Address matching method, apparatus and storage medium

Abstract

This application provides an address matching method, apparatus, and storage medium, including obtaining a target service address to be matched and multiple first standard addresses; determining the target service address body corresponding to the target service address and multiple standard address bodies corresponding to the multiple first standard addresses; determining the target standard address body that matches the target service address body; and determining the target standard address that matches the target service address based on the standard address corresponding to the target standard address body. In this application, an address body that can represent the address is determined for each address text. Matching is first performed using the address body, and then matching is performed using the complete address. Since the character length of the address body is less than the character length of the complete address, the number of characters to be matched is reduced, effectively improving the efficiency of address matching. A second matching is also performed using the complete address, which improves the efficiency of address matching while ensuring the accuracy of address matching.

Description

Technical Field

[0001] This application relates to the field of address identification and matching, specifically to an address matching method, apparatus, and storage medium. Background Technology

[0002] In scenarios such as logistics and mapping, addresses obtained from different sectors (water, electricity, social security, hotels, postal services, etc.) are essentially addresses collected and used by users within their specific business scope. For example, a water usage address is the address saved when the water company collects information. A postal address is the address generated when a user sends or receives a package. Through certain matching algorithms, addresses from different data sources can be associated with the same address, thereby fusing the attributes or features carried by multiple addresses from different sources and generating significant value. Due to the sheer number of addresses, existing address matching methods require matching all addresses sequentially, resulting in massive computational overhead and low efficiency. Summary of the Invention

[0003] This application aims to provide an address text matching method, apparatus, and storage medium based on address text, in order to solve the problem of low address matching efficiency in the prior art.

[0004] In a first aspect, embodiments of this application provide an address matching method, the method comprising:

[0005] Obtain the target service address to be matched and multiple first standard addresses. The target service address comes from an external service platform. Among the multiple first standard addresses, there is a first standard address that has at least some address information that is the same as the target service address. The first standard address is an address that conforms to a preset address specification.

[0006] The target service address subject corresponding to the target service address and the multiple standard address subjects corresponding to the multiple first standard addresses are determined respectively;

[0007] Among the plurality of standard address subjects, a target standard address subject that matches the target service address subject is determined;

[0008] Based on the standard address corresponding to the target standard address subject, determine the target standard address that matches the target service address;

[0009] Wherein, the character length of the subject of the target service address is less than the character length of the target service address, and the character length of the subject of each standard address is less than the character length of the corresponding first standard address.

[0010] In one possible embodiment, before obtaining the target service address to be matched and multiple first standard addresses, the method further includes:

[0011] The special characters and erroneous characters in the initial business address and multiple initial standard addresses to be matched are replaced to obtain the first business address and multiple second standard addresses;

[0012] Remove useless characters from the first service address and the plurality of second standard addresses to obtain the second service address and the plurality of third standard addresses;

[0013] Replace the uppercase and lowercase characters in the second service address and the plurality of third standard addresses to obtain the third service address and the plurality of fourth standard addresses;

[0014] If a string containing a Chinese character and multiple zeros appears in the third service address and the plurality of fourth standard addresses, the multiple zeros are converted into a single zero to obtain the target service address and the plurality of first standard addresses.

[0015] In one possible embodiment, determining the target service address body corresponding to the target service address and the multiple standard address bodies corresponding to the multiple first standard addresses includes:

[0016] According to the preset address classification strategy, the target business address is split into multiple levels of addresses to obtain multiple levels of business address word segmentation;

[0017] According to the preset address hierarchical strategy, the multiple first standard addresses are split into multiple levels of address segments to obtain multiple levels of standard address segmentation sets. Each standard address segmentation set corresponds to a first standard address, and each standard address segmentation set includes multiple levels of standard address segments.

[0018] Based on the word segmentation of the multiple hierarchical business addresses and the multiple standard address word segmentation sets, the target business address subject corresponding to the target business address and the multiple standard address subjects corresponding to the multiple first standard addresses are determined.

[0019] In one possible embodiment, determining the target business address body corresponding to the target business address and the multiple standard address bodies corresponding to the multiple first standard addresses based on the multiple hierarchical business address word segmentation and the multiple standard address word segmentation sets includes:

[0020] Remove the first number of business address words from the first preset address level among the multiple levels of business address word segments, and extract the second number of business address words from the second preset address level to obtain the target business address body corresponding to the target business address;

[0021] Among the multiple standard address word segmentation sets, any one of the standard address word segmentation sets shall be the target standard address word segmentation set;

[0022] Remove the third number of standard address segments at the third preset address level from the target standard address segmentation set, and extract the fourth number of standard address segments at the fourth preset address level to obtain the target standard address body corresponding to the target standard address segmentation set, so as to obtain multiple standard address bodies.

[0023] In one possible embodiment, determining the target standard address subject that matches the target service address subject among the plurality of standard address subjects includes:

[0024] The similarity between the multiple standard address subjects and the target business address subject is determined to obtain the similarity of multiple address subjects;

[0025] The target address subject similarity is determined by any address subject similarity among the multiple address subject similarities.

[0026] Determine whether the subject similarity of the target address is not less than a preset subject similarity threshold;

[0027] If the similarity of the target address subject is not less than the preset address subject similarity threshold, then it is determined that the target standard address subject matches the target business address subject.

[0028] In one possible embodiment, determining the target standard address matching the target service address based on the standard address corresponding to the target standard address subject includes:

[0029] The step of determining the target standard address matching the target service address based on the standard address corresponding to the target standard address subject includes:

[0030] Determine multiple candidate first standard addresses corresponding to the target standard address subject;

[0031] Each candidate first standard address is compressed separately to obtain multiple compressed fifth standard addresses, with one candidate first standard address corresponding to one fifth standard address;

[0032] The plurality of fifth standard addresses are split into a plurality of parent standard addresses and a plurality of child standard addresses. Each fifth standard address corresponds to a parent standard address and a child standard address, and the plurality of parent standard addresses correspond to the same address level.

[0033] Among the multiple parent standard addresses, determine the target parent standard address that matches the target service address;

[0034] Determine multiple first-level child standard addresses corresponding to the target parent standard address;

[0035] Based on address similarity, determine the target sub-level standard address that matches the target service address from among the plurality of first sub-level standard addresses;

[0036] Based on the target parent standard address and the target child standard address, determine the target fifth standard address that matches the target service address.

[0037] In one possible embodiment, determining the target first standard address that matches the target service address from the plurality of candidate first standard addresses further includes:

[0038] The partial candidate first standard addresses before compression corresponding to the target fifth standard address are determined to be multiple candidate sixth standard addresses;

[0039] Among the multiple candidate sixth standard addresses, a target first standard address that matches the target service address is determined according to a precise matching method.

[0040] In one possible embodiment, determining the target first standard address that matches the target service address from the plurality of candidate sixth standard addresses according to a precise matching method includes:

[0041] Perform AOI-level precise matching on the target service address and the multiple candidate sixth standard addresses to determine multiple candidate seventh standard addresses that match the target service address;

[0042] Perform precise building-level matching between the target service address and the multiple candidate seventh standard addresses to determine multiple candidate eighth standard addresses that match the target service address;

[0043] The target service address and the plurality of candidate eighth standard addresses are matched exactly to determine the target candidate eighth standard address that matches the target service address exactly. The target candidate eighth standard address is the target first standard address.

[0044] Secondly, embodiments of this application provide an address matching device, the device comprising:

[0045] The address acquisition module is used to acquire the target business address to be matched and multiple first standard addresses. The first business address comes from an external business platform. Among the multiple first standard addresses, there is a first standard address that has at least some address information that is the same as the first business address. The multiple first standard addresses are addresses that conform to a preset address specification.

[0046] The address subject determination module is used to determine the target service address subject corresponding to the target service address and the multiple standard address subjects corresponding to the multiple first standard addresses respectively;

[0047] The address subject matching module is used to determine, among the plurality of standard address subjects, the target standard address subject that matches the target service address subject;

[0048] The address matching module is used to determine the target standard address that matches the target service address based on the standard address corresponding to the target standard address subject.

[0049] Wherein, the character length of the subject of the target service address is less than the character length of the target service address, and the character length of the subject of each standard address is less than the character length of the corresponding first standard address.

[0050] Thirdly, this application also provides an electronic device, the electronic device comprising:

[0051] One or more processors;

[0052] Memory; and

[0053] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the address matching method described in any one of the first aspects.

[0054] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, the computer program being loaded by a processor to perform the steps in the address matching method described in any of the first aspects.

