etc.). Named entities related to the content of scientific data naming fields are identified from the supporting information carriers, generally including professional terms, identifiers, task numbers, data types, etc. Named entities and associated contextual semantic features are extracted from the above supporting information carriers to obtain a naming information reference set consisting of named entities and associated contextual semantic features, for example:

[0131] [{"Named Entity":"Task Number T2024-RP-01",

[0132] "Context":"The task book stipulates that the data sources corresponding to task number T2024-RP-01 include sensor A and sensor B.",

[0133] "Associated semantic features": ["Data collection", "Sensor A", "Sensor B"]},

[0134] {"Named Entity":"Data Type: Time Series Data",

[0135] "Context": "The data management plan states that the format of time series data is CSV and the sampling timestamp is every second.",

[0136] "Associated semantic features": ["CSV format", "samples per second"]}.

[0137] S212, mapping the naming field of the reference naming rule to the naming entity of the naming information reference set, and obtaining key naming information of the naming field of the reference naming rule from the naming information reference set;

[0138] S213, constructing a semantic network, and performing concept clustering on the key naming information based on the semantic network to generate a reference concept cluster, wherein the reference concept cluster includes a concept-named entity feature set and a concept-semantic feature set;

[0139] When constructing a semantic network, each named entity in the naming information reference set is used as a node, and the connection between nodes is established using associated semantic features. Graph structure analysis technology is used to identify key nodes and important relationships in the semantic network. For example, named entities extracted from scientific data supporting information carriers, such as data field: "meteorological data", collection method: "satellite remote sensing", data source: "Gaofen Satellite 1", application field: "environmental monitoring", etc., constitute nodes of the semantic network. According to the association between named entities and the contextual semantic features of named entities, a semantic network is established to reflect the logical relationship between named entities of scientific data, thereby obtaining high-dimensional feature information of scientific data naming. Concept clusters can then be obtained through semantic vector clustering analysis.

[0140] For example, an ecological and environmental research project submitted meteorological data, soil data, and animal and plant resource distribution data. In order to unify the naming standards, the research team needs to extract high-dimensional semantic features from complex supporting information carriers. The extracted named entities include "soil sensor", "humidity sensor", "satellite remote sensing data", "soil data", "meteorological data", "meteorological station", "field observation station", etc. Each named entity is used as a semantic network node, and edges are added according to the relationship between named entities and contexts, such as associating data collection equipment with "'soil sensor', 'meteorological station', 'field observation station'", and associating data applications with "'soil data', 'meteorological data', 'environmental monitoring'".

[0141] Through concept clustering, we can get concept clusters, which include concept-concept semantic feature sets and concept-named entity feature sets, thus providing guidance in the naming process of scientific data sets. For example, after concept clustering, we get the concept cluster "satellite remote sensing data and ecological monitoring data", which includes the concept-semantic feature set:

[0142] "Satellite remote sensing data: large-scale surface change data collected by satellite sensors"

[0143] “Ecological monitoring data: data used to analyze and record the status of ecosystems, which may include vegetation coverage, distribution of animal habitats, etc.”

[0144] Includes concept-named entity feature set: "data source (satellite platform), data type (spectral data, image data), spatial resolution, data application scenario (ecological research, environmental protection)".

[0145] S214, acquiring naming feature distribution according to the reference concept clustering, and generating the naming semantic information according to the naming feature distribution;

[0146] For each concept cluster, the value with the highest frequency on the concept cluster dimension is calculated, which is the naming feature distribution of the concept cluster. For example, for the concept clusters "weather station and meteorological data", both contain temperature data, but their descriptive semantic features are different, including application fields, data sources, reasons for data collection, observation tasks, research tasks and other information. The descriptive semantic features (such as application fields, data sources, reasons for data collection, etc.) are converted into vector representations, and then these features are clustered using a hierarchical clustering algorithm to form more detailed sub-clusters. These sub-clusters show the naming feature distribution of different scientific data sets. For example, the high-frequency clustering features of concept cluster 1 may include: "urban meteorological monitoring", "urban meteorological station", "urban heat island effect", "heat distribution features", "dynamic temperature monitoring". Based on the high-frequency features of the sub-clusters, the naming feature distribution can be obtained, such as city A-temperature distribution B-monitoring station C-time range D-sampling frequency E.

