A multi-source smart metering device database construction method, system, device and medium

Abstract

The application discloses a multi-source intelligent measurement equipment database construction method, system, device and medium, and belongs to the technical field of semantic database construction. The method is as follows: obtaining the standard clause text of the multi-source intelligent measurement equipment and splitting the standard clause text into an equipment semantic word set; obtaining the correlation degree and co-occurrence conditional probability between the equipment attribute words and the equipment technical index words to screen target index entities; obtaining the technical attribute correlation degree between any two target index entities and classifying the target index entities; obtaining corresponding technical index parameter data and mapping the technical index parameter data to a preset probability interval to obtain a standardized measurement result; and taking the technical index categories as indexes, correlatively storing the standardized measurement result, the target index entities and corresponding standard clause texts to obtain a standard database. Therefore, by implementing the application, semantic alignment and unified measurement of regional technical standards can be realized, and a quantifiable and reproducible database support is provided for equipment adaptation analysis and compliance decision-making.

Description

Technical Field

[0001] This invention relates to the field of semantic database construction technology, and in particular to a method, system, device and medium for constructing a database for multi-source intelligent measurement devices. Background Technology

[0002] With the increasing demand for multi-source applications of intelligent measurement equipment in fields such as the power Internet of Things, energy metering, and industrial monitoring, standardization organizations in different regions have successively formulated technical standards for communication methods, measurement accuracy, power consumption levels, and security constraints. These standards are usually published in natural language text form and have significant differences in the classification of technical indicators, the expression of evaluation systems, and the conditions for the constraints to take effect. This makes semantic alignment and unified measurement across standard systems a key prerequisite for equipment adaptation analysis and compliance decisions.

[0003] Existing standard database construction methods mostly focus on keyword extraction, structuring indicator fields, or classifying and storing hard and soft indicators in standard clause texts. While these methods can achieve basic retrieval and information aggregation, they have the following shortcomings: First, they lack semantic alignment capabilities across standard systems. Different standards do not completely agree on the sets of technical indicators they focus on for similar equipment, and the same technical indicator may be expressed using different terms in different standards. Existing methods struggle to identify the semantic equivalence between such heterogeneous expressions, leading to ineffective alignment of the database when comparing multiple standards. Second, they lack a unified measurement mechanism for heterogeneous evaluation systems. Various standards commonly use multiple heterogeneous evaluation systems, such as numerical, hierarchical, or rule-based systems, to describe the same technical indicator. Existing methods only formally store the indicators without establishing a unified measurement mechanism across evaluation systems, resulting in incomparability between indicators in various standards during engineering comparisons. Third, they lack the ability to recognize the contextual effects of constraint semantics. Technical constraints in standard clauses usually depend on the applicable environment or supplementary clauses to determine their effective scope. Existing methods fail to effectively identify the contextual conditions for the effectiveness of constraint semantics, easily misjudging conditional constraints as universally applicable requirements, leading to information distortion in the database.

[0004] Therefore, there is an urgent need for a solution that can achieve semantic alignment and unified measurement across standard systems, and provide quantifiable and reproducible database support for multi-source adaptation analysis and compliance decision-making of intelligent measurement devices. Summary of the Invention

[0005] This invention provides a method, system, device, and medium for constructing a database of multi-source intelligent measurement devices. By implementing this invention, semantic alignment and unified measurement of regional technical standards can be achieved, providing quantifiable and reproducible database support for device adaptation analysis and compliance decision-making.

[0006] This invention provides a method for constructing a database of multi-source intelligent measurement devices, comprising: Obtain the standard clause text of the multi-source intelligent measurement device, and decompose the standard clause text into a device semantic term set; Obtain the correlation degree and co-occurrence conditional probability between device attribute words and device technical indicator words in the device semantic word set, so as to filter a number of target indicator entities from the device semantic word set; Obtain the technical attribute correlation degree between any two target indicator entities, and classify several target indicator entities according to the technical attribute correlation degree to obtain several technical indicator categories and a subset of target indicator entities under each technical indicator category. Obtain the technical indicator parameter data corresponding to each target indicator entity in the target indicator entity subset, and map the technical indicator parameter data to a preset probability interval to obtain a standardized measurement result; Using the technical indicator category as an index, the standardized measurement results, the target indicator entity, and the corresponding standard clause text are associated and stored to obtain a standard database of the multi-source intelligent measurement equipment.

[0007] This invention, through obtaining the standard clause text of intelligent measurement equipment and splitting it into a device semantic term set, preserves the complete contextual structure of the technical constraints in the standard clauses, avoiding semantic distortion caused by isolated keyword extraction. It obtains the correlation degree and co-occurrence conditional probability between device attribute words and device technical indicator words in the device semantic term set to filter target indicator entities from the set. This enables accurate identification of technical indicators limited by constraints in the standard text, improving the accuracy of standard indicator extraction. Furthermore, it obtains the technical attribute correlation degree between target indicator entities and classifies them based on this correlation, allowing the same technical attribute expressed in different terms in standards from different sources to be automatically identified and aggregated into a technical indicator category. Finally, it obtains the technical indicator parameter data corresponding to each target indicator entity and maps this data to a preset probability interval. This invention unifies heterogeneous evaluation systems into measurable probability interval expressions, achieving semantic alignment and unified measurement across standard systems. Finally, using technical indicator categories as indexes, it associates and stores standardized measurement results, target indicator entities, and corresponding standard clause texts, constructing a quantifiable and reproducible multi-source standard database indexed by technical indicator categories. Compared to existing standard databases that only enable keyword retrieval and field-based storage, failing to express the correspondence between technical indicator categories across standard systems and comparable measurements under heterogeneous evaluation systems, this invention, through fully automated processing from semantic word set splitting, target indicator entity screening, technical category aggregation to probability interval mapping, constructs a standard database capable of supporting equipment adaptation analysis, compliance decision-making, and multi-source indicator difference quantification calculation, providing an objective and reproducible data foundation for the multi-source engineering applications of intelligent measurement equipment.

[0008] Furthermore, obtain the standard terms and conditions for multi-source intelligent measurement equipment, specifically: The standard source data, historical standard sample library data, and equipment capability description data of multi-source intelligent measurement equipment are obtained as standard clause texts; wherein, the standard source data includes equipment standard data and measurement operation specification data of intelligent measurement equipment; the historical standard sample library data includes historical indicator clauses and historical parameter records; and the equipment capability description data includes product manuals, test reports, and equipment technical white papers.