[0055] This application provides an address matching method, apparatus, and storage medium, comprising: acquiring a target service address to be matched and multiple first standard addresses, wherein the first service addresses are derived from an external service platform, and the multiple first standard addresses share at least some address information with the first service addresses; determining a target service address body corresponding to the target service address and multiple standard address bodies corresponding to the multiple first standard addresses; determining a target standard address body that matches the target service address body from among the multiple standard address bodies; and determining a target standard address that matches the target service address based on the standard address corresponding to the target standard address body. In this application, an address body that can represent the address is determined for each address text. Matching is first performed using the address body, and then matching is performed using the complete address. Since the character length of the address body is less than the character length of the complete address, the number of characters to be matched is reduced, effectively improving the efficiency of address matching. Furthermore, since the address body can represent the address, a secondary matching is performed using the complete address, thus improving the efficiency of address matching while ensuring the accuracy of address matching. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a schematic diagram of a scenario for the address matching system provided in an embodiment of this application;

[0058] Figure 2 This is a schematic flowchart of an embodiment of the address matching method provided in this application;

[0059] Figure 3 A schematic flowchart of an embodiment of address normalization processing provided in this application;

[0060] Figure 4 This application provides a schematic flowchart of an embodiment for determining the address subject.

[0061] Figure 5 This application provides a schematic flowchart of an embodiment for determining the target standard address subject;

[0062] Figure 6 A schematic flowchart illustrating an embodiment of determining the first standard address of a target provided in this application;

[0063] Figure 7 A schematic flowchart illustrating an embodiment of determining the first standard address of a target provided in this application;

[0064] Figure 8 This is a schematic diagram illustrating the complete process of the address matching method provided in the embodiments of this application;

[0065] Figure 9 A schematic diagram of an embodiment of the address matching device provided in this application;

[0066] Figure 10 This is a schematic diagram of an embodiment of the electronic device provided in this application. Detailed Implementation

[0067] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0068] In the description of this application, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more features. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0069] In this application, the term "exemplary" is used to mean "used as an example, illustration, or description." Any embodiment described as "exemplary" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use this application. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that this application can be made without using these specific details. In other instances, well-known structures and processes are not described in detail to avoid obscuring the description of this application with unnecessary detail. Therefore, this application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0070] It should be noted that since the method in this application embodiment is executed in an electronic device, the processing objects of each electronic device exist in the form of data or information, such as time, which is essentially time information. It can be understood that if size, quantity, position, etc. are mentioned in subsequent embodiments, they are all corresponding data that exist so that the electronic device can process them. Specific details will not be elaborated here.

[0071] This application provides an address matching method, apparatus, and storage medium, which will be described in detail below.

[0072] Please see Figure 1 , Figure 1 This is a schematic diagram of a scenario for an address matching system provided in an embodiment of this application. The address matching system may include an electronic device 100, which integrates an address matching device, such as... Figure 1 Electronic devices in the system.

[0073] In this embodiment, the electronic device 100 can be a standalone server, a server network, or a server cluster. For example, the electronic device 100 described in this embodiment includes, but is not limited to, a computer, a network host, a single network server, a set of multiple network servers, or a cloud server composed of multiple servers. The cloud server is composed of a large number of computers or network servers based on cloud computing.

[0074] Those skilled in the art will understand that Figure 1 The application environment shown is merely one application scenario of the present application solution and does not constitute a limitation on the application scenario of the present application solution. Other application environments may include more than one application scenario. Figure 1 The number of more or fewer electronic devices shown, for example Figure 1 Only one electronic device is shown in the image. It is understood that the address matching system may also include one or more other servers, which are not specified here.

[0075] In addition, such as Figure 1 As shown, the address matching system may also include a storage module 200 for storing data, such as storing address text data to be matched.

[0076] It should be noted that, Figure 1 The schematic diagram of the address matching system shown is merely an example. The address matching system and scenario described in this application are intended to more clearly illustrate the technical solutions of this application and do not constitute a limitation on the technical solutions provided in this application. As those skilled in the art will know, with the evolution of address matching systems and the emergence of new business scenarios, the technical solutions provided in this application are also applicable to similar technical problems.

[0077] First, this application provides an address matching method. The execution subject of the address matching method is an address matching device, which is applied to an electronic device. The address matching method includes: obtaining a target service address to be matched and multiple first standard addresses. The first service addresses are derived from an external service platform. Among the multiple first standard addresses, there exists a first standard address that has at least some address information identical to the first service address. The multiple first standard addresses are addresses that conform to a preset address specification; determining the target service address subject corresponding to the target service address and multiple standard address subjects corresponding to the multiple first standard addresses; determining the target standard address subject that matches the target service address subject among the multiple standard address subjects; and determining the target standard address that matches the target service address based on the standard address corresponding to the target standard address subject.

[0078] See Figure 2 , Figure 2This is a schematic flowchart of an embodiment of the address matching method provided in this application. Figure 2 As shown, the address matching method includes the following steps:

[0079] 21. Obtain the target business address to be matched and multiple first standard addresses.

[0080] In the embodiments of this application, different users use different ways to write the same specific address, making it difficult to combine the commonly used business address with the current standard address, resulting in unclear address identification. Therefore, this application needs to combine the business address determined according to the specific business with the current standard address.

[0081] The business address can be understood as the address collected or used by the user within a specific business scope, such as the Sunshine Community Express Station; the actual standard address corresponding to the Sunshine Community Express Station is Unit 1, Building 2, Sunshine Community, Room 101. This application needs to combine the Sunshine Community Express Station and Unit 1, Building 2, Sunshine Community, Room 101. By identifying the Sunshine Community Express Station, we can know that it is located in Unit 1, Building 2, Sunshine Community, Room 101. Similarly, by identifying Unit 1, Building 2, Sunshine Community, we can know that an express station exists there.

[0082] A standard address can be a location address obtained from a map service provider, etc. Generally speaking, the addresses provided by map service providers are relatively standard, and anyone can obtain a relatively precise location corresponding to certain addresses from map service providers. In the embodiments of this application, the target business address to be matched comes from an external business platform, and among the multiple first standard addresses, there is a first standard address that has at least some address information identical to the target business address. For example, the business address can be XX Textile City, XX Road, XX District, Suzhou City, while the standard address can be No. XX, XX Road, XX District, Suzhou City; there is a commonality between the business address and the standard address, namely "XX Road, XX District, Suzhou City". At the same time, since the multiple first standard addresses are provided by map service providers, the multiple first standard addresses are usually addresses that conform to preset address specifications; for example, the standard XX Building, XX Community, XX District, XX City, XX Province.

[0083] Since business addresses can be written in various ways, this embodiment of the application also requires standardization of the business addresses to make them usable by the address matching method provided in this embodiment. While standard addresses are already the current standard way of writing and generally do not require further standardization, the address standards corresponding to standard addresses are not entirely the same in different address matching scenarios. Therefore, this embodiment of the application also requires standardization of standard addresses.

[0084] like Figure 3As shown, which is a schematic flow diagram of an embodiment of address normalization processing provided by the embodiments of the present application, the method may include:

[0085] 31. Replacing special characters and error characters in the initial service address to be matched and a plurality of initial standard addresses to obtain a first service address and a plurality of second standard addresses.

[0086] In the embodiments of the present application, the address normalization processing method provided in the present application is described in detail by taking the initial service address to be matched as an example, which is also applicable to the normalization processing of initial standard addresses; the specific processing method is as follows:

[0087] Since actual service addresses are determined by different users, there may be cases of special characters and error characters. The special characters may be brackets, irregular Arabic numerals, etc. Therefore, it is necessary to first replace, i.e., correct, the special characters and error characters in the initial service address.

[0088] In the embodiments of the present application, there are multiple different methods for processing different special characters. Specifically, Chinese brackets in the initial service address may be replaced with English brackets; for example, "幸福路(102)号" is replaced with "幸福路(102)号"; pictographic Arabic numerals are replaced with regular Arabic numerals.

[0089] If special characters such as "#" and "-" appear in the initial service address, completion or replacement is also required. Judging what word the special character needs to be replaced with is performed based on the text before or after the special character in the initial service address. For example, for "二村50#", since the text before the special character "#" is "村", "#" is corrected to "组".

[0090] 32. Removing useless characters from the first service address and the plurality of second standard addresses to obtain a second service address and a plurality of third standard addresses.

[0091] For useless characters, they are usually useless characters located after the room level in the first service address. Since the room level has been accurate to a precision that cannot be further reduced, the position cannot be further narrowed according to the characters after the room level and they have no significance for consideration, so these useless characters are removed.

[0092] For example, the first service address has been accurate to Unit 2, Room 101, which is already the smallest and most accurate address from the perspective of addressing. If there are other characters after Unit 2, Room 101, they are directly removed and only Unit 2, Room 101 is retained.

[0093] 33. Replacing uppercase and lowercase characters in the second service address and the plurality of third standard addresses to obtain a third service address and a plurality of fourth standard addresses.

[0094] For case replacement, generally, Chinese capital numerals with three digits or less in the second service address are converted into lowercase Arabic numerals; single-digit lowercase Arabic numerals in the second service address are converted into Chinese capital numerals.