[0147] like Figure 4 As shown, Figure 4 A flowchart for obtaining a reference naming structure provided in this embodiment includes:

[0148] S221, initializing the named feature network graph;

[0149] S222, using the concept words in the concept-named entity feature set and the concept-semantic feature set in the reference concept cluster as nodes of the naming feature network diagram, and defining a semantic label and feature description for each of the nodes;

[0150] S223, establishing connection edges of the naming feature network diagram according to the naming feature distribution corresponding to the reference concept cluster, and logically or semantically associating the concept words;

[0151] S224, mapping the reference naming rule to the naming feature network diagram to obtain a reference naming structure that complies with the reference naming rule;

[0152] The reference naming structure includes the reference naming format of the scientific dataset to be checked, the acceptable content of the reference naming field, the reference hierarchical relationship of the folder where it is located, and the reference supporting information carrier.

[0153] It is understandable that the naming feature network diagram is composed of the concept-named entity feature set and the concept-semantic feature set in the reference concept cluster as nodes, each node represents a concept, and is accompanied by a semantic label and feature description. The connecting edges in the naming feature network diagram are used to describe the semantic or logical association between concepts.

[0154] Specifically, the mapping process of the reference naming rule is as follows: after the naming feature network diagram is constructed, the named entities and semantic associations of the reference naming rule are mapped to the naming feature network diagram, and the nodes corresponding to each named entity are identified. For example, the data domain node can represent "genomic data", and the collection time node can represent the creation time of a specific data set.

[0155] It can be understood that by analyzing the node arrangement order of the naming feature network diagram, the combination of different fields is extracted to generate a unified naming format;

[0156] It is understandable that each node contains the optional value or content type of the node. In the named feature network diagram, the supporting information carrier pointed to by the edge connecting the nodes can provide a reference value for the named field, thereby obtaining the desirable content of the named field;

[0157] It is understandable that the named feature network graph consists of nodes and edges, where nodes represent named entities and edges represent semantic associations between named entities. In the folder hierarchy, higher-level folders usually contain broader categories or concepts, while lower-level folders represent more specific subsets. The folder hierarchy relationship of the scientific data set can be obtained based on the conceptual hierarchy of the nodes in the named feature network graph. As shown below, the folder hierarchy mapping rules may include:

[0158] Upper level nodes (root folders): Broader and more general entities in named features are usually mapped to the upper level of folders. For example, "data type" can be a root folder because it distinguishes different types of scientific datasets.

[0159] Lower level nodes (subfolders): More specific named entities will be mapped to lower level folders. For example, information such as collection time and data source can be further divided into specific subfolders under the root folder.

[0160] Leaf nodes (files): The most specific named entity combination, usually mapped to the actual file level.

[0161] On the other hand, for data sets with more data files and finer classification, they can be divided according to semantic association. In the named feature network diagram, semantic association determines the order of the folder hierarchy. Determine which entity should be located at which level based on the strength of semantic association or dependency. For example, if "data type" has a direct association with "acquisition time", then a subfolder of "acquisition time" can be created under the "data type" folder.

[0162] In the naming feature network diagram, nodes often point to related supporting information carrier nodes. For example, a node (such as a data type) may be connected to a project book, indicating that the acceptable content and reference basis of this field come from the description in the project book. By analyzing these supporting information carrier nodes, it is possible to determine which external documents or information sources the content of the named field comes from, and further standardize the value range and naming basis of the field.

[0163] S130, obtaining an actual supporting information carrier corresponding to the scientific data set to be checked, obtaining an actual naming requirement according to the actual supporting information carrier, and obtaining an actual naming rule according to the actual naming requirement;

[0164] S140, acquiring a prompt word template according to the actual supporting information carrier, the actual naming requirement and the actual naming rule, and generating description information of the scientific data set to be checked based on the prompt word template;

[0165] like Figure 5 As shown, step S140 can be subdivided into the following steps: Figure 5 A flow chart for obtaining the prompt word template provided in this embodiment includes:

[0166] S410, defining an actual naming requirement input framework according to the actual matching information carrier;

[0167] S420, defining an actual concept clustering input framework according to actual naming rules;

[0168] S430, determining the information extraction task according to the actual supporting information carrier;

[0169] S440: Determine the description information generation task of the scientific data set to be checked according to the actual naming requirement, the association between the actual concept cluster and the actual supporting information carrier.

