Resource alignment method and device between heterogeneous systems
By using a domain regularization engine and a standardized terminology library to process resource names from heterogeneous systems, the problem of difficult resource association in cross-system integration is solved, achieving efficient and accurate resource alignment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING ZHUOYIDA TECH CO LTD
- Filing Date
- 2025-11-27
- Publication Date
- 2026-05-01
AI Technical Summary
Existing technologies suffer from low accuracy in extracting key entities and difficulty in associating resources due to a lack of domain knowledge during cross-system integration and data consolidation. They also fail to identify domain abbreviations, model variations, etc. For example, Chinese word segmentation tools may incorrectly segment structured names.
We employ a domain-specific regular expression engine and library. By extracting key fields from resources and standardizing them using a standardized terminology library, we calculate the matching degree between resources based on field type matching weights and similarity. This process is then combined with manual review to ensure accuracy.
It improves the accuracy of key entity extraction, reduces the difficulty and mismatch rate of resource alignment, and achieves efficient resource alignment, making it suitable for cross-system data integration tasks in multiple fields.
Smart Images
Figure CN121959037A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the fields of data governance and intelligent matching technology, and in particular to a method and apparatus for resource alignment between heterogeneous systems. Background Technology
[0002] During cross-system integration and data consolidation, different business systems have significantly different naming rules for the same entity (such as data center, equipment, assets, etc.). Therefore, cross-system integration and data consolidation require associating the same resources.
[0003] Existing technologies mainly utilize general word segmentation methods for key entity extraction, but they lack domain knowledge, resulting in low extraction accuracy and difficulties in resource association. For example, Chinese word segmentation tools may incorrectly segment structured names and fail to recognize domain-specific abbreviations and model variations. For instance, "manufacturer switch model" is incorrectly split into "manufacturer," "switch," and "machine model." They also fail to recognize that "OLT" and "optical line terminal" refer to the same entity, and that "V3.0" and "Version3.0" have the same meaning. Summary of the Invention
[0004] This application provides a method and apparatus for resource alignment between heterogeneous systems, aiming to solve the problem that the existing technology uses general word segmentation methods to extract key entities, which leads to difficulties in resource association.
[0005] Firstly, this application provides a method for resource alignment between heterogeneous systems, including: Obtain the resource sets of two target alignment systems, where each resource in the resource set has a resource name; Based on the domain regular expression engine, obtain the first key field among all resources in two resource sets; Based on the first key field, determine the first matching degree between any first resource in the first resource set and all second resources in the second resource set; The correspondence between the first resource in the first resource set and the second resource in the second resource set is determined based on all the first matching degrees.
[0006] As an example, the first key field among all resources in two resource sets is obtained based on a domain regularization engine, specifically including: The domain to which the target alignment system belongs is determined based on the data characteristics of the resource set, and the domain's domain regularization engine and domain regularization library are invoked. Extract key field extraction rules from the domain regular expression library; Based on the key field extraction rules, the second key field is extracted from all resources in the two resource sets using a domain regularization engine; The first key field is determined based on the second key field.
[0007] As one example, determining the first key field based on the second key field specifically includes: Call the domain's standardized terminology database, which stores the correspondence between domain-specific terms and their synonyms and aliases, the correspondence between pinyin and the domain's Chinese name, the correspondence between domain abbreviations and their full names, and the correspondence between the domain's simplified and complete expressions; The first key field is obtained by standardizing the second key field based on a standardized terminology database.
[0008] As one embodiment, determining the first matching degree between any first resource in the first resource set and all second resources in the second resource set based on a first key field specifically includes: Extract matching rules for each field type in the domain from the domain regular expression library. The matching rules include the matching weights of the field types. Determine the similarity between each third key field in the first resource and the fourth key field in the second resource, where the field types of the third key fields are the same as those of the fourth key fields; The second matching degree between the first resource and the second resource is determined based on the matching weight of each field type and the similarity of the field types. The first matching degree between the first resource and the second resource is determined based on the second matching degree.
