Address attribute identification method and device

Abstract

The invention discloses an address attribute identification method and device, and relates to the technical field of computers. According to one specific embodiment, the method comprises the steps that a to-be-recognized target address is obtained, interest information is extracted from the target address, and the interest information comprises interest points and / or interest surfaces; and searching for an individual attribute mark of the interest information, determining an address attribute mark of the target address according to the individual attribute mark of the interest information and a preset address attribute identification strategy, and further obtaining an address attribute of the target address, the address attribute mark of the target address having a credibility level. According to the embodiment, real-time and accurate address attribute identification is realized, the address attribute mark of the target address is determined in real time based on the individual attribute mark and the address attribute identification strategy, the address attribute mark of the target address can accurately represent the credibility of attribute identification, and further application of the address attribute is facilitated.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for identifying address attributes. Background Art

[0002] Address data, as the core element of location information data, its work-residence attribute constitutes an indispensable feature of basic data, and provides strong data support for residential customer profiles, commercial area profiles, regional data statistics, public service facility planning, and transportation planning. Existing technical solutions generally identify address attributes by mining keywords in addresses or through third-party logistics collection and delivery information.

[0003] In the process of implementing the present invention, the inventors found that the prior art has the following problems:

[0004] The address attribute recognition method using keyword mining has a certain misjudgment rate. For example, directly marking keywords such as "company" and "development zone" as workplace company attributes through text recognition may result in misjudgment. The address attribute recognition method based on logistics collection and delivery information is an offline recognition method and cannot guarantee real-time and rapid address attribute recognition. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a method and device for identifying address attributes, which realizes real-time and accurate identification of address attributes. Based on individual attribute tags and address attribute identification strategies, not only the address attribute tags of the target address are determined in real time, but also the address attribute tags of the target address can accurately characterize the credibility of attribute identification, which is more conducive to the further application of address attributes.

[0006] To achieve the above object, according to one aspect of an embodiment of the present invention, a method for identifying address attributes is provided, comprising:

[0007] Acquire a target address to be identified, and extract interest information from the target address, wherein the interest information includes interest points and / or interest surfaces;

[0008] The individual attribute tag of the interest information is searched, and the address attribute tag of the target address is determined according to the individual attribute tag of the interest information and a preset address attribute identification strategy, thereby obtaining the address attribute of the target address, wherein the address attribute tag of the target address has a trust level.

[0009] Optionally, in response to not finding the individual attribute tag of the interest information, the method further includes: based on a pre-constructed feature word library with address attributes and matching priorities, matching the interest information with feature words in the feature word library in order of priority, and obtaining the individual attribute tag of the interest information according to the matching result.

[0010] Optionally, the feature word library is constructed in the following manner: obtaining each address text information, performing word segmentation processing on each address text information, and filtering out each interest information; counting the word frequency of each word contained in each interest information, and filtering out feature words that exceed a preset word frequency threshold from the various words; performing address attribute marking and matching priority marking on the feature words, and grouping feature words with the same matching priority into the same group to obtain the feature word library.

[0011] Optionally, the individual attribute tags include precise attribute tags and inaccurate attribute tags, and the precise attribute tags include, in order of priority: a residential ending attribute tag, a company ending attribute tag, a pure residential attribute tag, a residential attribute tag, and a company attribute tag; the inaccurate attribute tags include, in order of priority: a comprehensive functional attribute tag, a similar residential ending attribute tag, and a similar company ending attribute tag; the address attribute tag of the target address is determined according to the individual attribute tags of the interest information and a preset address attribute identification strategy, including: in response to the interest information being a precise attribute tag, obtaining the first hit result of the individual attribute tag according to the priority order of the precise attribute tags, determining the corresponding trust level according to the priority corresponding to the first hit individual attribute tag, and obtaining the address attribute tag of the target address; in response to the interest information being an inaccurate attribute tag, obtaining the first hit result of the individual attribute tag according to the priority order of the inaccurate attribute tag, determining the corresponding trust level according to the priority corresponding to the first hit individual attribute tag, and obtaining the address attribute tag of the target address.

[0012] Optionally, in response to the first hit individual attribute tag being the pure residential attribute tag, and the target address also having the company attribute tag, the method further includes: lowering the trust level in the address attribute tag of the target address, and generating the address attribute tag of the target address based on the company attribute tag and the lowered trust level.

[0013] Optionally, in response to the first hit individual attribute tag being the comprehensive functional attribute tag, and the target address also having the similar residential ending attribute tag, the method further includes: lowering the trust level in the address attribute tag of the target address, and generating the address attribute tag of the target address based on the similar residential ending attribute tag and the lowered trust level.

[0014] Optionally, in response to the fact that the address attribute tag of the target address is not determined based on the individual attribute tag of the interest information and a preset address attribute identification strategy, the method also includes: defining the target address with a house number and no interest information as a residential address attribute; defining the target address with a point of interest as a company address attribute.

[0015] Optionally, extracting interest information from the target address includes: inputting the target address into a pre-built address model, obtaining a word segmentation result of the target address output by the address model; and extracting words marked as interest information from the word segmentation result.

[0016] According to a second aspect of an embodiment of the present invention, there is provided a device for identifying address attributes, comprising:

[0017] An interest information extraction module, used to obtain a target address to be identified, and extract interest information from the target address, wherein the interest information includes interest points and / or interest surfaces;

[0018] The address attribute identification module is used to find the individual attribute mark of the interest information, determine the address attribute mark of the target address according to the individual attribute mark of the interest information and a preset address attribute identification strategy, and then obtain the address attribute of the target address, wherein the address attribute mark of the target address has a trust level.