[0170] It is understandable that the function of the prompt word template focuses on generating descriptive information, while the reference naming structure is generated by the system based on the reference naming rules and the naming feature network diagram. The reference naming structure is a naming format that complies with the specifications and is generated by the system according to standardized rules and structures, and is determined according to the nodes and semantic labels in the naming feature network diagram. This structure defines the specific form of naming of scientific data sets that complies with the specifications and is a naming standard model. The core role of the reference naming structure is to serve as a reference for comparison during the naming inspection process. When checking the scientific data set, the system will extract the actual naming structure and then compare it with the reference naming structure. Through this comparison, the non-compliant or incomplete parts of the actual naming structure can be identified, thereby generating difference data. These difference data reflect the degree of deviation of the naming of the scientific data set from the reference naming structure.

[0171] The prompt word template dynamically generates descriptive information of the scientific data set according to the actual situation, which is an expression of the actual naming information. The function of the prompt word template is to generate descriptive information of the scientific data set when the system automatically processes the actual naming supporting information carrier, the actual naming requirements and the actual naming rules. This is significantly different from the reference naming structure. The reference naming structure is a structured naming format that conforms to the standard naming rules, while the descriptive information generated by the prompt word template is generated based on the actual naming features, providing a basis for comparing and updating the naming.

[0172] It is understandable that the structural design of the prompt word template is mainly centered around the source of the naming requirement information, the naming rules, and the supporting information carriers of the scientific data set to be checked. The template consists of three key parts: the naming requirement field input framework, the concept clustering field framework, and the information extraction task definition. The naming requirement field input framework is used to capture the core information of the naming requirements, such as data type, data source, and usage scenario; the concept clustering field framework helps to capture the contextual information related to the naming rules in a structured manner, such as the data collection method, application field, etc. The information extraction task definition ensures that the key content related to naming is accurately obtained from the supporting information carriers.

[0173] The prompt word template framework covers multiple aspects required for naming scientific datasets, including data type and format, collection method, task background, application field, data source, data processing steps, and research purpose. Through these standardized prompt words, the template can guide users to systematically enter the key information of scientific datasets.

[0174] The task of generating the description information of the scientific dataset is accomplished by semantic analysis, corpus extraction, and information combination of the information input by the user. First, the system extracts key corpus from the naming requirement field of the prompt word template, and combines it with the concept clustering field to generate preliminary description information of the scientific dataset. These description information are then compared and optimized with the existing naming rules and supporting information carriers to ensure that the description information meets the requirements of the naming specification. Finally, the generated scientific dataset description information is used to construct the actual naming structure, and the results of the naming specification check are provided after comparison with the actual naming information.

[0175] The above steps introduce a prompt word template construction process, which is a dynamic process. Specifically, in step S410 and step S420, the design of the actual naming requirement input framework and the actual concept clustering input framework needs to dynamically reflect the requirements and rules that users actually encounter in the process of naming scientific data sets. Therefore, the framework should be adjusted multiple times according to real-time feedback and data changes.

[0176] Step S430 identifies information extraction tasks related to actual naming requirements by analyzing the actual naming supporting information carriers. This means that in this process, the system needs to perform semantic analysis, extract and organize information, so as to provide substantial support for subsequent actual naming requirements.

[0177] Specifically, the user refers to not only the scientific dataset creation / submission user, but also the data management / review user and system automation program. The user references are different at different stages of dataset submission, review and acceptance. The actual naming includes but is not limited to the following:

[0178] 1. In the acceptance of scientific and technological projects in a certain niche field, the data set submission and acceptance process has been standardized. Data file classification and naming will be carried out only after the data supporting information carrier associated with the data set has been reviewed and approved. The supporting information carrier will record all data resource descriptions in the scientific data set. According to the description, the naming of the data file and the hierarchical relationship of the data folder can be clarified. Data submitters can use prompt word templates to obtain naming information. However, in reality, data submitters will name and classify data files according to their own understanding. At this time, data management users or data receiving agencies need to obtain the actual description information of the scientific data set based on the naming information created by the data submitter;

[0179] 2. In non-standard data submission and acceptance, such as paper data submission and personal scientific data submission, the dataset file compression package and supporting information carrier may be submitted together. The naming, folder hierarchy, and supporting information carrier may not be standardized. In this case, it is necessary to obtain the dataset description information based on the specific naming requirements and the data archiving intentions of the data receiving organization.

[0180] It is understandable that the description information includes multiple fields, such as data type, data source, collection method, application field, etc. According to the structure and characteristic keywords of the prompt word template, description information with clear classification and naming reference can be obtained.

[0181] It is understandable that the generated structured description information will be compared with the reference naming rules to ensure that the generated description information meets the requirements of the reference naming rules. The key to this step is to emphasize "conformity" during the comparison process, using the reference naming rules as a reference to check the completeness and accuracy of the description information.

