Address verification method and device, equipment and storage medium

Abstract

The invention discloses an address verification method and device, electronic equipment and a storage medium. According to the technical scheme, the address components and the combination rule are defined, the components of the address information to be verified are combined according to the preset address combination rule, the candidate address information is obtained, the candidate address information is verified through the pre-trained address verification model, and the address verification efficiency is improved. According to the address checking method, the problem that a large number of address databases are traversed in the address checking process is solved, the address checking efficiency is improved, the preset address combination rule can adapt to new place name standards and changes, various industry applications are supported, and the address checking efficiency is improved. And the method has good expandability.

Description

Technical Field

[0001] The present invention relates to the field of address verification technology, and in particular to an address verification method, device, equipment and storage medium. Background Art

[0002] Due to the non-standard nature of address information, addresses are expressed in a variety of ways in different regions and scenarios. Users often use a variety of different expressions when entering addresses. This diverse format lacks a unified standard, which makes it difficult for the system to parse and understand the address when it is actually used. Manual screening and confirmation are required, which greatly increases the workload of user intervention.

[0003] Traditional address parsing algorithms are usually unable to handle complex address structures and have difficulty coping with combinations and sequence changes of different formats, which also creates difficulties in the process of data cleaning and standardization. At the same time, existing address standards often lack unified composition rules, and different industries or application scenarios (such as logistics, navigation, real estate, etc.) may adopt different standards, further increasing the complexity of address information processing.

[0004] Traditional address verification algorithms rely on rule matching, which usually requires traversing a large number of address databases. Due to the optimized search and matching algorithm, it takes a long time to verify the matching address information, especially when the data volume is large, which leads to problems such as poor user experience, delayed processing and waste of resources. Summary of the invention

[0005] The present invention provides an address verification method, device, equipment and storage medium, which solve the problem of traversing a large number of address databases during the address verification process to achieve address verification.

[0006] In a first aspect, an embodiment of the present invention provides an address verification method, the method comprising:

[0007] Determine first address element information, second address element information and third address element information of the address information to be verified; the first address element information includes an address element representing an administrative division, the second address element information includes an address element representing a basic area, and the third address element information includes an address element representing a local point location;

[0008] The address elements in the first address element information, the address elements in the second address element information, and the address elements in the third address element information are combined according to a preset address combination rule to obtain at least one candidate address information of the address information to be verified; the preset address combination rule includes an address priority rule for the first address element information, an address priority rule for the second address element information, and an address priority rule for the third address element information;

[0009] The candidate address information is verified using a pre-trained address verification model to determine the target address information, and the address information to be verified is verified using the target address information. The address verification model is obtained by training a neural network model based on sample address data and address annotation information.

[0010] In a second aspect, an embodiment of the present invention further provides an address verification device, the device comprising:

[0011] An address element information determination module, used to determine first address element information, second address element information and third address element information of the address information to be verified; the first address element information includes an address element representing an administrative division, the second address element information includes an address element representing a basic area, and the third address element information includes an address element representing a local point location;

[0012] An address element determination module, configured to combine the address elements in the first address element information, the address elements in the second address element information, and the address elements in the third address element information according to a preset address combination rule to obtain at least one candidate address information of the address information to be verified;

[0013] The address verification module is used to verify the candidate address information using a pre-trained address verification model, determine the target address information, and use the target address information to verify the address information to be verified. The address verification model is obtained by training a neural network model based on sample address data and address annotation information.

[0014] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, an address verification method as described in any one of the embodiments of the present invention is implemented.

[0015] In a fourth aspect, an embodiment of the present invention further provides a storage medium storing computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, are used to execute the address verification method as described in any one of the embodiments of the present invention.

[0016] The technical solution of the embodiment of the present invention defines address components and combination rules, combines the various components of the address information to be verified according to preset address combination rules, obtains candidate address information, uses a pre-trained address verification model to verify the candidate address information, obtains the target address information and performs address verification on the address information to be verified based on it, implements the address verification method, solves the problem of traversing a large number of address databases during the address verification process, improves the address verification efficiency, and the preset address combination rules can adapt to new place name standards and changes, support a variety of industry applications, and have good scalability.

