Multi-source object identity normalization and evolution management method based on ontology world model

Abstract

The application provides a multi-source object identity normalization and evolution management method based on an ontology world model, and relates to the technical field of multi-source object management. The method comprises the following steps: processing multi-source observation data based on a normalization processing rule of the ontology world model and outputting a representation object; comparing the representation object with the category of a system object, screening a system object with the same category as the representation object as a candidate object, and outputting a similarity comparison object group; matching a similarity comparison calculation rule according to the category of the similarity comparison object group, calculating the object similarity between the representation object and the candidate object, generating an object information view of the candidate object, and outputting an object identity normalization result; managing the identity evolution event of the candidate object, outputting an object identity evolution management result, and synchronizing the object identity evolution management result to an upper application activity center, so that the data specification is unified through the ontology world model, the object identity determination accuracy is improved, accurate evolution management is realized, and the upper application data is consistent and reliable.

Description

Technical Field

[0001] This invention relates to the field of multi-source object governance technology, and in particular to a method for the identity normalization and evolution governance of multi-source objects based on an ontology world model. Background Technology

[0002] Currently, various business and observation data come from diverse sources, with significant differences in structural definitions, attribute standards, and representation formats across different data sources. This makes direct object comparison and association matching impossible for heterogeneous data from different sources. Existing object identity recognition methods often employ generalized similarity calculation logic, failing to design corresponding matching rules for different types of object attributes. This results in insufficient adaptability, limited matching accuracy, and a tendency for object mismatches and omissions, making it difficult to guarantee the accuracy of identity determination results.

[0003] Furthermore, business objects undergo various evolutionary states throughout their lifecycle, such as merging, splitting, and continuing, as business scenarios change. Existing technologies lack standardized governance for these evolutionary events, making it impossible to track changes in the lineage of object identifiers and to automatically synchronize evolution results to upper-layer application systems. This leads to inconsistencies in object referencing and data definitions between different business systems, severely impacting the credibility of data applications and the accuracy of business decisions.

[0004] Therefore, it is necessary to provide a method for unifying and governing the identity of multi-source objects based on ontology world models to solve the above-mentioned technical problems. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for normalizing and governing the identity of multi-source objects based on an ontology world model. This method addresses the issues of existing multi-source heterogeneous data lacking unified standards, having a single method for calculating object similarity, making it difficult to accurately determine object identity, and lacking an evolutionary event governance mechanism.

[0006] The present invention provides a method for unifying and governing the identity of multi-source objects based on an ontology world model, the method comprising: Receive multi-source observation data, process the multi-source observation data based on the normalized processing rules defined by the ontology world model, and output a representation object that satisfies ontology attribute constraints; Obtain system objects, and based on the object category comparison rules defined in the ontology world model, compare the object categories of the description object with those of the system objects, filter out the system objects with the same object category as the description object as candidate objects, and output a similarity comparison object group composed of the description object and the candidate objects with the same object category; wherein, the object category includes text category, number category and location category; According to the similarity comparison calculation rules corresponding to the object category of the similarity comparison object group, calculate the object similarity between the described object and the candidate object in the similarity comparison object group, generate the object information view of the candidate object, and output the object identity normalization result; Based on the object identity normalization result, the identity evolution events of the candidate objects are governed, and the object identity evolution governance result is output and automatically synchronized to the upper-layer application activity center; wherein, the identity evolution events include merge evolution events, split evolution events and successive evolution events.

[0007] Preferably, the step of receiving multi-source observation data and processing the multi-source observation data based on the normalized processing rules defined by the ontology world model to output a representation object that satisfies ontology attribute constraints includes: The system receives the multi-source observation data, including radar detection data, electronic reconnaissance data, and manual message data, and simultaneously collects observation tags, including source identifiers, collection time, and data source credibility level. Based on the concept mapping rules defined in the ontology world model, the system performs concept mapping between the multi-source observation data and the observation tags, and outputs the first observation data. According to the concept attribution verification rules defined in the ontology world model, verify the concept attribution compatibility of the first observation data, map heterogeneous fields to ontology standard attributes, remove invalid observation data with missing required attributes, and output the second observation data. Based on the attribute type semantic rules defined in the ontology world model, the second observation data is subjected to spatiotemporal coordinate system unification, numerical unit normalization, enumeration value semantic mapping and string format standardization, and the representation object that satisfies the ontology attribute constraints is output. The normalization processing rules include the concept mapping rules, the concept attribution verification rules, and the attribute type semantic rules.

[0008] Preferably, the ontology attribute constraints are attribute semantic limitation constraints bound to the ontology world model, including concept inheritance compatibility constraints, attribute value range legal interval constraints, attribute type semantic classification constraints, and discrimination weight configuration constraints.