[0182] In actual naming, scientific data submitters will provide non-standard data classification and data naming. When the data receiving agency verifies the integrity and consistency of the descriptive information in the scientific data set based on the project task book, data management plan, etc., it will search for the corresponding related content of each data file from the supporting information carrier based on the naming information provided by the data submitter, so that the data reviewer can understand the data file.

[0183] For example, this embodiment provides a naming case:

[0184] A certain island ecological survey project submitted a set of infrared camera photos and related data. The data submitter named the folder "Island Survey Infrared Camera Photos" and named the data file "2021.5.8-Island Survey Data". The supporting information carriers such as the project task book and the data management plan specify the survey area, time, equipment used (such as infrared camera model), specific methods of data collection, and recorded content (such as climate conditions, species, etc.). Data auditors need to use large language models (LLMs) and prompt word templates to obtain descriptive information of scientific data sets from supporting information carriers (such as project task books, data management plans, etc.) and verify the standardization and consistency of file naming.

[0185] Generate precise prompts based on the file name to directly point to specific chapters or fields in the project task book, such as "Please find the 'Data Collection Methods and Equipment' section in the project task book to obtain relevant information about the infrared camera equipment and monitoring targets used in the island survey."

[0186] The large language model used the prompt words to search for the "Data Collection Methods and Equipment" section in the project brief and extracted the following key information: "Surveys of species that are rare in number, have special activity patterns, and are difficult to see in the wild should use infrared automatic digital cameras to obtain the distribution of mammals on various islands in a certain area. Monitoring content should include animal activity trajectories and changes in temperature and humidity."

[0187] Further information such as the observed object and observed content corresponding to the data set can be obtained.

[0188] Assuming that the naming rules require that the data file name contains the observation content and the data folder contains the observation object, then the data folder name "Island Survey Infrared Camera Photos" can be corrected to "Monitoring / Infrared Photos of the Distribution of Animals on Island in a Certain Region".

[0189] It must be noted that the comparison with the reference naming rules is to obtain the real and complete descriptive information corresponding to the actual naming provided by the data submitter. This process emphasizes "compliance", "completeness" and "availability", and does not ensure "accuracy" and "consistency". That is, it ensures that the actual naming structure generated based on the extracted data descriptive information can be compared with the reference naming structure. Because the reference naming structure is also established based on naming rules and supporting information carriers, the difference is that the reference naming structure is extracted and established from the supporting information carrier according to the naming rules; the actual naming structure extracts similar or related descriptive information from the supporting information carrier according to the actual naming characteristics, and classifies and structures the descriptive information according to the naming rules. The actual naming structure reflects the naming requirements of the data submitter, data classification methods and other contents. The reference naming structure reflects the reference naming rules. Through comparison, the actual naming structure can achieve a balance between accuracy and consistency, ensuring that the final actual naming structure can meet the specific needs of users while complying with the specifications.

[0190] Understandably, the establishment of prompt word templates can achieve the following effects:

[0191] Generate auxiliary information for actual naming: The prompt word template provides relevant prompts to users or systems to help generate descriptive information for scientific data sets and ensure the consistency of information extraction.

[0192] Simplify user input and operation: Based on actual naming requirements, actual concept clusters and actual supporting information carriers, the prompt word template provides an input framework for users or systems. This framework makes data input and operation processes more convenient and intuitive.

[0193] Ensure that the actual naming is in compliance with the requirements: The prompt word template can refer to the reference naming rules and generate targeted prompts based on the actual naming requirements to ensure that the actual naming of scientific data sets conforms to the established naming rules to a certain extent. This helps to reduce human errors and ensure the consistency and standardization of data naming.

[0194] Guiding the creation of the actual naming structure: During the creation of the actual naming structure, the prompt word template generates the required naming information for the user or system according to the actual needs and actual naming rules, helping the system or user to effectively guide the creation of the actual naming structure.

[0195] S150, extracting actual naming information features from the scientific data set to be checked, and acquiring an actual naming structure based on the actual naming information features and the description information;

[0196] like Figure 6 As shown, step S150 can be subdivided into the following steps: Figure 6 A flowchart for obtaining an actual naming structure provided in this embodiment includes:

[0197] S510, extracting the actual naming information features according to the folder where the scientific data set to be checked is located, other file name keywords in the folder, and key fields in the description information;

[0198] S520, acquiring the actual naming structure according to the hierarchy of the folder, the structure of the other file names and the key field;

[0199] The actual naming structure includes the naming format of the scientific data set, the acceptable content of the naming field, the hierarchical relationship of the folder where it is located, and the supporting information carrier on which the naming is based.