[0019] According to a third aspect of an embodiment of the present invention, there is provided an electronic identification device for address attributes, comprising:

[0020] one or more processors;

[0021] a storage device for storing one or more programs,

[0022] When the one or more programs are executed by the one or more processors, the one or more processors implement the method provided in the first aspect of the embodiment of the present invention.

[0023] According to a fourth aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method provided by the first aspect of the embodiment of the present invention is implemented.

[0024] According to a fifth aspect of an embodiment of the present invention, there is provided a computer program product, including a computer program, wherein when the computer program is executed by a processor, the method provided in the first aspect of the embodiment of the present invention is implemented.

[0025] An embodiment of the invention has the following advantages or beneficial effects: by acquiring a target address to be identified, interest information is extracted from the target address, wherein the interest information includes points of interest and / or surfaces of interest; individual attribute tags of the interest information are searched, and the address attribute tags of the target address are determined according to the individual attribute tags of the interest information and a preset address attribute recognition strategy, thereby obtaining a technical solution for the address attributes of the target address, wherein the address attribute tags of the target address have a trust level, thereby realizing real-time and accurate recognition of address attributes, and based on the individual attribute tags and the address attribute recognition strategy, not only the address attribute tags of the target address are determined in real time, but also the address attribute tags of the target address can accurately represent the credibility of attribute recognition, which is more conducive to the further application of address attributes. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0027] Figure 1 is a schematic diagram of the main process of the method for identifying address attributes according to an embodiment of the present invention;

[0028] Figure 2 is a schematic diagram of extracting interest information according to an embodiment of the present invention;

[0029] Figure 3 is a schematic diagram of real-time tagging of interest information based on company or residence feature words according to an embodiment of the present invention;

[0030] Figure 4 is a schematic diagram of real-time tagging of interest information based on characteristic words of similar companies or similar residences according to an embodiment of the present invention;

[0031] Figure 5 is an overall schematic diagram of individual attribute marking of interest information according to an embodiment of the present invention;

[0032] Figure 6 Schematic diagram of a method for constructing a feature word library according to an embodiment of the present invention;

[0033] Figure 7 A method for determining an address attribute tag when a target address is a precise attribute tag according to an embodiment of the present invention;

[0034] Figure 8 A method for determining an address attribute tag when a target address is an imprecise attribute tag according to an embodiment of the present invention;

[0035] Fig. 9 is an overall schematic diagram of a method for identifying address attributes according to an embodiment of the present invention;

[0036] Fig.10is a schematic diagram of main modules of an address attribute identification device according to an embodiment of the present invention;

[0037] Fig.11 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0038] Fig.12 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0039] It should be noted that in the technical solution of the present invention, the collection / collection, update, analysis, use, transmission, storage and other aspects of user personal information involved are in compliance with the provisions of relevant laws and regulations, are used for legal and reasonable purposes, are not shared, leaked or sold outside of these legal uses, and are subject to supervision and management by national regulatory authorities. Necessary measures should be taken for user personal information to selectively block the use or access to personal information data to prevent illegal access to such personal information data, ensure that persons who have access to personal information data comply with the provisions of relevant laws and regulations, and ensure the security of user personal information. In addition, once these user personal information data are no longer needed, the risks should be minimized by limiting or even prohibiting data collection and / or deleting data.

[0040] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0041] The existing scheme uses keyword mining to identify address attributes, which has a certain misjudgment rate. For example, directly marking keywords such as "company" and "development zone" as workplace company attributes through text recognition may result in misjudgment; and the address attribute recognition method based on logistics collection and delivery information is an offline recognition method, which cannot guarantee real-time and fast address attribute recognition and cannot meet practical applications well.

[0042] In order to solve the above problems existing in the prior art, the present invention proposes a method for identifying address attributes, which obtains address attribute tags with trust levels of the target address by searching the individual attribute tags of the interest information in the target address online in real time, combined with a preset address attribute identification strategy, thereby realizing real-time and accurate identification of address attributes. Based on the individual attribute tags and the address attribute identification strategy, not only the address attribute tags of the target address are determined in real time, but also the address attribute tags of the target address can accurately characterize the credibility of attribute identification, which is more conducive to the further application of address attributes.

[0043] Figure 1 is a schematic diagram of the main process of the method for identifying address attributes according to an embodiment of the present invention, such as Figure 1 As shown, the method for identifying address attributes in the embodiment of the present invention includes the following steps S101 to S103.

[0044] Step S101: Acquire a target address to be identified, and extract interest information from the target address, wherein the interest information includes interest points and / or interest surfaces.

[0045] Specifically, the target address to be identified is parsed from the received address attribute identification task or address attribute identification request. The specific address attribute identification task can be manually sent or scheduled. Of course, the target address to be identified can be one target address or multiple target addresses. When the target address data volume is large and is not suitable for packaging into the address attribute identification task or address attribute identification request, it can be stored separately as an address file. The identification system parses the address attribute identification task or the address attribute identification request to obtain the storage location of the address file, and then obtains the target address to be identified.

[0046] Although address data is difficult to achieve complete consistency in the description of the same location due to subjective reasons, there are general rules in address data. Generally, it will contain some information of elements such as province (or autonomous region), city, district, county, administrative district, township street, village community, group (or team), AOI name, road name, directional word, house number, POI name, building, express cabinet, unit, floor, house number, etc. Among them, the AOI name of the interest surface and the POI name of the point of interest generally have clearer work and residence attributes, such as "X Company" and "Y Residential Building", etc. Therefore, the embodiments of the present invention use the points of interest and / or interest surfaces in the target address as the main basis for judging the work and residence attributes of the identified address.

[0047] Furthermore, an embodiment of the present invention extracts organizational rules of address text from massive address text, establishes an address text extraction template based on the organizational rules, matches the target address text with a pre-configured address text extraction template, and extracts interest points and / or interest surfaces information from the target address according to the positions of the interest points and / or interest surfaces identified in the successfully matched template in the address text.