[0009] Preferably, the step of calculating the object similarity between the described object and the candidate object in the similarity comparison object group according to the object category corresponding to the similarity comparison object group and the corresponding similarity comparison calculation rule includes: If the object category corresponding to the similarity comparison object group is the text category, then the normalized edit distance calculation rule corresponding to the text category is matched; Extract the strings corresponding to the description object and the candidate object in the similarity comparison object group respectively, and perform case unification, invalid character removal and standard alias mapping on the strings corresponding to the description object and the candidate object according to the ontology attribute constraints to generate a similarity comparison string group; For the similarity comparison string group, based on the normalized edit distance calculation rule, the string similarity between the representation object and the string corresponding to the candidate object is calculated. as follows: In the formula, The string representing the object being described String corresponding to the candidate object The Lewinstein edit distance between them; max(m,n) represents the maximum value between the string length m corresponding to the representation object and the string length n corresponding to the candidate object; Obtain a semantic equivalence term library. If the string corresponding to the described object and the candidate object does not belong to the same semantic equivalence term, then the string similarity is used as the object similarity. If the string corresponding to the described object and the candidate object belongs to the same semantic equivalence term, then the string similarity is raised to a preset semantic matching degree and used as the object similarity.

[0010] Preferably, if the object category corresponding to the similarity comparison object group is the numeric category, then the corresponding tolerance normalization difference calculation rule is matched for the numeric category; Extract the numerical values ​​corresponding to the description object and the candidate object respectively, and perform measurement unit unification, numerical precision unification and outlier filtering on the numerical values ​​corresponding to the description object and the candidate object according to the ontology attribute constraints to generate similarity comparison numerical groups. For the similarity comparison numerical group, calculate the absolute difference between the numerical values ​​corresponding to the described object and the candidate object. And retrieve the preset allowable tolerance threshold. ; Based on the tolerance-normalized difference calculation rule, the numerical similarity between the numerical values ​​corresponding to the described object and the candidate object is calculated. As the object similarity, the numerical similarity The calculation formula is as follows: .

[0011] Preferably, if the object category corresponding to the similarity comparison object group is the location category, then the spatiotemporal joint distance decay calculation rule corresponding to the location category is matched; Extract the position coordinates corresponding to the described object and the candidate object, as well as the acquisition time corresponding to the described object and the candidate object, respectively; According to the ontological attribute constraints, the location coordinates corresponding to the description object and the candidate object are subjected to geographic coordinate system one, and the collection time corresponding to the description object and the candidate object is unified with time reference, and the similarity comparison spatiotemporal data group is generated. For the similarity comparison spatiotemporal data set, the spatial distance d is calculated based on the position coordinates corresponding to the described object and the candidate object, and the acquisition time difference is calculated based on the acquisition time corresponding to the described object and the candidate object. ; Retrieve preset space allowance threshold With time allowable threshold ; Based on the spatiotemporal joint distance attenuation calculation rule, the positional similarity between the described object and the candidate object is calculated using a Gaussian attenuation model. As the object similarity, the positional similarity The calculation formula is as follows: .

[0012] Preferably, based on the object identity normalization result, the identity evolution events of the candidate objects are governed, and the object identity evolution governance result is output and automatically synchronized to the upper-layer application activity center; wherein, the identity evolution events include merge evolution events, split evolution events, and successive evolution events, including: Based on the object identity normalization result, identity evolution trigger signals are identified, and the source credibility and attribute compliance of the identity evolution trigger signals are verified based on the ontology attribute constraints. Invalid evolution trigger signals are eliminated, and the identity evolution events corresponding to the verified identity evolution trigger signals are output. The identity evolution events include the merge evolution event, the split evolution event, and the successive evolution event. For the merging evolution event, all involved object referencing relationships are merged and written into the redirection mapping function; for the splitting evolution event, the original object referencing relationships are split and clustered and reorganized according to the new features; for the successive evolution event, successive relationships between objects are established and selective merging is performed, outputting the referencing relationship set and the redirection table. Based on the reference relationship set and the redirection table, each evolution operation is registered as the identity evolution event, the triggering evidence, the operation subject and the operation time are recorded, the object identification lineage is constructed, and the evolution governance file is output. Based on the redirection table and the evolution governance archive, object references in the upper-layer application activities are automatically mapped to surviving object identifiers through a unified fact source, and the object identity evolution governance results are output and automatically synchronized to the upper-layer application activity center.

[0013] A multi-source object identity normalization and evolution governance system based on an ontology world model, the system comprising: The representation object output module is used to receive multi-source observation data, process the multi-source observation data based on the normalized processing rules defined by the ontology world model, and output a representation object that satisfies the ontology attribute constraints. The object category comparison module is used to obtain system objects, compare the object categories of the description object with those of the system objects based on the object category comparison rules defined in the ontology world model, filter out the system objects with the same object category as the description object as candidate objects, and output a similarity comparison object group composed of the description object and the candidate objects with the same object category; wherein, the object category includes text category, number category and location category; The identity normalization module is used to match the corresponding similarity comparison calculation rules according to the object category corresponding to the similarity comparison object group, calculate the object similarity between the described object and the candidate object in the similarity comparison object group, generate the object information view of the candidate object, and output the object identity normalization result. The identity evolution governance module is used to govern the identity evolution events of the candidate objects based on the identity normalization results of the objects, output the object identity evolution governance results and automatically synchronize them to the upper-layer application activity center; wherein, the identity evolution events include merge evolution events, split evolution events and successive evolution events.

[0014] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the steps of the multi-source object identity normalization and evolution governance method based on the ontology world model as described in any of the above.