[0009] This approach utilizes standard source data, historical standard sample library data, and equipment capability description data from multi-source intelligent measurement equipment as standard clause text. The standard source data includes equipment standard data and measurement operation specification data for intelligent measurement equipment, covering mandatory requirements and recommended guidelines at different levels and application scenarios from both the perspectives of equipment technical specifications and operational process specifications. This ensures that the subsequently extracted target indicator entities have a complete source foundation across standard systems and operational stages. The historical standard sample library data includes historical indicator clauses and historical parameter records, effectively utilizing the indicator descriptions and parameter value information accumulated in historical standard texts. This provides rich prior knowledge for calculating the correlation between equipment attribute terms and equipment technical indicator terms, and estimating parameter distribution. The equipment capability description data includes product manuals, test reports, and equipment technical white papers, introducing real technical parameter expressions at the actual engineering capability level of the equipment. This ensures that the standard clause text not only covers the normative requirements from the source but also encompasses descriptive data at the actual capability level of the equipment, providing supplementary support from actual engineering data for subsequent target indicator entity screening and technical indicator parameter data extraction, thus enhancing the correspondence between the standard indicator extraction results and engineering practice.

[0010] Furthermore, the correlation degree and co-occurrence conditional probability between device attribute words and device technical indicator words in the device semantic word set are obtained, specifically as follows: Obtain the index positions of equipment attribute terms and equipment technical indicator terms in the standard clause text, and obtain the maximum number of attribute terms between equipment attribute terms and equipment technical indicator terms; The relevance is calculated based on the index position and the maximum number of attribute words. The number of times that equipment attribute terms and equipment technical indicator terms appear together in the same paragraph in the standard clause text is obtained, and the co-occurrence conditional probability is calculated by combining the number of times that equipment technical indicator terms appear alone.

[0011] By obtaining the index positions of equipment attribute terms and equipment technical indicator terms in the standard clause text, and by determining the maximum number of attribute terms between them, the word spacing between attribute terms and indicator terms can be accurately located at the text position level. This ensures that the calculation of the correlation is based on the objective and quantifiable foundation of the relative position of words. The correlation is calculated based on the index positions and the maximum number of attribute terms. By combining the word position difference with the maximum number of attribute terms, the semantic tightness of attribute terms and indicator terms in the text context can be effectively measured. The closer the positions, the higher the correlation, fully reflecting... The strength of the modifying relationship between the attributive constraints and the central indicator in the standard clauses was measured, so that word pairs that co-occur at a distance would not receive excessively high association scores. The number of times equipment attribute words and equipment technical indicator words co-occur in the same paragraph in the text of the standard clauses was obtained. Combined with the number of times equipment technical indicator words appeared alone, the co-occurrence conditional probability was calculated. Through paragraph-level co-occurrence statistics and conditional probability modeling, the dependence of equipment technical indicator words on the occurrence of equipment attribute words can be measured from the discourse structure level. This makes the co-occurrence conditional probability not only reflect the co-occurrence frequency between words, but also characterize the conditional dependency relationship between attribute constraints and indicators.

[0012] Furthermore, the technical attribute correlation degree between any two of the target indicator entities is obtained, and the target indicator entities are classified according to the technical attribute correlation degree, specifically as follows: For any two target indicator entities, obtain the semantic vectors corresponding to the two target indicator entities respectively, and obtain the semantic similarity between the two target indicator entities based on the semantic vectors; Obtain the common device attribute words of the two target indicator entities under the same constraints, and calculate the association confidence difference between the two target indicator entities based on the common device attribute words, so as to obtain the association similarity between the two target indicator entities based on the association confidence difference. The technical attribute correlation degree of the two target indicator entities is calculated based on the semantic similarity and the association similarity. When the technical attribute correlation degree is greater than the preset category determination threshold, the two target indicator entities are classified into the same technical indicator category.

[0013] By acquiring the semantic vectors corresponding to any two target indicator entities and calculating their semantic similarity, this approach captures the inherent connections between different indicator entities at the semantic level, avoiding misjudgments caused by relying solely on surface character matching and enhancing the robustness of identifying the same technical attribute across standard systems. Furthermore, by acquiring the common device attribute words of the two target indicator entities under the same constraints and calculating the difference in association confidence based on these common device attribute words to obtain association similarity, the co-occurrence information of attribute words can effectively correct the bias of pure semantic similarity under the same technical constraints, improving the engineering interpretability of association judgments. Finally, by calculating the technical attribute association degree of the two target indicator entities based on semantic similarity and association similarity, and classifying the two target indicator entities into the same technical indicator category when the technical attribute association degree is greater than the category determination threshold, this approach achieves complementary fusion of semantic information and constraint information. This ensures that classification decisions simultaneously consider the consistency of indicator semantics and the matching of engineering constraints, avoiding classification bias or overly coarse-grained aggregation caused by single-dimensional judgments.

[0014] Furthermore, after obtaining several technical indicator categories and a subset of target indicator entities under each of the technical indicator categories, the method further includes: Obtain the source set of each target indicator entity under each technical indicator category, and combine it with the total number of sources in the standard clause text to obtain the coverage breadth of each technical indicator category; The constraint strength of each technical indicator category is obtained by obtaining the number of times each target indicator entity and equipment attribute word co-occurs in the standard clause text under each technical indicator category, and by combining the total number of times each target indicator entity occurs in the standard clause text. The generality score of each technical indicator category is obtained based on the coverage breadth and the constraint strength, so as to filter out general technical indicator categories based on the generality score and the preset generality threshold, and the general technical indicator categories are used as the updated technical indicator categories.

[0015] By obtaining the source set of each target indicator entity under each technical indicator category and combining it with the total number of sources in the standard clause text, the coverage breadth of each technical indicator category can be obtained. This allows for an objective quantification of the distribution range of each technical indicator category across different standard sources, enabling a numerical representation of the degree of consensus on indicators across standard systems. Furthermore, by obtaining the number of times each target indicator entity and equipment attribute term co-occurs with each technical indicator category in the standard clause text and combining it with the total number of occurrences of each target indicator entity in the standard clause text, the constraint strength of each technical indicator category can be obtained. This effectively measures the relationship between indicator categories and technical constraints on the co-occurrence frequency dimension. The tightness between conditions ensures that the constraint strength reflects both the standard attention and the strictness of the technical constraints of the indicator. A generality score is obtained for each technical indicator category based on the coverage breadth and the constraint strength. General technical indicator categories are then selected based on the generality score and a preset generality threshold, and these general technical indicator categories are used as updated technical indicator categories. This achieves a weighted fusion evaluation of coverage breadth and constraint strength. Through a dual-dimensional comprehensive scoring and threshold screening mechanism, general indicator categories that are widely adopted in multi-source standards and have strong technical constraint consensus can be automatically identified from all technical indicator categories, while secondary categories with single sources or sparse constraints are filtered out.