[0095] 34. If a character string of one Chinese character plus multiple zeros appears in the third service address and multiple fourth standard addresses, multiple zeros are changed to one zero, so as to obtain a target service address and multiple first standard addresses.

[0096] Specifically, in the embodiment of the present application, as long as the above situation exists in the third service address and each fourth standard address, the above processing needs to be performed regardless of whether it is the third service address or the fourth standard address, so as to complete address normalization processing, and obtain the normalized target service address and multiple first standard addresses.

[0097] In the embodiment of the present application, there are also some special cases: for example, if a character string of one Chinese character plus multiple zeros appears in the initial service address, the multiple zeros are converted into one zero. In a specific embodiment, for Xingfu Community (幸福小区) 00017 Building, after normalization processing, it becomes Xingfu Community 017 Building; if it is Taihu Xin 0000 City (太湖新0000城), after normalization processing, it becomes Taihu Xin 0 City.

[0098] In addition to the above normalization processing method, there are other processing methods. For example, if the initial service address contains the situation of "village + numeral", the characters are normalized to normal characters according to the difference of the last character. For example, "embankment + numeral + number" is replaced with "village + numeral + group"; for "ordinal word + Chinese capital numeral + group", the ordinal word is deleted and replaced with "Arabic numeral + group".

[0099] Specifically, for example, "embankment 3 number" in "Xinying embankment 3, Songling Town, Wujiang District" is replaced with "village 3 group", so as to obtain "Xinying Village 3 Group, Songling Town, Wujiang District". Or, delete "ordinal word" in "Ercun Third Group, Songling Town, Wujiang District" and perform case replacement to obtain "Ercun 3 Group, Songling Town, Wujiang District". There is also a case of repeated characters, such as "lou lou" (two repeated Chinese characters for building), the repeated characters are replaced with a single character, and only one "lou" is retained.

[0100] It should be noted that, in the embodiment of the present application, for the same initial service address, all the above address normalization methods need to be used to perform normalization processing on the address, instead of only using one or several of them; and there is no sequence among the multiple address normalization processing methods provided above. At the same time, there are other methods for normalization processing of address texts in other embodiments. The multiple normalization processing methods in the embodiments of the present application are only for illustration, and do not constitute a limitation to the present application.

[0101] 22. Determine the target business address subject corresponding to the target business address and the multiple standard address subjects corresponding to the multiple first standard addresses respectively.

[0102] After normalizing the initial business address to obtain the target business address, it is necessary to further determine the address body of the target business address. The address body consists of multiple characters that can accurately identify an address; however, the number of characters in the address body should not be too many or too few.

[0103] In this embodiment of the application, the target service address and multiple first standard addresses are split according to a preset address hierarchy strategy, and the address subject is determined according to the different levels of the split addresses.

[0104] In one specific embodiment, a preset 18-level address hierarchy strategy can be used to split the target business address and multiple first standard addresses. The 18-level address hierarchy strategy divides the address into 18 different address levels, which can be: Province 1, City 2, District 3, Industrial Park 4, Street 5, Village 6, Group 7, Business District 9, Road 10, Branch Road 11, House Number 12, Branch Number 13, Industrial Park 13, Building 14, Unit 15, Floor 16, Room Number 17, and other information 18, totaling 18 levels. Different address levels actually represent different address ranges of varying sizes; for example, the address range included in the province level is larger than that included in the city level.

[0105] The target business address and multiple primary standard addresses typically include addresses at multiple address levels. For example, the target business address might be "XX Province, XX City, XX District, XXX Industrial Park, XX Road, XX Number," which includes address segments at six different address levels. The address body requires extracting a portion of these address segments and combining them to obtain an address body that can accurately identify the address. For example, "XX District, XXX Industrial Park, XX Road" can be extracted as the address body; based on this address body, the approximate location of the address can be determined. It should be noted that the extracted address body is accurate to the "XX Road" level, but not to the "XX Number," so the exact location of the address cannot be precisely confirmed based on the address body alone. However, among multiple different "XX District, XX Road" addresses, a subset of addresses that are accurate to the "XX Road" level can be selected, and further matching and confirmation only needs to be performed on the selected "XX Road" addresses. This significantly reduces the number of addresses to be matched and improves the efficiency of address matching.

[0106] 23. Among the multiple standard address subjects, determine the target standard address subject that matches the target service address subject.

[0107] This application requires matching business addresses and standard addresses, and the first step is to match the business address body and the standard address body. This is because both the business address body and the standard address body are the smallest strings that can represent an address. Matching the address body first can effectively reduce the number of characters that need to be matched, thereby improving the efficiency of address matching.

[0108] For example, the business address subject can be No. 100, Xingfu Road, while multiple first standard addresses correspond to multiple standard address subjects such as No. 100, No. 102, No. 109, and No. 122, Xingfu Road. Then, we only need to find No. 100, Xingfu Road among No. 100, No. 102, No. 109, and No. 122, which matches the business address subject. There's no need to match No. 102, No. 109, and No. 122. This significantly reduces the number of standard addresses that need to be matched subsequently. Furthermore, since only No. 100, Xingfu Road is matched, there's no need to match the complete business address subject, "No. 100, Xingfu Road, XX Town, XX City, XX District," greatly reducing the length of the characters to be matched and improving the efficiency of address matching.

[0109] In the above embodiments, if an address level not present in the preset address hierarchy strategy appears in the actual address, the missing address level can be inferred to find the corresponding address level in the preset address hierarchy strategy. For example, if the address level "town" appears in the actual address, but the preset address hierarchy strategy does not have the "town" address level, it can be inferred that the "town" address level is one of the multiple levels in the preset address hierarchy strategy. If the inference determines that it is the "village" level, then the "town" level in the actual address can be classified as the "village" level in the address hierarchy strategy.

[0110] It should be noted that, in the embodiments of this application, the character length of the target service address body is usually shorter than the character length of the target service address, and the character length of multiple standard address bodies is also usually shorter than the character length of the multiple first standard addresses. Thus, when matching address bodies, it is unnecessary to match longer complete addresses; instead, matching is performed on shorter address bodies. Furthermore, the address body can accurately determine the location of the address. Therefore, while ensuring matching accuracy, the efficiency of address matching can be improved.

[0111] 24. Based on the standard address corresponding to the target standard address subject, determine the target standard address that matches the target service address.

[0112] Because the address subject is not the most precise address, there may be situations where one standard address subject corresponds to multiple business address subjects. For example, multiple standard addresses may be Building 1, Building 2...10 of Xingfu Community, and the standard address subject corresponding to these standard addresses may all be Xingfu Community.

[0113] Therefore, in this embodiment of the application, it is necessary to further determine multiple standard addresses corresponding to the target standard address subject, and then determine the target standard address that matches the target service address among the multiple standard addresses.

[0114] Since this application matches the business address to the standard address, it is necessary to confirm the target standard address subject corresponding to the target business address subject. Even if multiple standard address subjects correspond to the same standard address subject, matching can be performed again based on the complete target business address and standard address. At this time, performing a second matching based on the target business address and standard address has greatly reduced the number of addresses to be matched, thus effectively improving the efficiency of address matching.

[0115] The address matching method provided in this application includes: obtaining a target service address to be matched and multiple first standard addresses, wherein the first service addresses originate from an external service platform, and the multiple first standard addresses share at least some address information with the first service addresses; determining the target service address body corresponding to the target service address and multiple standard address bodies corresponding to the multiple first standard addresses; determining the target standard address body that matches the target service address body among the multiple standard address bodies; and determining the target standard address that matches the target service address based on the standard address corresponding to the target standard address body. In this application, an address body that can represent the address is determined for each address text. Matching is first performed using the address body, and then matching is performed using the complete address. Since the character length of the address body is less than the character length of the complete address, the number of characters to be matched is reduced, effectively improving the efficiency of address matching. Furthermore, since the address body can represent the address, a second matching is performed using the complete address, thus improving the efficiency of address matching while ensuring the accuracy of address matching.

[0116] like Figure 4 The diagram shown is a flowchart illustrating an embodiment of determining the address subject provided in this application. In the embodiments of this application, determining the service address subject corresponding to the target service address and the multiple standard address subjects corresponding to multiple first standard addresses may include:

[0117] 41. Based on the preset address hierarchy strategy, the target business address is split into multiple levels of addresses to obtain multiple levels of business address word segments.

[0118] 42. Based on the preset address hierarchy strategy, multiple first standard addresses are split into multiple levels of addresses to obtain a set of standard address word segments at multiple levels.

[0119] This application provides an 18-level address hierarchy strategy. Using this preset strategy, the target service address and multiple first standard addresses can be split to obtain multiple service address segments corresponding to the target service address, and multiple standard address segment sets corresponding to the multiple first standard addresses. Each standard address segment set corresponds to one first standard address, and each standard address segment set includes standard address segments at multiple levels.