[0017] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 is a flow chart of an address verification method provided by Embodiment 1 of the present invention;

[0020] Figure 2 is a schematic diagram of obtaining candidate address information by combining according to a preset address combination rule provided in the first embodiment of the present invention;

[0021] Figure 3 is a structural diagram of an address verification device provided by Embodiment 2 of the present invention;

[0022] Figure 4 It is a schematic diagram of the structure of an electronic device for implementing the address verification method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] Embodiment 1

[0026] Figure 1 A flowchart of an address verification method is provided for the first embodiment of the present invention. This embodiment is applicable to address verification situations. The method can be performed by an address verification device. The address verification device can be implemented in the form of hardware and / or software. The address verification device can be configured in any electronic device with network communication and computing capabilities. Figure 1 As shown, the method includes:

[0027] S110, determining first address element information, second address element information, and third address element information of address information to be verified.

[0028] In this embodiment, the first address element information includes an address element representing an administrative division, the second address element information includes an address element representing a basic area, and the third address element information includes an address element representing a local point location.

[0029] Specifically, the first address element information can be address elements such as province, city, district, town, exclusive area, village, etc., the second address element information can be address elements such as road, area, villagers' group, etc., and the third address element information can be address elements such as house number, subsidiary house number, building address, landmark, point of interest, etc.

[0030] In this embodiment, the composition of the address information to be verified is divided into first address element information, second address element information and third address element information, and each element of an address is separated out to make the address structure clear and improve the address verification efficiency.

[0031] As an optional but non-limiting implementation, determining the first address element information, the second address element information and the third address element information of the address information to be verified includes steps A1-A2:

[0032] Step A1: pre-process the address information to be verified to determine candidate address elements of the address information to be verified.

[0033] Step A2: classify the candidate address elements to determine first address element information, second address element information and third address element information of the address information to be verified.

[0034] In this embodiment, the preprocessing includes at least address denoising and address segmentation of the address information to be verified. Among them, address denoising includes removing irrelevant characters and filtering advertisements. There may be extra spaces, commas, periods and other punctuation marks in the address information to be verified. The string processing function in the programming language can be used to remove the spaces at both ends of the string and replace specific punctuation marks with empty strings to clean up irrelevant characters and make the text of the address information to be verified more regular. For example, "Chaoyang District, Beijing" is processed as "Chaoyang District, Beijing". In addition, meaningless special symbols such as "#" and "&" or garbled characters (which may be caused by data entry errors, etc.) may be mixed in the address information to be verified. Regular expressions can be written to match and delete these characters that do not meet the address specifications. For example, for addresses such as "Chaoyang District, Beijing", remove "%" and "#".

[0035] Secondly, the text of the address information to be verified may be interspersed with some advertising slogans (such as "high-quality housing, located in Chaoyang District, Beijing") or notes (such as "this address is close to the subway station, Haidian District, Beijing"). You can pre-set some keyword lists such as "high-quality housing" and "close to", and determine whether the text contains these keywords, then remove the keywords and the irrelevant content behind them, and only retain the key address part. Normal addresses often have a certain reasonable length range and format characteristics. If a certain section of the address information to be verified is too long and does not conform to the conventional address structure (for example, it contains a large section of descriptive text), it can be determined that it may contain more noise, and it can be appropriately discarded or some content that conforms to the address format characteristics can be extracted.

[0036] It should be noted that the segmentation of the address information to be verified may be based on the administrative region level, based on the universal separator, or based on the address database. In this embodiment, the segmentation method of the address information to be verified is not specifically limited.

[0037] The candidate address element represents the address element obtained after the address information to be checked is denoised and segmented. It should be noted that the candidate address element in this embodiment can be, for example, address elements such as "Beijing City", "Jiangsu Province", "Heping District", "Central Avenue", "Daping Township", etc.

[0038] Specifically, it can be divided into provincial, municipal, district and county levels and below. Taking "No. 113, Xuanwu Street, Gulou District, Nanjing City, Jiangsu Province" as an example, you can first extract the provincial administrative district name by searching or matching the position of the keyword "province" through a string, and use this as the starting point for subsequent segmentation. For example, identify "Jiangsu Province" as the first level part of the address. Then use the keyword "city" to find its position and separate the city-level name. Like the above example, the city-level part "Nanjing City" is obtained, which is usually after the provincial name. Similarly, based on keywords such as "district", "county", "street", "town", and "township" that represent different administrative levels or grassroots regional units, gradually segment the corresponding district and county level, street and township level, and specific house numbers and other more detailed address components, and finally obtain a segmentation result list such as "Jiangsu Province", "Nanjing City", "Gulou District", "Xuanwu Street", and "No. 113" as candidate address elements.