[0200] The system will scan the folder hierarchy and file name structure of the scientific dataset to identify key naming features in the file system. These features may include folder hierarchical relationships, specific identifiers contained in file names (such as task number, sampling time, data type, etc.), and file extension patterns. By scanning the file system, the system can preliminarily determine the logical organization structure of the scientific dataset. The extracted key naming features are matched with the information in the supporting information carrier of the scientific dataset. The information is usually stored in the metadata file or supporting description documents in the supporting information carrier of the scientific dataset, such as the data management plan, data submission plan, or project task book. The system will extract the corresponding fields from the information in the supporting information carrier based on the keywords in the folder and file names, such as task number, data type, data source, sampling time, etc. After extracting the corresponding fields, the system maps the fields with the fields in the actual naming to obtain the actual naming structure of the scientific dataset. This mapping process involves organizing the corresponding fields in the information in the supporting information carrier according to the hierarchical relationship of the actual naming structure, thereby forming a naming rule corresponding to the actual data structure. For example, the folder name may correspond to the task number or project phase, and the parts in the file name may correspond to fields such as data type and acquisition time.

[0201] Exemplarily, the folder hierarchy reflects the logical organization of the scientific data set to be examined, and is layered according to projects, tasks, or data types. For example, a project's parent folder may represent the summary data of the entire project, and the subfolders represent specific tasks or phases of data. Example:

[0202] The parent folder is named "ProjectData_2024" and the subfolder is named "Task_T2024-RP-01". The system recognizes that the parent folder is at the project level and the subfolder is at the task level;

[0203] The file name structure reflects the information within the data file, such as the task number, sensor type, data type, etc., to ensure that the file name accurately describes the content and source of the file. Example:

[0204] The file name is "T2024-RP-01_SensorA_TempData_20240501.csv", and the system extracts "Task number T2024-RP-01", "Sensor type SensorA", "Data type TempData", and "Collection time 20240501".

[0205] The supporting information carriers (such as project task books, data management plans, etc.) provide field definitions, field contents, field association information and formats in the naming. The system needs to ensure that the fields in the actual naming are consistent with these specifications. Example:

[0206] The supporting information carrier specifies that the task number format is "T+year+task number", and the data type can be "temperature data, time series data", etc. The system checks whether the task number "T2024-RP-01" and the data type "TempData" in the file name meet these requirements.

[0207] Based on the above information, generate the actual naming structure, for example:

[0208] Parent folder: project level, such as "ProjectData_2024"

[0209] Subfolder: Task level, such as "Task_T2024-RP-01"

[0210] The file name format is: "Task number_Sensor type_Data type_Date.csv", such as "T2024-RP-01_SensorA_TempData_20240501.csv"

[0211] The actual naming structure of scientific datasets generated by this process will ensure that the names of each file and folder clearly reflect its content and source, which is convenient for project management, data sharing and subsequent data analysis. This structured naming method will help quickly locate and identify datasets during project acceptance and data review.

[0212] S160: Compare the actual naming structure with the reference naming structure to obtain difference data, update the naming of the scientific data set to be checked according to the difference data, and record a naming update log.

[0213] like Figure 7 As shown, step S160 can be subdivided into the following steps: Figure 7 A flow chart of obtaining difference data by comparing an actual naming structure with a reference naming structure provided in this embodiment includes:

[0214] S610, comparing the differences between the actual naming structure and the reference naming structure in terms of naming format, possible content of the naming fields, hierarchical relationship of the folders, and supporting information carriers, and obtaining difference data;

[0215] S620: Update the name of the scientific data set to be checked according to the difference data, and record a name update log.

[0216] For example, in the actual naming structure, the folder name is "ProjectData_2024" and the subfolder name is "Task_T2024-RP-01", while the reference naming structure requires the folder name to be "2024_ProjectData" and the subfolder name to be "RP_01_Task". The differences here include different naming formats (such as the use of underscores and hyphens), changes in hierarchical relationships (such as the naming order of folders and subfolders), and the accuracy of field content (such as the format of task numbers).

[0217] During the difference comparison process, special attention is paid to the following features: whether the naming format is consistent and meets the preset specifications; whether the hierarchical relationship meets the hierarchical requirements in the reference naming structure; whether the field content is complete and accurate; whether the description information can clearly reflect the data type and collection source; and whether the associated data files are named in accordance with the standardized format and structure. Through detailed comparison of these aspects, inconsistencies in the actual naming structure can be identified.