[0009] As one example, determining the similarity between each third key field in the first resource and the fourth key field in the second resource specifically includes: Extract the matching priority order of field types in the domain from the domain regular expression library; Based on priority from high to low, perform the following steps for the specified field type: If both the first and second resources have key fields of a specified field type, determine the first similarity between the third key field of the specified field type and the fourth key field of the specified field type. If the first similarity is greater than or equal to the similarity threshold of the specified field type, calculate the second similarity between the third key field of the specified field type of the next priority in the first resource and the fourth key field of the specified field type of the next priority in the second resource. If the first similarity is lower than the similarity threshold of the specified field type, set the similarity of all key fields between the first resource and the second resource to zero; The specified field type is the field type with the highest matching priority among the preset number of field types.
[0010] As one example, determining the similarity between each third key field in the first resource and the fourth key field in the second resource specifically includes: For text fields of the same type, the similarity between the third and fourth key fields is determined based on the edit distance between the third and fourth key fields; For coded fields of the same type, the similarity between the third and fourth key fields is determined based on the exact matching principle.
[0011] As one embodiment, determining the first matching degree between the first resource and the second resource based on the second matching degree specifically includes: If the second matching degree is within the matching degree threshold range, a manual review request is output, which indicates that the second matching degree between the first resource and the second resource should be manually reviewed. Receive the results of manual review; The first match degree is determined based on the results of manual review.
[0012] Secondly, this application provides a resource alignment device between heterogeneous systems, comprising: The resource acquisition module is used to acquire the resource sets of two target alignment systems, and each resource in the resource set has a resource name; The key field retrieval module is used to obtain the first key field among all resources in two resource sets based on the domain regular expression engine; The matching degree determination module is used to determine the first matching degree between any first resource in the first resource set and all second resources in the second resource set based on the first key field. The correspondence determination module is used to determine the correspondence between the first resource in the first resource set and the second resource in the second resource set based on all first matching degrees.
[0013] Thirdly, this application also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-mentioned resource alignment methods between heterogeneous systems.
[0014] Fourthly, this application also provides a non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements any of the above-mentioned resource alignment methods between heterogeneous systems.
[0015] Fifthly, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements any of the above-described resource alignment methods between heterogeneous systems. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 This is an architecture diagram of the resource alignment system between heterogeneous systems provided in this application; Figure 2 This is one of the flowcharts illustrating the resource alignment method between heterogeneous systems provided in this application; Figure 3 This application provides an instance of the first key field among all resources in two resource sets obtained using a domain regularization engine. Figure 4 This is one of the schematic diagrams of the matching score calculation model provided in this application; Figure 5 This is one of the flowcharts of the feedback optimization layer optimization rule base provided in this application; Figure 6 This is a schematic diagram of the interactive data display interface of the matching comparison panel provided in this application; Figure 7 This is one of the structural schematic diagrams of the resource alignment device between heterogeneous systems provided in this application; Figure 8 This is a schematic diagram of the structure of the electronic device provided in this application. Detailed Implementation
[0018] 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. 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.
[0019] It should be noted that, in the description of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.
[0020] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects have an "or" relationship.
[0021] It should be noted that this application is for automated data integration and alignment across systems and platforms, and is particularly suitable for the automated alignment of data containing Chinese characters. This application can be widely applied to various alignment tasks in multiple fields, such as telecommunications, finance, government affairs, and industrial IoT, including cross-network management system device alignment tasks, device ledger and IoT data association tasks, and cross-departmental building and asset data integration tasks.
[0022] The following is combined Figures 1 to 8 This application describes a method and apparatus for resource alignment between heterogeneous systems.
[0023] Figure 1 This is an architecture diagram of the resource alignment system between heterogeneous systems provided in this application. For example... Figure 1 As shown, the resource alignment system between heterogeneous systems provided in this application includes a data input layer, a processing core layer, an output layer, and a feedback optimization layer.