[0048] According to one embodiment of the present invention, extracting interest information from the target address includes: inputting the target address into a pre-built address model, obtaining a word segmentation result of the target address output by the address model; and extracting words marked as interest information from the word segmentation result.

[0049] Specifically, considering that the accuracy of interest information extraction plays a key role in the identification of address attributes, in order to ensure the accuracy of interest information extraction, an embodiment of the present invention uses a pre-built address model after deep learning, and uses the address model to accurately segment the target address. The segmentation results output by the address model are accompanied by AOI and POI tags, and then the segments marked as points of interest and / or areas of interest are extracted from the segmentation results.

[0050] Figure 2 Schematic diagram of extracting interest information in an embodiment of the present invention. The requester of address attribute recognition applies to use the word segmentation processing service in the business service. The word segmentation processing service calls the word segmentation service encapsulated by the address model, and sends the target address to the word segmentation interface of the word segmentation service. The word segmentation interface sends the received target address to the address model, and the address model performs word segmentation and AOI / POI tagging, outputs the word segmentation result with AOI / POI tag, and returns it to the word segmentation processing service. For example, the target address is Unit 1, Building 1, Z Community, Economic and Technological Development Zone, Y City, X Province, and the word segmentation result output by the address model in the word segmentation service is: Province: X; City: Y; Virtual District: Economic and Technological Development Zone; AOI: Z Community; Building Number: 1; House Number: 101. It can be seen that although the target address contains the Economic and Technological Development Zone, the address model can still accurately identify the AOI as Z Community, ensuring that the subsequent recognition of address attributes is mainly based on AOI, and the target address will not be misjudged as a company workplace attribute because of the Economic and Technological Development Zone.

[0051] Step S102, searching for the individual attribute tag of the interest information, determining the address attribute tag of the target address according to the individual attribute tag of the interest information and a preset address attribute identification strategy, and then obtaining the address attribute of the target address, wherein the address attribute tag of the target address has a trust level.

[0052] Specifically, based on the AOI and / or POI in the target address obtained above, the individual attribute tags of the AOI and / or POI of the target address are searched from the AOI and POI cache that have been marked with individual attribute tags. Considering that there is sometimes more than one AOI or POI in a target address, the embodiment of the present invention sets individual attribute tags of high-level interest information based on experience, and uses the individual attribute tags of high-level interest information as the address attribute tags of the target address. This logic of determining the address attribute tag is the address attribute identification strategy mentioned in the embodiment of the present invention, thereby ensuring the accuracy of the identification, and retaining the level of the individual attribute tag in the address attribute tag, that is, the trust level, to characterize the credibility of the address attribute of the currently obtained target address.

[0053] According to one embodiment of the present invention, in response to the individual attribute tag of the interest information and the preset address attribute identification strategy, the address attribute tag of the target address is not determined, and the method also includes: defining the target address with a house number and no interest information as a residential address attribute; defining the target address with a point of interest as a company address attribute.

[0054] Inevitably, it may happen that the address attribute tag of the target address cannot be determined, for example, there is no information of interest in the target address, or an individual attribute tag of the POI cannot be determined or found. In this case, the embodiment of the present invention sets a fallback strategy to determine the target address containing a house number but no information of interest as a residential address attribute, and of course the corresponding address attribute tag has the lowest level of trust; the target address with a point of interest POI is defined as a company address attribute, and of course the corresponding address attribute tag has the lowest level of trust.

[0055] The above-mentioned method for identifying address attributes is performed on the premise that individual attribute tags of the information of interest can be found. If the individual attribute tags of the AOI and / or POI of the target address are not found in the AOI and POI cache that have been marked with individual attribute tags, then the present invention provides another embodiment.

[0056] According to another embodiment of the present invention, in response to failure to find the individual attribute tag of the interest information, the method further includes: based on a pre-constructed feature word library with address attributes and matching priorities, matching the interest information with feature words in the feature word library in order of priority, and obtaining the individual attribute tag of the interest information according to the matching result.

[0057] Specifically, when the individual attribute tag of the interest information is not found, the interest information needs to be marked in real time. The interest information and the feature words in the feature word library are matched in order according to the priority defined in the feature word library. The feature words in the feature word library of the embodiment of the present invention have a priority order, and the specific priority order is: feature words representing companies or residences, feature words representing pure residential communities with water and electricity for the people, feature words representing similar companies or residences ("apartments", "buildings", "nursing homes", etc.), feature words representing large-scale comprehensive functional places ("universities", "resorts"), feature words representing express lockers or post stations, and feature words representing well-known real estate.

[0058] In addition, for the feature words representing companies or residences, which are key feature words of the embodiment of the present invention, the embodiment of the present invention performs a finer-grained division, and distinguishes according to the position of the feature words representing companies or residences at the target address, such as at the end of the target address, or at the middle of the target address, and gives a higher priority to the one at the end of the target address than the one in the middle of the target address. It is understandable that, usually, the AOI and POI at the end of the destination address are the most accurate. For example, "family building of a certain construction group" contains the hit company feature words, but it is actually more inclined to residential attributes. Therefore, in addition to the names of the AOI and POI, the interest information of the embodiment of the present invention also has the location information of the interest information at the target address, such as the end or the middle.

[0059] Specifically, based on a feature word library with priority, the highest priority ending feature word representing a company or residence is first used to match the interest information. If the match is successful, the company or residence represented by the feature word is used as the individual attribute tag of the interest information, and the tail position information is appended to the end of the individual attribute tag; if the highest priority feature word is not matched successfully, the next priority feature word containing a company or residence is matched with the interest information. If the match is successful, the company or residence represented by the feature word is used as the individual attribute tag of the interest information, and the middle position information is appended to the end of the individual attribute tag.