[0039] Furthermore, according to the first address element information being address elements such as province, city, district, town, exclusive area, village, etc., the second address element information being address elements such as road, district, villagers' group, etc., and the third address element information being address elements such as house number, subsidiary house number, building address, landmark, point of interest, etc., the candidate address elements are classified to determine the first address element information, the second address element information and the third address element information of the address information to be verified.

[0040] S120. Combine the address elements in the first address element information, the address elements in the second address element information, and the address elements in the third address element information according to a preset address combination rule to obtain at least one candidate address information of the address information to be verified.

[0041] In this embodiment, the preset address combination rule includes an address priority rule of the first address element information, an address priority rule of the second address element information, and an address priority rule of the third address element information. The candidate address information represents the address information after combining the address element in the first address element information, the address element in the second address element information, and the address element in the third address element information.

[0042] Specifically, the address priority rule of the first address element information may be that in the place names limited by administrative divisions, if provinces, cities, districts and counties can all be used, districts and counties take precedence over cities and provinces. The address priority rule of the second address element information may be that in the place names limited by basic areas, if both district names and street names can be used for description, street names take precedence over district names. The address priority rule of the third address element information may be that in the local point location, if landmark names, point of interest names and door (building) addresses can all be used for description, door (building) addresses take precedence over landmark names, and landmark names take precedence over point of interest names. This application does not specifically limit the address priority rule of the first address element information, the address priority rule of the second address element information, and the address priority rule of the third address element information.

[0043] The preset address combination rule can be to combine in order of administrative levels or address ranges from large to small, such as following the order of provincial level - municipal level - district and county level - street and specific house number. According to this rule, in the above example, "Jiangsu Province" should first be combined with "Nanjing City", "Gulou District" and other subsequent level elements in sequence, and "Zhejiang Province" should similarly be combined with "Hangzhou City", "Xihu District" and so on.

[0044] Secondly, we can try to ensure that the candidate address information is complete, that is, it covers the necessary address levels. We cannot just combine to the city level, unless partial address presentation is allowed in a specific scenario. Generally, we need to include smaller address units that can be located (such as streets and house numbers, etc.).

[0045] In addition, the address naming standards and existence conditions must be met. For example, some provinces do not have a specific city name, or a city does not have a corresponding district or county name. The existing standard address database can be used to verify whether the combination of each element is legal, avoiding the combination of "Guangzhou City, Jiangsu Province" that does not conform to the actual address.

[0046] As an optional but non-limiting implementation, the address elements in the first address element information, the address elements in the second address element information, and the address elements in the third address element information are combined according to a preset address combination rule to obtain at least one candidate address information of the address information to be verified, including steps B1-B4:

[0047] Step B1: Determine the first address element from the first address element information according to the address priority rule of the first address element information.

[0048] Step B2: Determine the second address element from the second address element information according to the address priority rule of the second address element information.

[0049] Step B3: Determine the third address element from the third address element information according to the address priority rule of the third address element information.

[0050] Step B4: Combine the first address element, the second address element, and the third address element to obtain candidate address information.

[0051] In this embodiment, the first address element represents the address element determined from the first address element information according to the address priority rule of the first address element information, the second address element represents the address element determined from the second address element information according to the address priority rule of the second address element information, and the third address element represents the address element determined from the third address element information according to the address priority rule of the third address element information.

[0052] Among them, the first address element can be "Province A", "City B" or "District / County C", the second address element can be "Avenue a", "Area b" or "C Villagers Group (Brigade)", and the third address element can be "Building 2", "501", "Forbidden City", "Square" or "Library", etc.

[0053] Furthermore, according to the address priority rules of address elements such as province, city, district and county, district and county take precedence over city and province; street and lane names take precedence over area names, door (building) addresses take precedence over landmark names, and landmark names take precedence over points of interest names, at least one candidate address information can be determined by combining the first address element, the second address element, and the third address element in sequence, thereby ensuring the correct regional level of the candidate address information and improving the accuracy of the verified address.

