WO2022100154A1 - 基于人工智能的地址标准化方法、装置、设备和存储介质 - Google Patents

基于人工智能的地址标准化方法、装置、设备和存储介质 Download PDF

Info

Publication number
WO2022100154A1
WO2022100154A1 PCT/CN2021/109461 CN2021109461W WO2022100154A1 WO 2022100154 A1 WO2022100154 A1 WO 2022100154A1 CN 2021109461 W CN2021109461 W CN 2021109461W WO 2022100154 A1 WO2022100154 A1 WO 2022100154A1
Authority
WO
WIPO (PCT)
Prior art keywords
level
address
field
processed
hierarchical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/109461
Other languages
English (en)
French (fr)
Inventor
李超
徐国强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
OneConnect Smart Technology Co Ltd
Original Assignee
OneConnect Smart Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by OneConnect Smart Technology Co Ltd filed Critical OneConnect Smart Technology Co Ltd
Publication of WO2022100154A1 publication Critical patent/WO2022100154A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking

Definitions

  • the present application relates to the field of big data technologies, and in particular, to an artificial intelligence-based address standardization method, apparatus, device, and storage medium.
  • an artificial intelligence-based address standardization method, apparatus, device, and storage medium are provided.
  • An artificial intelligence-based approach to address normalization including:
  • the address to be processed is segmented to obtain segmented fields
  • a standardized address is obtained according to the area level corresponding to the address to be processed.
  • An artificial intelligence-based address normalization device comprising:
  • the first obtaining module is used to obtain the hierarchical keyword list corresponding to the address to be processed
  • the segmentation module is used to segment the address to be processed based on the hierarchical field value in the hierarchical keyword list to obtain the segmented field;
  • the search module is used to search the area level corresponding to the segment field from the level keyword list, and use the segment field that fails to find the area level as the segment field to be processed;
  • the second acquisition module is used to acquire the address configuration library corresponding to the address to be processed, and obtain the area level corresponding to the segmentation field to be processed from the address configuration library;
  • the standardization module is used to obtain a standardized address according to the area level corresponding to the address to be processed.
  • a computer device comprising a memory and one or more processors, the memory having computer-readable instructions stored therein, the computer-readable instructions, when executed by the processor, cause the one or more processors to execute The following steps:
  • the address to be processed is segmented to obtain segmented fields
  • a standardized address is obtained according to the area level corresponding to the address to be processed.
  • One or more computer-readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the following steps:
  • the address to be processed is segmented to obtain segmented fields
  • a standardized address is obtained according to the area level corresponding to the address to be processed.
  • the above-mentioned artificial intelligence-based address standardization method, device, computer equipment and storage medium obtain a hierarchical keyword list corresponding to the address to be processed; perform segmentation processing on the address to be processed based on the hierarchical field value in the hierarchical keyword list to obtain segmented fields; Find the regional level corresponding to the segmented field in the hierarchical keyword list, and use the segmented field that cannot be found at the regional level as the segmented field to be processed; obtain the address configuration library corresponding to the address to be processed, from the address configuration library Obtain the area level corresponding to the segment field to be processed; obtain the standardized address according to the area level corresponding to the address to be processed.
  • the corresponding information can be directly obtained according to the pre-configured library, and then the segmentation processing of the addresses to be processed can be realized according to the obtained information, and then the standardized addresses can be obtained.
  • the standardization processing efficiency of the address is improved.
  • FIG. 1 is an application environment diagram of an artificial intelligence-based address normalization method according to one or more embodiments.
  • FIG. 2 is a schematic flowchart of an artificial intelligence-based address normalization method according to one or more embodiments.
  • FIG. 3 is a schematic flowchart of obtaining a segmented field by segmenting an address to be processed based on a level field value in a level keyword list according to one or more embodiments.
  • FIG. 4 is a structural block diagram of an artificial intelligence-based address normalization apparatus according to one or more embodiments.
  • FIG. 5 is a diagram of the internal structure of a computer device in accordance with one or more embodiments.
  • the artificial intelligence-based address standardization method provided in this application can be applied to the application environment shown in FIG. 1 .
  • the terminal 102 communicates with the server 104 through the network.
  • the server 104 obtains a hierarchical keyword list corresponding to the address to be processed; performs segmentation processing on the address to be processed based on the hierarchical field value in the hierarchical keyword list to obtain a segmentation field; searches the hierarchical keyword list for the regional level corresponding to the segmentation field , and take the segmented field that fails to find the regional level as the segmented field to be processed; obtain the address configuration library corresponding to the address to be processed, and obtain the regional level corresponding to the segmented field to be processed from the address configuration library;
  • the area level corresponding to the processing address gets a standardized address.
  • the terminal 102 can be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, and the server 104 can be implemented by an independent server or a server cluster composed of multiple servers.
  • an artificial intelligence-based address standardization method is provided, and the artificial intelligence-based address standardization method can be applied to a computer device, wherein the computer device can be a server or a terminal, The following steps can be included:
  • Step 202 Obtain a hierarchical keyword list corresponding to the address to be processed.
  • the address to be processed is the address that needs to be standardized, and the standardized process refers to the process of knowing the regional level included in the address to be processed.
  • the regional level is the administrative level corresponding to the administrative region, such as "province", "city” or “county”, etc., and the regional level may also include "region", "unit” and so on.
  • the hierarchical keyword list corresponds to the regional level. Specifically, a corresponding hierarchical keyword list may be constructed for each regional level in advance.
  • the computer device acquires the addresses to be processed, matches the addresses to be processed with the area levels in the hierarchical keyword list, and uses the hierarchical keyword list corresponding to the successfully matched area levels as corresponding to the addresses to be processed A list of hierarchical keywords.
  • the computer device may obtain a list of hierarchical keywords corresponding to each area level from the address to be processed. For example, when the address to be processed is "Donggang District, Rizhao City, Shandong province****", the area levels included in the address to be processed at this time are "province", "city” and “district", and the computer equipment can obtain A list of level keywords corresponding to each area level.
  • the level keyword list may include level information corresponding to a region level, and the level information may include but not limited to one or more of level field values, level keywords, and level location information corresponding to the region level. indivual.
  • the computer device obtains the region level corresponding to the address to be processed, and obtains the corresponding level keyword list according to the region level, so as to find the level field value, level keyword and level corresponding to the region level in the level keyword list location relationship, etc.
  • a specific address to be processed is: "FLT 11 22/F BLK 33 AAA HSE BBB EST CCC STREET SHAM SHUI PO KLN"
  • the region level in the address to be processed includes “region” and "unit”. Therefore, the computer device can obtain the hierarchical keyword lists corresponding to "region” and “unit” respectively. Specifically, it can be obtained from the "region” keyword list that the level field value of the region level "region” is "KLN", and the corresponding level keyword is "KLN”, and the corresponding level position information is "the location is at the same time before and after”.
  • the level field value of the area level "unit” obtained by the computer equipment from the "unit” keyword list is "FLT 11". Among them, the level keyword is "FLT”, the corresponding level position information is "position first”, and the same is true for others.
  • the previous keywords representing the regional level “unit” in the level keyword list are: “FLAT”, “FLT”, “RM”, “ROOM”, “SHOP”, “UNIT”, etc., No other keywords.
  • the following keywords that indicate the area level “street” are: “AVENUE”, “LANE”, “RD”, “ROAD”, “ST”, “STREET”, “TERRACE”, etc., and no other keywords.
  • the preceding keywords indicating the regional level “floor” are: “FLOOR”, the latter keywords are: “/F”, and the other keywords are not.
  • the keywords before and after the region level "region” include “KOWLOON”, “KLN”, “KOWLOON”, “NT”, etc., and no other keywords.
  • the computer device may construct a region-level keyword list corresponding to each region-level respectively based on each region-level, so that the obtained number of region-level keyword lists is consistent with the number of region-levels.
  • the computer device may also construct all the area levels in the same area level keyword list, which is not limited herein.
  • Step 204 segment the address to be processed based on the level field value in the level keyword list to obtain segment fields.
  • the segmentation field is a field in the address to be processed that needs to be standardized, that is, a field that needs to obtain regional level information.
  • the computer device may match the address field in the address to be processed with the value of the hierarchical field in the hierarchical keyword list, and perform segmentation processing on the address to be processed according to the matching result to obtain the segmented field.
  • the successfully matched address fields can be segmented from the addresses to be processed and used as segmented fields.
  • the number of segmentation fields is one or more, which is not limited here.
  • each segmentation field can correspond to one or more area levels. It should be noted that, when the address to be processed corresponds to multiple hierarchical keyword lists, it may also include more than one segmented field obtained by dividing the addresses to be processed according to the multiple hierarchical keyword lists respectively.
  • Step 206 Search for the region level corresponding to the segmented field from the level keyword list, and use the segmented field for which the region level cannot be found as the segmented field to be processed.
  • the level keyword list may include the correspondence between the area level and the level field value. Since the segmented sub-segment corresponds to the value of the level field, the area level corresponding to the segmented field can also be found in the level keyword list.
  • each segmented field obtained by the computer device has a corresponding level field value, and in this case, the area level corresponding to each segmented field can be found based on the level keyword list.
  • the area level corresponding to the segmented field cannot be found from the level keyword list.
  • Step 208 Acquire an address configuration library corresponding to the address to be processed, and acquire the area level corresponding to the segmented field to be processed from the address configuration library.
  • the address configuration library may include address identifiers corresponding to addresses to be processed, and further configuration information corresponding to the address identifiers may be acquired from the address configuration library according to the address identifiers.
  • the address configuration library is pre-built, and the address configuration library corresponding to different provinces can be constructed, and the corresponding province is associated with the address configuration library, so that the address configuration library associated with the corresponding province can be quickly obtained in the specific implementation. , and obtain the address information of the corresponding province according to the corresponding address configuration library.