[0060] Figure 3It is a schematic diagram of real-time marking of interest information based on company or residential feature words according to an embodiment of the present invention. The highest priority residential ending feature words, company ending feature words and interest information in the feature word library are matched. If the residential ending feature words and interest information are successfully matched, the interest information is marked as C1E, where C represents residential, 1 symbolically represents priority, and E represents the end; accordingly, if the company ending feature words and interest information are successfully matched, the interest information is marked as B1E. If the above matching fails, then the next priority residential feature words, company feature words and interest information are matched. If the residential feature words are successfully matched, the interest information is marked as C1C, where the last C represents the middle; of course, if the company feature words and interest information are successfully matched, the interest information is marked as B1C.

[0061] For the feature words representing similar companies or similar residences, the embodiment of the present invention is referred to as company-like or residence-like feature words. Similar to the feature words representing companies or residences, the embodiment of the present invention mainly considers the recognition contribution brought by the interest information at the end. That is, the similar company feature words mentioned above are essentially feature words ending with similar companies, and the similar residence feature words are essentially feature words ending with similar residences. For example, Figure 4 This is a schematic diagram of real-time tagging of interest information based on feature words of similar companies or similar residences according to an embodiment of the present invention. Feature words ending with similar companies or similar residences in the feature word library are matched with the interest information. If the feature words ending with similar residences and the interest information are matched successfully, the interest information is marked as C2E. If the feature words ending with similar companies and the interest information are matched successfully, the interest information is marked as B2E.

[0062] By analogy, according to the above method, the interest information is matched in order of priority, and the individual attribute tags of the interest information are obtained online and in real time according to the matching results.

[0063] Figure 5: is an overall schematic diagram of the individual attribute marking of the interest information of an embodiment of the present invention. Based on the interest information in the target address, the interest information is first matched with the feature words representing residences or companies according to the priority order of the feature word library. If the match is successful, the individual attribute of the interest information is marked as C1 or B1. It should be noted that this includes the above-mentioned C1E, C1C, B1E and B1C. If the match fails, the interest information is matched with the feature words representing pure residential areas. If the match is successful, the individual attribute of the interest information is marked as CZZ. If the match still fails, the interest information is matched with the feature words representing similar residences or similar residences. If the match is successful, the individual attribute of the interest information is marked as C2 or B2. Similarly, C2E and B2E are included. If the match still fails, the interest information is matched with the feature words representing large-scale comprehensive functional places. If the match is successful, the individual attribute of the interest information is marked as BBC. If the match fails, the interest information is matched with the feature words representing express cabinets or post stations. If the match is successful, the individual attribute of the interest information is marked as KDG. If the match still fails, the interest information is finally matched with the characteristic words representing well-known real estate. If the match is successful, the individual attribute mark of the interest information is CFC. If the match is still not successful, the information marking the failure is returned, and the address attribute of the target address is determined according to the above-mentioned fallback strategy.

[0064] According to another embodiment of the present invention, the feature word library is constructed in the following manner: obtaining each address text information, performing word segmentation processing on each of the address text information, and filtering out each interest information; counting the word frequency of each word contained in each of the interest information, and filtering out feature words that exceed a preset word frequency threshold from the various words; performing address attribute marking and matching priority marking on the feature words, and grouping feature words with the same matching priority into the same group to obtain the feature word library.

[0065] Specifically, the construction of the feature word library of the embodiment of the present invention mainly relies on obtaining a large amount of address text information from the address library, and after the word segmentation processing of the word segmentation tool, preferably using the address model in the above-mentioned word segmentation service, the interest information is filtered out from the word segmentation results. Then the word frequency of each word contained in the massive interest information is counted, and high-frequency feature words exceeding the word frequency threshold are screened out, such as pharmacies, research institutes, family compounds, post stations, etc. Finally, these high-frequency feature words are marked with address attributes and matching priorities, and high-frequency feature words with the same matching priority are grouped together to obtain a feature word library with matching priority and address attributes. For example, the address attribute of the high-frequency feature words of the above-mentioned pharmacies and research institutes is company, and its matching priority is the highest priority company ending feature word and the next level priority containing company feature word; while the address attribute of the post station high-frequency feature word is express cabinet or post station, and its matching priority is lower than the company ending feature word, containing company feature word and feature words of pure residential areas.

[0066] Figure 6 It is a schematic diagram of a method for constructing a feature word library of an embodiment of the present invention. Based on a large amount of address text information, after being processed by the address model in the word segmentation service, a word segmentation result is obtained; the interest information and each word element contained in the interest information are filtered and extracted from the word segmentation result; the word frequency analysis is performed on each word element, the word frequency of each word contained in the interest information is counted, and high-frequency feature words exceeding the word frequency threshold are screened out; finally, these high-frequency feature words are marked with address attributes and matching priorities, and high-frequency feature words with the same matching priority are grouped into the same group, such as feature words representing companies or residences, feature words representing pure residential communities for civil water and electricity, feature words representing similar companies or similar residences ("apartment", "building", "nursing home", etc.), feature words representing large comprehensive functional places ("university", "resort"), feature words representing express lockers or post stations, and feature words representing well-known real estate.

[0067] By marking the interest information in real time and determining its individual attribute tags without searching for the individual attribute tags of the AOI and / or POI of the target address, the application scenarios of the address attribute recognition method of the present invention are expanded and the adaptability of the method is improved to be suitable for more address attribute recognition scenarios.

[0068] Considering that the embodiment of the present invention is mainly used to identify the work-residence attribute of the target address, in order to obtain accurate identification results, the present invention provides another embodiment for determining the address attribute tag of the target address.