[0015] A readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of the multi-source object identity normalization and evolution governance method based on the ontology world model as described above.

[0016] Compared with related technologies, the multi-source object identity normalization and evolution governance method based on ontology world model provided by this invention has the following beneficial effects: This invention receives multi-source observation data and processes it based on standardized processing rules defined by an ontology world model, outputting representation objects that satisfy ontology attribute constraints. It then acquires system objects and, based on object category comparison rules defined by the ontology world model, compares the object categories of the representation objects with those of the system objects, selecting system objects with the same object category as candidate objects, and outputting a similarity comparison object group consisting of representation objects and candidate objects with the same object category. The object categories include text, numbers, and locations. Finally, it matches objects according to the object category corresponding to the similarity comparison object group. The system employs corresponding similarity comparison calculation rules to calculate the object similarity between the described object and the candidate object in the similarity comparison object group, generates an object information view of the candidate object, and outputs the object identity normalization result. Based on the object identity normalization result, it governs the identity evolution events of the candidate objects, outputs the object identity evolution governance result, and automatically synchronizes it to the upper-layer application activity center. Among them, identity evolution events include merged evolution events, split evolution events, and continuous evolution events, thereby unifying data specifications through the ontology world model, improving the accuracy of object identity determination, achieving precise evolution governance, and ensuring consistent and reliable upper-layer application data.

[0017] This invention establishes unified and standardized processing rules through an ontology world model, enabling concept mapping, attribution verification, and semantic standardization of multi-source heterogeneous data. It outputs unified object representations that conform to ontology attribute constraints, thereby ensuring data consistency and compliance. Different similarity calculation rules are matched to different object categories, adapting to features of different attribute types, significantly improving the accuracy of object identity determination, reducing mismatches and omissions, and outputting a clear view of object information and normalized results. Simultaneously, by constructing an object evolution governance mechanism, it can accurately identify various evolution trigger signals, verify the credibility and compliance of these signals, perform corresponding governance operations on different types of evolution events, establish object identification lineage links, retain evolution governance archives, and achieve traceability management of the entire object lifecycle. Furthermore, the governance results can be automatically synchronized to the upper-layer application activity center, achieving automatic mapping of object references through a unified fact source, ensuring consistent data standards in upper-layer applications, and effectively improving the reliability of data applications and the efficiency of business operations. Attached Figure Description

[0018] Figure 1 A flowchart illustrating the method for unifying and evolving the identity of multi-source objects based on an ontology world model, provided in an embodiment of the present invention. Figure 2 A system block diagram of a multi-source object identity normalization and evolution governance system based on an ontology world model provided in this embodiment of the invention; Figure 3 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention; Among them, 30-electronic device, 31-processor, 32-memory, 33-bus. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] like Figure 1 The diagram shown is a flowchart of the multi-source object identity normalization and evolution governance method based on the ontology world model provided in this embodiment of the invention. Figure 1 The execution subject of the method shown can be a software and / or hardware device. The execution subject of this invention can include, but is not limited to, at least one of the following: user equipment, network equipment, etc. User equipment can include, but is not limited to, computers, smartphones, personal digital assistants (PDAs), and the aforementioned electronic devices. Network equipment can include, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers. Cloud computing is a type of distributed computing, consisting of a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. Steps S1 to S4 are detailed as follows: S1 receives multi-source observation data and processes the multi-source observation data based on the normalized processing rules defined by the ontology world model, outputting a representation object that satisfies the ontology attribute constraints.

[0021] The process of receiving multi-source observation data, processing the multi-source observation data based on the normalized processing rules defined by the ontology world model, and outputting a representation object that satisfies ontology attribute constraints includes: The system receives the multi-source observation data, including radar detection data, electronic reconnaissance data, and manual message data, and simultaneously collects observation tags, including source identifier, collection time, and data source credibility level. Based on the concept mapping rules defined by the ontology world model, the system performs concept mapping between the multi-source observation data and the observation tags, and outputs the first observation data. According to the concept attribution verification rules defined in the ontology world model, verify the concept attribution compatibility of the first observation data, map heterogeneous fields to ontology standard attributes, remove invalid observation data with missing required attributes, and output the second observation data. Based on the attribute type semantic rules defined in the ontology world model, the second observation data is subjected to spatiotemporal coordinate system unification, numerical unit normalization, enumeration value semantic mapping, and string format standardization, and the representation object that satisfies the ontology attribute constraints is output. The normalization processing rules include the concept mapping rules, the concept attribution verification rules, and the attribute type semantic rules.

[0022] The ontology attribute constraints are attribute semantic constraints bound to the ontology world model, including concept inheritance compatibility constraints, attribute value range legal interval constraints, attribute type semantic classification constraints, and discrimination weight configuration constraints.

[0023] In practical applications, multi-source observation data, including radar detection data, electronic reconnaissance data, and manual message data, are received simultaneously. Corresponding observation tags are also collected, including data source identifiers, acquisition times, and data source reliability levels. Based on predefined concept mapping rules of the ontology world model, various heterogeneous observation data and their observation tags are mapped and aligned with the ontology concept system. Custom field descriptions from different data sources are mapped to general concept items defined in the ontology, resolving heterogeneity in field naming and description dimensions among different data sources, and outputting the first observation data.