[0054] See also Figure 2 The figure shows a schematic diagram of combining candidate address information according to preset address combination rules. In this embodiment, different candidate address information can be combined according to administrative divisions + basic areas. For example, [province] [city] <district and county> [town] [village] <road> [road] [area] [villager group], if the address elements in the square brackets exist, they are combined, if not, they are not combined, but the address elements in the angle brackets must exist, if not, they cannot be combined, after confirming the combination form, at least one candidate address information is obtained.

[0055] S130: Use a pre-trained address verification model to verify the candidate address information, determine the target address information, and use the target address information to verify the address information to be verified.

[0056] In this embodiment, the address verification model is obtained by training a neural network model based on sample address data and address annotation information, and the target address information is address information determined from candidate address information according to the address verification model.

[0057] It should be noted that in this embodiment, the candidate address information is verified through the trained address verification model, and it can be determined whether the candidate address information contains the necessary address information of the administrative division, basic area, and local point location, and whether it is a correct address, so that the candidate address information that best matches the actual address is used as the target address information and the address information to be verified is verified.

[0058] As an optional but non-limiting implementation, the process of constructing the address verification model includes steps C1-C2:

[0059] Step C1, obtaining sample address data and address annotation information, wherein the address annotation information is annotation information of the first address element information, the second address element information and the third address element information of the sample data, and matching information of the first address element information and the second address element information, and matching information of the second address element information and the third address element information.

[0060] Step C2: training a neural network model according to the sample address data and address annotation information to obtain an address verification model.

[0061] In this embodiment, the sample address data is used as input data for training the neural network model, and the address annotation information includes annotations of the first address element information, the second address element information, and the third address element information of the sample data, as well as matching information of the first address element information and the second address element information, and matching information of the second address element information and the third address element information. The matching information of the first address element information and the second address element information is that the first address element in the first address element information matches the second address element in the second address element information, or the first address element information does not match the second address element information, the second address element in the second address element information matches the third address element in the third address element information, or the second address element in the second address element information does not match the third address element in the third address element information.

[0062] In actual applications, sample address data covering different types, regions and formats can be obtained from local government service websites and geographic information system public platforms. These websites may contain address data specific to the street and community levels to support local urban management, public services and other work, or cover detailed administrative division addresses, important landform addresses and other information, integrating more refined address resources in the city, involving detailed addresses of residential communities, commercial outlets, municipal facilities, etc., and can be collected with the help of the interface or data download function provided by the platform.

[0063] Furthermore, the obtained sample address data is preprocessed, and then manually labeled or automatically matched with address data in a public geographic information system to obtain labeling information of the first address element information, labeling information of the second address element information, and labeling information of the third address element information of the sample address data, as well as matching information of the first address element information and the second address element information, and matching information of the second address element information and the third address element information.

[0064] The annotation information of the first address element information of the sample address data indicates the administrative division annotation in the sample address data, the annotation information of the second address element information of the sample address data indicates the basic area annotation in the sample address data, and the annotation information of the third address element information of the sample address data indicates the local point position annotation in the sample address data. The matching information of the first address element information and the second address element information indicates whether the address between the administrative division and the basic area in the sample address data matches, and the matching information of the second address element information and the third address element information indicates whether the address between the basic area and the local point position in the sample address data matches.

[0065] For example, if "Guangzhou City (first address element information) Tianma Avenue (second address element information)" does not exist in the geographic information system, the matching information between the first address element information and the second address element information of "Guangzhou City Tianma Avenue" is that the first address element information and the second address element information do not match.

[0066] Furthermore, the neural network model is trained according to the sample address data and address annotation information to obtain an address verification model. Commonly used neural network models include recurrent neural networks, convolutional neural networks, and decision trees.

[0067] In this embodiment, the decision tree can be trained by using the sample address data and address annotation information, so that the decision tree learns the features and rules in the sample address data to obtain an address verification model.

[0068] When the address information to be verified needs to be verified, the address information to be verified is input into the address verification model based on the decision tree, and the judgment result of whether the address is valid is output, thereby realizing the address verification function.

[0069] As an optional but non-limiting implementation, a neural network model is trained based on the sample address data and its address annotation information to obtain an address verification model, including steps D1-D4:

[0070] Step D1: Use the sample address data and address annotation information as sample data, and divide the sample data into a training set and a test set.

[0071] Step D2: Use the training set to train the neural network model to obtain a trained neural network model.

[0072] Step D3: input the sample address data of the test set into the trained neural network model to obtain the address verification result.