[0069] According to another embodiment of the present invention, the individual attribute tags include precise attribute tags and imprecise attribute tags, and the precise attribute tags include, in order of priority: residential ending attribute tags, company ending attribute tags, pure residential attribute tags, residential attribute tags, and company attribute tags; the imprecise attribute tags include, in order of priority: comprehensive functional attribute tags, similar residential ending attribute tags, and similar company ending attribute tags; according to the individual attribute tags of the interest information and a preset address attribute identification strategy, the address attribute tags of the target address are determined, including: in response to the interest information being a precise attribute tag, according to the priority order of the precise attribute tags, the first hit result of the individual attribute tags is obtained, and according to the priority corresponding to the first hit individual attribute tag, the corresponding trust level is determined to obtain the address attribute tag of the target address; in response to the interest information being an imprecise attribute tag, according to the priority order of the imprecise attribute tags, the first hit result of the individual attribute tags is obtained, and according to the priority corresponding to the first hit individual attribute tag, the corresponding trust level is determined to obtain the address attribute tag of the target address.

[0070] Specifically, based on the individual attribute tags in the above-mentioned feature word library, considering that the key attribute tags of the embodiment of the present invention mainly revolve around the company attribute tags and the residential attribute tags, and there may be multiple individual attribute tags of interest information in a target address, then it is necessary to set a strategy for determining the address attributes, that is, the address attribute recognition strategy mentioned in the embodiment of the present invention, and the recognition priority of the individual attribute tags will be defined in the address attribute recognition strategy. Specifically, the individual attribute tags involving the company attribute tags and the residential attribute tags are summarized and classified, and first divided into two categories according to the accuracy: precise attribute tags and inaccurate attribute tags, among which the precise attribute tags include the attribute tags ending with residential, the attribute tags ending with company, the pure residential attribute tags, the attribute tags containing residential, and the attribute tags containing company, and the recognition priority is determined according to the credibility of these tags, and the recognition priority is ordered from high to low as follows: the attribute tags ending with residential, the attribute tags ending with company, the pure residential attribute tags, the attribute tags containing residential, and the attribute tags containing company. Similarly, the comprehensive functional attribute tags, attribute tags ending with "similar to residence" and attribute tags ending with "similar to company" that are non-precise attribute tags also determine their respective recognition priorities. The recognition priority is in the order of high to low: comprehensive functional attribute tags, attribute tags ending with "similar to residence" and attribute tags ending with "similar to company".

[0071] Further, when the individual attribute tag of the interest information of the target address is a precise attribute tag, the first hit attribute tag of the individual attribute tag of the interest information is obtained first according to the order of the recognition priority agreed upon above, and the trust level of this address attribute recognition is determined according to the recognition priority corresponding to the first hit attribute tag, and the address attribute tag of the target address is generated. For example, the individual attribute tags of the interest information in the target address are searched in the order of the residential ending attribute tag, the company ending attribute tag, the pure residential attribute tag, the residential attribute tag, and the company attribute tag. If the individual attribute tags of the interest information in the target address have the residential ending attribute tag and the company attribute tag, then the residential ending attribute tag is hit first according to the order, and the residential ending attribute tag is used as the address attribute tag of the target address. At the same time, since the residential ending attribute tag has the highest priority, the highest trust level is attached to the address attribute tag to obtain the final address attribute tag of the target address.

[0072] When the individual attribute tag of the target address's interest information is an imprecise attribute tag, the identification priority order of the imprecise attribute tag agreed upon above is followed, and the first hit attribute tag of the individual attribute tag of the interest information is first obtained, and the trust level of this address attribute identification is determined according to the identification priority corresponding to the first hit attribute tag, and the address attribute tag of the target address is generated. For example, the individual attribute tags of the interest information in the target address are searched in the order of comprehensive functional attribute tags, similar residential ending attribute tags, and similar company ending attribute tags. If the individual attribute tags of the interest information in the target address have comprehensive functional attribute tags and similar company ending attribute tags, then the first hit is the comprehensive functional attribute tag, and the comprehensive functional attribute tag is used as the address attribute tag of the target address, and the order of the identification priority corresponding to the comprehensive functional attribute tag is attached to the address attribute tag as the trust level, and the final address attribute tag of the target address is obtained.

[0073] In addition, if the final address attributes of the target address are derived from comprehensive functional attribute tags, then the individual attribute tags of the express cabinet or post station in the target address and the individual attribute tags of well-known real estate can be combined to jointly determine the work-residence attributes of the target address.

[0074] According to another embodiment of the present invention, in response to the first hit individual attribute tag being the pure residential attribute tag, and the target address also having the included company attribute tag, the method further includes: lowering the trust level in the address attribute tag of the target address, and generating the address attribute tag of the target address based on the included company attribute tag and the lowered trust level.

[0075] Specifically, for the case where the interest information is a precise attribute tag, in order to further improve the accuracy of address attribute recognition, a special case needs to be considered: a pure residential address will inevitably contain company attributes. Therefore, when the first hit individual attribute tag is a pure residential attribute tag, and the target address also has a company attribute tag, such as Hairdressing Studio 102 on the 1st floor of Building 2 of a family compound, in this case, the address attribute tag is determined to be a company attribute based on the company attribute tag. In addition, considering that this is a company attribute in a pure residential address, its trust level needs to be downgraded, and finally a complete address attribute tag is generated.