  • the address information of multiple provinces can also be constructed into the same address configuration database, and the address configuration information corresponding to the province can be searched in the same address configuration database.
  • the province is the highest-level administrative unit, therefore, the address configuration database corresponding to each province can be constructed with the province as the smallest unit.
  • it may also include constructing an address configuration database corresponding to the Hong Kong administrative region, and specifically, the English address of Hong Kong is used as an example for description.
  • the field names may include field identifiers (IDs), address identifiers, area levels, level depths, and hierarchical relationships between different regions.
  • IDs field identifiers
  • the address identifier of Hong Kong may be simultaneously or separately the English name of the address "Hong Kong” and the Chinese name of the address "Hong Kong”. Since Hong Kong is the highest administrative region in the geographical administrative division, the level of the previous regional level of Hong Kong is recorded as "", that is, it is recorded as empty, the regional level of Hong Kong can be recorded as "region”, and the level of the depth of Hong Kong can be recorded as "region”.
  • the field identification (ID) of Hong Kong in the address allocation database can be automatically recorded as 1. It should be noted that, in other embodiments, the field value corresponding to the Hong Kong administrative region may also be marked in other forms.
  • the regional level represents the level of the administrative region.
  • the higher the level of the corresponding regional level (for example, the level of 1 is greater than the level of 2), it means that the corresponding administrative region has a larger geographic range, and the corresponding address range is also higher. bigger.
  • address information corresponding to different address identifiers is preconfigured in the address configuration library, and the address information may specifically include area levels, so that the corresponding area levels can be obtained from the address configuration library according to the address identifiers.
  • the address to be processed may include multiple administrative regions, such as an administrative region with a level depth of 1, and an administrative region with a level depth of 2, etc., and the region levels corresponding to different level depths are different. of.
  • the corresponding level depth of Hong Kong is 1, the corresponding regional level is "region”, the level corresponding to Hong Kong Island is 2, the corresponding regional level is “district”, and the corresponding level depth of Kowloon is 2 , the corresponding area level is "district”.
  • the computer device may search the area level corresponding to the segmentation field from the address configuration library. Since there is pre-configured area level information in the address configuration library, the area level corresponding to the segmentation field can be obtained according to the address configuration library. By combining the address configuration library and the hierarchical keyword list, the regional level corresponding to the segmented field is determined, which makes the acquisition of the regional level more accurate and fast.
  • Step 208 Obtain a standardized address according to the area level corresponding to the address to be processed.
  • the computer device searches the address configuration library for the region level corresponding to each segmented field according to the pre-built address configuration library and the hierarchical keyword list, thereby realizing the standardized processing of the to-be-processed fields in the addresses to be standardized. , to obtain the region level corresponding to the field to be processed, so as to obtain the region level corresponding to the field in the address to be processed.
  • the corresponding information can be directly obtained according to the pre-configured library, and then the segmentation processing of the address to be processed can be realized according to the obtained information, A normalized address is obtained to improve the efficiency of the normalized processing of the address.
  • acquiring a list of hierarchical keywords corresponding to addresses to be processed may include: acquiring historical address data, and marking the area levels in the historical address data to obtain a hierarchical labeling value; obtaining a hierarchical labeling value from the historical address data The level field value corresponding to the value, the level keyword corresponding to the level field value, and the level position information of the level keyword; according to the level label value, level field value, level keyword and level position information of the level keyword corresponding to each regional level , establish a hierarchical keyword list corresponding to each regional level, and the hierarchical keyword list is stored in the blockchain.
  • the computer device randomly selects about 200 pieces of address data, and marks "region”, “district”, “street”, “estate”, “building”, “block”, “floor”, “unit” the level field value of the eight regional levels, the level keyword corresponding to the level field value, and the level position information corresponding to the level keyword (the position is in the front, the position is in the back, the position is in the front and back at the same time, or none).
  • the computer device may establish a corresponding hierarchical keyword list according to each regional level, or may form a column list document from the hierarchical keyword lists of multiple regional levels, which is not limited herein.
  • the above-mentioned hierarchical keyword list may also be stored in a node of a blockchain.
  • the hierarchical keyword list is constructed by acquiring a large amount of address data, so that the data information contained in the hierarchical keyword list is more comprehensive, and the accuracy of subsequent determination of the regional level corresponding to the address to be processed according to the hierarchical keyword list is further improved. and efficiency.
  • FIG. 3 a schematic flowchart of segmenting an address to be processed based on a hierarchical field value in a hierarchical keyword list to obtain a segmented field is provided. Specifically, based on the hierarchical field value in the hierarchical keyword list, the address to be processed is segmented to obtain segmented fields, which may include:
  • Step 302 Match the address field in the address to be processed with the hierarchical field value in the hierarchical keyword list, extract the successfully matched hierarchical field value as the segmented field value, and extract the successfully matched address field as the segmented field .
  • the level keyword list includes level label values, level field values, level keywords, and level position information of the level keywords corresponding to the region level.
  • the address to be processed may include one or more address fields.
  • the computer device can respectively match the address field in the address to be processed with the hierarchical field value in the keyword list of each level, and extract the successfully matched level field value as the segmented field value, and the matching is successful.
  • the address field is extracted as the field to be segmented.
  • Step 304 Obtain the hierarchical keywords corresponding to the segmented field values and the hierarchical location information corresponding to the hierarchical keywords from the hierarchical keyword list.
  • the to-be-segmented field refers to a field with matching configuration information in the hierarchical keyword list, so the computer device can obtain corresponding configuration information from the corresponding hierarchical keyword list according to the to-be-segmented field.
  • the configuration information may include, but is not limited to, a hierarchical label value corresponding to a segmented field value, a hierarchical field value, and hierarchical location information of a hierarchical keyword, and the like.
  • Step 306 Acquire the level position replacer corresponding to the level position information, and perform replacement processing on the field to be segmented according to the level position replacer to obtain the address to be replaced.
  • the computer device may pre-configure corresponding hierarchical position replacers for different hierarchical position information. For example, to set the position replacer whose hierarchical position corresponds to the previous keyword to “
  • the position replacement character of the word setting is "after keyword
  • the computer device determines that the corresponding information is the preceding keyword according to the hierarchical keyword list, it executes to replace the "preceding keyword" with "
  • the computer device can also replace a plurality of consecutive "
  • the address to be replaced is segmented according to the position replacement character "
  • Step 308 Perform segmentation processing on the address to be replaced according to the position replacement character in the address to be replaced to obtain a plurality of segmentation fields.
  • the address to be processed is "FLT 11 22/F BLK 33 AAA HSE BBB EST CCC STREET SHAM SHUI PO KLN"
  • the field to be processed is replaced according to the position replacement character "
  • the address to be replaced is "FLT11 22/F
  • the address information in each quotation mark is used as a segmented field.
  • the segmentation processing of the addresses to be processed is implemented according to the hierarchical keyword list, and the processing process is simple and feasible, and the segmentation efficiency of the addresses to be processed can be improved. Moreover, the configuration information in the hierarchical keyword list can be maintained at any time, which also makes the segmentation processing based on the hierarchical keyword list more flexible and applicable to more scenarios.
  • the method may further include: searching for the number of corresponding level keywords in the segment field; When the one digit of the keyword is one, search the area level corresponding to the segmented field from the hierarchical keyword list; when the number of level keywords is more than one, continue to segment the field to be processed until the inclusion of A segmented field for hierarchical keywords.
  • further processing may be performed for a segmented field in which two levels of keywords or more appear in the segmented field.
  • the level field values corresponding to the area levels “unit” and “block” include two segmentation fields
  • punctuation marks such as ",” and “ ⁇ n” will be used between the two segmentation fields. symbols are separated.
  • the punctuation symbols such as “,” and “ ⁇ n” in the address to be processed can be replaced with "
  • segmented fields when there are still two-level keywords or more segmented fields, it is also possible to judge whether the segmented fields contain fields consisting of only two words such as "unit” and "block” , and replace it with a regular pattern according to the hierarchical position information of the corresponding hierarchical keyword (such as the case of front/back). For example, for the segmentation field "FLT 11 22/F”, the regular expression r"%s.+?”% (the previous keyword) can be used to extract “FLT 11" and replace it with "FLT 11
  • the processing method of the segmentation field is determined by detecting the number of hierarchical keywords included in the segmentation field, so that the segmentation fields including different number of hierarchical keywords can match the corresponding processing methods, and according to The corresponding processing method processes the segmented fields, which makes the processing of the segmented fields more flexible and fast.
  • the method may further include: acquiring adjacent segmentation fields adjacent to the segmentation field to be processed in the address to be processed ; Obtain the adjacent area level corresponding to the adjacent segmented field to determine the area level of the segmented field to be processed according to the adjacent area level; obtain the standardized address corresponding to the to-be-processed address according to the area level corresponding to each segmented field.
  • the matched segmented field can be recorded (each segmented field can be performed in the form of an address line text) display) at the regional level.
  • each segmented field can be performed in the form of an address line text) display
  • the regional level corresponding to the segmentation field (address line text) after the segmentation field “AAA HSE” can be obtained, and the regional level corresponding to the segmentation field “KLN” can be obtained as "region", the regional level corresponding to the segmentation field "SHAM SHUI PO" is "district”, etc.
  • the segment field whose region level cannot be determined can also be used as the segment field to be processed, and the segment to be processed in other ways.
  • the field is processed again to determine the area level corresponding to the segmented field to be processed.
  • it may include deriving the region level where the remaining to-be-processed segmentation fields are located.
  • the region level corresponding to the segment to be processed can also be determined according to the location information, for example, the segment to be processed is "A B C", and the region level of A is "unit” , the regional level of C is "block”, then it can be considered that the regional level of B is "floor”.