[0024] Based on the concept attribution verification rules defined in the ontology world model, the first observation data undergoes concept attribution compatibility verification to ensure that the data concepts are compatible and consistent with the ontology classification system. Simultaneously, various heterogeneous data fields are uniformly mapped to standard ontology attribute items, achieving standardized alignment of data attribute dimensions. Invalid observation data with missing required attributes are filtered out, and redundant data that does not conform to ontology specifications is filtered out. The verified second observation data is then output, ensuring the compliance and validity of the data at the concept and attribute levels.

[0025] Based on the semantic rules of attribute types defined by the ontology world model, the second observation data is standardized, including unifying the spatiotemporal coordinate system of different data sources to eliminate differences in spatial reference benchmarks; unifying the measurement units and precision paradigms of various numerical attributes to achieve dimensional consistency; performing semantic mapping on enumeration values ​​to unify the semantic representation of enumeration items; and standardizing string format to eliminate differences in character format and expression style, outputting representation objects that satisfy ontology attribute constraints.

[0026] Among them, ontology attribute constraints are semantic constraints of attributes bound to the ontology world model. Specifically, they include concept inheritance compatibility constraints, legal range constraints of attribute value domains, semantic classification constraints of attribute types, and discrimination weight configuration constraints. These constraints set unified standards from the dimensions of concept level compliance, attribute value range, attribute semantic category, and discrimination weight configuration to ensure that the output representation objects conform to the unified specifications of the ontology system in terms of semantics, format, and value.

[0027] S2, Obtain system objects. Based on the object category comparison rules defined in the ontology world model, compare the object categories of the description object with those of the system objects, filter out the system objects with the same object category as the description object as candidate objects, and output a similarity comparison object group composed of the description object and the candidate objects with the same object category; wherein, the object category includes text category, number category, and location category.

[0028] The system retrieves all stored system objects and, based on the object category comparison rules defined in the ontology world model, compares and matches the normalized representation objects with the object categories of the system objects. Specifically, object categories are divided into text, number, and location categories, each corresponding to different attribute types of object data dimensions.

[0029] After category comparison, system objects that completely match the category of the described object are selected as candidate objects, while system objects with mismatched categories are excluded, thus accurately narrowing down the comparison scope. The described object is then paired with each of the selected candidate objects to form a similarity comparison object group.

[0030] By using a category-based pre-screening mechanism, it can be ensured that subsequent similarity calculations are all performed within the same attribute dimension, avoiding invalid comparisons across attribute categories. This not only guarantees the adaptability of similarity calculations and the rationality of the judgment results, but also effectively reduces the data scale of similarity comparison calculations and improves the operational efficiency of object identity normalization processing.

[0031] S3. According to the similarity comparison calculation rules corresponding to the object category of the similarity comparison object group, calculate the object similarity between the described object and the candidate object in the similarity comparison object group, generate the object information view of the candidate object, and output the object identity normalization result.

[0032] The step of calculating the object similarity between the described object and the candidate object in the similarity comparison object group according to the object category corresponding to the similarity comparison object group and the corresponding similarity comparison calculation rule includes: If the object category corresponding to the similarity comparison object group is the text category, then the normalized edit distance calculation rule corresponding to the text category is matched; Extract the strings corresponding to the description object and the candidate object in the similarity comparison object group respectively, and perform case unification, invalid character removal and standard alias mapping on the strings corresponding to the description object and the candidate object according to the ontology attribute constraints to generate a similarity comparison string group; For the similarity comparison string group, based on the normalized edit distance calculation rule, the string similarity between the representation object and the string corresponding to the candidate object is calculated. as follows: In the formula, The string representing the object being described String corresponding to the candidate object The Lewinstein edit distance between them; max(m,n) represents the maximum value between the string length m corresponding to the representation object and the string length n corresponding to the candidate object; Obtain a semantic equivalence term library. If the string corresponding to the described object and the candidate object does not belong to the same semantic equivalence term, then the string similarity is used as the object similarity. If the string corresponding to the described object and the candidate object belongs to the same semantic equivalence term, then the string similarity is raised to a preset semantic matching degree and used as the object similarity.

[0033] Based on the text categories corresponding to the similarity comparison object groups, the corresponding normalized edit distance calculation rules are matched. Since the character sequences and semantic representations of text category objects are different, the normalized calculation method based on edit distance can accurately quantify the degree of matching at the character level and adapt to the comparison features of text attributes.

[0034] The strings corresponding to the description objects and candidate objects in the similarity comparison object group are extracted separately. Based on the ontology attribute constraints, the case format of the strings is unified to eliminate the matching deviation caused by case differences. Invalid and redundant characters in the strings are removed, and meaningless format symbols and interference content are filtered out. Standard alias mapping conversion is performed to uniformly map the category names and abbreviations to the standard description items defined in the ontology, and finally, similarity comparison string groups are generated.

[0035] Based on the normalized edit distance calculation rule, the Lewinstein edit distance between the corresponding strings of the representation object and the candidate object is calculated. The Lewinstein edit distance represents the minimum number of edit operations required to convert between two strings, and is used to quantify the degree of difference between character sequences. Then, the Lewinstein edit distance is converted into string similarity using the maximum length of the two strings as the normalization benchmark.