[0024] The data input layer includes a database interface, an API gateway, and a file parser. The database interface is responsible for interacting with various databases and performing CRUD operations (Create, Read, Update, Delete). The API gateway handles the routing, authentication, and rate limiting of external requests. The file parser supports the parsing and conversion of formats such as CSV, JSON, and XML.
[0025] The core processing layer comprises a parsing engine and a matching core. The parsing engine includes a rule base and a regular expression processor. The rule base stores domain-specific regular expression libraries, which contain domain-specific parsing rules (e.g., key field extraction rules, standardized terminology libraries, etc.) and matching rules. These rules can be modified or configured as needed. The regular expression processor includes domain-specific regular expression engines for extracting and matching key fields from complex text. The matching core includes a weight calculator and a conflict detection module. The weight calculator calculates the weights of various field types within each domain, while the conflict detection module identifies rule conflicts and provides solutions when matching key fields across different systems.
[0026] The output layer includes a matching relationship table, a visualization interface, and a manual review interface. The matching relationship table is used for structured storage and output of the final matching results (e.g., the matching relationship table itself). The visualization interface provides interactive data display. Figure 6 An interactive data display interface for the matching and comparison panel is shown. A manual review interface is provided to support human intervention and result correction.
[0027] The feedback optimization layer is equipped with a rule optimization engine, which is used to continuously optimize the rule base through machine learning.
[0028] In one possible implementation, the system adopts a microservice architecture, with the parsing engine and the matching core communicating via an API (Application Programming Interface). The output layer supports matching relationship tables in XML / JSON / Excel formats and a visual comparison interface.
[0029] It should be noted that the resource alignment method between heterogeneous systems provided in this application embodiment is implemented based on the resource alignment device between heterogeneous systems. The resource alignment method between heterogeneous systems integrates domain knowledge into the domain regularity library and the domain regularity engine, obtains key fields applicable to the domain to which the resource belongs through structured parsing, avoids errors generated by general word segmentation, improves extraction accuracy, and reduces the difficulty and accuracy of resource alignment.
[0030] This application describes the resource alignment method between heterogeneous systems using a resource alignment device between heterogeneous systems as an example.
[0031] Figure 2 This is one of the flowcharts illustrating the resource alignment method between heterogeneous systems provided in this application.
[0032] like Figure 2 As shown, the resource alignment method between heterogeneous systems provided in this application includes: S210: Obtain the resource set for the two target alignment systems. Each resource in the resource set has a resource name.
[0033] Combination Figure 1 It can receive raw data from two resource sets through JDBC (Java Database Connectivity), API gateway, or file interface. Each resource in the resource set has a unique identifier and resource name.
[0034] In one possible implementation, the target alignment system identifies resources based on a unique identifier, with the resource name used to assist in resource identification.
[0035] S220: Obtain the first key field among all resources in two resource sets based on the domain regularization engine.
[0036] The domain-specific regular expression engine is used to break down the resource name of each resource into key fields belonging to its domain. For example, for resource names in the data center / building domain, key fields include region, campus, building number, and room number; for resource names in the network equipment domain, key fields include manufacturer, equipment type, model, and version; and for resource names in the industrial equipment domain, key fields include equipment category, brand, and specifications.
[0037] Understandably, data preprocessing of the resource collection is necessary before extracting key fields, such as using regular expressions to filter stop words (e.g., filtering "model" and "version" in the industrial equipment field).
[0038] S230: Determine the first matching degree between any first resource in the first resource set and all second resources in the second resource set based on the first key field.
[0039] Specifically, the key fields of all field types in the first resource are matched with the key fields of the same field type in the second resource, and then the first matching degree is obtained based on the matching results.
[0040] S240: Determine the correspondence between the first resource in the first resource set and the second resource in the second resource set based on all first matching degrees.