  • the remaining segmented fields to be processed are displayed in the form of address lines. If the segmented fields to be processed are more than 2 consecutive lines, it can also be determined by determining that the remaining address lines are in the total The upper half or the lower half of the address line is determined by the method of top-down/bottom-up reasoning until all address lines are determined at the region level to achieve standardized processing of the addresses to be processed. In other embodiments, because some addresses to be processed may not be standardized, or the administrative division may be updated, or the preconfigured database may contain incomplete information, there may be a case where only the last "KLN" is matched. Case.
  • the corresponding information can be directly obtained according to the pre-configured library, and then the segmentation processing of the address to be processed can be realized according to the obtained information, Further, a standardized address is obtained, which improves the efficiency of standardized processing of the address.
  • acquiring the address configuration library corresponding to the address to be processed may include: acquiring web page data corresponding to the address identifier, the address identifier being determined according to the administrative region corresponding to the maximum level of hierarchical depth; The sub-administrative area corresponding to the address identifier, the area level corresponding to the address identifier, and the sub-area level corresponding to each sub-administrative area; an address configuration library corresponding to the address identifier is constructed according to the address identifier, sub-administrative area, area level, and sub-area level.
  • the sub-administrative area is a subordinate area of the administrative area corresponding to the address identifier. For example, when the administrative region corresponding to the address identifier is Hong Kong, then the sub-administrative regions are Hong Kong Island, Kowloon and the New Territories.
  • Hong Kong includes three administrative regions, “Hong Kong Island”, “Kowloon Peninsula” and “New Territories”, and continue to visit the introduction of these three administrative regions respectively.
  • the address data acquisition requirements are met, stop the access, and continue to store the accessed data in the database to obtain the address configuration library.
  • the field identification is implemented in the address configuration library in a self-increasing manner. Specifically, the field identifier of Hong Kong Island is automatically increased to "2" in the address configuration database, the Chinese name of the address corresponding to the address identifier is recorded as "Hong Kong Island”, and the English name of the address corresponding to the address identifier is recorded as "Hong Kong Island”, The previous level depth is recorded as “1” (Hong Kong's level depth is "1”), the district level is recorded as “district”, and the level depth is recorded as "2".
  • the field identifier of "Kowloon Peninsula” is also in the way of ID self-increasing, and is recorded as "3"
  • the Chinese name of the address in the address identifier is recorded as "Kowloon Peninsula”
  • the English name of the address in the address identifier is recorded as "Kowloon”.
  • Peninsula and the previous level depth is recorded as "1”
  • the regional level is recorded as "district”
  • the level depth is recorded as "2”.
  • the regional level is like “province/municipality”, “prefecture-level city”, “county-level city/district”, and should be recorded in combination with the country's administrative divisions, here because Hong Kong There is no clear hierarchical division of addresses, so the actual implementation can be combined with the actual needs of address standardization, artificially divided into "region”, “district”, “street”, “estate”, “building”, “block”, “floor” , “unit” eight regional levels. In other embodiments, the determination of the area level may also be performed by setting other rules.
  • the address configuration database may include relevant information of one or more administrative regions, for example, the administrative region may specifically be one or more provinces, urban areas, and counties.
  • the relevant information may specifically be the regional level to which the corresponding administrative region belongs, the hierarchical depth, and the hierarchical relationship between different administrative regions.
  • the regional level can be divided into multiple levels such as “province”, “city”, “county”, “town”, and “village”.
  • Level size for example, you can set the level depth of the regional level corresponding to "province” to 1, set the level depth of the regional level corresponding to "city” to 2, and set the level of 1 to be greater than 2.
  • the hierarchical relationship is used to represent the hierarchical relationship between different administrative regions. For example, an administrative region with a hierarchical depth of 1 is the previous level of an administrative region with a hierarchical depth of 2.
  • Shandong province is an administrative region category "province”
  • Shandong province includes a plurality of urban areas
  • the administrative region category corresponding to each urban area is "regional” A county-level city”
  • a prefecture-level city also includes multiple counties
  • the county's corresponding administrative division category is "county”.
  • the level of "province” is higher than that of "prefecture-level city”
  • the level of prefecture-level city is higher than that of county, so the level depth of "province” can be set to 1, and the level of prefecture-level city can be set to 2.
  • the county has a hierarchy depth of 3.
  • the prefecture-level cities in Shandong province include Jinan City, Qingdao City, Heze City, etc.
  • the counties under Heze City include Cao County, Juye, and Dingtao, etc. Therefore, you can also set the regional level of Shandong province to " province”, Jinan, Qingdao, Heze and other regional levels are set to “city”, and regional levels such as Caoxian, Juye and Dingtao are set to “county”. Further, you can also set the hierarchical relationship between different regional levels.
  • the address configuration database corresponding to the Hong Kong administrative region when constructing the address configuration database corresponding to the Hong Kong administrative region, first take Hong Kong with the highest administrative level as an example, first store the first piece of data in the database, and each field is as follows: The growth method is determined, firstly determined as "1", and the Chinese name in the address identification is recorded as "Hong Kong", the English name of the address in the address identification is recorded as "Hong Kong”, the depth of the previous level is recorded as "", the regional level Denote it as "region”, and the level depth as "1". Then continue to obtain the prefecture-level area data corresponding to Hong Kong, and continue to configure the data in the database according to the relationship between different prefecture-level area data and between the prefecture-level area data and Hong Kong to obtain the final address configuration library.
  • the address configuration library can be automatically generated according to the preset rules in the specific implementation, which improves the generation efficiency of the address configuration library.
  • the method of crawling data also ensures that the data in the address configuration library is the latest data, and improves the accuracy of the address configuration library.
  • the above method may further include: acquiring an address configuration library, and extracting personalized configuration information from the address configuration library; and performing generalization processing on the personalized configuration information according to preset rules, so as to convert the personalized configuration information into the address configuration library.
  • the personalized configuration information is converted into general configuration information.
  • the administrative division database can also be optimized. Combined with the actual situation, some data can be modified manually. For example “Kowloon Peninsula”, in practical applications, “Kowloon” is more often used, and English is generally written as “Kowloon” instead of “Kowloon Peninsula”.
  • the above method may further include: acquiring area update information, and performing update processing on the address configuration library according to the area update. Specifically, it is possible to supplement unassigned area levels in the address configuration library, or perform error correction processing or completion processing on addresses in the address configuration library. For example, in real address writing, there may be situations where regional levels such as "district" are omitted or incorrectly written, resulting in the lack of regional level information even in standardized addresses.
  • the address configuration information is pre-stored in the address configuration library, so the address information that is not filled in the address to be processed can be acquired according to the address configuration library. For example, it is possible to supplement/correct the level field values of these area levels by querying the address configuration library.
  • all the superior administrative regions to which the current administrative region belongs can be obtained by continuously querying the ID of the upper level of the administrative region in the address configuration database.
  • the data of the address configuration database is independent, so that the query efficiency is very high in the specific query process like an administrative division tree. For example, as long as "building" appears in the address to be processed and can be queried in the address configuration library, then each level of administrative division (to "region") from "building" upwards is determined, then it can be carried out. Completion and error correction. For the case of the same name, it is only necessary to judge whether other fields of an original address exist in the current administrative division.
  • an address configuration library may be set first, in which the regional level, level depth and hierarchical relationship corresponding to the corresponding administrative regions and sub-administrative regions are pre-stored, and then the corresponding corresponding administrative regions and sub-administrative regions can be directly obtained through the address Regional level and other data to improve the efficiency of data acquisition.
  • the corresponding configuration information can be directly obtained according to the preconfigured hierarchical keyword list, so that the segmentation processing of the address to be processed can be achieved according to the acquired configuration information. field, and obtain the regional level corresponding to each segmented field, and then obtain a standardized address including regional level information, which improves the standardization processing efficiency of the address.
  • the address standardization can be implemented relatively accurately by adopting the combination of the address configuration library and the hierarchical keyword list, and the construction of the address configuration library is simple and fast. Moreover, regular expressions are rarely used, which is convenient for code modification and maintenance. Further, for addresses that are not particularly complete, standardization can also be achieved, which has a certain robustness, and also supports address completion and error correction.
  • AI artificial intelligence
  • digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.
  • the basic technologies of artificial intelligence generally include technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation/interaction systems, and mechatronics.
  • Artificial intelligence software technology mainly includes computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning/deep learning.
  • steps in the flowcharts of FIGS. 2-3 are shown in sequence according to the arrows, these steps are not necessarily executed in the sequence shown by the arrows. Unless explicitly stated herein, the execution of these steps is not strictly limited to the order, and the steps may be executed in other orders. Moreover, at least a part of the steps in FIGS. 2-3 may include multiple steps or multiple stages. These steps or stages are not necessarily executed and completed at the same time, but may be executed at different times. The execution of these steps or stages The order is also not necessarily sequential, but may be performed alternately or alternately with other steps or at least a portion of the steps or phases within the other steps.
  • an artificial intelligence-based address standardization device including:
  • the first obtaining module 402 is configured to obtain the hierarchical keyword list corresponding to the address to be processed.
  • the first segmenting module 404 is configured to segment the address to be processed based on the level field value in the level keyword list to obtain segment fields.
  • the search module 406 is configured to search the area level corresponding to the segment field from the level keyword list, and use the segment field that fails to find the area level as the segment field to be processed.
  • the second obtaining module 408 is configured to obtain the address configuration library corresponding to the address to be processed, and obtain the area level corresponding to the segmentation field to be processed from the address configuration library.
  • the standardization module 410 is configured to obtain a standardized address according to the area level corresponding to the address to be processed.
  • the first obtaining module 402 is further configured to obtain historical address data, mark the area level in the historical address data to obtain the level label value; obtain the level field value corresponding to the level label value from the historical address data , the level keyword corresponding to the level field value and the level position information of the level keyword; according to the level label value, level field value, level keyword and level position information of the level keyword corresponding to each regional level, establish each regional level The corresponding hierarchical keyword list.