[0016] Furthermore, the technical indicator parameter data is mapped to a preset probability interval to obtain standardized measurement results, specifically as follows: The evaluation system type of the technical indicator parameter data is obtained, and the mapping features of the technical indicator parameter data are extracted according to the evaluation system type; wherein, for numerical technical indicator parameter data, the numerical range and constraint direction are extracted as mapping features; for graded technical indicator parameter data, grade terms are extracted as mapping features; and for rule-based technical indicator parameter data, applicable conditions are extracted as mapping features. Based on the mapping features, the posterior probability of the technical indicator parameter data being mapped to a preset probability interval is calculated using Bayesian classification. The probability interval with the largest posterior probability is then used as the mapping result of the technical indicator parameter data. Based on the mapping result, the technical indicator parameter data is converted into the interval identifier or interval value range of the corresponding probability interval to obtain a standardized measurement result.

[0017] By acquiring the evaluation system type of the technical indicator parameter data, and extracting the mapping features of the technical indicator parameter data according to the evaluation system type, the most representative mapping features can be extracted for different evaluation system types, ensuring that the original constraint semantics of various types of data are fully preserved in the subsequent mapping process. Based on the mapping features, the posterior probability of the technical indicator parameter data being mapped to a preset probability interval is calculated using Bayesian classification, and the probability interval with the highest posterior probability is taken as the mapping result. Then, based on the mapping result, the technical indicator parameter data is converted into the interval identifier or interval numerical range of the corresponding probability interval, resulting in a standardized measurement result. Thus, under the Bayesian framework, the optimal interval matching is achieved through the maximum posterior probability criterion, uniformly mapping heterogeneous original expressions to a standardized probability interval system and directly outputting the interval identifier or numerical range. This not only preserves the quantitative support of the posterior probability calculation process for the reliability of the mapping, but also makes the standardized results more intuitive and operable, significantly improving the direct comparability and engineering practicality of measurement results across standard systems.

[0018] Furthermore, using the technical indicator category as an index, the standardized measurement results, the target indicator entity, and the corresponding standard clause text are associated and stored to obtain a multi-source standard database of the intelligent measurement equipment, specifically: Each of the obtained technical indicator categories will be used as the primary key of the database index; For each of the aforementioned technical indicator categories, obtain the corresponding standard clause text, target indicator entity, and standardized measurement results; The index primary key is associated with and stored in relation to the standard clause text, the target indicator entity, and the standardized metric to obtain a multi-source standard database for the intelligent measurement equipment.

[0019] By using each of the obtained technical indicator categories as the primary key of the database index, and associating the primary key with the standard clause text, the target indicator entity, and the standardized metric, a complete association link is constructed from the source of the original standard clause to the result of indicator entity extraction and then to the unified probability interval metric value. This achieves an organic unity between the accurate tracing of the standard clause, the accurate positioning of the target indicator entity, and the reproducibility of the standardized metric result, so that any subsequent query operation can simultaneously obtain the original semantic information, entity representation information, and quantitative comparison information.

[0020] Another embodiment of the present invention provides a multi-source intelligent measurement equipment database construction system, including: a device semantic word set acquisition module, a target indicator entity screening module, a technical indicator category classification module, and a standardized measurement result acquisition and database construction module; The device semantic term set acquisition module is used to acquire the standard clause text of multi-source intelligent measurement devices and split the standard clause text into device semantic term sets. The target indicator entity filtering module is used to obtain the correlation degree and co-occurrence conditional probability between equipment attribute words and equipment technical indicator words in the equipment semantic word set, so as to filter a number of target indicator entities from the equipment semantic word set. The technical indicator category classification module is used to obtain the technical attribute correlation degree between any two target indicator entities, so as to classify several target indicator entities according to the technical attribute correlation degree, and obtain several technical indicator categories and a subset of target indicator entities under each technical indicator category. The standardized measurement result acquisition module is used to acquire the technical indicator parameter data corresponding to each target indicator entity in the target indicator entity subset, and map the technical indicator parameter data to a preset probability interval to obtain the standardized measurement result. The database construction module is used to associate and store the standardized measurement results, the target indicator entities and the corresponding standard clause texts using the technical indicator categories as indexes, so as to obtain a standard database of the multi-source intelligent measurement devices.

[0021] This invention, through obtaining the standard clause text of intelligent measurement equipment and splitting it into a device semantic term set, preserves the complete contextual structure of the technical constraints in the standard clauses, avoiding semantic distortion caused by isolated keyword extraction. It obtains the correlation degree and co-occurrence conditional probability between device attribute words and device technical indicator words in the device semantic term set to filter target indicator entities from the set. This enables accurate identification of technical indicators limited by constraints in the standard text, improving the accuracy of standard indicator extraction. Furthermore, it obtains the technical attribute correlation degree between target indicator entities and classifies them based on this correlation, allowing the same technical attribute expressed in different terms in standards from different sources to be automatically identified and aggregated into a technical indicator category. Finally, it obtains the technical indicator parameter data corresponding to each target indicator entity and maps this data to a preset probability interval. This invention unifies heterogeneous evaluation systems into measurable probability interval expressions, achieving semantic alignment and unified measurement across standard systems. Finally, using technical indicator categories as indexes, it associates and stores standardized measurement results, target indicator entities, and corresponding standard clause texts, constructing a quantifiable and reproducible multi-source standard database indexed by technical indicator categories. Compared to existing standard databases that only enable keyword retrieval and field-based storage, failing to express the correspondence between technical indicator categories across standard systems and comparable measurements under heterogeneous evaluation systems, this invention, through fully automated processing from semantic word set splitting, target indicator entity screening, technical category aggregation to probability interval mapping, constructs a standard database capable of supporting equipment adaptation analysis, compliance decision-making, and multi-source indicator difference quantification calculation, providing an objective and reproducible data foundation for the multi-source engineering applications of intelligent measurement equipment.

[0022] Another embodiment of the present invention provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of the multi-source intelligent measurement device database construction method of the present invention.

