Power customer address accuracy studying and judging method and system

Abstract

The invention relates to a power customer address accuracy research and judgment method and system, and belongs to the technical field of address research and judgment, and the method comprises the following steps: obtaining latest standard administrative division data and division codes, and building a standard address database; performing structured preprocessing on the addresses of the power customers, performing step-by-step matching with the addresses in the standard address database according to five-level address systems of provinces, cities, districts, counties, streets and communities, and marking the addresses as abnormal if unmatched levels exist; verifying the power customer addresses marked to be abnormal by adopting an intelligent algorithm, and classifying the power customer address problems according to a verification result; automatically generating a work order according to an address problem classification result, issuing the work order to a power supply unit for field check, and establishing a feedback mechanism to track a governance progress; and intelligent algorithm parameters and verification rules are adjusted according to treatment feedback to improve the research and judgment accuracy.

Description

Technical Field

[0001] This application relates to the technical field of address research and judgment, and more specifically, to a method and system for judging the accuracy of power customer addresses. Background Art

[0002] In the customer management of the power industry, the accuracy of address data is crucial and involves core operations such as electricity bill settlement, fault repair, and customer service. In the existing technology, the judgment of address accuracy mainly relies on rule-based string matching or basic edit distance algorithms. For example, a Chinese patent application with the publication number CN114168705A discloses an address standardization method, which includes extracting elements through regular expressions and matching a standard database; a patent application with the publication number CN114048797A uses an edit distance algorithm to detect address similarity. However, these methods have significant problems:

[0003] High misjudgment rate: The existing methods lack hierarchical processing and cannot distinguish spelling mistakes, semantic deviations, and administrative division changes (such as historical changes in street names), resulting in a large number of false alarms. For example, simple string matching may misjudge "Haidian District, Beijing" and "Haidian District, Beijing" (traditional Chinese) as incorrect, while in fact, it is just a difference in expression.

[0004] Poor adaptability: Rule-based systems (such as regular expressions) are difficult to process unstructured addresses (such as missing levels or colloquial expressions), and cannot dynamically adapt to administrative division changes, requiring frequent manual maintenance.

[0005] Insufficient intelligence: Existing algorithms (such as a single edit distance) do not combine semantic parsing, cannot capture context dependencies (such as the semantic similarity between "community" and "residential area"), and lack a feedback mechanism, resulting in a decline in the judgment accuracy over time.

[0006] Low efficiency: The preprocessing stage relies on manual intervention (such as filling in missing levels). In a large-scale power system (such as tens of millions of customer addresses), the processing is time-consuming and error-prone.

[0007] Therefore, there is an urgent need for an intelligent and hierarchical method for judging address accuracy to reduce the misjudgment rate, improve the automation level, and support dynamic feedback optimization. Summary of the Invention

[0008] To solve the above technical problems, the present invention proposes a method and system for judging the accuracy of power customer addresses.

[0009] The technical solution of the present invention is as follows:

[0010] The present invention proposes a method for judging the accuracy of power customer addresses, including the following steps:

[0011] Step S1: Obtain the latest standard administrative division data and division codes, and establish a standard address database. The standard address structure is a five-level system of province, city, district / county, street, and community.

[0012] Step S2: After the electricity customer's address is preprocessed in a structured manner, it is matched with the address in the standard address database according to the five-level address system of province, city, district / county, street and community. If there is a mismatch at a certain level, it is marked as abnormal.

[0013] Step S3: Verify the marked abnormal power customer addresses using an intelligent algorithm, and classify the power customer address problems according to the verification results;

[0014] Step S4: Based on the address problem classification results, work orders are automatically generated and sent to the power supply unit for on-site verification, and a feedback mechanism is established to track the progress of the governance; based on the governance feedback, the intelligent algorithm parameters and verification rules are adjusted to improve the accuracy of the judgment.

[0015] Preferably, the structured processing of electricity customer address data specifically involves:

[0016] The electricity customer address data is structured according to the format of "province + city + district / county + street + community + detailed address";

[0017] For electricity customers' address data, which is unstructured, regular expression matching is used to extract administrative division elements;

[0018] Automatically completes the superior administrative division code for missing hierarchical addresses.

[0019] Preferably, the intelligent algorithm includes: an edit distance algorithm and a deep semantic address parsing matching algorithm;

[0020] The edit distance between the marked abnormal power customer address and the standard address is calculated using the Edit Distance algorithm to filter out the candidate set of spelling errors;

[0021] A semantic consistency verification algorithm based on deep learning is used to perform semantic parsing and matching on the addresses in the spelling candidate set;

[0022] When the semantic similarity reaches a preset threshold, it is judged as an anomaly of expression difference; when it is below the threshold, it is marked as a substantive error.