[0073] Step D4: According to the address verification result and the address labeling information of the test set, the loss function of the neural network model is optimized until the address verification model is obtained by satisfying the preset conditions, wherein the preset conditions are that the address verification result and the address labeling information of the test set are the same or the preset training rounds are reached.

[0074] In this embodiment, the training set represents sample address data and corresponding address labeling information used to train the neural network model, and the test set represents sample address data and corresponding address labeling information used to optimize and adjust the loss function of the trained neural network model.

[0075] Specifically, the sample address data and address annotation information are used as sample data, the sample data is divided into a training set and a test set, and the training set is input into the decision tree model to train the decision tree model. The decision tree algorithm will build a decision tree based on the selected features and partitioning criteria (such as minimizing the Gini coefficient). At each node, the decision tree algorithm will evaluate all available features in the sample data, select the feature that can best distinguish between valid addresses and invalid addresses as the partition node, and then recursively build subtrees to obtain the trained decision tree model.

[0076] For example, the root node may be divided according to whether the address contains a legal country name, and then further divided in the child node according to whether it contains a legal provincial administrative region name, and so on, gradually building a complete decision tree model.

[0077] Furthermore, the sample address data of the test set is input into the trained decision tree model to obtain the address verification result. According to the address verification result and the accuracy of the address labeling information of the test set, the loss parameter of the decision tree is adjusted. The decision tree model can be adjusted by reducing the depth of the tree, increasing the minimum number of samples required for node splitting, etc., until the preset conditions are met, and the address verification model is obtained, so that the decision tree model can output the target verification result. The preset conditions can be that the address verification result is the same as the address labeling information of the test set or the preset training rounds are reached.

[0078] As an optional but non-limiting implementation, the candidate address information is verified using a pre-trained address verification model to determine the target address information, and the target address information is used to verify the address information to be verified, including steps E1-E3:

[0079] Step E1: Use the pre-trained address verification model to verify the candidate address information and determine the verification result of the candidate address information.

[0080] Step E2: determine the target address information according to the verification result, wherein the target address information is address information including first address element information, second address element information, and third address element information, and the first address element information matches the second address element information, and the second address element information matches the third address element information.

[0081] Step E3: Verify the target address information with the address information to be verified.

[0082] In this embodiment, the verification result is the marking information of the first address element information, the marking information of the second address element information, and the marking information of the third address element information of the candidate address information, as well as the matching information of the first address element information and the second address element information, and the matching information of the second address element information and the third address element information. The target address information is the actual address information that matches the address information to be verified, and the target address information is the address information that includes the first address element information, the second address element information, and the third address element information, and the matching of the first address element information and the second address element information, and the matching of the second address element information and the third address element information.

[0083] Furthermore, based on whether the address information contains necessary information such as administrative divisions, local point locations, basic areas, and whether the various levels match accurately, the candidate address information containing the first address element information, the second address element information, and the third address element information, and the first address element information matches the second address element information, and the second address element information matches the third address element information, is determined from the verification results as the target address information of the address information to be verified.

[0084] Furthermore, the address information to be verified is verified according to the target address information, including correcting the address information of the wrong address elements such as the province, city, district, street, etc. in the first address element information of the address information to be verified, supplementing the missing address information, or correcting the change of administrative divisions, so as to restore the address to a complete address service. For example, the address information of the missing administrative divisions (province / city / district / street), road names, and road numbers is supplemented.

[0085] In practical applications, the verified address information can also be stored based on administrative divisions, basic areas, and local point locations, and the address information and the location of the point of interest can be established and stored independently, which can meet the needs of address search and matching of the point of interest location. By entering the location of the point of interest and returning the relevant address information, you can use pinyin search (full spelling + first letter) and Chinese character search (including typos) to search for addresses. In addition, the address information verified by this submission verification method can continuously expand the standard address information of the place name address library, and continuously iterate and optimize the place name address library through feedback, thereby improving the accuracy and completeness of the geographic information system.

[0086] The technical solution of the embodiment of the present invention defines address components and combination rules, combines the various components of the address information to be verified according to preset address combination rules, obtains candidate address information, uses a pre-trained address verification model to verify the candidate address information, obtains the target address information and performs address verification on the address information to be verified based on it, implements the address verification method, solves the problem of traversing a large number of address databases during the address verification process, improves the address verification efficiency, and the preset address combination rules can adapt to new place name standards and changes, support a variety of industry applications, and have good scalability.