  • the first segmenting module 404 is further configured to match the address field in the address to be processed with the level field value in the level keyword list, and extract the successfully matched level field value as the segment field value , extract the successfully matched address field as the field to be segmented; obtain the hierarchical keyword corresponding to the segmented field value from the hierarchical keyword list, and the hierarchical location information corresponding to the hierarchical keyword; obtain the hierarchical location corresponding to the hierarchical location information Replacement character, perform replacement processing on the field to be segmented according to the hierarchical position replacement character to obtain the address to be replaced; perform segmentation processing on the address to be replaced according to the position replacement character in the address to be replaced to obtain a plurality of segmented fields.
  • the apparatus further includes a second segmentation module, and the second segmentation module is configured to search for the number of corresponding hierarchical keywords in the segmentation field; Search the region level corresponding to the segmented field in the keyword list; when the number of level keywords is more than one, continue segmenting the field to be processed until a segmented field including one level keyword is obtained.
  • the apparatus further includes a third segmentation module, and the third segmentation module is configured to obtain the adjacent segmentation fields adjacent to the to-be-processed segmentation fields in the address to be processed; obtain the adjacent segmentation fields The corresponding adjacent area level is used to determine the area level of the segmented field to be processed according to the adjacent area level; the standardized address corresponding to the to-be-processed address is obtained according to the area level corresponding to each segmented field.
  • the second obtaining module 408 is further configured to obtain web page data corresponding to the address identifier, the address identifier is determined according to the administrative region corresponding to the maximum level of hierarchical depth; the sub-administrative region corresponding to the address identifier is searched from the web page data , and the area level corresponding to the address identifier, and the sub area level corresponding to each sub-administrative area; an address configuration library corresponding to the address identifier is constructed according to the address identifier, the sub-administrative area, the area level, and the sub-area level.
  • Each module in the above-mentioned artificial intelligence-based address standardization apparatus may be implemented in whole or in part by software, hardware, and combinations thereof.
  • the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
  • a computer device is provided, and the computer device may be a server or a terminal, and its internal structure diagram may be as shown in FIG. 5 .
  • the computer device includes a processor, memory, and a network interface connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities.
  • the memory of the computer device includes non-volatile or volatile storage media, internal memory.
  • the non-volatile or volatile storage medium stores an operating system, computer readable instructions and a database.
  • the internal memory provides an environment for the execution of the operating system and computer-readable instructions in the non-volatile storage medium.
  • the computer device's database is used to store address normalization data.
  • the network interface of the computer device is used to communicate with an external terminal through a network connection.
  • the computer-readable instructions when executed by a processor, implement an artificial intelligence-based address normalization method.
  • FIG. 5 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied. Include more or fewer components than shown in the figures, or combine certain components, or have a different arrangement of components.
  • a computer device comprising a memory and one or more processors, the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, causes one or more processors to perform the following steps: obtaining an address to be processed Corresponding hierarchical keyword list; segment the address to be processed based on the hierarchical field value in the hierarchical keyword list to obtain the segmented field; find the regional level corresponding to the segmented field from the hierarchical keyword list, and will fail to find The segmented field to the area level is used as the segmented field to be processed; the address configuration library corresponding to the address to be processed is obtained, the area level corresponding to the segmented field to be processed is obtained from the address configuration library; and the area corresponding to the address to be processed is obtained Hierarchies get standardized addresses.
  • the processor when the processor executes the computer-readable instructions, the processor is further configured to implement the following steps: acquiring historical address data, marking the area levels in the historical address data to obtain a level marking value; The level field value corresponding to the level label value, the level keyword corresponding to the level field value, and the level position information of the level keyword; and the level label value, level field value, level keyword and level keyword corresponding to each regional level Hierarchical location information, establish a hierarchical keyword list corresponding to each regional level.
  • the processor when the processor executes the computer-readable instructions, it is further configured to implement the following steps: match the address field in the address to be processed with the value of the hierarchy field in the hierarchy keyword list, and match the successfully matched hierarchy field The value is extracted as the segmented field value, and the successfully matched address field is extracted as the field to be segmented; the hierarchical keyword corresponding to the segmented field value and the hierarchical location information corresponding to the hierarchical keyword are obtained from the hierarchical keyword list; The hierarchical position replacer corresponding to the hierarchical position information, perform replacement processing on the field to be segmented according to the hierarchical position replacer to obtain the address to be replaced; and perform segmentation processing on the address to be replaced according to the position replacer in the address to be replaced, to obtain a plurality of segments field.
  • the processor when the processor executes the computer-readable instructions, the processor is further configured to implement the following steps: searching for the number of corresponding level keywords in the segmentation field; Find the region level corresponding to the segmented field in the keyword list; and when the number of level keywords is more than one, continue segmenting the field to be processed until a segmented field including one level keyword is obtained.
  • the processor when the processor executes the computer program, the processor is further configured to implement the following steps: obtaining the adjacent segmented fields adjacent to the segmented fields to be processed in the address to be processed; The adjacent area level is used to determine the area level of the segmented field to be processed according to the adjacent area level; and obtain the standardized address corresponding to the to-be-processed address according to the area level corresponding to each segmented field.
  • the processor when the processor executes the computer-readable instructions, the processor is further configured to implement the following steps: acquiring web page data corresponding to an address identifier, where the address identifier is determined according to the administrative area corresponding to the maximum level of hierarchy depth; from the web page data Find the sub-administrative area corresponding to the address identifier, the area level corresponding to the address identifier, and the sub-area level corresponding to each sub-administrative area; and construct the address configuration corresponding to the address identifier based on the address identifier, sub-administrative area, area level, and sub-area level library.
  • the processor when the processor executes the computer-readable instructions, it is further configured to implement the following steps: acquiring an address configuration library, extracting personalized configuration information from the address configuration library; and performing the personalized configuration information according to preset rules. Generalization processing to convert the personalized configuration information in the address configuration library into general configuration information.
  • One or more computer-readable storage media storing computer-readable instructions, when the computer-readable instructions are executed by one or more processors, cause the one or more processors to perform the following steps: obtaining the level key corresponding to the address to be processed word list; segment the address to be processed based on the hierarchical field value in the hierarchical keyword list to obtain the segmented field; find the regional level corresponding to the segmented field from the hierarchical keyword list, and will fail to find the regional level.
  • the segmentation field is used as the segmentation field to be processed; the address configuration library corresponding to the address to be processed is obtained, and the area level corresponding to the segmentation field to be processed is obtained from the address configuration library; and the area level corresponding to the address to be processed is standardized. address.
  • the computer-readable instructions are further used to implement the following steps when executed by the processor: obtaining historical address data, marking the area level in the historical address data to obtain a level marking value; The level field value corresponding to the level label value, the level keyword corresponding to the level field value, and the level position information of the level keyword; and the level label value, level field value, level keyword and level keyword corresponding to each regional level Hierarchical location information, establish a hierarchical keyword list corresponding to each regional level.
  • the computer-readable instructions are further configured to implement the following steps when executed by the processor: match the address field in the address to be processed with the value of the hierarchical field in the hierarchical keyword list, and match the successfully matched hierarchical field The value is extracted as the segmented field value, and the successfully matched address field is extracted as the field to be segmented; the hierarchical keyword corresponding to the segmented field value and the hierarchical location information corresponding to the hierarchical keyword are obtained from the hierarchical keyword list; The hierarchical position replacer corresponding to the hierarchical position information, perform replacement processing on the field to be segmented according to the hierarchical position replacer to obtain the address to be replaced; and perform segmentation processing on the address to be replaced according to the position replacer in the address to be replaced, to obtain a plurality of segments field.
  • the computer-readable instructions are further configured to implement the following steps when executed by the processor: search for the number of corresponding level keywords in the segmented field; Find the region level corresponding to the segmented field in the keyword list; and when the number of level keywords is more than one, continue segmenting the field to be processed until a segmented field including one level keyword is obtained.
  • the computer-readable instructions are further used to implement the following steps when executed by the processor: obtaining adjacent segmentation fields adjacent to the segmentation field to be processed in the address to be processed; obtaining adjacent segmentation fields The corresponding adjacent area levels are used to determine the area levels of the segmented fields to be processed according to the adjacent area levels; and the standardized addresses corresponding to the addresses to be processed are obtained according to the area levels corresponding to each segmented field.
  • the computer-readable instructions are further used to implement the following steps when executed by the processor: obtaining web page data corresponding to an address identifier, where the address identifier is determined according to the administrative region corresponding to the maximum level of hierarchical depth; from the web page data Find the sub-administrative area corresponding to the address identifier, the area level corresponding to the address identifier, and the sub-area level corresponding to each sub-administrative area; and construct the address configuration corresponding to the address identifier based on the address identifier, sub-administrative area, area level, and sub-area level library.
  • the computer-readable instructions are further used to implement the following steps when executed by the processor: acquiring an address configuration library, extracting personalized configuration information from the address configuration library; and performing the personalized configuration information according to preset rules.
  • Generalization processing to convert the personalized configuration information in the address configuration library into general configuration information.
  • the blockchain referred to in the present invention is a new application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm.
  • Blockchain essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information to verify its Validity of information (anti-counterfeiting) and generation of the next block.
  • the blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.
  • any reference to memory, storage, database or other media used in the various embodiments provided in this application may include at least one of non-volatile and volatile memory.
  • Non-volatile memory may include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, or optical memory, and the like.
  • Volatile memory may include random access memory (RAM) or external cache memory.