[0023] Preferably, the classification of electricity customer address issues based on the verification results specifically includes:

[0024] Address issues are categorized into five types: spelling errors, non-standard formats, semantic discrepancies, changes in administrative divisions, and differences in expression.

[0025] Set confidence score thresholds for different types of anomalies;

[0026] Automatically generate difference analysis reports for suspected errors that exceed the threshold.

[0027] Preferably, the Edit Distance algorithm is a hierarchy-aware dynamic edit distance formula, specifically as follows:

[0028]

[0029] In the formula: ED layer The value of the dynamic edit distance; S and T are the hierarchical sequences of the address to be verified and the standard address, respectively; n is the total number of address levels; i is the address level index; w i D represents the weight coefficients of the i-th layer; α, β, and γ are the basic cost coefficients for insertion, deletion, and replacement operations, respectively; insert (i), D delete (i) and D substitute (i) represents the character difference at level i for insertion, deletion, and replacement operations, respectively.

[0030] Preferably, the deep semantic address parsing and matching algorithm adopts a hierarchical address encoding model, which combines a hierarchical dedicated BERT and a bidirectional GRU network. Its matching degree calculation function is as follows:

[0031]

[0032] In the formula: Sim(A,B) is the matching degree value between address A and address B; k is the address level index; σ(·) is the activation function; λ k The weight representing the contribution of the k-th level to the matching degree; Attn(·) is used to calculate the attention similarity. , respectively, are the k-th level encoding vectors generated by the hierarchical BERT model for addresses A and B; cos(·) is the cosine similarity calculation; , respectively, are the k-th level cross-level dependency feature vectors captured by the bidirectional GRU network for addresses A and B; μ is the weight parameter.

[0033] Preferably, the feedback mechanism includes:

[0034] Collect GPS positioning data from on-site verification and verify its spatial coordinates against the work order address;

[0035] Statistics on the work order response time and error correction accuracy of each power supply unit;

[0036] Generate regional address anomaly heatmaps to guide key governance areas.

[0037] On the other hand, the present invention also provides a power customer address accuracy assessment system, comprising:

[0038] The standard address database construction module obtains the latest standard administrative division data and division codes, and establishes a standard address database. The standard address structure is a five-level system of province, city, district / county, street, and community.

[0039] The hierarchical matching module performs structured preprocessing on the addresses of electricity customers and then matches them with addresses in the standard address database according to a five-level address system: province, city, district / county, street, and community. If a mismatch exists, it is marked as an anomaly.

[0040] The verification and filtering module uses an intelligent algorithm to verify the addresses of electricity customers marked as abnormal, and filters the electricity customer data with address problems based on the verification results.

[0041] The verification and feedback module automatically generates work orders based on the address problem classification results and sends them to the power supply unit for on-site verification. It also establishes a feedback mechanism to track the progress of the remediation. Based on the remediation feedback, it adjusts the intelligent algorithm parameters and verification rules to improve the accuracy of the judgment.

[0042] In another aspect, the present invention also provides an electronic device having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements a method for determining the accuracy of an electricity customer's address as described in any embodiment of the present invention.

[0043] In another aspect, the present invention also provides a computer-readable medium for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement a method for determining the accuracy of power customer addresses as described in any embodiment of the present invention.

[0044] The present invention has the following beneficial effects:

[0045] The five-level address system (province, city, district / county, street, community) is matched step by step. Combined with hierarchical perception editing distance and deep semantic parsing, it accurately distinguishes spelling errors, semantic deviations, and administrative division changes, reducing the misjudgment rate. The preprocessing module automates the structuring of non-standard addresses, the efficiency of extracting elements using regular expressions is improved, and missing hierarchical completion reduces manual intervention. This method can also be transferred to fields that rely on address data, such as logistics and e-commerce, to support the construction of smart city infrastructure. Attached Figure Description

[0046] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0048] 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.

[0049] It should be understood that the step numbers used in the text are for ease of description only and are not intended to limit the order in which the steps are performed.

[0050] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0051] The terms “comprising” and “including” indicate the presence of the described feature, whole, step, operation, element and / or component, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0052] The term “and / or” refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0053] Example 1:

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, specific embodiments of this application will be described below, with reference to the accompanying drawings. Figure 1 The technical solution of the present invention will be clearly and completely described.