[0218] According to the difference data, the specific differences between the reference naming structure and the actual naming structure can be found. Preferably, a naming specification check report can be generated. The generated naming specification check report includes two main parts: naming specification suggestions and scientific data set file structure description table. The naming specification suggestions describe in detail the specific differences between the actual naming structure and the reference naming structure, and put forward improvement suggestions, such as changing the folder naming format, adjusting the hierarchical structure, etc.; the scientific data set file structure description table lists the file structure of the scientific data set to be checked, including the naming of each folder and file and its corresponding data description, to help reviewers understand and adjust the naming of scientific data sets to ensure that they comply with the prescribed naming specifications. The report is automatically generated by the system, combining the difference analysis results and specification suggestions, and provides detailed guidance for the review and acceptance of scientific data.

[0219] In this embodiment, there is the following situation: a research institution finds during the process of checking the naming specifications of scientific data sets that the existing naming requirement information source and structure, naming rule structure and naming entity library fail to fully cover details such as folder hierarchy, field content, data description, etc., resulting in actual data naming not meeting the specifications. The research institution can provide user feedback. Based on user feedback and specification inspection reports, the naming requirement information source structure, naming rule structure and naming entity library need to be optimized and corrected.

[0220] Preferably, regarding the optimization of the naming requirement information source structure, if it is found during the scientific data set review process that the original naming requirement structure fails to cover the actual results acceptance and data sharing requirements, the naming requirement information source and structure will be optimized, and new information fields such as "results acceptance requirements", "data sharing standards" and "project management guidelines" will be added. Specifically, the newly added fields include "results acceptance requirements" to describe the acceptance criteria that scientific data must meet, such as "data integrity verification" and "data quality assessment"; "data sharing standards" specify the data sharing format and protocol, such as "data should comply with ISO 19115 standards"; "project management guidelines" list project-specific naming requirements, such as "folder hierarchy should include 'project name / dataset / sensor type'".

[0221] Preferably, for the optimization of the naming rule structure, if the audit finds that the original naming rule structure is not detailed enough and fails to cover the actual project acceptance and data dissemination requirements, the naming rule structure document is updated during the optimization process, including detailed provisions for the folder structure and specific requirements for the data field content. For example, a new "folder structure" rule is added, stipulating that "folders should be organized in the format of 'project name / data category / collection method / timestamp'"; and "field content" requirements, such as "data fields must contain information such as 'collection device model', 'collection time' and 'data type'". These modifications ensure that the naming rules can accurately reflect the actual naming needs and standards.

[0222] Preferably, for the optimization of the named entity library, if based on the specification inspection report, it is found that specific named entities related to project management and acceptance of results are missing in the entity library. In response to this, new named entities such as "project number", "data quality level" and "acceptance person in charge" are added, and existing named entities are revised. For example, "project number" is added as a named entity and defined as "unique identifier in project acceptance"; "data quality level" is added as a named entity, and the quality categories of labeled data are specified as "high", "medium" and "low"; "acceptance person in charge" is added as a named entity to indicate the data management personnel responsible for the review. Through these optimizations, the named entity library can more comprehensively support the needs of the actual data review and acceptance process, and improve the effectiveness and accuracy of the system.

[0223] Based on the same inventive concept, the present application embodiment also provides a scientific data set naming standard checking system, such as Figure 8 As shown, a scientific dataset naming standard checking system model diagram is provided, including:

[0224] A reference naming structure construction module 311, an actual naming structure construction module 312, and a comparison module 313. Among them:

[0225] The reference naming structure construction module 311 is used to obtain the reference supporting information carrier of the scientific data set to be checked, obtain the reference naming requirement according to the reference supporting information carrier, and obtain the reference naming rule according to the reference naming requirement; obtain the naming semantic information according to the reference supporting information carrier, construct the naming feature network diagram based on the naming semantic information, and obtain the reference naming structure from the naming feature network diagram based on the reference naming rule;

[0226] In this embodiment, the reference naming structure building module 311 can be used to execute Figure 1 As shown in step S110 and step S120, for the specific description of the reference naming structure building module 311, reference can be made to the description of step S110 and step S120.

[0227] The actual naming structure building module 312 is used to obtain the actual supporting information carrier corresponding to the scientific data set to be checked, obtain the actual naming requirement according to the actual supporting information carrier, and obtain the actual naming rule according to the actual naming requirement; obtain the prompt word template according to the actual supporting information carrier, the actual naming requirement and the actual naming rule, and generate the description information of the scientific data set to be checked based on the prompt word template; extract the actual naming information features from the scientific data set to be checked, and obtain the actual naming structure based on the actual naming information features and the description information;

[0228] In this embodiment, the actual naming structure building module 312 can be used to perform Figure 1 As shown in step S130, step S140 and step S150, for the specific description of the construction parameter actual naming structure module 312, reference can be made to the description of step S130, step S140 and step S150.