  • the RAM may be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Artificial Intelligence (AREA)
  • Remote Sensing (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种基于人工智能的地址标准化方法，涉及大数据技术领域，包括：获取待处理地址对应的层级关键词列表（S202）；基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段（S204）；从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段（S206）；获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级（S208）；根据待处理地址对应的区域层级得到标准化的地址（S210）。

Description

基于人工智能的地址标准化方法、装置、设备和存储介质
相关申请的交叉引用
本申请要求于2020年11月16日提交中国专利局，申请号为202011278707.0，申请名称为“地址标准化方法、装置、计算机设备和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及大数据技术领域，特别是涉及一种基于人工智能的地址标准化方法、装置、设备和存储介质。
背景技术
随着物流技术的发展，快递在我们生活中越来越普遍，在寄快递的过程中都需要填写地址。而在一些情况下，地址是由人工填写，存在地址填写不规范的问题。
在传统方式中，可以通过人工的方式对不规范的地址进行标准化处理。
但是，发明人意识到，通过人工进行地址标准化的处理，使得地址的标准化处理效率低下。
发明内容
根据本申请公开的各种实施例，提供一种基于人工智能的地址标准化方法、装置、设备和存储介质。
一种基于人工智能的地址标准化方法，包括：
获取待处理地址对应的层级关键词列表；
基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段；
从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段；
获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级；及
根据待处理地址对应的区域层级得到标准化的地址。
一种基于人工智能的地址标准化装置，包括：
第一获取模块，用于获取待处理地址对应的层级关键词列表；
切分模块，用于基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段；
查找模块，用于从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找 到区域层级的切分字段作为待处理切分字段；
第二获取模块，用于获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级；及
标准化模块，用于根据待处理地址对应的区域层级得到标准化的地址。
一种计算机设备，包括存储器和一个或多个处理器，所述存储器中储存有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述一个或多个处理器执行以下步骤：
获取待处理地址对应的层级关键词列表；
基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段；
从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段；
获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级；及
根据待处理地址对应的区域层级得到标准化的地址。
一个或多个存储有计算机可读指令的计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：
获取待处理地址对应的层级关键词列表；
基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段；
从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段；
获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级；及
根据待处理地址对应的区域层级得到标准化的地址。
上述基于人工智能的地址标准化方法、装置、计算机设备和存储介质，获取待处理地址对应的层级关键词列表；基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段；层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段；获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级；根据待处理地址对应的区域层级得到标准化的地址。通过预先配置地址配置库以及层级关键词列表，然后在具体实施中就可以直接根据预配置库获取对应的信息，进而根据获取到的信息实现对待处理地址的切分处理，进而得到标准化地址，提高了对地址的标准化处理效率。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例中所需要使用的附图 作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1为根据一个或多个实施例中基于人工智能的地址标准化方法的应用环境图。
图2为根据一个或多个实施例中基于人工智能的地址标准化方法的流程示意图。
图3为根据一个或多个实施例中提供了一种基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段的流程示意图。
图4为根据一个或多个实施例中基于人工智能的地址标准化装置的结构框图。
图5为根据一个或多个实施例中计算机设备的内部结构图。
具体实施方式
为了使本申请的技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本申请提供的基于人工智能的地址标准化方法，可以应用于如图1所示的应用环境中。其中，终端102通过网络与服务器104进行通信。服务器104获取待处理地址对应的层级关键词列表；基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段；从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段；获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级；根据待处理地址对应的区域层级得到标准化的地址。并将标准化地址推送至终端102。其中，终端102可以但不限于是各种个人计算机、笔记本电脑、智能手机、平板电脑和便携式可穿戴设备，服务器104可以用独立的服务器或者是多个服务器组成的服务器集群来实现。
在其中一个实施例中，如图2所示，提供了一种基于人工智能的地址标准化方法，该基于人工智能的地址标准化方法具体可以应用于计算机设备，其中计算机设备具体可以是服务器或者终端，可以包括以下步骤：
步骤202，获取待处理地址对应的层级关键词列表。
其中，待处理地址是需要进行标准化处理的地址，标准化处理是指获知待处理地址中包括的区域层级的处理过程，如将待处理地址中的从属于不同区域层级的地址字段进行切分处理得到切分单元，并得到每一个切分单元对应的区域层级的过程。其中，区域层级是行政区域对应的行政等级，如可以是“省”、“市”或者“县”等，区域层级还可以包括“region”、“unit”等。
其中，层级关键词列表与区域层级对应。具体地，可以预先为每一个区域层级构建一个对应的层级关键词列表。
在其中一个实施例中，计算机设备获取待处理地址，并将待处理地址与层级关键词列表中的区域层级进行匹配，并将匹配成功的区域层级对应的层级关键词列表作为与待处理地址对应的层级关键词列表。需要说明的是，当待处理地址中包括一个或者多个区域层级 时，此时计算机设备可以从待处理地址中获取与每一个区域层级分别对应的层级关键词列表。例如，当待处理地址为“山东省日照市东港区****”时，此时待处理地址中包括的区域层级为“省”、“市”以及“区”，并且，计算机设备可以获取与每一个区域层级分别对应的层级关键词列表。
在本实施例中，层级关键词列表中可以包括区域层级对应的层级信息，层级信息中可以包括但不限于该区域层级对应的层级字段值、层级关键词、以及层级位置信息中的一个或者多个。
在本实施例中，计算机设备获取待处理地址对应的区域层级，并根据区域层级获取对应的层级关键词列表，以在层级关键词列表中查找区域层级对应的层级字段值、层级关键词以及层级位置关系等。
例如，一个具体的待处理地址为：“FLT 11 22/F BLK 33 AAA HSE BBB EST CCC STREET SHAM SHUI PO KLN”，该待处理地址中的区域层级包括“region”、“unit”。故而计算机设备可以分别获取与“region”以及“unit”对应的层级关键词列表。具体可以在“region”关键词列表中获取到区域层级“region”的层级字段值为“KLN”，并且对应的层级关键词为“KLN”，对应的层级位置信息是“位置同时在前后”。计算机设备从“unit”关键词列表中获取到的区域层级“unit”的层级字段值为“FLT 11”。其中，层级关键词为“FLT”，对应的层级位置信息是“位置在前”，其他同理。
在其中一个实施例中，层级关键词列表中表示区域层级“unit”的在前关键词有：“FLAT”、“FLT”、“RM”、“ROOM”、“SHOP”、“UNIT”等，其他关键词无。表示区域层级“street”的在后关键词有：“AVENUE”、“LANE”、“RD”、“ROAD”、“ST”、“STREET”、“TERRACE”等，其他关键词无。表示区域层级“floor”的在前关键词有：“FLOOR”，在后关键词有：“/F”，其他关键词无。表示区域层级“region”的同时在前后的关键词有“KOWLOON”、“KLN”、“KOWLOON”、“NT”等，其他关键词无。需要说明的是，计算机设备可以基于每一个区域层级分别构建与每一个区域层级分别对应的区域层级关键词列表，这样，得到的区域层级关键词列表的数量与区域层级的数量是一致的。计算机设备还可以将所有的区域层级构建在同一个区域层级关键词列表中，在此不作限制。
步骤204，基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段。
其中，切分字段是待处理地址中的需要进行标准化处理的字段，即是需要获取区域层级信息的字段。