[0087] Embodiment 2

[0088] Figure 3 This is a schematic diagram of the structure of an address verification device provided in the second embodiment of the present invention. This embodiment is applicable to address verification situations. The address verification device can be implemented in the form of hardware and / or software. The address verification device can be configured in any electronic device with network communication and computing capabilities. Figure 3 As shown, the device comprises:

[0089] An address element information determination module 310 is used to determine first address element information, second address element information and third address element information of the address information to be verified; the first address element information includes an address element representing an administrative division, the second address element information includes an address element representing a basic area, and the third address element information includes an address element representing a local point location;

[0090] An address element determination module 320, configured to combine the address elements in the first address element information, the address elements in the second address element information, and the address elements in the third address element information according to a preset address combination rule to obtain at least one candidate address information of the address information to be verified;

[0091] The address verification module 330 is used to verify the candidate address information using a pre-trained address verification model, determine the target address information, and use the target address information to verify the address information to be verified. The address verification model is obtained by training a neural network model based on sample address data and address annotation information.

[0092] Optionally, determining first address element information, second address element information, and third address element information of address information to be verified includes:

[0093] Preprocessing the address information to be verified to determine candidate address elements of the address information to be verified; the preprocessing at least includes denoising and segmenting the address information with verification;

[0094] The candidate address elements are classified to determine first address element information, second address element information, and third address element information of the address information to be verified.

[0095] Optionally, combining the address elements in the first address element information, the address elements in the second address element information, and the address elements in the third address element information according to a preset address combination rule to obtain at least one candidate address information of the address information to be verified includes:

[0096] Determining the first address element from the first address element information according to the address priority rule of the first address element information;

[0097] determining the second address element from the second address element information according to the address priority rule of the second address element information;

[0098] determining a third address element from the third address element information according to an address priority rule of the third address element information;

[0099] The first address element, the second address element, and the third address element are combined in sequence to obtain candidate address information.

[0100] Optionally, the address verification model construction process includes:

[0101] Acquire sample address data and address annotation information, wherein the address annotation information is annotation information of first address element information, annotation information of second address element information, and annotation information of third address element information of the sample data, as well as matching information of the first address element information and the second address element information, and matching information of the second address element information and the third address element information;

[0102] The neural network model is trained according to the sample address data and address annotation information to obtain an address verification model.

[0103] Optionally, a neural network model is trained according to the sample address data and address annotation information thereof to obtain an address verification model, including:

[0104] Taking the sample address data and address annotation information as sample data, and dividing the sample data into a training set and a test set;

[0105] Using the training set to train the neural network model, to obtain a trained neural network model;

[0106] Inputting the sample address data of the test set into the trained neural network model to obtain the address verification result;

[0107] According to the address verification results and the address labeling information of the test set, the loss function of the neural network model is optimized until the address verification model is obtained by satisfying the preset conditions, wherein the preset conditions are that the address verification results and the address labeling information of the test set are the same or the preset training rounds are reached.

[0108] Optionally, using a pre-trained address verification model to verify the candidate address information, determining the target address information, and using the target address information to verify the address information to be verified, includes:

[0109] Verifying the candidate address information using a pre-trained address verification model to determine a verification result of the candidate address information;

[0110] Determine target address information according to the verification result, wherein the target address information is address information including first address element information, second address element information, and third address element information, and the first address element information matches the second address element information, and the second address element information matches the third address element information;

[0111] The target address information is verified against the address information to be verified.

[0112] Optionally, the first address element information includes at least one province, city, district, town, exclusive area, and village; the second address element information includes at least one road, area, and villagers' group; and the third address element information includes at least one house number, subsidiary house number, building address, landmark, and point of interest.

[0113] The technical solution of the embodiment of the present invention defines address components and combination rules, combines the various components of the address information to be verified according to preset address combination rules, obtains candidate address information, uses a pre-trained address verification model to verify the candidate address information, obtains the target address information and performs address verification on the address information to be verified based on it, implements the address verification method, solves the problem of traversing a large number of address databases during the address verification process, improves the address verification efficiency, and the preset address combination rules can adapt to new place name standards and changes, support a variety of industry applications, and have good scalability.

[0114] The address verification device provided in the embodiment of the present invention can execute the address verification method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0115] Embodiment 3

[0116] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0117] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0118] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0119] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 performs the various methods and processes described above, such as an address verification method.