[0229] The comparison module 313 is used to compare the actual naming structure with the reference naming structure to obtain difference data, update the naming of the scientific data set to be checked according to the difference data, and record a naming update log;

[0230] In this embodiment, the comparison module 313 can be used to perform Figure 1 As shown in step S160, for the detailed description of the comparison module 313, reference may be made to the description of step S160.

[0231] Based on the same inventive concept, this embodiment also provides an electronic device, Fig. 9 The structure diagram of the electronic device of this embodiment is shown, which includes a memory 21 and a processor 22. The memory 21 stores computer-readable instructions, and the processor 22 executes the computer-readable instructions to implement the scientific data set naming specification checking method of this embodiment.

[0232] Preferably, the electronic device further includes a bus 23 and a communication interface 24 , and the processor 22 , the communication interface 24 and the memory 21 are connected via the bus 23 .

[0233] The memory 21 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 24 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used. The bus 23 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus 23 may be divided into an address bus, a data bus, a control bus, etc. (not fully drawn in the figure).

[0234] The processor 22 may be an integrated circuit chip with signal processing capability. In the specific implementation process, each step in the above method embodiment may be completed by the hardware integrated logic circuit or software instructions in the processor 22. The above processor 22 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, which may implement or execute the methods, steps, and logic diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or the processor 22 may also be any conventional processor 22, etc. The steps of the method disclosed in conjunction with the embodiments of the present invention may be directly embodied as being executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in a decoding processor. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 21, and the processor 22 reads the information in the memory 21 and completes the steps of the method of the above embodiment in combination with its hardware.

[0235] An embodiment of the present invention further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by the processor 22, the computer-executable instructions prompt the processor 22 to implement the above-mentioned optical cable buried depth detection method. The specific implementation can be found in the embodiment, which will not be repeated here.

[0236] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the 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, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0237] Obviously, the above embodiments of the present application are merely examples for clearly illustrating the technical solution of the present application, and are not intended to limit the specific implementation methods of the present application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the claims of the present application shall be included in the protection scope of the claims of the present application.

Claims

1. A method for checking the naming convention of scientific data sets, characterized in that: include: Obtaining a reference supporting information carrier of the scientific data set to be checked, obtaining a reference naming requirement according to the reference supporting information carrier, and obtaining a reference naming rule according to the reference naming requirement; Acquire naming semantic information according to the reference supporting information carrier, construct a naming feature network diagram based on the naming semantic information, and acquire a reference naming structure from the naming feature network diagram based on the reference naming rule; Acquire an actual supporting information carrier corresponding to the scientific data set to be checked, acquire an actual naming requirement according to the actual supporting information carrier, and acquire an actual naming rule according to the actual naming requirement; Acquire a prompt word template according to the actual supporting information carrier, the actual naming requirement and the actual naming rule, and generate description information of the scientific data set to be checked based on the prompt word template; Extracting actual naming information features from the scientific data set to be checked, and acquiring an actual naming structure based on the actual naming information features and the description information; The actual naming structure and the reference naming structure are compared to obtain difference data, the naming of the scientific data set to be checked is updated according to the difference data, and a naming update log is recorded.

2. A method for checking scientific dataset naming standards according to claim 1, characterized in that: The obtaining of the reference supporting information carrier of the scientific data set to be checked, obtaining the reference naming requirement according to the reference supporting information carrier, and obtaining the reference naming rule according to the reference naming requirement specifically includes: Acquire a reference naming requirement according to the reference matching information carrier, and extract the reference naming requirement related corpus from the reference naming requirement; Acquire text content related to the naming of the scientific data set to be checked from the reference naming requirement related corpus to form a rule description text set; Extracting keywords and topics from the rule description text set, obtaining associated semantic features of the keywords and topics through vector clustering analysis, and optimizing the rule description text set; The optimized rule description text set is identified to obtain reference naming rule features, the structure of the reference naming rule is defined, and then the reference naming rule features are formatted into reference naming rules according to the structure of the reference naming rules; wherein the reference naming rules include naming fields and rules that the naming fields need to follow.