[0023] Another embodiment of the present invention provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to perform the steps of the multi-source intelligent measurement device database construction method of the present invention. Attached Figure Description

[0024] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating an embodiment of the multi-source intelligent measurement device database construction method provided by the present invention; Figure 2 This is a schematic diagram of another embodiment of the multi-source intelligent measurement device database construction system provided by the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0028] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0031] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0032] See Figure 1 To address the problem of constructing a multi-source intelligent measurement device database in existing technologies, an embodiment of the present invention provides a method for constructing a multi-source intelligent measurement device database, comprising steps S1 to S5, the specific steps of which are as follows: S1. Obtain the standard clause text of the multi-source intelligent measurement device, and split the standard clause text into a device semantic term set; Specifically, based on the inherent chapter numbers, clause numbers, and punctuation marks in the standard clause text, the entire text is segmented and divided into multiple independent character blocks. Based on semantic boundary features such as verbs, modal verbs, and conjunctions, each character block is further divided into the smallest indivisible semantic character blocks. For each semantic character block obtained from the segmentation, a domain dictionary-enhanced sequence labeling algorithm is used, combined with a domain terminology dictionary and a constraint trigger word dictionary constructed from historical standard sample libraries, to perform semantic role labeling, identify equipment attribute words, equipment technical indicator words, and behavioral verbs, and finally summarize them to obtain the equipment semantic word set.

[0033] S2. Obtain the correlation degree and co-occurrence conditional probability between equipment attribute words and equipment technical indicator words in the equipment semantic word set, so as to select several target indicator entities from the equipment semantic word set; Specifically, the index positions of equipment attribute terms and equipment technical indicator terms in the standard clause text are obtained, and the maximum number of attribute words between the equipment attribute terms and equipment technical indicator terms is obtained. The correlation degree is calculated based on the index positions and the maximum number of attribute words. The closer the index positions and the fewer the number of attribute words, the greater the correlation degree. The number of times the equipment attribute terms and equipment technical indicator terms co-occur in the same paragraph in the standard clause text is obtained. Combined with the number of times the equipment technical indicator terms appear alone, the co-occurrence conditional probability is calculated. Based on the correlation degree and the co-occurrence conditional probability, several target indicator entities are selected from the equipment semantic word set.

[0034] S3. Obtain the technical attribute correlation degree between any two target indicator entities, and classify several target indicator entities according to the technical attribute correlation degree to obtain several technical indicator categories and a subset of target indicator entities under each technical indicator category. Specifically, for any two target indicator entities, the semantic vectors of the two target indicator entities are obtained to determine the semantic similarity between them; common device attribute words of the two target indicator entities under the same constraints are obtained, and the association confidence difference between the two target indicator entities is calculated based on the common device attribute words to determine the association similarity between them, with a smaller association confidence difference indicating a higher association similarity; the technical attribute correlation degree of the two target indicator entities is calculated based on the semantic similarity and the association similarity, and when the technical attribute correlation degree is greater than a preset category determination threshold, the two target indicator entities are classified into the same technical indicator category; S4. Obtain the technical indicator parameter data corresponding to each target indicator entity in the target indicator entity subset, and map the technical indicator parameter data to a preset probability interval to obtain a standardized measurement result. The process involves obtaining the evaluation system type of the technical indicator parameter data, and extracting mapping features of the technical indicator parameter data based on the evaluation system type. Specifically, for numerical technical indicator parameter data, the numerical range and constraint direction are extracted as mapping features; for graded technical indicator parameter data, grade terms are extracted as mapping features; and for rule-based technical indicator parameter data, applicable conditions are extracted as mapping features. These mapping features are then input into a Bayesian classifier, and, combined with reference samples from a historical standard sample library, the posterior probability of the technical indicator parameter data mapped to each preset probability interval is calculated. The probability interval with the highest posterior probability is determined as the mapping result for that technical indicator parameter data. Based on the mapping result, the technical indicator parameter data is converted into interval identifiers or interval numerical ranges corresponding to the probability intervals, resulting in a standardized measurement result.

[0035] S5. Using the technical indicator category as an index, the standardized measurement results, the target indicator entity and the corresponding standard clause text are associated and stored to obtain a standard database of the multi-source intelligent measurement equipment. The obtained technical indicator categories are used as the primary keys of the database index. For each technical indicator category, the standard clause text corresponding to the category, the target indicator entity belonging to the category, and the standardized measurement results obtained after mapping of each target indicator entity are obtained. The primary keys of the index are associated with the standard clause text, the target indicator entity, and the standardized measurement results for storage, resulting in a standard database with technical indicator categories as the core index, supporting quantitative calculation of technical indicator differences between multiple source standards and regional compliance analysis.

[0036] This invention, through obtaining the standard clause text of intelligent measurement equipment and splitting it into a device semantic term set, preserves the complete contextual structure of the technical constraints in the standard clauses, avoiding semantic distortion caused by isolated keyword extraction. It obtains the correlation degree and co-occurrence conditional probability between device attribute words and device technical indicator words in the device semantic term set to filter target indicator entities from the set. This enables accurate identification of technical indicators limited by constraints in the standard text, improving the accuracy of standard indicator extraction. Furthermore, it obtains the technical attribute correlation degree between target indicator entities and classifies them based on this correlation, allowing the same technical attribute expressed in different terms in standards from different sources to be automatically identified and aggregated into a technical indicator category. Finally, it obtains the technical indicator parameter data corresponding to each target indicator entity and maps this data to a preset probability interval. This invention unifies heterogeneous evaluation systems into measurable probability interval expressions, achieving semantic alignment and unified measurement across standard systems. Finally, using technical indicator categories as indexes, it associates and stores standardized measurement results, target indicator entities, and corresponding standard clause texts, constructing a quantifiable and reproducible multi-source standard database indexed by technical indicator categories. Compared to existing standard databases that only enable keyword retrieval and field-based storage, failing to express the correspondence between technical indicator categories across standard systems and comparable measurements under heterogeneous evaluation systems, this invention, through fully automated processing from semantic word set splitting, target indicator entity screening, technical category aggregation to probability interval mapping, constructs a standard database capable of supporting equipment adaptation analysis, compliance decision-making, and multi-source indicator difference quantification calculation, providing an objective and reproducible data foundation for the multi-source engineering applications of intelligent measurement equipment.