[0036] The system retrieves a pre-set semantic equivalence terminology. If two sets of strings do not belong to the same semantic equivalence term, the calculated string similarity is directly used as the final object similarity. If two sets of strings belong to the same semantic equivalence term, the string similarity is increased to the preset semantic matching degree and used as the final object similarity. This compensates for the deficiency of pure character calculation in recognizing synonyms and further improves the accuracy of the matching results.

[0037] If the object category corresponding to the similarity comparison object group is the numeric category, then the corresponding tolerance normalization difference calculation rule is matched for the numeric category; Extract the numerical values ​​corresponding to the description object and the candidate object respectively, and perform measurement unit unification, numerical precision unification, and outlier filtering on the numerical values ​​corresponding to the description object and the candidate object according to the ontology attribute constraints to generate similarity comparison numerical groups. For the similarity comparison numerical group, calculate the absolute difference between the numerical values ​​corresponding to the described object and the candidate object. And retrieve the preset allowable tolerance threshold. ; Based on the tolerance-normalized difference calculation rule, the numerical similarity between the numerical values ​​corresponding to the described object and the candidate object is calculated. As the object similarity, the numerical similarity The calculation formula is as follows: .

[0038] When the object category corresponding to the similarity comparison object group is a numeric category, the corresponding tolerance-normalized difference calculation rule is matched for it. Since there are mainly numerical magnitude and precision deviations between numeric category objects, the normalized difference calculation method based on tolerance threshold is adopted, which can simultaneously meet the requirements of the allowable range of measurement error of the observation data and the quantitative requirements of the matching degree.

[0039] The numerical values ​​corresponding to the described object and the candidate object are extracted separately. The unit of measurement of the numerical values ​​is unified based on the ontology attribute constraints to eliminate numerical deviations caused by differences in units. The effective precision and number of digits retained of the numerical values ​​are unified to eliminate differences caused by inconsistent precision formats. Abnormal values ​​that exceed the reasonable value range are filtered out to eliminate the interference of invalid data on the calculation results and generate similarity comparison numerical groups.

[0040] For similarity comparison numerical groups, the absolute difference between the corresponding values ​​of the described object and the candidate object is first calculated to quantify the degree of numerical difference between the two. Then, the system's preset tolerance threshold is retrieved as the allowable boundary for numerical matching.

[0041] Numerical similarity is calculated based on the tolerance-normalized difference calculation rule and used as the object similarity of the group of objects. When the absolute difference of the numerical values ​​does not exceed the allowable tolerance threshold, the numerical similarity is calculated by combining the ratio of the absolute difference of the numerical values ​​to the allowable tolerance threshold. The smaller the absolute difference of the numerical values, the higher the numerical similarity. When the absolute difference of the numerical values ​​exceeds the allowable tolerance threshold, the numerical matching degree between the two is determined to be zero, and the numerical similarity is zero. This can fully adapt to the numerical fluctuations of multi-source data and effectively improve the accuracy of numerical attribute comparison.

[0042] If the object category corresponding to the similarity comparison object group is the location category, then the spatiotemporal joint distance decay calculation rule corresponding to the location category is matched; Extract the position coordinates corresponding to the described object and the candidate object, as well as the acquisition time corresponding to the described object and the candidate object, respectively; According to the ontological attribute constraints, the location coordinates corresponding to the description object and the candidate object are subjected to geographic coordinate system one, and the collection time corresponding to the description object and the candidate object is unified with time reference, and the similarity comparison spatiotemporal data group is generated. For the similarity comparison spatiotemporal data set, the spatial distance d is calculated based on the position coordinates corresponding to the described object and the candidate object, and the acquisition time difference is calculated based on the acquisition time corresponding to the described object and the candidate object. ; Retrieve preset space allowance threshold With time allowable threshold ; Based on the spatiotemporal joint distance attenuation calculation rule, the positional similarity between the described object and the candidate object is calculated using a Gaussian attenuation model. As the object similarity, the positional similarity The calculation formula is as follows: .

[0043] When the object category corresponding to the similarity comparison object group is the location category, the corresponding spatiotemporal joint distance attenuation calculation rule is matched for this group of objects. Since the identity determination of location category objects needs to consider the matching degree of spatial distribution and observation time series, the spatiotemporal joint attenuation calculation method is adopted, which can comprehensively quantify the spatiotemporal fit of this group of objects.

[0044] The location coordinates of the described object and the candidate object, as well as the collection time of the two, are extracted separately. The geographic coordinate system of the location coordinates of the group of objects is unified to eliminate the coordinate deviation caused by the difference in spatial reference benchmarks of different data sources. The time benchmark of the collection time of the group of objects is unified to eliminate the time sequence deviation caused by the inconsistency of time zone and timing. Finally, the similarity comparison spatiotemporal data group is generated.

[0045] For similarity comparison spatiotemporal data groups, spatial distance is calculated based on the location coordinates of the objects in the group, and acquisition time difference is calculated based on the acquisition time of the objects in the group. At the same time, the system's preset spatial and temporal tolerance thresholds are retrieved. These thresholds are pre-configured according to the observation accuracy and motion characteristics of the business scenario, serving as the allowable boundaries for spatiotemporal matching.