[0076] Figure 7 This is a method for determining an address attribute tag when the target address of an embodiment of the present invention is a precise attribute tag. Based on the individual attribute tags of each interest information in the target address, the first hit result of the individual attribute tag is obtained in the order of the residential attribute tag ending, the company attribute tag ending, the pure residential attribute tag, the residential attribute tag, and the company attribute tag. If the residential attribute tag ending is hit, the address attribute tag of the target address is C1, where 1 represents the trust level, and the smaller the value, the stronger the trust, and 1 is the strongest trust. If the company attribute tag ending is hit, the address attribute tag of the target address is B1. If the pure residential attribute tag is hit and the target address does not contain the company attribute tag, the address attribute tag of the target address is C2. If the target address contains the company attribute tag, the address attribute tag of the target address is B3B, where 3B is the trust level after downgrade processing (from 2 to 3B). Similarly, if the residential attribute tag or the company attribute tag is hit, the address attribute tag of the target address is C3 or B3.

[0077] According to another embodiment of the present invention, in response to the first hit individual attribute tag being the comprehensive functional attribute tag, and the target address also having the similar residential ending attribute tag, the method further includes: lowering the trust level in the address attribute tag of the target address, and generating the address attribute tag of the target address based on the similar residential ending attribute tag and the lowered trust level.

[0078] Specifically, for the case where the interest information is a non-precise attribute tag, in order to further improve the accuracy of address attribute recognition, a special case needs to be considered: large comprehensive addresses will inevitably have residential attributes. Therefore, when the first hit individual attribute tag is a comprehensive functional attribute tag, and the target address also contains a similar residential ending attribute tag, such as Dormitory 102, 1st Floor, Building 2, Area A of a university, in this case, the address attribute tag is determined to be a residential attribute based on the similar residential ending attribute tag. In addition, considering that this is a residential attribute in a large comprehensive address, its trust level needs to be downgraded, and finally a complete address attribute tag is generated.

[0079] Figure 8 This is a method for determining an address attribute tag when the target address of an embodiment of the present invention is an imprecise attribute tag. Based on the individual attribute tags of each piece of interest information in the target address, the first hit result of the individual attribute tag is obtained in the order of the comprehensive functional attribute tag, the similar residential ending attribute tag, and the similar company ending attribute tag. If the comprehensive functional attribute tag is hit, and the target address does not contain the similar residential ending attribute tag, then the address attribute tag of the target address is B3A. If the target address contains the similar residential ending attribute tag, then the address attribute tag of the target address is C4E, where 4E is the trust level after downgrade processing (from 3A to 4E). Similarly, if the similar residential ending attribute tag or the similar company ending attribute tag is hit, then the address attribute tag of the target address is C4 or B4.

[0080] It should be noted that in the case where the target address's segmentation results include fourth-level administrative regions such as townships and towns, it is necessary to perform serial judgment based on the house number, unit and other information in the segmentation results of the target address, and determine the address attributes of the target address based on the preset company attribute segmentation. For example, "village committee" is judged as a company attribute, or when the target address does not contain a POI, according to the exclusive judgment logic, it is determined whether the target address has a specified residential attribute segmentation, such as "a certain township" or "a certain village". For target addresses with residential attribute segmentations, they are judged and identified as residential attributes, and the fourth-level administrative region logo is attached to clearly indicate the user of the subsequent address attributes.

[0081] In order to improve the recognition efficiency of address attributes, the embodiment of the present invention adds a cache strategy in each link. For example, after obtaining the address attributes of the target address, the target address and its corresponding address attributes are stored in the address attribute cache, so that when the target address is received later, the corresponding address attributes can be directly queried from the address attribute cache, saving recognition time. Or after obtaining the individual attribute tag of the interest information according to the matching result, the individual attribute tag of the interest surface marked in real time is stored in the AOI and POI cache of the individual attribute tag, so that the next time the same interest information is encountered, it can be directly obtained from the AOI and POI cache of the individual attribute tag.

[0082] Fig. 9 The figure is an overall schematic diagram of the address attribute recognition method of the embodiment of the present invention. To obtain the target address, first query the address attribute cache to see if there is a hit cache record for the target address. If there is a hit, directly return the cached address attribute of the hit, without the need for subsequent recognition. If there is no hit, use the word segmentation service to segment the target address and extract the interest information therein. Then search for the individual attribute tag of the interest information from the AOI and POI caches of the individual attribute tags. If found, then determine the address attribute tag of the target address according to the individual attribute tag of the interest information and the preset address attribute recognition strategy; if not found, then match the interest information and the feature words in the feature word library in the priority order based on the pre-built feature word library with address attributes and matching priorities, obtain the individual attribute tag of the interest information according to the matching result, perform online real-time tagging on the individuals of the interest information, and then determine the address attribute tag of the target address according to the individual attribute tag of the interest information and the preset address attribute recognition strategy, and then obtain the address attribute of the target address. Finally, cache the result of the address attribute of the target address in the address attribute cache, and return the recognition result (address attribute).

[0083] The embodiment of the present invention obtains the address attribute tag of the target address with a trust level by real-time online search for the individual attribute tag of the interest information in the target address, combined with a preset address attribute identification strategy, thereby realizing real-time and accurate identification of address attributes. Based on the individual attribute tag and the address attribute identification strategy, not only the address attribute tag of the target address is determined in real time, but also the address attribute tag of the target address can accurately characterize the credibility of attribute identification, which is more conducive to further application of address attributes in map-based software operation service SaaS service.

[0084] Fig.10 FIG. 2 is a schematic diagram of the main modules of the device for identifying address attributes according to an embodiment of the present invention. Fig.10 As shown, the device 1000 for identifying address attributes mainly includes an interest information extraction module 1001 and an address attribute identification module 1002 .

[0085] The interest information extraction module 1001 is used to obtain a target address to be identified and extract interest information from the target address, wherein the interest information includes interest points and / or interest surfaces;

[0086] The address attribute identification module 1002 is used to find the individual attribute tag of the interest information, determine the address attribute tag of the target address according to the individual attribute tag of the interest information and a preset address attribute identification strategy, and then obtain the address attribute of the target address, wherein the address attribute tag of the target address has a trust level.