[0037] In one embodiment, obtaining the standard terms and conditions text of a multi-source intelligent measurement device includes step S201, as follows: S201. Obtain standard source data, historical standard sample library data, and equipment capability description data of multi-source intelligent measurement equipment as standard clause text; wherein, the standard source data includes equipment standard data and measurement operation specification data of intelligent measurement equipment; the historical standard sample library data includes historical indicator clauses and historical parameter records; the equipment capability description data includes product manuals, test reports, and equipment technical white papers; When acquiring the standard clause text, the system first identifies region tags based on the issuing organization, standard number, language type, and document metadata in the standard source data, and stores the region tags along with the standard number, clause number, and original text content. Then, under each region tag, the standard clause text is parsed to obtain a structured record with region fields. Data from different regions is stored using the same field format, including region, standard number, clause number, target indicator entity, constraint description, evaluation type, and probability interval. Specifically, when splitting the standard clause text into a device semantic term set, the entire text is first segmented and divided according to the inherent chapter number, clause number, and punctuation marks in the standard document, splitting the standard clause text into multiple independent character blocks; then, the character blocks are further refined and extracted according to semantic boundaries; based on the semantic features of verbs, modal words, and conjunctions, the character blocks are split into multiple indivisible semantic words to form the initial device semantic term set; A domain-enhanced sequence labeling algorithm is used to perform constraint semantic segmentation on character blocks to filter target indicator entities from the device semantic vocabulary set. Based on a historical standard sample library (containing historical indicator terms and parameter records), a domain terminology dictionary, constraint trigger word dictionary, and constraint strength dictionary for intelligent measurement devices are constructed. A maximum matching algorithm is used to identify candidate technical phrases from the character blocks. The candidate technical phrases and their context are input into the sequence labeling model, and each candidate phrase is labeled with the semantic roles of action verbs, constraint conditions, and target indicator entities. According to constraint template matching rules and the positional relationships between phrases, action verbs, target indicator entities, and constraint conditions belonging to the same technical constraint are bound into a complete technical indicator description. For cases where the same phrase may correspond to multiple semantic roles, the final semantic role is determined based on its occurrence probability in the historical standard sample library, its distance from constraint trigger words, and its co-occurrence frequency with adjacent target indicator entities. Thus, technical indicator descriptions with clearly defined semantic roles are output as candidate target indicator entities. In practical applications, such as the statement "supports broadband carrier communication" in standard clause texts: the action verb is "support", the equipment technical specification is "broadband", and the equipment attribute is "carrier communication".

[0038] This invention employs standard source data, historical standard sample library data, and equipment capability description data from multi-source intelligent measurement equipment as standard clause text. The standard source data includes equipment standard data and measurement operation specification data for intelligent measurement equipment, covering mandatory requirements and recommended guidelines at different levels and application scenarios from both the equipment's technical specifications and operational process specifications dimensions. This ensures that the subsequently extracted target indicator entities have a complete source foundation across standard systems and operational stages. The historical standard sample library data includes historical indicator clauses and historical parameter records, effectively utilizing the accumulated data in historical standard texts. The descriptions of indicators and parameter values ​​provide rich prior knowledge for subsequent calculations of the correlation between equipment attribute terms and equipment technical indicator terms, as well as for parameter distribution estimation. The equipment capability description data includes product manuals, test reports, and equipment technical white papers, which can introduce real technical parameter expressions at the level of actual engineering capabilities of the equipment. This ensures that the standard clause text not only covers the normative requirements in the normative sources but also includes descriptive data at the level of actual equipment capabilities. This provides supplementary support for subsequent selection of target indicator entities and extraction of technical indicator parameter data, enhancing the correspondence between the standard indicator extraction results and engineering practice.

[0039] In one embodiment, obtaining the correlation degree and co-occurrence conditional probability between device attribute words and device technical indicator words in the device semantic word set includes steps S301 to S303, each step of which is as follows: S301. Obtain the index positions of equipment attribute words and equipment technical indicator words in the standard clause text respectively, and obtain the maximum number of attribute words between equipment attribute words and equipment technical indicator words; In this context, for any equipment attribute term and equipment technical indicator term, there are no other objects between them. Therefore, the distance between the two terms is analyzed; the closer the distance, the greater the correlation. Thus, the index positions of the equipment attribute term and equipment technical indicator term in the standard clause text are obtained, and the number of each attribute term between the equipment attribute term and equipment technical indicator term is obtained, with the largest number being recorded as the maximum number of attribute terms. The correlation is negatively correlated with the difference in index positions between the equipment attribute term and equipment technical indicator term, and with the maximum number of attribute terms between the equipment attribute term and equipment technical indicator term.

[0040] S302. Calculate the relevance degree based on the index position and the maximum number of attribute words; S303. Obtain the number of times that equipment attribute terms and equipment technical indicator terms appear together in the same paragraph in the standard clause text, and calculate the co-occurrence conditional probability by combining the number of times that equipment technical indicator terms appear alone. The specific formula for calculating the degree of correlation is as follows: in, The probability of equipment technical specifications and equipment attribute terms appearing in the same paragraph; The probability of technical specifications of equipment appearing in the text; To prevent zero constant; This represents the conditional probability of equipment technical indicator terms and equipment attribute terms appearing in the same paragraph in historical standard sample database data.

[0041] For any equipment technical indicator term and equipment attribute term, the probability function for determining whether the equipment technical indicator term belongs to the judgment condition of the equipment attribute term is as follows: in, The degree of correlation between equipment attribute terms and equipment technical indicator terms; The conditional probability of equipment technical indicator terms and equipment attribute terms appearing in the same paragraph in historical standard sample database data; Let the probability that a technical indicator term belongs to a condition for judging equipment attribute terms be denoted as the first probability. For the probability function, a constraint threshold is set. The number of equipment technical indicator terms belonging to the target indicator entity and the number of equipment technical indicator terms in the historical standard database are counted. The ratio of the two is used as the constraint threshold. If the probability that an equipment technical indicator term belongs to the equipment attribute term judgment condition is greater than the constraint threshold, it means that the corresponding equipment technical indicator term belongs to the target indicator entity.

[0042] This invention, through obtaining the index positions of equipment attribute terms and equipment technical indicator terms in the standard clause text, and obtaining the maximum number of attribute terms between them, can accurately locate the word interval distance between attribute terms and indicator terms at the text position level. This ensures that the calculation of the correlation is based on the objective and quantifiable foundation of the relative position of words. The correlation is calculated based on the index positions and the maximum number of attribute terms. By combining the word position difference with the maximum number of attribute terms, the semantic tightness of attribute terms and indicator terms in the text context can be effectively measured; the closer the positions, the higher the correlation. This reflects the strength of the modifying relationship between the attributive constraints and the central indicator in the standard clause, ensuring that word pairs that co-occur accidentally at a distance do not receive excessively high association scores. The number of times equipment attribute words and equipment technical indicator words co-occur in the same paragraph of the standard clause text is obtained, and combined with the number of times equipment technical indicator words appear alone, the co-occurrence conditional probability is calculated. Through paragraph-level co-occurrence statistics and conditional probability modeling, the dependence of equipment technical indicator words on being mentioned under the condition of the appearance of equipment attribute words can be measured from the discourse structure level. This makes the co-occurrence conditional probability not only reflect the co-occurrence frequency between words, but also characterize the conditional dependency relationship between attribute constraints and indicators.