[0046] Based on the spatiotemporal joint distance attenuation calculation rule, the positional similarity is calculated using a Gaussian attenuation model and used as the object similarity for this group of objects. When the spatial distance is less than or equal to the spatial allowable threshold and the acquisition time difference is less than or equal to the temporal allowable threshold, the similarity is calculated according to the formula. The location similarity is calculated, and the smaller the spatial distance and the time difference of collection, the higher the location similarity. When the spatial distance is greater than the spatial allowable threshold or the time difference of collection is greater than the time allowable threshold, the location matching degree of the group of objects is determined to be zero and the location similarity is zero, thereby effectively improving the reliability of the identity determination of location category objects.

[0047] S4. Based on the object identity normalization result, the identity evolution events of the candidate objects are governed, and the object identity evolution governance result is output and automatically synchronized to the upper-layer application activity center; wherein, the identity evolution events include merge evolution events, split evolution events, and successive evolution events.

[0048] Based on the object identity normalization result, the identity evolution events of the candidate objects are governed, and the object identity evolution governance result is output and automatically synchronized to the upper-layer application activity center; wherein, the identity evolution events include merge evolution events, split evolution events, and successive evolution events, including: Based on the object identity normalization result, identity evolution trigger signals are identified, and the source credibility and attribute compliance of the identity evolution trigger signals are verified based on the ontology attribute constraints. Invalid evolution trigger signals are eliminated, and the identity evolution events corresponding to the verified identity evolution trigger signals are output. Among them, the identity evolution events include the merge evolution event, the split evolution event, and the successive evolution event. For the merging evolution event, all involved object referencing relationships are merged and written into the redirection mapping function; for the splitting evolution event, the original object referencing relationships are split and clustered and reorganized according to the new features; for the successive evolution event, successive relationships between objects are established and selective merging is performed, outputting the referencing relationship set and the redirection table. Based on the reference relationship set and the redirection table, each evolution operation is registered as the identity evolution event, the triggering evidence, the operation subject, the operation time is recorded, the object identification lineage is constructed, and the evolution governance file is output. Based on the redirection table and the evolution governance archive, object references in the upper-layer application activities are automatically mapped to surviving object identifiers through a unified fact source, and the object identity evolution governance results are output and automatically synchronized to the upper-layer application activity center.

[0049] Based on the object identity normalization result, identity evolution trigger signals are identified, and the source credibility and attribute compliance of identity evolution trigger signals are verified based on ontology attribute constraints. Invalid evolution trigger signals with untrustworthy sources or attributes that do not conform to ontology specifications are eliminated, and identity evolution events corresponding to valid evolution trigger signals that have passed verification are selected, specifically including merged evolution events, split evolution events, and successive evolution events.

[0050] For merge evolution events, all object referencing relationships involved in the merge event are incorporated into the redirection mapping function, enabling mapping from multiple objects to a single surviving object. For split evolution events, the unified referencing relationships of the original objects are removed, and clustering and recombination are performed based on the newly added features of each split object to establish an independent object referencing system. For successive evolution events, the temporal succession relationships between preceding and succeeding objects are constructed, and selective merging is performed based on feature matching degrees, ultimately outputting a set of referencing relationships and a redirection table.

[0051] Based on the reference relationship set and the redirection table, each evolution operation is registered as an independent identity evolution event, and the triggering evidence, the operating subject and the operation time of the event are recorded. The object identification lineage is constructed in sequence to form an evolution governance file, realizing the traceability management of changes throughout the entire life cycle of the object.

[0052] Based on the redirection table and evolution governance archive, various object references in upper-layer application activities are automatically mapped to currently valid existing object identifiers through a unified fact source. The object identity evolution governance results are output and automatically synchronized to the upper-layer application activity center, thereby ensuring the consistency of object references and the accuracy of business data in the upper-layer business system.

[0053] like Figure 2 The diagram shown is a system block diagram of a multi-source object identity normalization and evolution governance system based on an ontology world model provided in an embodiment of the present invention. The system includes: The representation object output module is used to receive multi-source observation data, process the multi-source observation data based on the normalized processing rules defined by the ontology world model, and output a representation object that satisfies the ontology attribute constraints. The object category comparison module is used to acquire system objects, compare the object categories of the description object with those of the system objects based on the object category comparison rules defined in the ontology world model, filter out the system objects with the same object category as the description object as candidate objects, and output a similarity comparison object group composed of the description object and the candidate objects with the same object category; wherein, the object category includes text category, number category, and location category; The identity normalization module is used to match the corresponding similarity comparison calculation rules according to the object category corresponding to the similarity comparison object group, calculate the object similarity between the described object and the candidate object in the similarity comparison object group, generate the object information view of the candidate object, and output the object identity normalization result. The identity evolution governance module is used to govern the identity evolution events of the candidate objects based on the identity normalization results of the objects, output the object identity evolution governance results and automatically synchronize them to the upper-layer application activity center; wherein, the identity evolution events include merge evolution events, split evolution events and successive evolution events.