[0055] To address the problems in existing technologies, this invention provides a method for determining the accuracy of electricity customer addresses, comprising the following steps:

[0056] Step S1: Obtain the latest standard administrative division data and division codes, and establish a standard address database. The standard address structure is a five-level system of province, city, district / county, street, and community.

[0057] Obtain the latest standard administrative division data and division codes through the API interface, and obtain GPS coordinates in batches through the administrative division codes to build a standard address database.

[0058] Step S2: After the electricity customer's address is preprocessed in a structured manner, it is matched with the address in the standard address database according to the five-level address system of province, city, district / county, street and community. If there is a mismatch at a certain level, it is marked as abnormal.

[0059] As a preferred embodiment of this example, the structured processing of the electricity customer address data specifically involves:

[0060] The electricity customer address data is structured according to the format of "province + city + district / county + street + community + detailed address";

[0061] For electricity customer address data, which is unstructured, regular expression matching is used to extract administrative division elements; the regular expression extraction rules include:

[0062] Provincial matching rules: end with "province", full name of municipality, full name of autonomous region, and abbreviation of province, etc.;

[0063] City-level matching rules: Ending with "city", followed by autonomous prefecture, region, league, etc.

[0064] Matching rules at the district / county level: those ending with "district", "county", "flag", or special administrative divisions (Dongguan → directly administered municipality), etc.

[0065] Street-level matching rules: end with "street", end with "town", end with "village", end with "road / main road", etc.

[0066] Community-level matching rules: end with "community", end with "village", end with "residential area / garden", end with "garden / park", etc.

[0067] Automatically complete the superior administrative division code for missing hierarchical addresses, for example:

[0068] Input address: XX Community, XX Street, Xihu District. The province / city level field is missing. The administrative division code for "Xihu District" is "330106", which belongs to Hangzhou City "330100", and Hangzhou City belongs to Zhejiang Province "33". The system will automatically complete the superior administrative division code.

[0069] Step S3: Verify the marked abnormal power customer addresses using an intelligent algorithm, and classify the power customer address problems according to the verification results;

[0070] As a preferred embodiment of this invention, the intelligent algorithm includes: an edit distance algorithm and a deep semantic address parsing matching algorithm;

[0071] The Edit Distance algorithm is used to calculate the edit distance between the marked abnormal power customer address and the standard address to filter the candidate set of spelling errors. Based on the edit distance, the candidate set of spelling errors can be divided into an automatic correction candidate set and a semantic verification candidate set. If the edit distance of the automatic correction candidate set is less than the preset edit distance, it is classified as a spelling error and is corrected using the automatic correction method. If the edit distance of the semantic verification candidate set is greater than or equal to the preset edit distance, further semantic verification is required.

[0072] A semantic consistency verification algorithm based on deep learning is used to perform semantic parsing and matching on the addresses in the spelling candidate set;

[0073] When the semantic similarity reaches a preset threshold, it is judged as an anomaly of expression difference and classified as expression difference; when it is below the threshold, it is marked as a substantial error and classified as semantic deviation.

[0074] In a preferred embodiment of this invention, the Edit Distance algorithm is a hierarchy-aware dynamic edit distance formula, specifically as follows:

[0075]

[0076] In the formula: ED layer The value of the dynamic edit distance; S and T are the hierarchical sequences of the address to be verified and the standard address, respectively; n is the total number of address levels, which is 5; i is the address level index; w i w represents the weight coefficient of the i-th layer. i =2 (n-i) The higher the level, the greater the weight; α, β, and γ are the basic cost coefficients for insertion, deletion, and replacement operations, respectively. Unlike traditional editing with a fixed unit cost (e.g., a cost of 1 per operation), these can be dynamically adjusted, for example, α = 1, β = 1, γ = 1.2, reflecting the actual differences in the impact of different error types; D insert (i), D delete (i) and D substitute (i) represents the character difference degree of insertion, deletion, and replacement operations at level i, calculated as follows: Where S i and T i Let be the string at level i, and lev be the Levenshtein distance; |S i| and | T i | represents the length of the string at the i-th level.

[0077] In a preferred embodiment of this invention, the deep semantic address parsing and matching algorithm employs a hierarchical address encoding model. This model combines a hierarchical dedicated BERT network with a bidirectional GRU network, and its matching degree calculation function is as follows:

[0078]

[0079] The attention mechanism employed is dot product attention.