[0041] In one possible implementation, the second resource with the highest matching degree is determined for each first resource, and this second resource is used as the corresponding second resource. The unique identifiers of the corresponding first resources and second resources are then associated. Subsequently, a matching relationship table is obtained based on the corresponding first and second resources.
[0042] This application's embodiments integrate domain knowledge into a domain regular expression library and a domain regular expression engine. Through structured parsing, key fields applicable to the domain to which the resource belongs are obtained, avoiding errors caused by general word segmentation, improving extraction accuracy, and reducing the difficulty and accuracy of resource alignment.
[0043] In one possible implementation, step S220 involves obtaining the first key field from all resources in the two resource sets based on a domain regularization engine, specifically including: S2210: Determine the domain to which the target alignment system belongs based on the data characteristics of the resource set, and call the domain regularization engine and domain regularization library of that domain.
[0044] Data characteristics include the organizational structure of the resource set, field structure characteristics, and data content characteristics.
[0045] S2220: Extract key field extraction rules from the domain regular expression library to determine the domain to which the resource belongs.
[0046] Key field extraction rules refer to which key fields need to be extracted from a given domain. These key fields are those that are of high importance within that domain. For example, for resource names in the data center / building domain, key fields include region, campus, building number, and room number; for resource names in the network equipment domain, key fields include manufacturer, equipment type, model, and version; and for resource names in the industrial equipment domain, key fields include equipment category, brand, and specifications.
[0047] S2230: Based on key field extraction rules, use the domain regularization engine to extract the second key field from all resources in two resource sets.
[0048] For example, for data center / computer room resources: key fields can be extracted from “^^(.*?)(Data Center|IDC)(.*?)(\d+)[Building No.]?”: region, campus, building number; For network device resources: Key fields can be extracted from “^^(Manufacturer A|Manufacturer B)(Switch|Router)?(\w+)(V?\d+\.?\d*)”: Manufacturer, Series (Type), Model, Version.
[0049] S2240: Determine the first key field based on the second key field.
[0050] This application's embodiments determine the domain-specific regular expression engine and library used based on the data characteristics of the resources, thereby accurately extracting key fields according to domain rules. On the one hand, domain rules enable correct word segmentation of resource names; on the other hand, they reduce the probability of extracting useless key fields. Furthermore, by configuring the domain regular expression library, the system can quickly adapt to resource alignment tasks in new industries.
[0051] Current resource alignment methods use simple matching algorithms, but the models lack domain adaptability and have poor generalization ability for technical terms. For example, the models cannot provide accurate matching results for resource names that use abbreviations such as "PTN (Packet Transport Network) equipment" and "5G BBU (Building Baseband Unit)". The abbreviation of the equipment model "PLC-3000" and the full expression "Programmable Controller 3000" do not match. The building name "Municipal Government Building No. 1" and its alias "Government Affairs Center Main Building" do not match. The names of the same equipment, "Manufacturer Series Model" and "Series-Model Switch", do not match because of the presence of a hyphen "-".
[0052] Based on the above considerations, in one possible implementation, step S2240, determining the first key field based on the second key field, specifically includes: S2241: Call the standardized terminology database of the domain to which the resource belongs. The standardized terminology database stores the correspondence between professional terms in the domain and their synonyms and aliases, the correspondence between pinyin and the Chinese name of the domain, the correspondence between the abbreviations and the full name of the domain, and the correspondence between the simplified expression and the complete expression of the domain.
[0053] For example: "CHANGSHANG" = "manufacturer" (correspondence between pinyin and Chinese name of the field); "S series" = "S57 series" (correspondence between abbreviation and full expression); "A building" = "building A" (correspondence between professional terminology and alias in the field); "V2.0" = "version 2.0" (correspondence between professional terminology and alias in the field); "switch" = "SW" = "Switch" (correspondence between Chinese name, abbreviation and full English name).