[0054] Figure 2 The apparatus of the illustrated embodiment can be used to perform corresponding actions. Figure 1 The steps in the method embodiments shown are implemented in a similar manner and have similar technical effects, and will not be repeated here.

[0055] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the steps of the multi-source object identity normalization and evolution governance method based on the ontology world model as described in any of the above.

[0056] like Figure 3 The diagram shown is a hardware structure schematic of an electronic device according to an embodiment of the present invention. The electronic device 30 includes: a processor 31, a memory 32, and a computer program; wherein... The memory 32 is used to store the computer program, and the memory may also be flash memory. The computer program is, for example, an application program or functional module that implements the above method.

[0057] Processor 31 is configured to execute the computer program stored in the memory to implement the various steps performed by the device in the above method. For details, please refer to the relevant descriptions in the preceding method embodiments.

[0058] Alternatively, the memory 32 can be either standalone or integrated with the processor 31.

[0059] When the memory 32 is a device independent of the processor 31, the device may further include: Bus 33 is used to connect the memory 32 and the processor 31.

[0060] A readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of the multi-source object identity normalization and evolution governance method based on the ontology world model as described in any of the above.

[0061] The readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application-Specific Integrated Circuit (ASIC). Alternatively, the ASIC can be located in a user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0062] The present invention also provides a program product including executable instructions stored in a readable storage medium. At least one processor of the device can read the executable instructions from the readable storage medium, and the at least one processor executes the executable instructions to cause the device to implement the methods provided in the various embodiments described above.

[0063] In the embodiments of the above-described device, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for unifying and evolving the identity of multi-source objects based on an ontology world model, characterized in that: The method includes: Receive multi-source observation data, process the multi-source observation data based on the normalized processing rules defined by the ontology world model, and output a representation object that satisfies ontology attribute constraints; Obtain system objects, and based on the object category comparison rules defined in the ontology world model, compare the object categories of the description object with those of the system objects, filter out the system objects with the same object category as the description object as candidate objects, and output a similarity comparison object group composed of the description object and the candidate objects with the same object category; wherein, the object category includes text category, number category and location category; According to the similarity comparison calculation rules corresponding to the object category of the similarity comparison object group, calculate the object similarity between the described object and the candidate object in the similarity comparison object group, generate the object information view of the candidate object, and output the object identity normalization result; Based on the object identity normalization result, the identity evolution events of the candidate objects are governed, and the object identity evolution governance result is output and automatically synchronized to the upper-layer application activity center; wherein, the identity evolution events include merge evolution events, split evolution events and successive evolution events.

2. The method for unifying and evolving the identity of multi-source objects based on an ontology world model according to claim 1, characterized in that, The process of receiving multi-source observation data, processing the multi-source observation data based on the normalized processing rules defined by the ontology world model, and outputting a representation object that satisfies ontology attribute constraints includes: The system receives the multi-source observation data, including radar detection data, electronic reconnaissance data, and manual message data, and simultaneously collects observation tags, including source identifiers, collection time, and data source credibility level. Based on the concept mapping rules defined in the ontology world model, the system performs concept mapping between the multi-source observation data and the observation tags, and outputs the first observation data. According to the concept attribution verification rules defined in the ontology world model, verify the concept attribution compatibility of the first observation data, map heterogeneous fields to ontology standard attributes, remove invalid observation data with missing required attributes, and output the second observation data. Based on the attribute type semantic rules defined in the ontology world model, the second observation data is subjected to spatiotemporal coordinate system unification, numerical unit normalization, enumeration value semantic mapping and string format standardization, and the representation object that satisfies the ontology attribute constraints is output. The normalization processing rules include the concept mapping rules, the concept attribution verification rules, and the attribute type semantic rules.

3. The method for unifying and evolving the identity of multi-source objects based on an ontology world model according to claim 1, characterized in that, The ontology attribute constraints are attribute semantic constraints bound to the ontology world model, including concept inheritance compatibility constraints, legal range constraints of attribute value domains, semantic classification constraints of attribute types, and discrimination weight configuration constraints.

4. The method for unifying and evolving the identity of multi-source objects based on an ontology world model according to claim 1, characterized in that, The step of calculating the object similarity between the described object and the candidate object in the similarity comparison object group according to the object category corresponding to the similarity comparison object group and the corresponding similarity comparison calculation rule includes: If the object category corresponding to the similarity comparison object group is the text category, then the normalized edit distance calculation rule corresponding to the text category is matched; Extract the strings corresponding to the description object and the candidate object in the similarity comparison object group respectively, and perform case unification, invalid character removal and standard alias mapping on the strings corresponding to the description object and the candidate object according to the ontology attribute constraints to generate a similarity comparison string group; For the similarity comparison string group, based on the normalized edit distance calculation rule, the string similarity between the representation object and the string corresponding to the candidate object is calculated. as follows: In the formula, The string representing the object being described String corresponding to the candidate object The Lewinstein edit distance between them; max(m,n) represents the maximum value between the string length m corresponding to the representation object and the string length n corresponding to the candidate object; Obtain a semantic equivalence term library. If the string corresponding to the described object and the candidate object does not belong to the same semantic equivalence term, then the string similarity is used as the object similarity. If the string corresponding to the described object and the candidate object belongs to the same semantic equivalence term, then the string similarity is raised to a preset semantic matching degree and used as the object similarity.