具体地，计算机设备可以将待处理地址中的地址字段与层级关键词列表中的层级字段值进行匹配，根据匹配结果对待处理地址进行切分处理，得到切分字段。例如，可以将匹配成功的地址字段从待处理地址中切分出来，并作为切分字段。
在本实施例中，切分字段的数量为一个或者多个，在此不做限制。一般来说，每一个切分字段都可以对应一个或者多个区域层级。需要说明的是，当待处理地址对应多个层级 关键词列表时，还可以包括根据多个层级关键词列表对待处理地址分别进行切分处理后，得到的多于一个的切分字段。
步骤206，从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段。
在本实施例中，层级关键词列表中可以包括区域层级与层级字段值的对应关系。由于切分子段是与层级字段值相对应的，故而也能在层级关键词列表中查找到与切分字段对应的区域层级。
在其中一个实施例中，计算机设备得到的每一个切分字段都有对应的层级字段值，此时可以基于层级关键词列表查找到与每一个切分字段对应的区域层级。
在其中一个实施例中，计算机设备得到的切分字段中存在未与层级字段值对应的字段，此时存在无法从层级关键词列表中查找到与该切分字段对应的区域层级。
步骤208，获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级。
在本实施例中，地址配置库中可以包括待处理地址对应的地址标识，进而可以根据地址标识从地址配置库中获取与地址标识对应的配置信息。
具体地，地址配置库是预先构建的，可以构建对应不同省份的地址配置库，并将对应的省份与地址配置库进行关联，以在具体实施中可以快速地获取与对应省份关联的地址配置库，并根据对应的地址配置库获取对应省份的地址信息。
可以理解的是，在其他实施例中，还可以将多个省份的地址信息构建到同一个地址配置库中，可以在同一个地址配置库中查找与对应省份的地址配置信息。
需要说明的是，由于在国内邮寄快递的地址中，是以省份为最高级别的行政单位，故而，可以以省份为最小单元构建每一个省份对应的地址配置库。
进一步，不论是网上购物、寄快递等，都需要填写地址。在一些情况下，地址由用户自己填写，免不了会有缺失和错误，即便完全填写正确，如何明确知道地址的各个字段也非易事。而除了中文地址，还有英文地址等其他语言的地址。对于“xxx省xx市”这样的中文地址，关键词都在后面，且有比较明确的行政区划，地址的标准化相对简单一些。而对于类似“FLT 11 22/F BLK 33 AAA HSE BBB EST CCC STREET SHAM SHUI PO KLN”这样的英文地址(香港)，则没有特别好的方法。
在其中一个实施例中，还可以包括构建香港行政区对应的地址配置库，具体以香港英文地址为例进行说明。
具体地，首先确定地址配置库对应的一个或者多个字段名，其中字段名可包括字段标识(ID)、地址标识、区域层级、层级深度以及不同地区之间的层级关系等。具体地，香港的地址标识可同时或者分别为地址英文名“Hong Kong”，以及地址中文名“香港”。由于香港在地理行政划分中是最高级的行政区域，故而，将香港的上一个区域层级的级别记作“”，即记作为空，香港的区域层级可以记作“region”，香港的层级深度记作“1”，并 且，由于香港为最高层级的行政区域，故而可以将香港在地址配置库中的字段标识(ID)自动记作为1。需要说明的是，在其他实施例中，还可以通过其他形式来标记香港行政区域对应的字段值。
在本实施例中，区域层级代表了行政区域的级别，对应的区域层级的等级越高(如1的等级大于2)，说明对应的行政区域对应的地理范更大，以及对应的地址范围也越大。
在本实施例中，地址配置库中预先配置了不同地址标识对应的地址信息，地址信息中具体可以包括区域层级，从而可以根据地址标识从地址配置库中获取与其对应的区域层级。需要说明的是，待处理地址中可包括多个的行政区域，如可包括层级深度为1的行政区域，还可以包括层级深度为2的行政区域等，并且不同层级深度对应的区域层级是不同的。例如，以香港为例，香港对应的层级深度为1，对应的区域层级为“region”，香港岛对应的层级深度为2，对应的区域层级为“district”，以及九龙对应的层级深度为2，对应的区域层级为“district”。
在本实施例中，计算机设备可以从地址配置库中查找切分字段对应的区域层级。由于地址配置库中存在预先配置的区域层级信息，故而可以根据地址配置库获取切分字段对应的区域层级。通过将地址配置库以及层级关键词列表进行结合，以确定切分字段对应的区域层级，使得区域层级的获取更加精准以及快速。
步骤208，根据待处理地址对应的区域层级得到标准化的地址。
具体地，计算机设备根据预先构建的地址配置库以及层级关键词列表，从地址配置库中查找每一个切分字段对应的区域层级，进而实现对待处理地址中的待标准化的待处理字段的标准化处理，得到待处理字段对应的区域层级，以实现获取待处理地址中的字段对应的区域层级。
上述实施例中，通过预先配置地址配置库以及层级关键词列表，然后在具体实施例中就可以直接根据预配置库获取对应的信息，进而根据获取到的信息实现对待处理地址的切分处理，得到标准化地址，以提高对地址的标准化处理的效率。
在其中一个实施例中，获取待处理地址对应的层级关键词列表，可以包括：获取历史地址数据，对历史地址数据中的区域层级进行标注得到层级标注值；从历史地址数据中获取与层级标注值对应的层级字段值、层级字段值对应的层级关键词以及层级关键词的层级位置信息；根据每一个区域层级对应的层级标注值、层级字段值、层级关键词以及层级关键词的层级位置信息，建立每一个区域层级对应的层级关键词列表，所述层级关键词列表存储于区块链中。
在本实施例中，考虑到地址信息的不规范，在不同的地址中，对同一个区域层级的表达方式存在区别，本申请方案还可以包括通过在计算机设备中获取大量的历史地址数据，通过对历史地址数据进行分析，得到不同区域层级对应的层级字段值、层级关键词以及层级关键词对应的层级位置信息，并根据层级字段值、层级关键词以及层级位置信息中的至少一个构建层级关键词列表。
在其中一个实施例中，计算机设备随机选择大约200份地址数据，并标注出地址里面“region”、“district”、“street”、“estate”、“building”、“block”、“floor”、“unit”八个区域层级的层级字段值、层级字段值对应的层级关键词以及层级关键词对应的层级位置信息(位置在前、位置在后、位置同时在前后或无)。在本实施例中，计算机设备可以根据每一个区域层级建立对应的层级关键词列表，也可以将多个区域层级的层级关键词列表形成一个列列表文档，在此不作限制。
需要强调的是，为进一步保证上述层级关键词列表的私密性和安全性，上述层级关键词列表还可以存储于一区块链的节点中。
上述实施例中，通过获取大量的地址数据构建层级关键词列表，使得层级关键词列表中包含的数据信息更加全面，进一步提高了后续根据层级关键词列表确定待处理地址对应的区域层级的准确性以及效率性。
在其中一个实施例中，如图3所示，提供了一种基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段的流程示意图。具体地，基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段，可以包括：
步骤302，将待处理地址中的地址字段与层级关键词列表中的层级字段值进行匹配，将匹配成功的层级字段值提取为切分字段值，将匹配成功的地址字段提取为待切分字段。
具体地，层级关键词列表中包括对应区域层级的层级标注值、层级字段值、层级关键词以及层级关键词的层级位置信息。待处理地址中可以包括一个或者多个的地址字段。
在本实施例中，计算机设备可以将待处理地址中的地址字段与各层级关键词列表中的层级字段值分别进行匹配，并将匹配成功的层级字段值提取为切分字段值，将匹配成功的地址字段提取为待切分字段。
步骤304，从层级关键词列表中获取与切分字段值对应的层级关键词，以及层级关键词对应的层级位置信息。
可以理解的是，待切分字段是指在层级关键词列表中具有相匹配的配置信息的字段，故而计算机设备可以根据待切分字段从对应的层级关键词列表中获取对应的配置信息。
在本实施例中，配置信息可以包括但不限于与切分字段值对应的层级标注值、层级字段值以及层级关键词的层级位置信息等。
步骤306，获取层级位置信息对应的层级位置替换符，根据层级位置替换符对待切分字段进行替换处理得到待替换地址。
具体地，计算机设备可以预先为不同的层级位置信息配置对应的层级位置替换符，如为将层级位置对应为在前关键词的位置替换符设置为“|在前关键词”，为在后关键词设置的位置替换符为“在后关键词|”等。当计算机设备根据层级关键词列表确定对应的信息为在前关键词时，执行将“在前关键词”替换为“|在前关键词”，当为在后关键词时，执行将“在后关键词”替换为“在后关键词|”，当为同时在前后关键词时，执行将“同时在前后关键词”替换为“|同时在前后关键词|”的步骤，进而得到待替换地址。
进一步地，在待处理地址中替换关键词并得到待替换地址之后，计算机设备还可以将连续的多个的“|”替换为单个的“|”，以及将首尾的“|”进行去除处理，并根据位置替换符“|”将待替换地址进行切分处理，即可得到待处理地址对应的切分字段。
步骤308，根据待替换地址中的位置替换符对待替换地址进行切分处理，得到多个切分字段。
例如，待处理地址为“FLT 11 22/F BLK 33 AAA HSE BBB EST CCC STREET SHAM SHUI PO KLN”，根据位置替换符“|”对待处理字段进行替换处理得到待替换地址为“FLT11 22/F|BLK 33 AAA HSE|BBB EST|CCC STREET|SHAM SHUI PO|KLN”，然后以“|”字符作为切分字符，对待处理地址进行切分处理，得到包含各切分字段的地址为[“FLT 11 22/F”，“BLK 33\n AAA HSE”，“BBB EST”，“CCC STREET”，“SHAM SHUI PO”，“KLN”]。其中，每一个引号中的地址信息作为一个切分字段。
上述实施例中，根据层级关键词列表实现对待处理地址的切分处理，处理过程简单易行，可以提高待处理地址的切分效率。并且，层级关键词列表中的配置信息是可以随时维护的，也使得根据层级关键词列表进行切分处理可以更加灵活，适用于更多的场景。
在其中一个实施例中，基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段之后，还可以包括：查找切分字段中对应的层级关键词的个数；当层级关键词的个位为一个时，从层级关键词列表中查找与切分字段对应的区域层级；当层级关键词的个数为多于一个时，继续对待处理字段进行切分处理，直至得到包括一个层级关键词的切分字段。
具体地，对于切分字段中出现两个层级关键词及以上的切分字段，可以做进一步的处理。
在其中一个实施例中，区域层级“unit”和“block”对应的层级字段值中若包括两个切分字段，这两个切分字段之间会通过“，”、“\n”等标点符号分开。利用这一特性，可以把待处理地址里面的“，”、“\n”等标点符号都替换为“|”，这时类似包括多个层级关键词的层级字段值“BLK 33\n AAA HSE”也被进一步进行切分处理了，得到“BLK 33”以及“AAA HSE”，进而使得一个切分字段中只包括一个层级关键词以及一个区域层级。