[0054] S2242: Standardize the second key field based on the standardized terminology library to obtain the first key field.
[0055] In one possible implementation, based on the correspondence in a standardized terminology database, the first key field with the corresponding relationship is unified into one form, such as unifying it into an abbreviation, like replacing "data center" and "Internet Data Center" with "IDC".
[0056] Figure 3 One example is shown. For example... Figure 3As shown, the original name of the resource is "Vendor A Model 1 Switch". After regularization, key fields are extracted. Then, it is determined whether the extracted key fields are standard abbreviations. If not, a standardized terminology library (such as a thesaurus) is used for standardization to obtain the field extraction results. If the extracted key fields are standard abbreviations, the field extraction results are obtained directly. Figure 3 As shown, the field extraction results are "Manufacturer (field type): Manufacturer A); Model (field type): Model 1; Type (field type): Switch", thus obtaining the standardized output of the key fields.
[0057] This application's embodiments standardize key fields using a standardized terminology library, reducing the difficulty of matching key fields subsequently and improving matching efficiency.
[0058] In one possible implementation, step S230, determining the first matching degree between any first resource in the first resource set and all second resources in the second resource set based on the first key field, specifically includes: S2310: Extract matching rules for each field type in the domain from the domain regular expression library.
[0059] The matching rules include the matching weight of field types. The more important the field type, the greater the matching weight.
[0060] In one possible implementation, the matching weights for different field types are consistent across the same domain or resource type. Understandably, this can be adjusted periodically or irregularly using a weight counter (see [link to relevant documentation]). Figure 1 The matching weights are dynamically adjusted and then synchronized to the domain regular expression library.
[0061] S2320: Determine the similarity between each third key field in the first resource and the fourth key field in the second resource, wherein the field type of the third key field is the same as that of the fourth key field.
[0062] This step matches key fields on a per-field-type basis.
[0063] The key field extraction process may encounter the following problems: all key fields are not standardized, or the standardization of some key fields fails. Based on this consideration, in one possible implementation, in step S2320, if the similarity between the third and fourth key fields of the same field type is lower than a similarity threshold, the third and / or fourth key fields can be expanded to their corresponding other key fields for matching based on a standardized terminology library, thereby obtaining multiple combinations. The highest similarity among all combinations is taken as the similarity between the third and fourth key fields.
[0064] S2330: Determine the second matching degree between the first resource and the second resource based on the matching weight of each field type and the similarity of the field types.
[0065] Taking network device resources as an example, the second matching degree = 40% × manufacturer similarity + 30% × model similarity + 20% × version similarity + 10% × type similarity, such as... Figure 4 As shown.
[0066] S2340: Determine the first matching degree between the first resource and the second resource based on the second matching degree.
[0067] This application embodiment increases the influence of important field types on the matching degree by designing matching weights for field types, thereby avoiding false matching.
[0068] In one possible implementation, step S2320, determining the similarity between each third key field in the first resource and the fourth key field in the second resource, specifically includes: P1: Extract the matching priority order of field types in the domain from the domain regular expression library.
[0069] The field type matching priority order refers to the order in which field types are calculated when calculating similarity. For example, if the similarity is calculated as manufacturer > model > version, then the similarity of the manufacturer will be calculated first, followed by the similarity of the model and version in turn.
[0070] The field type with the highest matching priority (pre-defined number) is designated as the specified field type. For example, in the above example, the manufacturer and model number are designated as the specified field types.
[0071] P2: Based on priority from high to low, perform the following steps for the specified field type: P21: In cases where both the first and second resources have key fields of a specified field type (e.g., manufacturer), determine the first similarity between the third key field of the specified field type and the fourth key field of the specified field type.
[0072] If either the first or second resource lacks a critical field of a specified field type (e.g., manufacturer), a manual review process is triggered.