5. The method for unifying and evolving the identity of multi-source objects based on an ontology world model according to claim 4, characterized in that, If the object category corresponding to the similarity comparison object group is the numeric category, then the corresponding tolerance normalization difference calculation rule is matched for the numeric category; Extract the numerical values ​​corresponding to the description object and the candidate object respectively, and perform measurement unit unification, numerical precision unification and outlier filtering on the numerical values ​​corresponding to the description object and the candidate object according to the ontology attribute constraints to generate similarity comparison numerical groups. For the similarity comparison numerical group, calculate the absolute difference between the numerical values ​​corresponding to the described object and the candidate object. And retrieve the preset allowable tolerance threshold. ; Based on the tolerance-normalized difference calculation rule, the numerical similarity between the numerical values ​​corresponding to the described object and the candidate object is calculated. As the object similarity, the numerical similarity The calculation formula is as follows: 。 6. The method for unifying and evolving the identity of multi-source objects based on an ontology world model according to claim 4, characterized in that, If the object category corresponding to the similarity comparison object group is the location category, then the spatiotemporal joint distance decay calculation rule corresponding to the location category is matched; Extract the position coordinates corresponding to the described object and the candidate object, as well as the acquisition time corresponding to the described object and the candidate object, respectively; According to the ontological attribute constraints, the location coordinates corresponding to the description object and the candidate object are subjected to geographic coordinate system one, and the collection time corresponding to the description object and the candidate object is unified with time reference, and the similarity comparison spatiotemporal data group is generated. For the similarity comparison spatiotemporal data set, the spatial distance d is calculated based on the position coordinates corresponding to the described object and the candidate object, and the acquisition time difference is calculated based on the acquisition time corresponding to the described object and the candidate object. ; Retrieve preset space allowance threshold With time allowable threshold ; Based on the spatiotemporal joint distance attenuation calculation rule, the positional similarity between the described object and the candidate object is calculated using a Gaussian attenuation model. As the object similarity, the positional similarity The calculation formula is as follows: 。 7. The method for unifying and evolving the identity of multi-source objects based on an ontology world model according to claim 1, characterized in that, Based on the object identity normalization result, the identity evolution events of the candidate objects are governed, and the object identity evolution governance result is output and automatically synchronized to the upper-layer application activity center; wherein, the identity evolution events include merge evolution events, split evolution events, and successive evolution events, including: Based on the object identity normalization result, identity evolution trigger signals are identified, and the source credibility and attribute compliance of the identity evolution trigger signals are verified based on the ontology attribute constraints. Invalid evolution trigger signals are eliminated, and the identity evolution events corresponding to the verified identity evolution trigger signals are output. The identity evolution events include the merge evolution event, the split evolution event, and the successive evolution event. For the merging evolution event, all involved object referencing relationships are merged and written into the redirection mapping function; for the splitting evolution event, the original object referencing relationships are split and clustered and reorganized according to the new features; for the successive evolution event, successive relationships between objects are established and selective merging is performed, outputting the referencing relationship set and the redirection table. Based on the reference relationship set and the redirection table, each evolution operation is registered as the identity evolution event, the triggering evidence, the operation subject and the operation time are recorded, the object identification lineage is constructed, and the evolution governance file is output. Based on the redirection table and the evolution governance archive, object references in the upper-layer application activities are automatically mapped to surviving object identifiers through a unified fact source, and the object identity evolution governance results are output and automatically synchronized to the upper-layer application activity center.

8. A multi-source object identity normalization and evolution governance system based on an ontology world model, characterized in that, The system, applied to the multi-source object identity normalization and evolution governance method based on the ontology world model as described in any one of claims 1-7, comprises: The representation object output module is used to receive multi-source observation data, process the multi-source observation data based on the normalized processing rules defined by the ontology world model, and output a representation object that satisfies the ontology attribute constraints. The object category comparison module is used to acquire system objects, compare the object categories of the description object with those of the system objects based on the object category comparison rules defined in the ontology world model, filter out the system objects with the same object category as the description object as candidate objects, and output a similarity comparison object group composed of the description object and the candidate objects with the same object category; wherein, the object category includes text category, number category and location category; The identity normalization module is used to match the corresponding similarity comparison calculation rules according to the object category corresponding to the similarity comparison object group, calculate the object similarity between the described object and the candidate object in the similarity comparison object group, generate the object information view of the candidate object, and output the object identity normalization result. The identity evolution governance module is used to govern the identity evolution events of the candidate objects based on the identity normalization results of the objects, output the object identity evolution governance results and automatically synchronize them to the upper-layer application activity center; wherein, the identity evolution events include merge evolution events, split evolution events and successive evolution events.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor runs the computer program stored in the memory, the processor executes the steps of the multi-source object identity normalization and evolution governance method based on the ontology world model as described in any one of claims 1-7.

10. A readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it is used to implement the steps of the multi-source object identity normalization and evolution governance method based on the ontology world model as described in any one of claims 1-7.

Images