[0043] In one embodiment, the technical attribute correlation degree between any two target indicator entities is obtained, and the target indicator entities are classified according to the technical attribute correlation degree, including steps S401 to S403, each step of which is as follows: S401. For any two target indicator entities, obtain the semantic vectors corresponding to the two target indicator entities respectively, so as to obtain the semantic similarity between the two target indicator entities based on the semantic vectors. For any two target indicator entities, the word2ve algorithm is used to vectorize them, and the cosine similarity between the two semantic vectors is used as the semantic similarity. S402. Obtain the common device attribute words of the two target indicator entities under the same constraints, and calculate the association confidence difference of the two target indicator entities based on the common device attribute words, so as to obtain the association similarity between the two target indicator entities based on the association confidence difference. S403. Calculate the technical attribute correlation degree of the two target indicator entities based on the semantic similarity and the association similarity. When the technical attribute correlation degree is greater than the preset category determination threshold, classify the two target indicator entities into the same technical indicator category. The expression for the correlation degree of technical attributes between the target indicator entities is as follows: in, The semantic similarity of semantic vectors between the target indicator entities; The confidence level of the association between the i-th target indicator entity and the device attribute term U; The confidence level of the association between the j-th target indicator entity and the device attribute term U; It is an exponential constant with the natural constant as its base; for Kernel functions, boldly normalizing; The degree of correlation between the technical attributes of the i-th target indicator entity and the j-th target indicator entity; A category determination threshold is set. Target indicator entities with a technical attribute correlation greater than the category determination threshold are classified into the same technical indicator category. This operation is performed on all target indicator entities. The category determination threshold ranges from 0.3 to 0.7. Dimensional merging is performed on the same batch of historical standard samples at category determination thresholds of 0.5, 0.6, and 0.7, respectively, and the merged results are compared with the manual classification results from domain experts. Experimental results show that when the threshold is 0.5, some target indicator entities with different technical meanings are over-merged; when the threshold is 0.7, different expressions under the same technical indicator category are overly fragmented; when the threshold is 0.6, the merged technical indicator category is closest to the manual classification result and has the highest consistency. Therefore, this invention selects 0.6 as the category determination threshold. This invention, through its embodiments, obtains the semantic vectors corresponding to any two target indicator entities to calculate semantic similarity. This captures the inherent connections between different indicator entities at the semantic level, avoiding misjudgments caused by relying solely on surface character matching and enhancing the robustness of identifying the same technical attribute across standard systems. It also obtains common device attribute words of the two target indicator entities under the same constraints and calculates the difference in association confidence based on these common device attribute words to obtain association similarity. This allows the co-occurrence information of attribute words to effectively correct the bias of pure semantic similarity under the same technical constraints, improving the engineering interpretability of association judgment. Finally, it calculates the technical attribute association degree of the two target indicator entities based on semantic similarity and association similarity. When the technical attribute association degree is greater than the category determination threshold, the two target indicator entities are classified into the same technical indicator category. This achieves complementary fusion of semantic information and constraint information, enabling classification decisions to simultaneously consider the consistency of indicator semantics and the matching of engineering constraints, avoiding classification bias or overly coarse-grained aggregation caused by single-dimensional judgment.

[0044] In one embodiment, after obtaining several technical indicator categories and a subset of target indicator entities under each technical indicator category, the method further includes steps S501 to S503, each of which is as follows: S501. Obtain the source set of each target indicator entity under each technical indicator category, and combine it with the total number of sources in the standard clause text to obtain the coverage breadth of each technical indicator category. S502. Obtain the number of times each target indicator entity and equipment attribute word under each technical indicator category co-occur in the standard clause text, and combine the total number of times each target indicator entity appears in the standard clause text to obtain the constraint strength of each technical indicator category. S503. Obtain the universality score of each technical indicator category based on the coverage breadth and the constraint strength, so as to filter out the universal technical indicator categories based on the universality score and the preset universality threshold, and use the universal technical indicator categories as the updated technical indicator categories. The generality score of the technical indicator category is obtained by multiplying the coverage breadth and the constraint strength.

[0045] This invention, through obtaining the source set of each target indicator entity under each technical indicator category and combining it with the total number of sources in the standard clause text, obtains the coverage breadth of each technical indicator category. This objectively quantifies the distribution range of each technical indicator category across different standard sources, enabling a numerical representation of the degree of consensus on indicators across standard systems. Furthermore, by obtaining the number of times each target indicator entity and equipment attribute term co-occurs in the standard clause text under each technical indicator category and combining it with the total number of occurrences of each target indicator entity in the standard clause text, the constraint strength of each technical indicator category is obtained. This effectively measures the relationship between indicator categories and technologies on the co-occurrence frequency dimension. The tightness between constraints allows the constraint strength to simultaneously reflect the standard attention and technical strictness of the indicator. A generality score is obtained for each technical indicator category based on the coverage breadth and constraint strength. General technical indicator categories are then selected based on the generality score and a preset generality threshold, and these general technical indicator categories are used as updated technical indicator categories. This achieves a weighted fusion evaluation of coverage breadth and constraint strength. Through a dual-dimensional comprehensive scoring and threshold screening mechanism, general indicator categories that are widely adopted in multi-source standards and have strong technical constraint consensus can be automatically identified from all technical indicator categories, while secondary categories with single sources or sparse constraints are filtered out.

[0046] In one embodiment, the technical indicator parameter data is mapped to a preset probability interval to obtain a standardized measurement result, including steps S601 to S602, each step of which is as follows: S601. Obtain the evaluation system type of the technical indicator parameter data, and extract the mapping features of the technical indicator parameter data according to the evaluation system type; wherein, for numerical technical indicator parameter data, extract the numerical range and constraint direction as mapping features; for graded technical indicator parameter data, extract grade words as mapping features; for rule-based technical indicator parameter data, extract applicable conditions as mapping features. S602. Based on the mapping features, calculate the posterior probability of the technical indicator parameter data mapped to a preset probability interval using Bayesian classification, and take the probability interval with the largest posterior probability as the mapping result of the technical indicator parameter data, so as to convert the technical indicator parameter data into the interval identifier or interval value range of the corresponding probability interval according to the mapping result, and obtain the standardized measurement result. In practical applications, taking the "measurement error" technical category as an example, samples with clear corresponding relationships are extracted from the historical standard sample library, such as "accuracy level 1", "error not exceeding ±1%", and "A-level accuracy". These samples are then uniformly classified into preset probability interval categories, such as high requirement interval, medium requirement interval, and low requirement interval. For the current clause to be processed, features such as evaluation type, constraint direction, numerical range, level words, and applicable conditions are extracted. These features are input into a Bayesian classifier to calculate the posterior probability of the clause belonging to each probability interval. The interval with the highest posterior probability is taken as the mapping result of the clause. Specifically, for example, in historical samples, "accuracy level 1" appeared 80 times, of which 62 times corresponded to the high requirement range of "error ≤ ±1%", 14 times corresponded to the medium requirement range, and 4 times corresponded to the low requirement range. Therefore, the probability of "accuracy level 1" falling into the high requirement range is the highest. When the current regional standard is written as "equipment should meet the accuracy requirements of level 1", this level-type expression is mapped to the high requirement probability range.