[0073] P22: If the first similarity is greater than or equal to the similarity threshold of the specified field type, calculate the second similarity between the third key field of the specified field type of the next priority in the first resource and the fourth key field of the specified field type of the next priority in the second resource.
[0074] P23: If the first similarity is lower than the similarity threshold of the specified field type, set the similarity of all key fields between the first resource and the second resource to zero.
[0075] P3: Having obtained the similarity of all specified field types, determine the similarity between the third key field of other field types in the first resource and the fourth key field of the same field type in the second resource.
[0076] Taking the example of Manufacturer > Model > Version, with Manufacturer and Model as the specified field types, the matching rule for Manufacturer is that the Manufacturers of the two resources must be the same, i.e., the similarity between Manufacturers is 0 (inconsistent) or 1 (consistent). If the similarity between Manufacturers is 0, then the first and second resources do not match, and the similarity of all key fields between the first and second resources is set to zero, ending the matching process between the first and second resources. This achieves automatic interception based on the principle of mandatory rule verification. If the similarity between Manufacturers is 1, then the similarity between the Model in the first resource and the Model in the second resource is calculated. If the similarity threshold for Model is 0.9, and the actual similarity is 0.93, then all specified field types have similarity, and the similarity calculation continues for non-specified field types (such as Version).
[0077] This application embodiment verifies the similarity determination process through mandatory rules, and automatically intercepts fields with higher importance when their similarity does not meet the requirements based on the priority order of field types, which can reduce the error rate of matching and improve matching efficiency.
[0078] In one possible implementation, step S2320, determining the similarity between each third key field in the first resource and the fourth key field in the second resource, specifically includes: For text fields of the same type (e.g., manufacturer, model), the similarity between the third and fourth key fields is determined based on the edit distance between them; for coded fields of the same type (e.g., version), the similarity between the third and fourth key fields is determined based on the exact matching principle.
[0079] In one possible implementation, please combine Figure 4 Each field type has its own specific matching rules. For example, in the vendor matching rule, key fields must be completely identical, and matching via an alias library is supported. In the type matching rule, matching via synonym substitution is supported. In the version matching rule, lenient matching is supported.
[0080] In one possible implementation, a lower similarity threshold is applied to text fields with lower priority. Please refer to... Figure 4In the model matching rules, the maximum allowed edit distance is 2, such as "Series A" ≈ "Series B". Furthermore, differences caused by hyphens "-", forward slashes " / ", and spaces are ignored during model matching.
[0081] The embodiments of this application determine the similarity calculation rules according to the classification of fields, thereby improving the accuracy of the calculation.
[0082] In one possible implementation, step S2340, determining the first matching degree between the first resource and the second resource based on the second matching degree, specifically includes: S2341: If the second matching degree is within the matching degree threshold range, output a manual review request. The manual review request indicates that the second matching degree between the first resource and the second resource be manually reviewed.
[0083] In one possible implementation, when the second matching degree is in the range of 0.7-0.9, the matching degree in this range cannot accurately determine whether two resources match, and at this time, the manual review process is triggered.
[0084] S2342: Receive the results of manual review.
[0085] S2343: Determine the first matching degree based on the results of manual review.
[0086] Based on this embodiment, please refer to Figure 5 In the resource alignment system between heterogeneous systems provided in this application, the rule optimization engine optimizes the rule base using the following method: For error cases marked after manual review, the error type is analyzed. If it is a parsing error, the regularization rules are adjusted using machine learning algorithms; otherwise, the weight parameters are modified using machine learning algorithms. Regression testing is then performed based on the adjustment results. If the accuracy improves, the rule base is updated using the adjustment results; otherwise, the error type is re-analyzed.
[0087] In this embodiment of the application, when the automatically obtained matching degree cannot accurately determine whether two resources match, confirmation is made through manual review. This fault-tolerance mechanism can improve the overall matching degree.