在其中一个实施例中，对于仍然存在两个层级关键词及以上的切分字段时，还可以通过判断切分字段中是否包含“unit”和“block”这种只由两个单词组成的字段，并根据对应层级关键词的层级位置信息(如在前/后的情况)通过正则进行替换。例如对于切分字段“FLT 11 22/F”，可以通过正则表达式r”％s.+？”％(在前关键词)，先提取“FLT 11”并替换其为“FLT 11|”，关键词在后的情况同理。这时就得到了只包含一个层级关键词的切分字段，如[“FLT 11”，“22/F”，“BLK 33”，“AAA HSE”，“BBB EST”，“CCC STREET”，“SHAM SHUI PO”，“KLN”]。
需要说明的是，上述实施例中以香港英文地址为例，但思路对于其他地址的标准化依然适用，在此不作限制。
上述实施例中，通过检测切分字段中包括的层级关键词个数确定对切分字段的处理方式，使得包括不同个数层级关键词的切分字段都能匹配到对应的处理方式，并根据对应的处理方式对切分字段进行处理，使得对切分字段处理的方式更加灵活、快速。
在其中一个实施例中，将未能查找到区域层级的切分字段作为待处理切分字段之后，还可以包括：在待处理地址中获取与待处理切分字段相邻的相邻切分字段；获取相邻切分字段对应的相邻区域层级，以根据相邻区域层级确定待处理切分字段的区域层级；根据各切分字段对应的区域层级得到待处理地址对应的标准化地址。
在本实施例中，当在对应的层级关键词列表中匹配到切分字段对应的区域层级时，可以记录匹配到的切分字段(可以将每一个切分字段以一个地址行文本的形式进行显示)所在区域层级。例如，对于包括多个切分字段的地址[“FLT 11”，“22/F”，“BLK 33”，“AAA HSE”，“BBB EST”，“CCC STREET”，“SHAM SHUI PO”，“KLN”]，根据对应的层级关键词列表可以获取到切分字段“AAA HSE”之后的切分字段(地址行文本)对应的区域层级，即可得切分字段“KLN”对应的区域层级为“region”，切分字段“SHAM SHUI PO”对应的区域层级为“district”等。
但是，当在对应的层级关键词列表中无法匹配到切分字段对应的区域层级时，还可以将未能确定区域层级的切分字段作为待处理切分字段，并按照其他方式对待处理切分字段进行再次处理，以确定待处理切分字段对应的区域层级。
在其中一个实施例中，可可以包括推导剩余的待处理切分字段所在的区域层级。具体的，当还存在待处理切分字段时，还可以根据位置信息确定待处理切分字段对应的区域层级，例如，待处理切分字段“A B C”，A的区域层级为“unit”，C的区域层级为“block”，那么可以认为B的区域层级为“floor”。
在其中一个实施例中，对于剩下的待处理切分字段以地址行的形式进行展示，若待处理切分字段是连续2行以上的情况，还可以通过确定剩下的地址行在总的地址行的上半部分还是下半部分，然后采取从上而下/从下而上推理的方法来确定，直到所有地址行都确定所在区域层级，实现对待处理地址的标准化处理。在其他实施例中，由于有的待处理地址不一定规范，或者行政区划可能会存在更新的情况，或者预配置的数据库存在收录信息不完全的情况，则可能存在只匹配到最后的“KLN”的情况。
上述实施例中，通过预先配置地址配置库以及层级关键词列表，然后在具体实施例中就可以直接根据预配置库获取对应的信息，进而根据获取到的信息实现对待处理地址的切分处理，进而得到标准化地址，提高了对地址的标准化处理效率。
在其中一个实施例中，获取与待处理地址对应的地址配置库，可以包括：获取地址标识对应的网页数据，地址标识是根据最大级别的层级深度对应的行政区域确定；从网页数据中查找与地址标识对应的子行政区域，以及地址标识对应的区域层级、各子行政区域对应的子区域层级；根据地址标识、子行政区域、区域层级以及子区域层级构建地址标识对应的地址配置库。
其中，子行政区域是地址标识对应的行政区域的下级区域。如当地址标识对应的行政区域为香港时，那么子行政区域为香港岛、九龙以及新界。
具体地，通过访问网页上的“香港”的公开介绍，可以得知香港包括三个行政区，“香港岛”、“九龙半岛”以及“新界”，并继续分别访问这三个行政区的介绍之后可继续得知这三个行政区的下级行政区，以及还可以继续访问下级行政区，直至达到地址数据获取需求为止，停止访问，并将访问到的数据继续存入数据库中，得到地址配置库。
在其中一个实施例中，以“香港岛”为例，字段标识(ID)在地址配置库中通过自增长的方式实现。具体地，在地址配置库中将香港岛的字段标识自增长为“2”，地址标识对应的地址中文名记作“香港岛”，地址标识对应的地址英文名记作“Hong Kong Island”，上一个层级深度记作“1”(香港的层级深度为“1”)，区域层级记作“district”，层级深度记作“2”。同理，“九龙半岛”的字段标识也以ID自增长的方式，并记作“3”，地址标识中的地址中文名记作“九龙半岛”，地址标识中的地址英文名记作“Kowloon Peninsula”，以及上一个层级深度记作“1”，区域层级记作“district”，层级深度记作“2”。并且，继续通过访问下一级行政区的公开介绍，并将访问到的数据按照以上存储的方式将数据存入数据库中，即可建立行政区划数据库即包括地址信息的地址配置库。
需要说明的是，以上的数据存储方式中，其中区域层级就像“省/直辖市”、“地级市”、“县级市/区”一样，理应结合国家的行政区划来记录，这里因为香港地址没有明确的层级划分，所以具体实施中可以结合地址标准化的实际需求，人为地分为“region”、“district”、“street”、“estate”、“building”、“block”、“floor”、“unit”八个区域层级。在其他实施例中，还可以通过设置其他规则来进行区域层级的确定。
在本实施例中，地址配置库中可以包括一个或者多个行政区域的相关信息，如行政区域具体可以是某一个或者多个省份、市区以及县等。相关信息具体可以是对应的行政区域所属的区域层级、层级深度以及不同行政区域之间的层级关系。
在本实施例中，区域层级可以分为“省”“市”“县”“镇”“村”等多个层级，层级深度可以是对区域层级的量化值，具体可用于表征对应区域层级的级别大小，如可以将对应为“省”的区域层级的层级深度设置为1，将对应为“市”的区域层级的层级深度设置为2，并且设置1的级别大于2。层级关系用于表征不同行政区域之间的级别关系，如层级深度为1的行政区域为层级深度为2的行政区域的上一个级别。
在其中一个实施例中，当地址配置库对应为山东省的配置库时，可知山东省为行政区类别“省”、山东省包括多个市区，并且每一个市区对应的行政区类别为“地级市”，以及地级市中还包括多个县，并且县对应的行政区类别为“县”。具体的，根据行政区划分规则可知，“省”的级别大于“地级市”，地级市的级别大于县，故而可以设定“省”的层级深度为1，地级市的层级深度为2，以及县的层级深度为3。以及，可知山东省的地级市包括济南市、青岛市、菏泽市等，菏泽市的下辖县包括曹县、巨野以及定陶等，故而，还可以设定山东省的区域层级设置为“省”、济南市、青岛市、菏泽市等区域层级设置为“市”、 曹县、巨野以及定陶等区域层级设置为“县”。进一步，还可以设置不同区域层级之间的层级关系，如设置区域层级为“省”的地区的上一个区域层级为“”(为空)，设置区域层级值为“市”的地区的上一个区域层级为1(为省的区域层级)，设置区域层级为“县”的地区的上一个区域层级为2(为地级市的区域层级)。
在其中一个实施例中，在构建香港行政区对应的地址配置库时，首先以最高行政级别的香港为例，在数据库中首先存入第一条数据，各字段如下：字段标识(ID)通过自增长的方式确定，首先确定为“1”，并且地址标识中的中文名记作“香港”，地址标识中的地址英文名记作“Hong Kong”，上一个层级深度记作“”，区域层级记作“region”，以及层级深度记作“1”。然后继续获取香港对应的地级区域数据，并根据不同地级区域数据之间以及地级区域数据与香港之间的关系，继续配置数据库中的数据，以得到最终的地址配置库。通过预先配置规则的方式，然后在具体实施中可以按照预设的规则自动生成地址配置库，提高了地址配置库的生成效率，以及由于网站上的数据都是实时更新的，故而通过自动从网站中爬取数据的方式也保证了地址配置库中的数据是最新的数据，以及提高了地址配置库的准确性。
在其中一个实施例中，上述方法还可以包括：获取地址配置库，从地址配置库中提取个性化配置信息；按照预设规则对个性化配置信息进行通用化处理，以将地址配置库中的个性化配置信息转换为通用配置信息。
具体地，还可以优化行政区划数据库。结合实际情况，可以通过人工对部分数据进行修改。例如“九龙半岛”，在实际应用中更多地使用“九龙”，英文一般也写成“Kowloon”而非“Kowloon Peninsula”。
在其中一个实施例中，上述方法还可以包括：获取区域更新信息，根据区域更新对地址配置库进行更新处理。具体可以实现在地址配置库中补充未赋值的区域层级，或者对地址配置库中的地址进行纠错处理或者补全处理等。例如，在真实的地址书写中，可能存在漏写或者错写“district”等区域层级的情况，导致即使是标准化后的地址中也会存在区域层级信息的缺失。
在其中一个实施例中，预先在地址配置库中存储了地址的配置信息，故而可以根据地址配置库获取待处理地址中未填写的地址信息。如可以通过查询地址配置库，以补充/更正这些区域层级的层级字段值。
具体地，对于任一个行政区域，可以通过在地址配置库中不断查询该行政区域的上一级的ID，进而可以获取当前行政区域所属的所有的上级的行政区域。并且，地址配置库的数据是分别独立的，使得在具体的查询过程中就像一个行政区划树一样，查询效率很高。如只要“building”在待处理地址中出现，且在地址配置库中可以查询到，那么从“building”往上的每一级行政区划(到“region”)都是确定的，那么就可以进行补全和纠错。而对于同名的情况，也只需要多判断一个原始地址的其他字段是否存在于当前行政区划即可。
在其中一个实施例中，可以首先设置地址配置库，地址配置库中预先存储了对应行政 区域以及子行政区域分别对应的区域层级、层级深度以及层级关系，进而可以通过地址配置库直接获取对应的区域层级等数据，提高数据获取的效率。
在本实施例中，通过预先配置层级关键词列表，使得可以直接根据预配置的层级关键词列表获取对应的配置信息，从而可以根据获取到的配置信息实现对待处理地址的切分处理得到切分字段，并得到每一个切分字段对应的区域层级，进而得到包含区域层级信息的标准化地址，提高了对地址的标准化处理效率。
上述实施例中，采用地址配置库和层级关键词列表相结合的方式，可以比较精确地实现地址标准化，同时地址配置库的构建简单、快捷。并且，极少采用正则表达式，便于代码修改和维护。进一步，对于填写不是特别完整的地址，也可以实现标准化，具有一定的鲁棒性，以及同时还支持地址补全和纠错。