[0120] For example, if the target business address is Xibang, Group 14, Yongxin Village, Xingfu Town, Wujiang District, then according to the preset 18-level address hierarchy strategy, the business address is divided into: Wujiang District 3 | Xingfu Town 5 | Yongxin Village 13 | Group 14 13 | Xibang 13; where Wujiang District, Xingfu Town, Yongxin Village, Group 14, and Xibang are address segments at different address levels, and the number after each address segment indicates the address level represented by the segment. Taking Wujiang District as an example, according to the 18-level address hierarchy strategy, "district" is the third level, therefore Wujiang District corresponds to 3.

[0121] Generally speaking, regardless of whether it's the target business address or the primary standard address, an address contains multiple address segments at different address levels. Multiple address segments corresponding to the same primary standard address form a standard address segment set. For example, "Wujiang District 3︱Xingfu Town 5︱Yongxin Village 13︱Group 14 13︱Xibang 13" is a standard address segment set. When there are multiple primary standard addresses, there are also multiple standard address segment sets corresponding to these primary standard addresses.

[0122] It should be noted that in the embodiments of this application, due to different ways of writing addresses, the preset address segmentation strategy does not match the actual address level. In this case, the address level in the actual address needs to be inferred and assigned to the address level in the preset address segmentation strategy. Therefore, there may be cases where different address segments correspond to the same address level. For example, in the aforementioned embodiments, Yongxin Village and Group 14 both correspond to the address level 13.

[0123] 43. Remove the first number of business address segments from the first preset address level among multiple levels of business address segments, and extract the second number of business address segments from the second preset address level to obtain the target business address body corresponding to the target business address.

[0124] Based on the word segmentation of the multiple hierarchical business addresses and the multiple standard address word segmentation sets, the target business address subject corresponding to the target business address and the multiple standard address subjects corresponding to the multiple first standard addresses can be determined.

[0125] Specifically, this can be done by: removing some business address segments with excessively large or small ranges, and then extracting and combining some of the remaining address segments to obtain the main body of the business address. In the embodiments of this application, the multiple address segments corresponding to the target business address are sorted according to the hierarchical order corresponding to the aforementioned 18-level hierarchical strategy. Therefore, the first number of business address segments at the first preset address level can be removed from multiple business address segments at different levels; and the second number of business address segments at the second preset address level can be extracted from the remaining multiple address segments.

[0126] The first preset address level of address segments to be removed typically includes address segments with excessively broad ranges, such as "XX province XX city"; simultaneously, address segments with excessively narrow ranges, such as "XX building XX number," also need to be removed. In the embodiments of this application, since not every address includes all 18 address levels of address segments, the first number of address levels to be removed can be set according to actual needs; however, it is usually the first and last address segments that are removed simultaneously. Furthermore, the second preset address level and the second number of business address segments extracted can also be set according to actual needs.

[0127] For example, address segmentation words corresponding to the five levels at the beginning and the five levels at the end can be removed; alternatively, address segmentation words corresponding to the three levels at the beginning and the two levels at the end can also be removed. In the embodiments of this application, typically, some address segmentation words at the beginning and some at the end are removed simultaneously; however, the number of address segmentation words removed at the beginning and the number removed at the end can be different.

[0128] In the above embodiments, address segments located at the beginning and end positions are removed because: the aforementioned preset address hierarchy strategy divides addresses according to the size of the address range. Address segments located at the beginning position, such as provinces and cities, represent too large a range, resulting in too many standard addresses (or business addresses), which is not conducive to matching. Therefore, they are removed to reduce the number of addresses to be matched. Address segments located at the end position are removed because the position represented at the end is too small, resulting in too few addresses to be matched, affecting the accuracy of matching.

[0129] The above embodiments removed some address word segments, and it is still necessary to extract some address word segments from the remaining address word segments to obtain the target business address body. It should be noted that, in the embodiments of this application, it is necessary to extract address word segments from specific locations or specific address levels. This is because the probability of a region-level word segment appearing is highest only at a specific address level, and the probability of a region-level word segment appearing is lower at other address levels.

[0130] Taking the aforementioned 18 levels as an example, in the embodiments of this application, address segmentation corresponding to the 6th level village, the 10th level road, the 11th level house number and the 13th level park can be extracted; while address segmentation corresponding to the 7th level group and the 9th level business district is not extracted.

[0131] In a specific embodiment, taking the target business address as Jiangsu Province 1|Suzhou City 2|Wujiang District 3|Yangguang Town 5|Xingfu Garden 13|Building 4, 14 as an example, after removing some address segmentation words, only the address segmentation word Xingfu Garden 13 remains. Therefore, Xingfu Garden 13 is taken as the target address subject.

[0132] Because business addresses can be written in various ways, there may be instances where the same business address contains multiple address segments at the same address level. In this case, it is necessary to extract the address segments at the same address level and use them as multiple different address bodies, meaning that the same business address may correspond to multiple address bodies.

[0133] In the above embodiments, there are several scenarios. The first scenario is that within the address hierarchy of the preset location to be extracted, only one address hierarchy contains multiple different address segments. Another scenario is that within the address hierarchy of the preset location to be extracted, multiple address hierarchies contain multiple different address segments. For the first scenario, the remaining address segments need to be further divided from the address segments at the same address hierarchy to obtain multiple address segments; each address segment includes at least one address segment. In this embodiment, address segments at the preset address hierarchy need to be extracted from the multiple address segments and combined to obtain multiple target service address bodies corresponding to the target service address.

[0134] In one specific embodiment, if the remaining address segments after removing some address segments are "Xingfu Village No. 26, Xingfu Village Group 26", and there are two "Xingfu Village" segments with the same address level, then the remaining address segments need to be split into two segments, namely "Xingfu Village No. 26" and "Xingfu Village Group 26", with "Xingfu Village" as the boundary. After determining the two segments, the address segments with the preset address level are extracted from each segment and combined to obtain two address entities, namely "Xingfu Village No. 26" and "Xingfu Village Group 26".

[0135] In the above embodiments, the specific address segments corresponding to two identical address levels are also the same. However, in other embodiments, there may be cases where the specific address segments corresponding to two identical address levels are different. The above address subject determination method is also applicable and will not be repeated here.

[0136] In the second scenario, where multiple address levels within the preset address hierarchy contain multiple different address segments, we can first extract the corresponding address body for each preset address level. Then, we can divide the address segments corresponding to the same address level into different address bodies to obtain a single address body. Additionally, we need to use the address segment corresponding to the last address level as a separate address body.

[0137] In one specific embodiment, if the extracted address segments are 10|10|13|13, then the address segments corresponding to the same level, such as the 10th level and the 13th level, can be divided into different address bodies; thus, the address body is 10|13. Simultaneously, the address segment corresponding to the last address level, 13, should be used as a separate address body.

[0138] 44. Use any one of the multiple standard address segmentation sets as the target standard address segmentation set.

[0139] 45. Remove the third number of standard address segments at the third preset address level from the target standard address segmentation set, and extract the fourth number of standard address segments at the fourth preset address level to obtain the target standard address body corresponding to the target business address segmentation set, so as to obtain multiple standard address bodies.

[0140] In the embodiments of this application, the same address subject determination method can be used to determine the address subject corresponding to the target service address and each of the multiple first standard addresses. Therefore, the method for determining the standard address subject corresponding to each of the multiple first standard addresses can refer to the foregoing content, and will not be repeated here.

[0141] It should be noted that since not every standard address and every business address corresponds to 18 address levels, the number of address segments to be removed and extracted can be determined according to actual needs; that is, the specific sizes of the first, second, third, and fourth numbers can be adjusted, and are not limited in this embodiment. Similarly, the address segments of the first, second, third, and fourth preset address levels to be removed and extracted can also be adjusted according to actual needs.

[0142] After determining the target address subject and multiple standard address subjects, it is necessary to match the target address subject and multiple standard address subjects to determine the target standard address subject that matches the target business address subject. For example... Figure 5 The diagram shown is a flowchart illustrating an embodiment of determining a target standard address subject according to this application, which may include:

[0143] 51. Determine the similarity between multiple standard address subjects and the target business address subject to obtain the similarity of multiple address subjects.

[0144] In the embodiments of this application, determining the target standard address subject that matches the target business address subject mainly involves calculating the address subject similarity and determining the target standard address subject based on the magnitude of the address subject similarity.

[0145] Specifically, each character in the target service address body and each character in the target standard address body can be determined separately. Then, the harmonic mean between the characters in the target service address body and the characters in the target standard address body at the same position can be calculated according to the character arrangement order (usually from left to right). This harmonic mean is used as the distance between the target address body and the standard address body. This distance is then used as the address body similarity. Of course, in other embodiments, other methods can be used to calculate the similarity between address bodies, which is not limited here.

[0146] 52. Take any address subject similarity among multiple address subject similarities as the target address subject similarity.