[0087] According to one embodiment of the present invention, in response to not finding the individual attribute tag of the interest information, the address attribute identification device 1000 also includes an individual attribute tag determination module (not shown in the figure), which is used to: based on a pre-constructed feature word library with address attributes and matching priorities, match the interest information and feature words in the feature word library in order of priority, and obtain the individual attribute tag of the interest information according to the matching result.

[0088] According to another embodiment of the present invention, the feature word library is constructed in the following manner: obtaining each address text information, performing word segmentation processing on each of the address text information, and filtering out each interest information; counting the word frequency of each word contained in each of the interest information, and filtering out feature words that exceed a preset word frequency threshold from the various words; performing address attribute marking and matching priority marking on the feature words, and grouping feature words with the same matching priority into the same group to obtain the feature word library.

[0089] According to another embodiment of the present invention, the individual attribute tag includes a precise attribute tag and an imprecise attribute tag, and the precise attribute tag includes, in order of priority: a residential ending attribute tag, a company ending attribute tag, a pure residential attribute tag, a residential attribute tag, and a company attribute tag; the imprecise attribute tag includes, in order of priority: a comprehensive functional attribute tag, a similar residential ending attribute tag, and a similar company ending attribute tag; the address attribute identification module 1002 is further used to: in response to the interest information being a precise attribute tag, obtain the first hit result of the individual attribute tag according to the priority order of the precise attribute tags, determine the corresponding trust level according to the priority corresponding to the first hit individual attribute tag, and obtain the address attribute tag of the target address; in response to the interest information being an imprecise attribute tag, obtain the first hit result of the individual attribute tag according to the priority order of the imprecise attribute tags, determine the corresponding trust level according to the priority corresponding to the first hit individual attribute tag, and obtain the address attribute tag of the target address.

[0090] According to another embodiment of the present invention, in response to the first hit individual attribute tag being the pure residential attribute tag, and the target address also having the company attribute tag, the address attribute identification device 1000 also includes a pure residential attribute supplementary judgment module (not shown in the figure), which is used to: reduce the trust level in the address attribute tag of the target address, and generate the address attribute tag of the target address according to the company attribute tag and the reduced trust level.

[0091] According to another embodiment of the present invention, in response to the first hit individual attribute mark being the comprehensive functional attribute mark, and the target address also having the similar residential ending attribute mark, the address attribute identification device 1000 also includes a comprehensive attribute supplementary judgment module (not shown in the figure), which is used to: reduce the trust level in the address attribute mark of the target address, and generate the address attribute mark of the target address according to the similar residential ending attribute mark and the reduced trust level.

[0092] According to another embodiment of the present invention, in response to the fact that the address attribute tag of the target address is not determined based on the individual attribute tag of the interest information and the preset address attribute identification strategy, the address attribute identification device 1000 also includes a fallback attribute identification module (not shown in the figure), which is used to: define the target address with a house number and no interest information as a residential address attribute; and define the target address with a point of interest as a company address attribute.

[0093] According to another embodiment of the present invention, the interest information extraction module 1001 is further used to: input the target address into a pre-constructed address model, obtain the word segmentation result of the target address output by the address model; and extract the word segmentation marked as interest information from the word segmentation result.

[0094] Fig.11 is an exemplary system architecture diagram to which embodiments of the present invention can be applied.

[0095] like Fig.11 As shown, system architecture 1100 may include terminal devices 1101, 1102, 1103, a network 1104, and a server 1105. Network 1104 is used to provide a medium for communication links between terminal devices 1101, 1102, 1103 and server 1105. Network 1104 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0096] Users can use terminal devices 1101, 1102, 1103 to interact with server 1105 through network 1104 to receive or send messages, etc. Terminal devices 1101, 1102, 1103 can be installed with various communication client applications, such as address attribute recognition applications, etc. (only as an example).

[0097] The terminal devices 1101 , 1102 , and 1103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc.

[0098] The server 1105 may be a server that provides various services, such as a background management server (only an example) that provides support for address attributes performed by users using terminal devices 1101, 1102, and 1103. The background management server may obtain a target address to be identified, extract interest information from the target address, wherein the interest information includes points of interest and / or surfaces of interest; search for individual attribute tags of the interest information, determine the address attribute tags of the target address based on the individual attribute tags of the interest information and a preset address attribute identification strategy, and then obtain the address attributes of the target address, wherein the address attribute tags of the target address have a trust level and other processing, and feed back the processing results (such as address attributes, etc. - only an example) to the terminal device.

[0099] It should be noted that the address attribute identification method provided in the embodiment of the present invention is generally executed by the server 1105 , and accordingly, the address attribute identification device is generally set in the server 1105 .

[0100] It should be understood that Fig.11 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0101] Reference below Fig.12 , Fig.12 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. Fig.12 The terminal device or server shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0102] like Fig.12As shown, the computer system 1200 includes a central processing unit (CPU) 1201, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1202 or a program loaded from a storage part 1208 into a random access memory (RAM) 1203. In the RAM 1203, various programs and data required for the operation of the system 1200 are also stored. The CPU 1201, the ROM 1202, and the RAM 1203 are connected to each other through a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0103] The following components are connected to the I / O interface 1205: an input section 1206 including a keyboard, a mouse, etc.; an output section 1207 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1208 including a hard disk, etc.; and a communication section 1209 including a network interface card such as a LAN card, a modem, etc. The communication section 1209 performs communication processing via a network such as the Internet. A drive 1210 is also connected to the I / O interface 1205 as needed. A removable medium 1211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1210 as needed, so that a computer program read therefrom is installed into the storage section 1208 as needed.