3. A method for checking scientific dataset naming standards according to claim 1, characterized in that: The obtaining of naming semantic information according to the reference supporting information carrier specifically includes: Extracting named entities and contextual semantic features associated with the named entities from the reference supporting information carrier to form a naming information reference set; Match the naming field of the reference naming rule to the naming entity of the naming information reference set, and obtain key naming information of the naming field of the reference naming rule from the naming information reference set; Constructing a semantic network, performing concept clustering on the key naming information based on the semantic network, and generating a reference concept cluster, wherein the reference concept cluster includes a concept-named entity feature set and a concept-semantic feature set; The naming feature distribution is acquired according to the reference concept clustering, and the naming semantic information is generated according to the naming feature distribution.

4. A method for checking scientific dataset naming standards according to claim 3, characterized in that: The step of constructing a naming feature network diagram based on the naming semantic information and acquiring a reference naming structure from the naming feature network diagram based on the reference naming rule specifically includes: Initialize the named feature network graph; Taking the concept words in the concept-named entity feature set and the concept-semantic feature set in the reference concept cluster as nodes of the naming feature network diagram, and defining a semantic label and a feature description for each of the nodes; According to the naming feature distribution corresponding to the reference concept cluster, establishing the connection edge of the naming feature network diagram, and logically or semantically associating the concept words; Mapping the reference naming rule to the naming feature network diagram to obtain a reference naming structure that complies with the reference naming rule; The reference naming structure includes the reference naming format of the scientific dataset to be checked, the acceptable content of the reference naming field, the reference hierarchical relationship of the folder where it is located, and the reference supporting information carrier.

5. A method for checking scientific dataset naming standards according to claim 1, characterized in that: The obtaining of the prompt word template according to the actual supporting information carrier, the actual naming requirement and the actual naming rule specifically includes: Defining an actual naming requirement input framework according to the actual matching information carrier; Define the actual concept clustering input framework according to the actual naming rules; According to the actual supporting information carrier information extraction task; According to the actual naming requirement, the association between the actual concept clusters and the actual supporting information carriers, the description information generation task of the scientific data set to be checked is determined.

6. A method for checking scientific dataset naming standards according to claim 1, characterized in that: The extracting of actual naming information features from the scientific data set to be checked and obtaining the actual naming structure based on the actual naming information features and the description information specifically includes: Extracting the actual naming information features according to the folder where the scientific data set to be checked is located, other file name keywords in the folder, and key fields in the description information; Acquire the actual naming structure according to the hierarchy of the folder, the structure of the other file names and the key field; The actual naming structure includes the actual naming format of the scientific data set to be checked, the actual content of the actual naming field, the actual hierarchical relationship of the folder where it is located, and the actual supporting information carrier.

7. A method for checking scientific dataset naming standards according to claim 1, characterized in that: Comparing the actual naming structure with the reference naming structure to obtain difference data, updating the naming of the scientific data set to be checked according to the difference data, and recording a naming update log, specifically including: Comparing the actual naming structure with the reference naming structure in terms of naming format, possible content of naming fields, hierarchical relationship of folders, and supporting information carriers, to obtain difference data; The name of the scientific data set to be checked is updated according to the difference data, and a name update log is recorded.

8. A scientific data set naming standard checking system, characterized in that: include: A reference naming structure module is constructed, which is used to obtain a reference supporting information carrier of a scientific data set to be checked, obtain a reference naming requirement according to the reference supporting information carrier, and obtain a reference naming rule according to the reference naming requirement; obtain naming semantic information according to the reference supporting information carrier, construct a naming feature network diagram based on the naming semantic information, and obtain a reference naming structure from the naming feature network diagram based on the reference naming rule; Constructing an actual naming structure module, used to obtain the actual supporting information carrier corresponding to the scientific data set to be checked, obtain the actual naming requirement according to the actual supporting information carrier, and obtain the actual naming rule according to the actual naming requirement; obtain a prompt word template according to the actual supporting information carrier, the actual naming requirement and the actual naming rule, and generate description information of the scientific data set to be checked based on the prompt word template; Extracting actual naming information features from the scientific data set to be checked, and acquiring an actual naming structure based on the actual naming information features and the description information; The comparison module compares the actual naming structure with the reference naming structure to obtain difference data, updates the naming of the scientific data set to be checked according to the difference data, and records a naming update log.

9. An electronic device, comprising a memory and a processor, characterized in that: The memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement a scientific data set naming specification checking method according to any one of claims 1 to 7.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, a method for checking the naming convention of a scientific data set according to any one of claims 1 to 7 is implemented.

Images