[0147] 53. Determine whether the subject similarity of the target address is not less than the preset subject similarity threshold.

[0148] 54. If the similarity between the target address and the subject is not less than the preset similarity threshold, then the target standard address subject is determined to match the target business address subject.

[0149] After determining the similarity of multiple address subjects, the relationship between the multiple address subject similarities and the preset address subject similarity threshold can be judged. The standard address subject corresponding to the address subject similarity that is not less than the preset address subject similarity threshold is used as the target standard address subject to match the target business address subject.

[0150] Since there are multiple addresses with similarity scores not less than a preset address threshold, in some embodiments of this application, the largest address similarity score among the multiple address similarities can be directly determined; and the standard address corresponding to the largest address similarity score can be used as the target standard address. This can further increase the number of standard addresses that may match the target service address.

[0151] Since multiple different standard addresses may correspond to the same standard address subject in this embodiment, after determining the target standard address subject through address subject matching, it is also necessary to determine multiple candidate first standard addresses corresponding to the target standard address subject; and among the multiple candidate first standard addresses, there is a first standard address that matches the target service address. In this embodiment, it is also necessary to further determine the target first standard address that truly matches the target service address from among the multiple candidate first standard addresses.

[0152] like Figure 6 The diagram shown is a flowchart illustrating an embodiment of determining a target first standard address provided in this application, which may include:

[0153] 61. Compress multiple candidate first standard addresses to obtain multiple compressed fifth standard addresses.

[0154] In real-world address matching scenarios, the number of addresses is enormous. In this embodiment, after reducing the number of addresses to be matched by address body matching, it is still necessary to perform matching again using complete addresses. Since the number of standard addresses for secondary matching is still large, this embodiment requires compressing each candidate first standard address to obtain multiple compressed fifth standard addresses; and one candidate first standard address corresponds to one fifth standard address.

[0155] It should be noted that the compression of multiple candidate first standard addresses in this embodiment differs from the aforementioned removal and extraction of partial address segments from the standard addresses. In this embodiment, the compression of multiple candidate first standard addresses mainly involves removing smaller ranges of address segments such as unit number and address number from each candidate first standard address, compressing multiple addresses that differ only in unit number and address number to obtain the same compressed fifth standard address.

[0156] For example, multiple candidate first standard addresses include: Building 2, Xingfu International, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province; Units 101-1403, Building 2, Xingfu International, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province; Units 101-1403, Building 3, Xingfu International, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province, etc., all accurate to the house number. In this embodiment, these candidate first standard addresses are compressed to the building level, resulting in two fifth standard addresses: Building 2, Xingfu International, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province, and Building 3, Xingfu International, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province.

[0157] It should be noted that the process of compressing multiple candidate first standard addresses separately here differs from the aforementioned process of determining the address subject. This is because address word segments were not removed or extracted when compressing candidate first standard addresses. Since the compressed fifth standard address is an address that differs only in building number and house number, it is possible to perform province, city, and district level matching on the compressed fifth standard address, followed by subsequent building level and house number precision level matching.

[0158] 62. Split multiple fifth standard addresses to obtain multiple parent standard addresses and multiple child standard addresses.

[0159] After obtaining multiple compressed fifth standard addresses, it is necessary to further split these fifth standard addresses to obtain multiple parent standard addresses and multiple child standard addresses. Each fifth standard address corresponds to one parent standard address and one child standard address, and the multiple parent standard addresses correspond to the same address level.

[0160] In one specific embodiment, the two fifth standard addresses, namely Building 2 of Xingfu International, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province, and Building 3 of Xingfu International, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province, can be split to obtain the parent standard address of No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province, and the child standard addresses of Building 2 and Building 3 of Xingfu International.

[0161] In the above embodiments, the criterion for dividing parent and child standard addresses is to retain addresses with the same AOI level as parent standard addresses. For example, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province, represents the common part between two fifth standard addresses. Child standard addresses are those with different AOI levels, such as Building 2 and Building 3 of Xingfu International, which are different addresses.

[0162] 63. Among multiple parent standard addresses, determine the target parent standard address that matches the target service address.

[0163] 64. Determine the multiple first-level child standard addresses corresponding to the target parent standard address.

[0164] 65. Among multiple first-level standard addresses, determine the target sub-level standard address that matches the target service address.

[0165] 66. Based on the target parent standard address and the target child standard address, determine the target fifth standard address that matches the target service address.

[0166] In this embodiment, after splitting the fifth standard address to obtain multiple parent standard addresses and multiple child standard addresses, it is necessary to determine the target parent standard address that matches the target service address from among the multiple parent standard addresses. After the first matching is completed, the child standard addresses are then attached through the target parent standard address, that is, multiple first child standard addresses corresponding to the target parent standard address are determined. Furthermore, the target child standard address that matches the target service address is determined from among the multiple first child standard addresses; based on the target parent standard address and the target child standard address, the target fifth standard address that matches the target service address can be determined.

[0167] In the above embodiments, firstly, the matching degree between multiple parent standard addresses and the service address is determined to obtain the target parent standard address with the highest matching degree; then, multiple child standard addresses corresponding to the target parent standard address are obtained. Next, the matching degree between each of the multiple child standard addresses corresponding to the target parent standard address and the service address is calculated to obtain the target child standard address with the highest matching degree. Therefore, based on the target parent standard address and the target child standard address, the target fifth standard address matching the target service address is determined.

[0168] Specifically, assuming the target parent standard address is No. 106, XX Road, XX Town, Suzhou City, Jiangsu Province, and the target child standard address is Building 2, Xingfu International, then the target fifth standard address can be determined as: Building 2, Xingfu International, No. 106, XX Road, XX Town, Suzhou City, Jiangsu Province.

[0169] Since the target fifth standard address is a compressed address, it is also necessary to determine multiple candidate sixth standard addresses before compression corresponding to the fifth standard address, and then determine the target first standard address that matches the target business address from among these candidate sixth standard addresses. For example, if the target fifth standard address is: Building 2, Xingfu International, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province, the corresponding multiple candidate sixth standard addresses before compression could be: Room 101-1403, Building 2, Xingfu International, No. 10610, XX Road, XX Town, Suzhou City, Jiangsu Province, etc., multiple addresses accurate to different house numbers.

[0170] In the above embodiments, since secondary matching is performed using the parent standard address and the child standard address, the number of standard addresses to be matched is further reduced, which greatly improves the efficiency of address matching compared with matching standard addresses one by one in the prior art.

[0171] Meanwhile, since there are multiple candidate sixth standard addresses, further matching is needed to determine the target first standard address that matches the target service address from among the multiple candidate sixth standard addresses. For example... Figure 7 The diagram shown is a flowchart illustrating an embodiment of determining a target first standard address provided in this application, which may include:

[0172] 71. Perform AOI-level precise matching between the target service address and multiple candidate sixth standard addresses to determine multiple candidate seventh standard addresses that match the target service address.

[0173] In the embodiments of this application, matching multiple candidate sixth standard addresses and target service addresses also achieves precise matching by continuously narrowing down the address range; specifically, it involves AOI-level matching - building matching - complete matching. For AOI-level matching, the specific matching process is as follows:

[0174] In real-world address matching scenarios, due to differing address subject extraction standards and the existence of multiple identical or similar address segments corresponding to the same address level, a single standard address or business address may correspond to multiple address subjects. For example, "Shengze Oriental Textile City, No. 100, Market Road, Shengze Town, Suzhou City, Jiangsu Province" could include two address subjects: "No. 699, Market Road" and "Shengze Oriental Textile City." With multiple address subjects, it becomes impossible to accurately determine the specific location of the address; therefore, AOI-level matching is required again based on the address subjects.

[0175] In some embodiments, if a candidate sixth standard address corresponds to only one address body among multiple candidate sixth standard addresses, and the target service address also corresponds to only one address body, then the address bodies are directly compared to determine whether they are the same for AOI-level matching. Specifically, if any candidate sixth standard address corresponds to only one address body, and is the same as the unique address body corresponding to the target service address, then the candidate sixth standard address is retained. If the address bodies are different, then the candidate sixth standard address is discarded. However, if there is an inclusion relationship between two address bodies, the candidate sixth standard address should be retained even if the address bodies are different.

[0176] In one specific embodiment, the address subject corresponding to the target business address can be No. 38 Gaoxin Road, while the address subject corresponding to the candidate sixth standard address is No. 928 Gaoxin Road. Since the two address subjects are different and there is no inclusion relationship, the candidate sixth standard address corresponding to No. 928 Gaoxin Road needs to be removed.

[0177] In the above embodiments, if the address subject of a candidate sixth standard address is "road number + house number", such candidate sixth standard addresses should also be removed. This is because in actual business addresses, not every business address includes 18 complete address levels, and not every address level is the same as the preset 18 address levels. If there are cases that do not belong to the preset 18 address levels, some addresses that do not belong to the 18 address levels will be divided according to the preset address level division method. This could result in situations where the same address level is actually divided into different address levels, leading to inaccurate identification of the address subject and subsequent inaccurate address matching.