[0104] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1209, and / or installed from the removable medium 1211. When the computer program is executed by the central processing unit (CPU) 1201, the above-mentioned functions defined in the system of the present invention are executed.

[0105] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination thereof.

[0106] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0107] The units involved in the embodiments of the present invention may be implemented by software or hardware. The units described may also be arranged in a processor, for example, it may be described as follows: a processor includes an interest information extraction module and an address attribute identification module.

[0108] The names of these modules do not, in some cases, constitute limitations on the modules themselves. For example, the interest information extraction module can also be described as "a module for obtaining a target address to be identified and extracting interest information from the target address."

[0109] On the other hand, the present invention also provides a computer-readable medium, which may be included in the device described in the embodiment; or it may exist independently without being assembled into the device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by a device, the device includes: obtaining a target address to be identified, extracting interest information from the target address, wherein the interest information includes points of interest and / or surfaces of interest; searching for individual attribute tags of the interest information, determining the address attribute tags of the target address according to the individual attribute tags of the interest information and a preset address attribute identification strategy, and then obtaining the address attributes of the target address, wherein the address attribute tags of the target address have a trust level.

[0110] According to the technical solution of the embodiment of the present invention, it has the following advantages or beneficial effects: by acquiring the target address to be identified, interest information is extracted from the target address, wherein the interest information includes interest points and / or interest surfaces; searching for individual attribute tags of the interest information, and determining the address attribute tags of the target address according to the individual attribute tags of the interest information and a preset address attribute identification strategy, thereby obtaining a technical solution for the address attributes of the target address, wherein the address attribute tags of the target address have a trust level, thereby realizing real-time and accurate identification of address attributes, and based on the individual attribute tags and the address attribute identification strategy, not only the address attribute tags of the target address are determined in real time, but also the address attribute tags of the target address can accurately represent the credibility of attribute identification, which is more conducive to the further application of address attributes.

[0111] The specific implementation methods described herein do not constitute limitations on the scope of protection of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A method for identifying address attributes, characterized in that: include: Acquire a target address to be identified, and extract interest information from the target address, wherein the interest information includes interest points and / or interest surfaces; The individual attribute tag of the interest information is searched, and the address attribute tag of the target address is determined according to the individual attribute tag of the interest information and a preset address attribute identification strategy, thereby obtaining the address attribute of the target address, wherein the address attribute tag of the target address has a trust level.

2. The method according to claim 1, characterized in that In response to not finding the individual attribute tag of the interest information, the method further includes: Based on a pre-built feature word library with address attributes and matching priorities, the interest information and feature words in the feature word library are matched in order of priority, and the individual attribute tag of the interest information is obtained according to the matching result.

3. The method according to claim 2, characterized in that The feature word library is constructed in the following way: Acquire each address text information, perform word segmentation processing on each address text information, and filter out each interest information; Counting the frequency of each word contained in each of the interest information, and selecting characteristic words exceeding a preset frequency threshold from the words; The feature words are marked with address attributes and matching priorities, and feature words with the same matching priority are grouped together to obtain the feature word library.

4. The method according to claim 1, characterized in that The individual attribute tags include precise attribute tags and imprecise attribute tags. The precise attribute tags include, in order of priority: an attribute tag ending in residence, an attribute tag ending in company, a pure residential attribute tag, an attribute tag including residential attributes, and an attribute tag including company attributes; the imprecise attribute tags include, in order of priority: a comprehensive functional attribute tag, an attribute tag ending in similar residence, and an attribute tag ending in similar company. Determining the address attribute tag of the target address according to the individual attribute tag of the interest information and a preset address attribute identification strategy includes: In response to the interest information being a precise attribute tag, obtaining a first hit result of the individual attribute tag according to the priority order of the precise attribute tag, determining a corresponding trust level according to the priority corresponding to the first hit individual attribute tag, and obtaining an address attribute tag of the target address; In response to the interest information being an imprecise attribute tag, the first hit result of the individual attribute tag is obtained according to the priority order of the imprecise attribute tag, and the corresponding trust level is determined according to the priority corresponding to the first hit individual attribute tag to obtain the address attribute tag of the target address.

5. The method according to claim 4, characterized in that In response to the first hit individual attribute tag being the pure residential attribute tag, and the target address also having the company-containing attribute tag, the method further includes: The trust level in the address attribute tag of the target address is reduced, and the address attribute tag of the target address is generated according to the company attribute tag and the reduced trust level.

6. The method according to claim 4, characterized in that In response to the individual attribute tag of the first hit being the comprehensive functional attribute tag, and the target address also having the similar residential ending attribute tag, the method further includes: The trust level in the address attribute tag of the target address is reduced, and the address attribute tag of the target address is generated according to the similar residential ending attribute tag and the reduced trust level.

7. The method according to claim 1, characterized in that In response to the fact that the address attribute tag of the target address is not determined according to the individual attribute tag of the interest information and the preset address attribute identification strategy, the method further includes: Define the target address with house number and no information of interest as residential address attribute; Define the target address with the point of interest as the business address attribute.

8. The method according to claim 1, characterized in that Extracting interest information from the target address includes: Input the target address into a pre-built address model, and obtain a word segmentation result of the target address output by the address model; The segmented words marked as interest information are extracted from the segmented words results.

9. A device for identifying address attributes, characterized in that: include: An interest information extraction module, used to obtain a target address to be identified, and extract interest information from the target address, wherein the interest information includes interest points and / or interest surfaces; The address attribute identification module is used to find the individual attribute mark of the interest information, determine the address attribute mark of the target address according to the individual attribute mark of the interest information and a preset address attribute identification strategy, and then obtain the address attribute of the target address, wherein the address attribute mark of the target address has a trust level.

10. A mobile electronic device terminal, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 8.

11. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images