[0120] In some embodiments, the address verification method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the address verification method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the address verification method in any other appropriate manner (e.g., by means of firmware).

[0121] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0122] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0123] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0124] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0125] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0126] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0127] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0128] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An address verification method, characterized in that: include: Determine first address element information, second address element information and third address element information of the address information to be verified; the first address element information includes an address element representing an administrative division, the second address element information includes an address element representing a basic area, and the third address element information includes an address element representing a local point location; The address elements in the first address element information, the address elements in the second address element information, and the address elements in the third address element information are combined according to a preset address combination rule to obtain at least one candidate address information of the address information to be verified; the preset address combination rule includes an address priority rule for the first address element information, an address priority rule for the second address element information, and an address priority rule for the third address element information; The candidate address information is verified using a pre-trained address verification model to determine the target address information, and the address information to be verified is verified using the target address information. The address verification model is obtained by training a neural network model based on sample address data and address annotation information.

2. The method according to claim 1, characterized in that Determining first address element information, second address element information, and third address element information of address information to be verified includes: Preprocessing the address information to be checked to determine candidate address elements of the address information to be checked; the preprocessing at least includes denoising and segmenting the address information to be checked; The candidate address elements are classified to determine first address element information, second address element information, and third address element information of the address information to be verified.

3. The method according to claim 2, characterized in that Combining the address elements in the first address element information, the address elements in the second address element information, and the address elements in the third address element information according to a preset address combination rule to obtain at least one candidate address information of the address information to be verified, including: Determining the first address element from the first address element information according to the address priority rule of the first address element information; determining the second address element from the second address element information according to the address priority rule of the second address element information; determining a third address element from the third address element information according to an address priority rule of the third address element information; The first address element, the second address element and the third address element are combined to obtain candidate address information.

4. The method according to claim 1, characterized in that The construction process of the address verification model includes: Acquire sample address data and address annotation information, wherein the address annotation information is annotation information of first address element information, annotation information of second address element information, and annotation information of third address element information of the sample data, as well as matching information of the first address element information and the second address element information, and matching information of the second address element information and the third address element information; The neural network model is trained according to the sample address data and address annotation information to obtain an address verification model.

5. The method according to claim 4, characterized in that The neural network model is trained according to the sample address data and the address annotation information to obtain an address verification model, including: The sample address data and the corresponding address annotation information are used as sample data, and the sample data is divided into a training set and a test set; Using the training set to train the neural network model, to obtain a trained neural network model; Inputting the sample address data of the test set into the trained neural network model to obtain the address verification result; According to the address verification result and the address labeling information of the test set, the loss function of the neural network model is optimized until the address verification model is obtained by satisfying the preset conditions.

6. The method according to claim 1, characterized in that Using a pre-trained address verification model to verify the candidate address information, determining the target address information, and using the target address information to verify the address information to be verified, including: Verifying the candidate address information using a pre-trained address verification model to determine a verification result of the candidate address information; Determine target address information according to the verification result, wherein the target address information is address information including first address element information, second address element information, and third address element information, and the first address element information matches the second address element information, and the second address element information matches the third address element information; The target address information is verified against the address information to be verified.

7. The method according to claim 1, characterized in that The first address element information includes at least one province, city, district, town, exclusive area, and village; the second address element information includes at least one road, area, and villagers' group; and the third address element information includes at least one house number, subsidiary house number, building address, landmark, and point of interest.

8. An address verification device, characterized in that: include: An address element information determination module, used to determine first address element information, second address element information and third address element information of the address information to be verified; the first address element information includes an address element representing an administrative division, the second address element information includes an address element representing a basic area, and the third address element information includes an address element representing a local point location; An address element determination module, configured to combine the address elements in the first address element information, the address elements in the second address element information, and the address elements in the third address element information according to a preset address combination rule to obtain at least one candidate address information of the address information to be verified; The address verification module is used to verify the candidate address information using a pre-trained address verification model, determine the target address information, and use the target address information to verify the address information to be verified. The address verification model is obtained by training a neural network model based on sample address data and address annotation information.

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, the address verification method as described in any one of claims 1 to 7 is implemented.

10. A storage medium storing computer executable instructions, characterized in that: The computer executable instructions are used to perform the address verification method as described in any one of claims 1 to 7 when executed by a computer processor.

Images