[0178] Specifically, the target business address could be Unit 101, Building 12, No. 928, Gaoxin Road, Songling Town, Suzhou City, Jiangsu Province, while the sixth candidate standard address is Unit 101, Building 12, Gaoxin Road, Songling Street, Suzhou City, Jiangsu Province. The "Building 12" in the target business address is a level 14 segmentation term, while the "No. 12" in the sixth candidate standard address is not actually a building, but rather a road corresponding to a level 14 address segmentation term. The two are not equivalent at the building level, so they need to be eliminated.

[0179] In other embodiments, if any one of the target service address or multiple candidate sixth standard addresses corresponds to a single subject, while other addresses correspond to multiple subject addresses, the following matching process needs to be performed:

[0180] Assuming the target service address has only one address subject, it's called a "single-subject" address; a candidate sixth standard address corresponds to multiple address subjects, called a "multi-address" subject. If the address subject of a single-subject address is the same as the first address subject among the multiple address subjects corresponding to a multi-subject address, then this candidate sixth standard address subject needs to be removed. This is because a multi-subject address also includes other address subjects with smaller ranges. The address subject of a single subject only shares the same address subject with the first address subject in a larger range, making it impossible to further confirm the specific address of the single subject within a smaller address range, and it cannot be matched with other address subjects with smaller ranges, therefore it needs to be removed.

[0181] In one specific embodiment, the address subject of the business address is No. 6, South Ring Road, while the address subjects of the candidate sixth standard address include No. 6, South Ring Road and Zhenxing Warp Knitting Factory. The address subject of the business address, No. 6, South Ring Road, is the same as the first address subject, No. 6, South Ring Road, in the candidate sixth standard address. However, the candidate sixth standard address also corresponds to an address subject of Zhenxing Warp Knitting Factory, and the business address cannot be matched within a smaller range of address subjects, so this candidate sixth standard address needs to be removed.

[0182] If the address subject of a single entity is the same as the last address subject of a multi-entity entity, then the candidate sixth standard address is retained. This is because the last address subject is already the address subject with the smallest range. For address subjects, a match has already been made within the smallest range, and the specific address of the single entity can be determined. Therefore, the candidate sixth standard address is retained.

[0183] For example, if the address subject of the business address is Zhenxing Warp Knitting Factory, and the address subjects of the candidate sixth standard address are No. 6-1 South Ring Road and Zhenxing Warp Knitting Factory respectively, and the last address subject in the candidate sixth standard address is Zhenxing Warp Knitting Factory, which is the same as the address subject in the business address, then the candidate sixth standard address needs to be retained.

[0184] In other embodiments, if the last address of a multi-subject is the suffix of a single-subject, these candidate sixth standard addresses should normally be discarded; however, if the first two characters of a single-subject are the starting characters of the first address of a multi-subject, these candidate sixth standard addresses are retained.

[0185] Specifically, the target business address is No. B5-1, Mingfeng Garden, Pubei Road, with the address subject being Mingfeng Garden, Pubei Road; the candidate sixth standard address is No. 1, Mingfeng Garden, No. 29, Pubei Road, Wujiang District, Suzhou City, Jiangsu Province, with the corresponding address subjects being No. 29, Pubei Road and Mingfeng Garden, respectively. The last address subject of the multi-subject is Mingfeng Garden, which is the suffix of the single-subject address; normally, the candidate sixth standard address of No. 1, Mingfeng Garden, No. 29, Pubei Road, Wujiang District, Suzhou City, Jiangsu Province should be eliminated. However, in this embodiment, since the first two characters of the single-subject address are Pubei, which is the starting character of the first address subject of the multi-subject address, the candidate sixth standard address of No. 1, Mingfeng Garden, No. 29, Pubei Road, Wujiang District, Suzhou City, Jiangsu Province needs to be retained.

[0186] It should be noted that in the above embodiment, the determination is made regarding whether the first two characters of a single entity are the starting characters of the first address entity of a multi-entity entity. This is because most AOI areas such as roads, villages, and towns are named with three characters, and the third character is the same, with only the first two characters differing; for example, Jiangsu Province and Zhejiang Province, or Happiness Road and Sunshine Road, etc. Therefore, it is only necessary to determine the first two characters. In other embodiments, the determination can also be made regarding whether the first three or four characters of a single entity are the starting characters of the first address entity of a multi-entity entity, and the specific determination can be adjusted according to actual needs.

[0187] In other embodiments, both the target service address and the candidate sixth standard address may include multiple address subjects; in this embodiment, the multiple address subjects in the target service address are referred to as "service multiple subjects," and the multiple address subjects corresponding to the candidate sixth standard address are referred to as "standard multiple subjects"; in this case, AOI-level matching of multiple candidate sixth standard addresses may include:

[0188] If the suffix of the first address subject of the business multi-subject is the same as the suffix of the last address subject of the standard multi-subject, and the suffix of the last address subject of the business multi-subject is the same as the suffix of the last address subject of the standard multi-subject, and the first two words of the first address subject of the business multi-subject are the same as the first two words of the first address subject of the standard multi-subject, then the candidate sixth standard address is retained.

[0189] Specifically, assume the target business address corresponds to the following multiple entities: "Changban Road Binhu Zhongnan Century City", "Binhu Zhongnan Century City North District", and "Binhu Zhongnan Century City"; while the candidate sixth standard address corresponds to the following standard entities: "Changban Road No. 1588" and "Zhongnan Century City". In this case, the suffix "Zhongnan Century City" of the first address entity of the multiple entities, "Changban Road Binhu Zhongnan Century City", is also the last address entity of the standard multiple entities. Furthermore, the suffix "Zhongnan Century City" of the last address entity of the multiple entities is the last address entity of the standard multiple entities; and the first two characters "Changban" of the first address entity of the multiple entities are the same as the first two characters "Changban" of the first address entity of the standard multiple entities. If all three conditions are met, the candidate sixth standard address is successfully matched, and this candidate sixth standard address is retained.

[0190] Similarly, in other embodiments, it can be determined whether the first three words of the first address subject of the business multi-subject are the same as the first three words of the first address subject of the standard multi-subject.

[0191] In other embodiments, if the second address body of a service multi-subject is also the suffix of the first address body of the service multi-subject and the suffix of the last address body of the standard multi-subject; and the remaining string after deleting the last address body of the service multi-subject is the prefix of the first address body of the standard multi-subject, then the candidate sixth standard address is retained.

[0192] Specifically, assume the business multi-entity is "Kaiping Road Junyue International" and "Junyue International"; while the standard multi-entity is "Kaiping Road 450" and "Hengda Junyue International". In this case, is the second address entity "Junyue International" in the business multi-entity a suffix of the first address entity in the business multi-entity, or a suffix of the last address entity in the standard multi-entity? Furthermore, after deleting the last address entity "Junyue International" from the first address entity "Kaiping Road Junyue International", the remaining string is "Kaiping Road", which is the prefix of the first address entity "Kaiping Road 450" in the standard multi-entity. Therefore, this candidate sixth standard address can be retained.

[0193] In other embodiments, if the first address subject of the service multi-subject is the same as the first address subject of the standard multi-subject, and the last address subject of the service multi-subject is the same as the last address subject of the standard multi-subject, then the candidate sixth standard address is retained.

[0194] Specifically, suppose the business entities are "No. 18 Tongxin West Road", "Hengda Hotel", and "Hengda Junyue International", while the standard entities are "No. 18 Tongxin West Road" and "Hengda Junyue International". In this case, the first address of both the business entity and the standard entity is "No. 18 Tongxin West Road"; and the last address of both the business entity and the standard entity is "Hengda Junyue International". Therefore, this candidate sixth standard address can be retained.

[0195] In other embodiments, if the intersection length of the business multi-subject and the standard multi-subject is equal to 1, and the last address subject of the standard multi-subject is a suffix of the last address subject of the business multi-subject, then the candidate sixth standard address is retained. Specifically, assume the business multi-subjects are "599 Market Middle Road" and "Shengze International Trade Center", and the standard multi-subjects are "599 Market Middle Road" and "International Trade Center". In this case, both the business multi-subject and the standard multi-subject include the address subject "599 Market Middle Road", meaning their intersection length is 1; and the last address subject of the standard multi-subject, "International Trade Center", is a suffix of the last address subject of the business multi-subject, "Shengze International Trade Center". Therefore, the candidate sixth standard address can be retained.

[0196] In the above embodiments, the candidate sixth standard address that matches the target service address is called the candidate seventh standard address; there can be multiple candidate seventh standard addresses. Furthermore, since the candidate sixth standard address is an AOI-level address match, it is necessary to further narrow down the scope to perform more precise address matching within a smaller range.