[0047] This invention, through obtaining the evaluation system type of the technical indicator parameter data, extracts mapping features of the technical indicator parameter data based on the evaluation system type. This allows for the targeted extraction of the most representative mapping features for different evaluation system types, ensuring that the subsequent mapping process fully preserves the original constraint semantics of various data types. Based on the mapping features, the posterior probability of the technical indicator parameter data mapping to a preset probability interval is calculated using Bayesian classification. The probability interval with the highest posterior probability is taken as the mapping result. Then, based on the mapping result, the technical indicator parameter data is converted into the interval identifier or interval numerical range of the corresponding probability interval, obtaining a standardized measurement result. Thus, under the Bayesian framework, optimal interval matching is achieved through the maximum posterior probability criterion, uniformly mapping heterogeneous original expressions to a standardized probability interval system and directly outputting the interval identifier or numerical range. This not only preserves the quantitative support for the reliability of the mapping through the posterior probability calculation process but also makes the standardized results more intuitive and operable, significantly improving the direct comparability and engineering practicality of measurement results across standard systems.

[0048] In one embodiment, the standardized measurement results, the target indicator entities, and the corresponding standard clause texts are associated and stored using the technical indicator category as an index to obtain a multi-source standard database of the intelligent measurement devices, including steps S701 to S703, each step of which is as follows: S701. Use each of the obtained technical indicator categories as the primary key of the database index; S702. For each of the aforementioned technical indicator categories, obtain the corresponding standard clause text, target indicator entity, and standardized measurement results; S703. Associate and store the index primary key with the standard clause text, the target indicator entity and the standardized metric to obtain a multi-source standard database of the intelligent measurement equipment. This invention constructs a complete association link from the original source of the standard clause to the extracted result of the indicator entity and then to the unified probability interval measurement value. This is achieved by using each of the obtained technical indicator categories as the primary key of the database and associating the primary key of the index with the standard clause text, the target indicator entity, and the standardized measurement. This realizes the organic unity between the accurate tracing of the standard clause, the accurate positioning of the target indicator entity, and the reproducibility of the standardized measurement result, so that any subsequent query operation can simultaneously obtain the original semantic information, entity representation information, and quantitative comparison information.

[0049] like Figure 2 As shown in the construction of the multi-source intelligent measurement device database, a corresponding system implementation is provided based on the above method implementation.

[0050] This invention provides a multi-source intelligent measurement equipment database construction system, including: an equipment semantic word set acquisition module 801, a target indicator entity screening module 802, a technical indicator category classification module 803, a standardized measurement result acquisition module 804, and a database construction module 805; The device semantic term set acquisition module 801 is used to acquire the standard clause text of the multi-source intelligent measurement device and split the standard clause text into a device semantic term set. The target indicator entity filtering module 802 is used to obtain the correlation degree and co-occurrence conditional probability between equipment attribute words and equipment technical indicator words in the equipment semantic word set, so as to filter a number of target indicator entities from the equipment semantic word set. The technical indicator category classification module 803 is used to obtain the technical attribute correlation degree between any two target indicator entities, so as to classify several target indicator entities according to the technical attribute correlation degree, and obtain several technical indicator categories and a subset of target indicator entities under each technical indicator category. The standardized measurement result acquisition module 804 is used to acquire the technical indicator parameter data corresponding to each target indicator entity in the target indicator entity subset, and map the technical indicator parameter data to a preset probability interval to obtain the standardized measurement result. The database construction module 805 is used to associate and store the standardized measurement results, the target indicator entities and the corresponding standard clause texts using the technical indicator categories as indexes, so as to obtain a standard database of the multi-source intelligent measurement devices.

[0051] This invention, through obtaining the standard clause text of intelligent measurement equipment and splitting it into a device semantic term set, preserves the complete contextual structure of the technical constraints in the standard clauses, avoiding semantic distortion caused by isolated keyword extraction. It obtains the correlation degree and co-occurrence conditional probability between device attribute words and device technical indicator words in the device semantic term set to filter target indicator entities from the set. This enables accurate identification of technical indicators limited by constraints in the standard text, improving the accuracy of standard indicator extraction. Furthermore, it obtains the technical attribute correlation degree between target indicator entities and classifies them based on this correlation, allowing the same technical attribute expressed in different terms in standards from different sources to be automatically identified and aggregated into a technical indicator category. Finally, it obtains the technical indicator parameter data corresponding to each target indicator entity and maps this data to a preset probability interval. This invention unifies heterogeneous evaluation systems into measurable probability interval expressions, achieving semantic alignment and unified measurement across standard systems. Finally, using technical indicator categories as indexes, it associates and stores standardized measurement results, target indicator entities, and corresponding standard clause texts, constructing a quantifiable and reproducible multi-source standard database indexed by technical indicator categories. Compared to existing standard databases that only enable keyword retrieval and field-based storage, failing to express the correspondence between technical indicator categories across standard systems and comparable measurements under heterogeneous evaluation systems, this invention, through fully automated processing from semantic word set splitting, target indicator entity screening, technical category aggregation to probability interval mapping, constructs a standard database capable of supporting equipment adaptation analysis, compliance decision-making, and multi-source indicator difference quantification calculation, providing an objective and reproducible data foundation for the multi-source engineering applications of intelligent measurement equipment.

[0052] It is understood that the above system embodiments correspond to the method embodiments of the present invention, and can implement the method for constructing a multi-source intelligent measurement device database provided by any of the above method embodiments of the present invention.

[0053] It should be noted that the system embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0054] For ease of description and brevity, the system embodiments of the present invention include all the implementation methods described in the above embodiments of the multi-source intelligent measurement device database construction method, and will not be repeated here.

[0055] Based on the above embodiments of the multi-source intelligent measurement device database construction method, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the multi-source intelligent measurement device database construction method of any embodiment of the present invention.

[0056] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0057] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0058] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0059] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the multi-source intelligent measurement device database construction method described in any of the above-described method embodiments of the present invention.