[0088] Based on the above, this application improves the accuracy of key field extraction through a domain regularization library and can achieve automatic alignment, which not only improves the accuracy of alignment but also improves the alignment efficiency. In addition, the domain regularization engine of this application is implemented based on pure rules, which eliminates the need for model training, achieves lightweight deployment, and is suitable for low-computing-power environments.
[0089] Actual testing shows that this application achieves an accuracy rate of over 99% in scenarios such as telecommunications and industrial IoT (compared to <80% for traditional methods), with a single match taking less than 0.1 seconds and a false match rate of less than 0.5%, significantly outperforming traditional matching schemes and effectively solving the problem of automated alignment of resources between heterogeneous systems.
[0090] Based on the above, this application also provides a resource alignment device between heterogeneous systems. The resource alignment device between heterogeneous systems and the aforementioned resource alignment method between heterogeneous systems can be referred to in correspondence with each other.
[0091] As an example, such as Figure 7 As shown, the resource alignment device between heterogeneous systems provided in this application includes: The resource acquisition module 710 is used to acquire resource sets of two target alignment systems, where each resource in the resource set has a resource name.
[0092] Key field retrieval module 720 is used to retrieve the first key field from all resources in two resource sets based on a domain regularization engine.
[0093] The matching degree determination module 730 is used to determine the first matching degree between any first resource in the first resource set and all second resources in the second resource set based on the first key field.
[0094] The correspondence determination module 740 is used to determine the correspondence between the first resource in the first resource set and the second resource in the second resource set based on all first matching degrees.
[0095] This application's embodiments integrate domain knowledge into a domain regular expression library and a domain regular expression engine. Through structured parsing, key fields applicable to the domain to which the resource belongs are obtained, avoiding errors caused by general word segmentation, improving extraction accuracy, and reducing the difficulty and accuracy of resource alignment.
[0096] Figure 8 This is a schematic diagram of the structure of the electronic device provided in this application, such as... Figure 8 As shown, the electronic device may include: a processor 810, a communications interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communications interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute a resource alignment method between heterogeneous systems, the method including: Obtain the resource sets of two target alignment systems, where each resource in the resource set has a resource name; Based on the domain regular expression engine, obtain the first key field among all resources in two resource sets; Based on the first key field, determine the first matching degree between any first resource in the first resource set and all second resources in the second resource set; The correspondence between the first resource in the first resource set and the second resource in the second resource set is determined based on all the first matching degrees.
[0097] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0098] On the other hand, this application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a computer, the computer is able to execute the resource alignment method between heterogeneous systems provided in the above embodiments. The method includes: Obtain the resource sets of two target alignment systems, where each resource in the resource set has a resource name; Based on the domain regular expression engine, obtain the first key field among all resources in two resource sets; Based on the first key field, determine the first matching degree between any first resource in the first resource set and all second resources in the second resource set; The correspondence between the first resource in the first resource set and the second resource in the second resource set is determined based on all the first matching degrees.
[0099] In another aspect, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, is implemented to perform the resource alignment method between heterogeneous systems provided in the above embodiments, the method comprising: Obtain the resource sets of two target alignment systems, where each resource in the resource set has a resource name; Based on the domain regular expression engine, obtain the first key field among all resources in two resource sets; Based on the first key field, determine the first matching degree between any first resource in the first resource set and all second resources in the second resource set; The correspondence between the first resource in the first resource set and the second resource in the second resource set is determined based on all the first matching degrees.
[0100] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0101] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.
[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.
Claims
1. A method for resource alignment between heterogeneous systems, characterized in that, include: Obtain resource sets for two target alignment systems, where each resource in the resource set has a resource name; Based on the domain regularization engine, obtain the first key field among all resources in the two resource sets; Based on the first key field, determine the first matching degree between any first resource in the first resource set and all second resources in the second resource set; The correspondence between the first resource in the first resource set and the second resource in the second resource set is determined based on all the first matching degrees.