[0197] 72. Perform precise building-level matching between the target business address and multiple candidate seventh standard addresses to determine multiple candidate eighth standard addresses that match the target business address.

[0198] Building-level precise matching is based on the aforementioned AOI-level matching. It is necessary to further determine the standard addresses with the same building number to obtain multiple candidate eighth standard addresses.

[0199] 73. Perform a perfect match between the target service address and multiple candidate eighth standard addresses to determine the target candidate eighth standard address that perfectly matches the target service address. The target candidate eighth standard address is the target first standard address.

[0200] After identifying multiple candidate eighth standard addresses, it is necessary to further determine the target candidate eighth standard address that matches at the household / room level. The target candidate eighth standard address is the target first standard address. In other words, the target first standard address that perfectly matches the target service address is determined, thus achieving matching between the service address and the standard address.

[0201] like Figure 8 The diagram shown is a complete flowchart of the address matching method provided in an embodiment of this application. Figure 8 In the embodiments shown, the address matching method mainly includes three parts: address normalization, nested parallel subject ablation matching, and precise address matching. Specifically, the target service address to be matched and multiple first standard addresses are normalized, including but not limited to handling special and erroneous characters, and case conversion, to obtain the processed target service address and multiple first standard addresses.

[0202] Next, it is necessary to determine the target business address body corresponding to the target business address, and the multiple standard address bodies corresponding to the multiple first standard addresses. Specifically, this can be done by using a preset address hierarchy strategy to classify the normalized addresses, obtaining address segments corresponding to multiple different address levels; and then extracting the address segments at preset positions from the multiple address segments as the address body.

[0203] After determining the address body, it is necessary to first perform address body-level matching using the target business address body and multiple standard address bodies to reduce the number of characters to be matched and improve matching efficiency. Simultaneously, after determining the target standard address body that matches the target business address body using address body matching, it is also necessary to determine multiple candidate first standard addresses corresponding to the target standard address body, and then select the target first standard address that matches the target business address from among these candidate first standard addresses.

[0204] In determining the target primary standard address from multiple candidate primary standard addresses, the multiple candidate primary standard addresses can be split into parent and child addresses. First, the parent address with the larger range is matched, then the child address with the smaller range is matched, achieving nested matching. Furthermore, since the number of addresses is large in real-world address matching scenarios, splitting the multiple candidate primary standard addresses allows parent and child address matching to be performed in parallel, improving matching efficiency.

[0205] After the aforementioned matching, there is still a problem of not finding a completely matching, precise address, requiring a second matching process. This second matching involves continuously narrowing down the range, from AOI-level precise matching to building-level precise matching to complete matching. By continuously narrowing the address range, the first target standard address that matches the target business address is finally determined, achieving a complete match between the business address and the standard address.

[0206] This application also provides an address matching device, such as... Figure 9 The diagram shown is a schematic representation of an embodiment of the address matching device provided in this application, which may include:

[0207] Address acquisition module 901 is used to acquire the target business address to be matched and multiple first standard addresses. The first business address comes from an external business platform, and the multiple first standard addresses and the first business address have at least some address information that is the same.

[0208] Address subject determination module 902 is used to determine the target service address subject corresponding to the target service address and the multiple standard address subjects corresponding to multiple first standard addresses.

[0209] Address subject matching module 903 is used to determine the target standard address subject that matches the target business address subject among multiple standard address subjects.

[0210] The address matching module 904 is used to determine the target standard address that matches the target service address based on the standard address corresponding to the target standard address subject.

[0211] The address matching apparatus provided in this application obtains a target service address to be matched and multiple first standard addresses. The first service addresses originate from an external service platform, and the multiple first standard addresses share at least some address information with the first service addresses. It determines the target service address body corresponding to the target service address and multiple standard address bodies corresponding to the multiple first standard addresses. Among the multiple standard address bodies, it determines the target standard address body that matches the target service address body. Based on the standard address corresponding to the target standard address body, it determines the target standard address that matches the target service address. In this application, an address body that can represent the address is determined for each address text. Matching is first performed using the address body, and then matching is performed using the complete address. Since the character length of the address body is less than the character length of the complete address, the number of characters to be matched is reduced, effectively improving the efficiency of address matching. Simultaneously, since the address body can represent the address, a second matching is performed using the complete address, thus improving the efficiency of address matching while ensuring the accuracy of address matching.

[0212] In some embodiments, the address subject determination module 902 may be specifically used to: perform multi-level address splitting on the target business address according to a preset address classification strategy to obtain multi-level business address word segmentation;

[0213] According to the preset address hierarchical strategy, multiple first standard addresses are split into multiple levels of address segments to obtain multiple levels of standard address segmentation sets. Each standard address segmentation set corresponds to a first standard address, and each standard address segmentation set includes multiple levels of standard address segments.

[0214] Based on word segmentation of multiple business addresses at multiple levels and word segmentation of multiple standard addresses, the target business address subject corresponding to the target business address and the multiple standard address subjects corresponding to multiple first standard addresses are determined.

[0215] In some embodiments, the address subject determination module 902 may be specifically used to: remove a first number of business address segments from multiple hierarchical business address segments, and extract a second number of business address segments at a preset address level to obtain the target business address subject corresponding to the target business address; take any standard address segment set from multiple standard address segment sets as the target standard address segment set; remove a third number of standard address segments from the target standard address segment set, and extract a fourth number of standard address segments at a preset address level to obtain the target standard address subject corresponding to the target business address segment set, so as to obtain multiple standard address subjects.

[0216] In some embodiments, the address subject matching module 903 can be specifically used to: determine the similarity between multiple standard address subjects and the target business address subject, and obtain multiple address subject similarities; take any address subject similarity among the multiple address subject similarities as the target address subject similarity; determine whether the target address subject similarity is not less than a preset address subject similarity threshold; if the target address subject similarity is not less than the preset address subject similarity threshold, then determine that the target standard address subject matches the target business address subject.

[0217] In some embodiments, the address matching module 904 may be specifically used to: determine a plurality of candidate first standard addresses corresponding to the target standard address subject; and determine the target first standard address that matches the target service address among the plurality of candidate first standard addresses.

[0218] In some embodiments, the address matching module 904 may be specifically used to: compress multiple candidate first standard addresses to obtain multiple compressed fifth standard addresses; split the multiple fifth standard addresses to obtain multiple parent standard addresses and multiple child standard addresses, wherein one fifth standard address corresponds to one parent standard address and one child standard address, and the multiple parent standard addresses correspond to the same address level; and determine, among the multiple parent standard addresses, a target parent standard address that matches the target first service address among the multiple first service addresses.

[0219] Determine multiple first-level child standard addresses corresponding to the target parent standard address; among the multiple first-level child standard addresses, determine the target child standard address that matches the target service address; based on the target parent standard address and the target child standard address, determine the target fifth standard address that matches the target service address.

[0220] In some embodiments, the address matching module 904 may be specifically used to: determine a plurality of candidate sixth standard addresses corresponding to the target fifth standard address; and determine the target first standard address that matches the target service address among the plurality of candidate sixth standard addresses.

[0221] In some embodiments, the address matching module 904 may be specifically used to: perform AOI-level precise matching on multiple candidate sixth standard addresses to determine multiple candidate seventh standard addresses that match the target service address; perform building-level precise matching on multiple candidate seventh standard addresses to determine multiple candidate eighth standard addresses that match the target service address; and determine the target first standard address that completely matches the target service address among the multiple candidate eighth standard addresses.

[0222] This application also provides an electronic device that integrates any of the address matching devices provided in this application. For example... Figure 10As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically:

[0223] The electronic device may include components such as a processor 1001 with one or more processing cores, a memory 1002 with one or more computer-readable storage media, a power supply 1003, and an input unit 1004. Those skilled in the art will understand that the electronic device structure shown in the figures does not constitute a limitation on the electronic device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0224] The processor 1001 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 1002, and by calling data stored in the memory 1002, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, the processor 1001 may include one or more processing cores; preferably, the processor 1001 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 1001.

[0225] The memory 1002 can be used to store software programs and modules. The processor 1001 executes various functional applications and data processing by running the software programs and modules stored in the memory 1002. The memory 1002 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 1002 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 1002 may also include a memory controller to provide the processor 1001 with access to the memory 1002.

[0226] The electronic device also includes a power supply 1003 that supplies power to various components. Preferably, the power supply 1003 can be logically connected to the processor 1001 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 1003 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0227] The electronic device may also include an input unit 1004, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0228] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 1001 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 1002 according to the following instructions, and the processor 1001 runs the applications stored in the memory 1002 to realize various functions, as follows:

[0229] Obtain the target business address to be matched and multiple first standard addresses. The target business address comes from an external business platform. The multiple first standard addresses have at least some address information that is the same as the target business address. The standard addresses are addresses that conform to the preset address specifications. Determine the target business address subject corresponding to the target business address and the multiple standard address subjects corresponding to the multiple first standard addresses. Among the multiple standard address subjects, determine the target standard address subject that matches the target business address subject. Based on the standard address corresponding to the target standard address subject, determine the target standard address that matches the target business address.