[0060] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0061] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for constructing a database of multi-source intelligent measurement devices, characterized in that, include: Obtain the standard clause text of the multi-source intelligent measurement device, and decompose the standard clause text into a device semantic term set; Obtain the correlation degree and co-occurrence conditional probability between device attribute words and device technical indicator words in the device semantic word set, so as to filter a number of target indicator entities from the device semantic word set; Obtain the technical attribute correlation degree between any two target indicator entities, and classify several target indicator entities according to the technical attribute correlation degree to obtain several technical indicator categories and a subset of target indicator entities under each technical indicator category. Obtain the technical indicator parameter data corresponding to each target indicator entity in the target indicator entity subset, and map the technical indicator parameter data to a preset probability interval to obtain a standardized measurement result; Using the technical indicator category as an index, the standardized measurement results, the target indicator entities, and the corresponding standard clause texts are associated and stored to obtain a standard database of the multi-source intelligent measurement equipment.

2. The method for constructing a multi-source intelligent measurement device database as described in claim 1, characterized in that, The standard terms and conditions for acquiring multi-source intelligent measurement equipment are as follows: The standard source data, historical standard sample library data, and equipment capability description data of multi-source intelligent measurement equipment are obtained as standard clause texts; wherein, the standard source data includes equipment standard data and measurement operation specification data of intelligent measurement equipment; the historical standard sample library data includes historical indicator clauses and historical parameter records; and the equipment capability description data includes product manuals, test reports, and equipment technical white papers.

3. The method for constructing a multi-source intelligent measurement device database as described in claim 1, characterized in that, The specific steps for obtaining the correlation degree and co-occurrence conditional probability between device attribute words and device technical indicator words in the device semantic word set are as follows: Obtain the index positions of equipment attribute terms and equipment technical indicator terms in the standard clause text, and obtain the maximum number of attribute terms between equipment attribute terms and equipment technical indicator terms; The relevance is calculated based on the index position and the maximum number of attribute words. The number of times that equipment attribute terms and equipment technical indicator terms appear together in the same paragraph in the standard clause text is obtained, and the co-occurrence conditional probability is calculated by combining the number of times that equipment technical indicator terms appear alone.

4. The method for constructing a multi-source intelligent measurement device database as described in claim 1, characterized in that, The step of obtaining the technical attribute correlation degree between any two target indicator entities, and classifying several target indicator entities according to the technical attribute correlation degree, specifically involves: For any two target indicator entities, obtain the semantic vectors corresponding to the two target indicator entities respectively, and obtain the semantic similarity between the two target indicator entities based on the semantic vectors; Obtain the common device attribute words of the two target indicator entities under the same constraints, and calculate the association confidence difference between the two target indicator entities based on the common device attribute words, so as to obtain the association similarity between the two target indicator entities based on the association confidence difference. The technical attribute correlation degree of the two target indicator entities is calculated based on the semantic similarity and the association similarity. When the technical attribute correlation degree is greater than the preset category determination threshold, the two target indicator entities are classified into the same technical indicator category.

5. The method for constructing a multi-source intelligent measurement device database as described in claim 1, characterized in that, After obtaining several technical indicator categories and a subset of target indicator entities under each technical indicator category, the method further includes: Obtain the source set of each target indicator entity under each technical indicator category, and combine it with the total number of sources in the standard clause text to obtain the coverage breadth of each technical indicator category; The constraint strength of each technical indicator category is obtained by obtaining the number of times each target indicator entity and equipment attribute word co-occurs in the standard clause text under each technical indicator category, and by combining the total number of times each target indicator entity occurs in the standard clause text. The generality score of each technical indicator category is obtained based on the coverage breadth and the constraint strength, so as to filter out general technical indicator categories based on the generality score and the preset generality threshold, and the general technical indicator categories are used as the updated technical indicator categories.

6. The method for constructing a multi-source intelligent measurement device database as described in claim 1, characterized in that, The step of mapping the technical indicator parameter data to a preset probability interval to obtain standardized measurement results specifically involves: The evaluation system type of the technical indicator parameter data is obtained, and the mapping features of the technical indicator parameter data are extracted according to the evaluation system type; wherein, for numerical technical indicator parameter data, the numerical range and constraint direction are extracted as mapping features; for graded technical indicator parameter data, grade terms are extracted as mapping features; and for rule-based technical indicator parameter data, applicable conditions are extracted as mapping features. Based on the mapping features, the posterior probability of the technical indicator parameter data being mapped to a preset probability interval is calculated using Bayesian classification. The probability interval with the largest posterior probability is then used as the mapping result of the technical indicator parameter data. Based on the mapping result, the technical indicator parameter data is converted into the interval identifier or interval value range of the corresponding probability interval to obtain a standardized measurement result.

7. The method for constructing a multi-source intelligent measurement device database as described in claim 1, characterized in that, The step of using the technical indicator category as an index to associate and store the standardized measurement results, the target indicator entities, and the corresponding standard clause texts to obtain a multi-source standard database of the intelligent measurement devices is as follows: Each of the obtained technical indicator categories will be used as the primary key of the database index; For each of the aforementioned technical indicator categories, obtain the corresponding standard clause text, target indicator entity, and standardized measurement results; The index primary key is associated with and stored in relation to the standard clause text, the target indicator entity, and the standardized metric to obtain a multi-source standard database for the intelligent measurement equipment.

8. A multi-source intelligent measurement device database construction system, characterized in that, include: The module includes a device semantic term set acquisition module, a target indicator entity filtering module, a technical indicator category classification module, and a standardized measurement result acquisition and database construction module. The device semantic term set acquisition module is used to acquire the standard clause text of multi-source intelligent measurement devices and split the standard clause text into device semantic term sets. The target indicator entity filtering module is used to obtain the correlation degree and co-occurrence conditional probability between equipment attribute words and equipment technical indicator words in the equipment semantic word set, so as to filter a number of target indicator entities from the equipment semantic word set. The technical indicator category classification module is used to obtain the technical attribute correlation degree between any two target indicator entities, so as to classify several target indicator entities according to the technical attribute correlation degree, and obtain several technical indicator categories and a subset of target indicator entities under each technical indicator category. The standardized measurement result acquisition module is used to acquire the technical indicator parameter data corresponding to each target indicator entity in the target indicator entity subset, and map the technical indicator parameter data to a preset probability interval to obtain the standardized measurement result. The database construction module is used to associate and store the standardized measurement results, the target indicator entities and the corresponding standard clause texts using the technical indicator categories as indexes, so as to obtain a standard database of the multi-source intelligent measurement devices.

9. A terminal device, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method for constructing a multi-source intelligent measurement device database as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the multi-source intelligent measurement device database construction method as described in any one of claims 1-7.

Images