2. The resource alignment method between heterogeneous systems according to claim 1, characterized in that, Based on the domain regularization engine, the first key field among all resources in the two resource sets is obtained, specifically including: Based on the data characteristics of the resource set, the domain to which the target alignment system belongs is determined, and the domain regularization engine and domain regularization library of the domain are invoked; Extract key field extraction rules from the domain regular expression library; Based on the key field extraction rules, the second key field is extracted from all resources in the two resource sets using the domain regularization engine; The first key field is determined based on the second key field.
3. The resource alignment method between heterogeneous systems according to claim 2, characterized in that, The first key field is determined based on the second key field, specifically including: The standardized terminology database of the field is invoked. The standardized terminology database stores the correspondence between the professional terms of the field and their synonyms and aliases, the correspondence between the pinyin and the Chinese name of the field, the correspondence between the abbreviations and the full name of the field, and the correspondence between the simplified expression and the complete expression of the field. The second key field is standardized based on the standardized terminology database to obtain the first key field.
4. The resource alignment method between heterogeneous systems according to claim 2, characterized in that, Determining the first matching degree between any first resource in the first resource set and all second resources in the second resource set based on the first key field specifically includes: Extract matching rules for each field type in the domain from the domain regular expression library, wherein the matching rules include the matching weights for the field types; Determine the similarity between each third key field in the first resource and the fourth key field in the second resource, wherein the field type of the third key field is the same as that of the fourth key field; The second matching degree between the first resource and the second resource is determined based on the matching weight of each field type and the similarity of the field types. A first matching degree is determined between the first resource and the second resource based on the second matching degree.
5. The resource alignment method between heterogeneous systems according to claim 4, characterized in that, Determining the similarity between each third key field in the first resource and the fourth key field in the second resource specifically includes: Extract the matching priority order of field types in the domain from the domain regular expression library; Based on priority from high to low, perform the following steps for the specified field type: If both the first resource and the second resource contain a key field of the specified field type, determine the first similarity between the third key field of the specified field type and the fourth key field of the specified field type. If the first similarity is greater than or equal to the similarity threshold of the specified field type, calculate the second similarity between the third key field of the specified field type of the next priority in the first resource and the fourth key field of the specified field type of the next priority in the second resource; If the first similarity is lower than the similarity threshold of the specified field type, the similarity of all key fields between the first resource and the second resource is set to zero. The specified field type is a preset number of field types with the highest matching priority.
6. The resource alignment method between heterogeneous systems according to claim 4 or 5, characterized in that, Determining the similarity between each third key field in the first resource and the fourth key field in the second resource specifically includes: For text fields of the same type, the similarity between the third key field and the fourth key field is determined based on the edit distance between the third key field and the fourth key field; For coded fields of the same type, the similarity between the third key field and the fourth key field is determined based on the exact matching principle.
7. The resource alignment method between heterogeneous systems according to claim 4, characterized in that, Determining the first matching degree between the first resource and the second resource based on the second matching degree specifically includes: If the second matching degree is within the matching degree threshold range, a manual review request is output, which indicates that the second matching degree between the first resource and the second resource is manually reviewed. Receive the results of manual review; The first matching degree is determined based on the results of the manual review.
8. A resource alignment device between heterogeneous systems, characterized in that, include: The resource acquisition module is used to acquire resource sets of two target alignment systems, wherein each resource in the resource set has a resource name; The key field acquisition module is used to obtain the first key field among all resources in the two resource sets based on the domain regularization engine; The matching degree determination module is used to determine the first matching degree between any first resource in the first resource set and all second resources in the second resource set based on the first key field; The correspondence determination module is used to determine the correspondence between the first resource in the first resource set and the second resource in the second resource set based on all first matching degrees.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the resource alignment method between heterogeneous systems as described in any one of claims 1 to 7.
10. A computer program product, the computer program product comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the resource alignment method between heterogeneous systems as described in any one of claims 1 to 7.