[0080]

[0081] In the formula: Sim(A,B) is the matching degree value between address A and address B. The closer the output value is to 1, the higher the address matching degree; the closer it is to 0, the lower the matching degree; k is the address level index, which calculates the similarity contribution layer by layer for the five levels of the address (province, city, district / county, street, community); σ(·) is the activation function; λ k The contribution weight of the k-th level to the matching degree is automatically learned through model training; Attn(·) is the attention similarity calculation, used to capture semantic consistency within the level; These are the k-th level encoding vectors generated by the hierarchical BERT model for addresses A and B, respectively; cos(·) is the cosine similarity calculation, used to measure the similarity of the address structure context; , respectively, are the k-th level cross-level dependency feature vectors captured by address A and address B through a bidirectional GRU network; μ is a trainable weight parameter.

[0082] In a preferred embodiment of this example, the classification of electricity customer address issues based on the verification results specifically involves:

[0083] Address issues are categorized into five types: spelling errors, non-standard formatting, semantic discrepancies, changes in administrative divisions, and differences in expression; specifically:

[0084] Spelling errors: These are classified as spelling errors based on calculations where the edit distance is less than a preset value and the semantic similarity is higher than a preset value.

[0085] Non-standard format: Failure to match using regular expressions during the data preprocessing stage is attributed to non-standard address format.

[0086] Semantic bias: Problems are classified as semantic bias based on whether the edit distance is greater than a preset value and the semantic similarity is lower than a preset value.

[0087] Administrative division change: The issue is identified as an administrative division change by matching the historical division codes in the standard address database with intelligent algorithms.

[0088] Expression difference: This is determined by calculating the edit distance, which is greater than a preset value, but the semantic similarity reaches a preset value.

[0089] Set confidence score thresholds for different types of anomalies;

[0090] Automatically generate difference analysis reports for suspected errors that exceed the threshold.

[0091] Step S4: Based on the address problem classification results, work orders are automatically generated and sent to the power supply unit for on-site verification, and a feedback mechanism is established to track the progress of the governance; based on the governance feedback, the intelligent algorithm parameters and verification rules are adjusted to improve the accuracy of the judgment.

[0092] As a preferred embodiment of this invention, the feedback mechanism includes:

[0093] The GPS positioning data collected for on-site verification is compared with the work order address for spatial coordinate verification. The error distance (such as Euclidean distance) of the GPS coordinate verification is used to automatically adjust the basic cost coefficients α, β and γ in the distance editing formula: if the average error is large, γ (replacement operation cost) is increased to more strictly handle substantive errors.

[0094] The response time and error correction accuracy of work orders for each power supply unit are statistically analyzed; the credibility score threshold (e.g., threshold = basic threshold × accuracy) is calculated using the response time and error correction accuracy of work orders, and dynamic adjustment is achieved.

[0095] Generate regional address anomaly heatmaps to guide key governance areas.

[0096] Example 2:

[0097] The standard address database construction module obtains the latest standard administrative division data and division codes, and establishes a standard address database. The standard address structure is a five-level system of province, city, district / county, street, and community.

[0098] The hierarchical matching module performs structured preprocessing on the addresses of electricity customers and then matches them with addresses in the standard address database according to a five-level address system: province, city, district / county, street, and community. If a mismatch exists, it is marked as an anomaly.

[0099] The verification and filtering module uses an intelligent algorithm to verify the addresses of electricity customers marked as abnormal, and filters the electricity customer data with address problems based on the verification results.

[0100] The verification and feedback module automatically generates work orders based on the address problem classification results and sends them to the power supply unit for on-site verification. It also establishes a feedback mechanism to track the progress of the remediation. Based on the remediation feedback, it adjusts the intelligent algorithm parameters and verification rules to improve the accuracy of the judgment.

[0101] Example 3:

[0102] This embodiment provides an electronic device that stores a computer program. When the computer program is executed by a processor, it implements a method for judging the accuracy of power customer addresses as described in any embodiment of the present invention.

[0103] Example 4:

[0104] This embodiment provides a computer-readable medium for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement a method for determining the accuracy of power customer addresses as described in any embodiment of the present invention.

[0105] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0106] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0107] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0108] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, essentially, or the part that contributes to the prior art, or a portion 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 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 described in 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.

[0109] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for judging the accuracy of electricity customer addresses, characterized in that, Includes the following steps: Step S1: Obtain the latest standard administrative division data and division codes, and establish a standard address database. The standard address structure is a five-level system of province, city, district / county, street, and community. Step S2: After the electricity customer's address is preprocessed in a structured manner, it is matched with the address in the standard address database according to the five-level address system of province, city, district / county, street and community. If there is a mismatch at a certain level, it is marked as abnormal. Step S3: Verify the marked abnormal power customer addresses using an intelligent algorithm, and classify the power customer address problems according to the verification results; Step S4: Based on the address problem classification results, work orders are automatically generated and sent to the power supply unit for on-site verification, and a feedback mechanism is established to track the progress of the governance; based on the governance feedback, the intelligent algorithm parameters and verification rules are adjusted to improve the accuracy of the judgment.