[0230] Among them, the character length of the target service address body is less than the character length of the target service address, and the character length of the multiple standard address bodies is less than the character length of the multiple first standard addresses.

[0231] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0232] Therefore, embodiments of this application provide a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, etc. A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps in any of the address matching methods provided in embodiments of this application. For example, the computer program loaded by the processor can execute the following steps:

[0233] Obtain the target business address to be matched and multiple first standard addresses. The target business address comes from an external business platform. The multiple first standard addresses have at least some address information that is the same as the target business address. The standard addresses are addresses that conform to the preset address specifications. Determine the target business address subject corresponding to the target business address and the multiple standard address subjects corresponding to the multiple first standard addresses. Among the multiple standard address subjects, determine the target standard address subject that matches the target business address subject. Based on the standard address corresponding to the target standard address subject, determine the target standard address that matches the target business address.

[0234] Among them, the character length of the target service address body is less than the character length of the target service address, and the character length of the multiple standard address bodies is less than the character length of the multiple first standard addresses.

[0235] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the detailed descriptions of other embodiments above, which will not be repeated here.

[0236] In practice, each of the above units or structures can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units or structures, please refer to the previous method embodiments, which will not be repeated here.

[0237] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0238] The above provides a detailed description of an address matching method, apparatus, and storage medium provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An address matching method, characterized in that, The method includes: Obtain the target business address to be matched and multiple first standard addresses. The target business address comes from an external business platform. Among the multiple first standard addresses, there is a first standard address that has at least some address information that is the same as the target business address. The multiple first standard addresses are addresses that conform to a preset address specification. The target service address subject corresponding to the target service address and the multiple standard address subjects corresponding to the multiple first standard addresses are determined respectively; Among the plurality of standard address subjects, a target standard address subject that matches the target service address subject is determined; Determine multiple candidate first standard addresses corresponding to the target standard address subject; Each candidate first standard address is compressed to obtain multiple compressed fifth standard addresses, with one candidate first standard address corresponding to one fifth standard address; wherein, the compression refers to removing address segments of unit number and house number from each candidate first standard address; The plurality of fifth standard addresses are split to obtain a plurality of parent standard addresses and a plurality of child standard addresses. Each fifth standard address corresponds to a parent standard address and a child standard address, and the plurality of parent standard addresses correspond to the same address level. The splitting criteria for the parent standard address and the child standard address are as follows: addresses with the same AOI level are retained as parent standard addresses, and the child standard addresses are different addresses below the AOI level. Among the multiple parent standard addresses, determine the target parent standard address that matches the target service address; Determine multiple first-level child standard addresses corresponding to the target parent standard address; Based on address similarity, determine the target sub-level standard address that matches the target service address from among the plurality of first sub-level standard addresses; Based on the target parent standard address and the target child standard address, determine the target fifth standard address that matches the target service address; Wherein, the character length of the subject of the target service address is less than the character length of the target service address, and the character length of the subject of each standard address is less than the character length of the corresponding first standard address.

2. The address matching method according to claim 1, characterized in that, Before obtaining the target service address to be matched and multiple first standard addresses, the method further includes: The special characters and erroneous characters in the initial business address and multiple initial standard addresses to be matched are replaced to obtain the first business address and multiple second standard addresses; Remove useless characters from the first service address and the plurality of second standard addresses to obtain the second service address and the plurality of third standard addresses; Replace the uppercase and lowercase characters in the second service address and the plurality of third standard addresses to obtain the third service address and the plurality of fourth standard addresses; If a string containing a Chinese character and multiple zeros appears in the third service address and the plurality of fourth standard addresses, the multiple zeros are converted into a single zero to obtain the target service address and the plurality of first standard addresses.

3. The address matching method according to claim 1, characterized in that, The step of determining the target service address subject corresponding to the target service address and the multiple standard address subjects corresponding to the multiple first standard addresses includes: According to the preset address classification strategy, the target business address is split into multiple levels of addresses to obtain multiple levels of business address word segmentation; According to the preset address hierarchical strategy, the multiple first standard addresses are split into multiple levels of address segments to obtain multiple levels of standard address segmentation sets. Each standard address segmentation set corresponds to a first standard address, and each standard address segmentation set includes multiple levels of standard address segments. Based on the word segmentation of the multiple hierarchical business addresses and the multiple standard address word segmentation sets, the target business address subject corresponding to the target business address and the multiple standard address subjects corresponding to the multiple first standard addresses are determined.

4. The address matching method according to claim 3, characterized in that, The step of determining the target business address body corresponding to the target business address and the multiple standard address bodies corresponding to the multiple first standard addresses based on the multiple hierarchical business address word segmentation and the multiple standard address word segmentation sets includes: Remove the first number of business address words from the first preset address level among the multiple levels of business address word segments, and extract the second number of business address words from the second preset address level to obtain the target business address body corresponding to the target business address; Among the multiple standard address word segmentation sets, any one of the standard address word segmentation sets shall be the target standard address word segmentation set; Remove the third number of standard address segments at the third preset address level from the target standard address segmentation set, and extract the fourth number of standard address segments at the fourth preset address level to obtain the target standard address body corresponding to the target standard address segmentation set, so as to obtain multiple standard address bodies.

5. The address matching method according to claim 1, characterized in that, The step of determining the target standard address subject that matches the target service address subject among the plurality of standard address subjects includes: The similarity between the multiple standard address subjects and the target business address subject is determined to obtain the similarity of multiple address subjects; The target address subject similarity is determined by any address subject similarity among the multiple address subject similarities. Determine whether the subject similarity of the target address is not less than a preset subject similarity threshold; If the similarity of the target address subject is not less than the preset address subject similarity threshold, then the target standard address subject is determined to match the target business address subject.

6. The address matching method according to claim 1, characterized in that, After determining the target fifth standard address that matches the target service address, the process further includes: The partial candidate first standard addresses before compression corresponding to the target fifth standard address are determined to be multiple candidate sixth standard addresses; Among the multiple candidate sixth standard addresses, a target first standard address that matches the target service address is determined according to a precise matching method.

7. The address matching method according to claim 6, characterized in that, The step of determining the target first standard address that matches the target service address from among the plurality of candidate sixth standard addresses according to a precise matching method includes: Perform AOI-level precise matching on the target service address and the multiple candidate sixth standard addresses to determine multiple candidate seventh standard addresses that match the target service address; Perform precise building-level matching between the target service address and the multiple candidate seventh standard addresses to determine multiple candidate eighth standard addresses that match the target service address; The target service address and the plurality of candidate eighth standard addresses are matched exactly to determine the target candidate eighth standard address that matches the target service address exactly. The target candidate eighth standard address is the target first standard address.

8. An address matching device, characterized in that, The device includes: The address acquisition module is used to acquire the target business address to be matched and multiple first standard addresses. The target business address comes from an external business platform. Among the multiple first standard addresses, there is a first standard address that has at least some address information that is the same as the target business address. The multiple first standard addresses are addresses that conform to a preset address specification. The address subject determination module is used to determine the target service address subject corresponding to the target service address and the multiple standard address subjects corresponding to the multiple first standard addresses respectively; The address subject matching module is used to determine, among the plurality of standard address subjects, the target standard address subject that matches the target service address subject; The address matching module is used to determine multiple candidate first standard addresses corresponding to the target standard address body; compress each candidate first standard address to obtain multiple compressed fifth standard addresses, with one candidate first standard address corresponding to one fifth standard address; the compression refers to removing address segments of unit number and house number from each candidate first standard address; the multiple fifth standard addresses are then split to obtain multiple parent standard addresses and multiple child standard addresses, with one fifth standard address corresponding to one parent standard address and one child standard address, and the multiple parent standard addresses corresponding to the same address level; the parent standard... The criteria for splitting the address and the child standard address are as follows: Addresses at the same AOI level are retained as the parent standard address, and the child standard addresses are addresses at different levels below the AOI level; among the multiple parent standard addresses, a target parent standard address matching the target service address is determined; multiple first child standard addresses corresponding to the target parent standard address are determined; based on address similarity, a target child standard address matching the target service address is determined among the multiple first child standard addresses; based on the target parent standard address and the target child standard address, a target fifth standard address matching the target service address is determined. Wherein, the character length of the subject of the target service address is less than the character length of the target service address, and the character length of the subject of each standard address is less than the character length of the corresponding first standard address.

9. A computer-readable storage medium, characterized in that, It stores a computer program, which is loaded by a processor to perform the steps of the address matching method according to any one of claims 1 to 7.

Images