2. The method for judging the accuracy of electricity customer addresses according to claim 1, characterized in that: The specific steps for structuring the address data of electricity customers are as follows: The electricity customer address data is structured according to the format of "province + city + district / county + street + community + detailed address"; For electricity customers' address data, which is unstructured, regular expression matching is used to extract administrative division elements; Automatically completes the superior administrative division code for missing hierarchical addresses.

3. The method for judging the accuracy of electricity customer addresses according to claim 1, characterized in that: The intelligent algorithms include: the Edit Distance algorithm and the Deep Semantic Address Resolution Matching algorithm; The edit distance between the marked abnormal power customer address and the standard address is calculated using the Edit Distance algorithm to filter out the candidate set of spelling errors; A semantic consistency verification algorithm based on deep learning is used to perform semantic parsing and matching on the addresses in the spelling candidate set; When the semantic similarity reaches a preset threshold, it is judged as an anomaly of expression difference; when it is below the threshold, it is marked as a substantive error.

4. The method for judging the accuracy of electricity customer addresses according to claim 1, characterized in that: The classification of electricity customer address issues based on the verification results is as follows: Address issues are categorized into five types: spelling errors, non-standard formats, semantic discrepancies, changes in administrative divisions, and differences in expression. Set confidence score thresholds for different types of anomalies; Automatically generate difference analysis reports for suspected errors that exceed the threshold.

5. The method for judging the accuracy of electricity customer addresses according to claim 3, characterized in that: The Edit Distance algorithm is a hierarchy-aware dynamic edit distance formula, specifically as follows: In the formula: ED layer The value of the dynamic edit distance; S and T are the hierarchical sequences of the address to be verified and the standard address, respectively; n is the total number of address levels; i is the address level index; w i D represents the weight coefficients of the i-th layer; α, β, and γ are the basic cost coefficients for insertion, deletion, and replacement operations, respectively; insert (i), D delete (i) and D substitute (i) represents the character difference at level i for insertion, deletion, and replacement operations, respectively.

6. The method for judging the accuracy of electricity customer addresses according to claim 3, characterized in that: The deep semantic address parsing and matching algorithm adopts a hierarchical address encoding model, which combines a hierarchical dedicated BERT and a bidirectional GRU network. Its matching degree calculation function is as follows: In the formula: Sim(A,B) is the matching degree value between address A and address B; k is the address level index; σ(·) is the activation function; λ k The weight representing the contribution of the k-th level to the matching degree; Attn(·) is used to calculate the attention similarity. , respectively, are the k-th level encoding vectors generated by the hierarchical BERT model for addresses A and B; cos(·) is the cosine similarity calculation; These are the k-th level cross-level dependency feature vectors captured by the bidirectional GRU network for addresses A and B, respectively. μ is the weighting parameter.

7. The method for judging the accuracy of electricity customer addresses according to claim 1, characterized in that: The feedback mechanism includes: Collect GPS positioning data from on-site verification and verify its spatial coordinates against the work order address; Statistics on the work order response time and error correction accuracy of each power supply unit; Generate regional address anomaly heatmaps to guide key governance areas.

8. A system for judging the accuracy of electricity customer addresses, characterized in that, include: The standard address database construction module obtains the latest standard administrative division data and division codes, and establishes a standard address database. The standard address structure is a five-level system of province, city, district / county, street, and community. The hierarchical matching module performs structured preprocessing on the addresses of electricity customers and then matches them with addresses in the standard address database according to a five-level address system: province, city, district / county, street, and community. If a mismatch exists, it is marked as an anomaly. The verification and filtering module uses an intelligent algorithm to verify the addresses of electricity customers marked as abnormal, and filters the electricity customer data with address problems based on the verification results. The verification and feedback module automatically generates work orders based on the address problem classification results and sends them to the power supply unit for on-site verification. It also establishes a feedback mechanism to track the progress of the remediation. Based on the remediation feedback, it adjusts the intelligent algorithm parameters and verification rules to improve the accuracy of the judgment.

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 program, it implements a method for judging the accuracy of power customer addresses as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements a method for determining the accuracy of power customer addresses as described in any one of claims 1 to 7.

Images