具体地，本申请实施例可以基于人工智能技术对相关的数据进行获取和处理。其中，人工智能(Artificial Intelligence，AI)是利用数字计算机或者数字计算机控制的机器模拟、延伸和扩展人的智能，感知环境、获取知识并使用知识获得最佳结果的理论、方法、技术及应用系统。
人工智能基础技术一般包括如传感器、专用人工智能芯片、云计算、分布式存储、大数据处理技术、操作/交互系统、机电一体化等技术。人工智能软件技术主要包括计算机视觉技术、机器人技术、生物识别技术、语音处理技术、自然语言处理技术以及机器学习/深度学习等几大方向。
应该理解的是，虽然图2-3的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，图2-3中的至少一部分步骤可以包括多个步骤或者多个阶段，这些步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤中的步骤或者阶段的至少一部分轮流或者交替地执行。
在其中一个实施例中，如图4所示，提供了一种基于人工智能的地址标准化装置，包括：
第一获取模块402，用于获取待处理地址对应的层级关键词列表。
第一切分模块404，用于基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段。
查找模块406，用于从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段。
第二获取模块408，用于获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级。
标准化模块410，用于根据待处理地址对应的区域层级得到标准化的地址。
在其中一个实施例中，第一获取模块402还用于获取历史地址数据，对历史地址数据 中的区域层级进行标注得到层级标注值；从历史地址数据中获取与层级标注值对应的层级字段值、层级字段值对应的层级关键词以及层级关键词的层级位置信息；根据每一个区域层级对应的层级标注值、层级字段值、层级关键词以及层级关键词的层级位置信息，建立每一个区域层级对应的层级关键词列表。
在其中一个实施例中，第一切分模块404还用于将待处理地址中的地址字段与层级关键词列表中的层级字段值进行匹配，将匹配成功的层级字段值提取为切分字段值，将匹配成功的地址字段提取为待切分字段；从层级关键词列表中获取与切分字段值对应的层级关键词，以及层级关键词对应的层级位置信息；获取层级位置信息对应的层级位置替换符，根据层级位置替换符对待切分字段进行替换处理得到待替换地址；根据待替换地址中的位置替换符对待替换地址进行切分处理，得到多个切分字段。
在其中一个实施例中，装置还包括第二切分模块，第二切分模块用于查找切分字段中对应的层级关键词的个数；当层级关键词的个位为一个时，从层级关键词列表中查找与切分字段对应的区域层级；当层级关键词的个数为多于一个时，继续对待处理字段进行切分处理，直至得到包括一个层级关键词的切分字段。
在其中一个实施例中，装置还包括第三切分模块，第三切分模块用于在待处理地址中获取与待处理切分字段相邻的相邻切分字段；获取相邻切分字段对应的相邻区域层级，以根据相邻区域层级确定待处理切分字段的区域层级；根据各切分字段对应的区域层级得到待处理地址对应的标准化地址。
在其中一个实施例中第二获取模块408还用于获取地址标识对应的网页数据，地址标识是根据最大级别的层级深度对应的行政区域确定；从网页数据中查找与地址标识对应的子行政区域，以及地址标识对应的区域层级、各子行政区域对应的子区域层级；根据地址标识、子行政区域、区域层级以及子区域层级构建地址标识对应的地址配置库。
关于基于人工智能的地址标准化装置的具体限定可以参见上文中对于基于人工智能的地址标准化方法的限定，在此不再赘述。上述基于人工智能的地址标准化装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。
在其中一个实施例中，提供了一种计算机设备，该计算机设备可以是服务器或者终端，其内部结构图可以如图5所示。该计算机设备包括通过系统总线连接的处理器、存储器和网络接口。其中，该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性或易失性存储介质、内存储器。该非易失性或易失性存储介质存储有操作系统、计算机可读指令和数据库。该内存储器为非易失性存储介质中的操作系统和计算机可读指令的运行提供环境。该计算机设备的数据库用于存储地址标准化数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机可读指令被处理器执行时以实现一种基于人工智能的地址标准化方法。
本领域技术人员可以理解，图5中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定，具体地计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
一种计算机设备，包括存储器和一个或者多个处理器，存储器中存储有计算机可读指令，计算机可读指令被该处理器执行时，使得一个或者多个处理器执行以下步骤：获取待处理地址对应的层级关键词列表；基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段；从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段；获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级；及根据待处理地址对应的区域层级得到标准化的地址。
在其中一个实施例中，该处理器执行计算机可读指令时还用于实现以下步骤：获取历史地址数据，对历史地址数据中的区域层级进行标注得到层级标注值；从历史地址数据中获取与层级标注值对应的层级字段值、层级字段值对应的层级关键词以及层级关键词的层级位置信息；及根据每一个区域层级对应的层级标注值、层级字段值、层级关键词以及层级关键词的层级位置信息，建立每一个区域层级对应的层级关键词列表。
在其中一个实施例中，该处理器执行计算机可读指令时还用于实现以下步骤：将待处理地址中的地址字段与层级关键词列表中的层级字段值进行匹配，将匹配成功的层级字段值提取为切分字段值，将匹配成功的地址字段提取为待切分字段；从层级关键词列表中获取与切分字段值对应的层级关键词，以及层级关键词对应的层级位置信息；获取层级位置信息对应的层级位置替换符，根据层级位置替换符对待切分字段进行替换处理得到待替换地址；及根据待替换地址中的位置替换符对待替换地址进行切分处理，得到多个切分字段。
在其中一个实施例中，该处理器执行计算机可读指令时还用于实现以下步骤：查找切分字段中对应的层级关键词的个数；当层级关键词的个位为一个时，从层级关键词列表中查找与切分字段对应的区域层级；及当层级关键词的个数为多于一个时，继续对待处理字段进行切分处理，直至得到包括一个层级关键词的切分字段。
在其中一个实施例中，该处理器执行计算机程序时还用于实现以下步骤：在待处理地址中获取与待处理切分字段相邻的相邻切分字段；获取相邻切分字段对应的相邻区域层级，以根据相邻区域层级确定待处理切分字段的区域层级；及根据各切分字段对应的区域层级得到待处理地址对应的标准化地址。
在其中一个实施例中，该处理器执行计算机可读指令时还用于实现以下步骤：获取地址标识对应的网页数据，地址标识是根据最大级别的层级深度对应的行政区域确定；从网页数据中查找与地址标识对应的子行政区域，以及地址标识对应的区域层级、各子行政区域对应的子区域层级；及根据地址标识、子行政区域、区域层级以及子区域层级构建地址标识对应的地址配置库。
在其中一个实施例中，该处理器执行计算机可读指令时还用于实现以下步骤：获取地 址配置库，从地址配置库中提取个性化配置信息；及按照预设规则对个性化配置信息进行通用化处理，以将地址配置库中的个性化配置信息转换为通用配置信息。
一个或多个存储有计算机可读指令的计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器执行以下步骤：获取待处理地址对应的层级关键词列表；基于层级关键词列表中的层级字段值对待处理地址进行切分处理得到切分字段；从层级关键词列表中查找与切分字段对应的区域层级，并将未能查找到区域层级的切分字段作为待处理切分字段；获取与待处理地址对应的地址配置库，从地址配置库中获取与待处理切分字段对应的区域层级；及根据待处理地址对应的区域层级得到标准化的地址。
在其中一个实施例中，计算机可读指令被处理器执行时还用于实现以下步骤：获取历史地址数据，对历史地址数据中的区域层级进行标注得到层级标注值；从历史地址数据中获取与层级标注值对应的层级字段值、层级字段值对应的层级关键词以及层级关键词的层级位置信息；及根据每一个区域层级对应的层级标注值、层级字段值、层级关键词以及层级关键词的层级位置信息，建立每一个区域层级对应的层级关键词列表。
在其中一个实施例中，计算机可读指令被处理器执行时还用于实现以下步骤：将待处理地址中的地址字段与层级关键词列表中的层级字段值进行匹配，将匹配成功的层级字段值提取为切分字段值，将匹配成功的地址字段提取为待切分字段；从层级关键词列表中获取与切分字段值对应的层级关键词，以及层级关键词对应的层级位置信息；获取层级位置信息对应的层级位置替换符，根据层级位置替换符对待切分字段进行替换处理得到待替换地址；及根据待替换地址中的位置替换符对待替换地址进行切分处理，得到多个切分字段。
在其中一个实施例中，计算机可读指令被处理器执行时还用于实现以下步骤：查找切分字段中对应的层级关键词的个数；当层级关键词的个位为一个时，从层级关键词列表中查找与切分字段对应的区域层级；及当层级关键词的个数为多于一个时，继续对待处理字段进行切分处理，直至得到包括一个层级关键词的切分字段。
在其中一个实施例中，计算机可读指令被处理器执行时还用于实现以下步骤：在待处理地址中获取与待处理切分字段相邻的相邻切分字段；获取相邻切分字段对应的相邻区域层级，以根据相邻区域层级确定待处理切分字段的区域层级；及根据各切分字段对应的区域层级得到待处理地址对应的标准化地址。
在其中一个实施例中，计算机可读指令被处理器执行时还用于实现以下步骤：获取地址标识对应的网页数据，地址标识是根据最大级别的层级深度对应的行政区域确定；从网页数据中查找与地址标识对应的子行政区域，以及地址标识对应的区域层级、各子行政区域对应的子区域层级；及根据地址标识、子行政区域、区域层级以及子区域层级构建地址标识对应的地址配置库。
在其中一个实施例中，计算机可读指令被处理器执行时还用于实现以下步骤：获取地址配置库，从地址配置库中提取个性化配置信息；及按照预设规则对个性化配置信息进行 通用化处理，以将地址配置库中的个性化配置信息转换为通用配置信息。
本发明所指区块链是分布式数据存储、点对点传输、共识机制、加密算法等计算机技术的新型应用模式。区块链(Blockchain)，本质上是一个去中心化的数据库，是一串使用密码学方法相关联产生的数据块，每一个数据块中包含了一批次网络交易的信息，用于验证其信息的有效性(防伪)和生成下一个区块。区块链可以包括区块链底层平台、平台产品服务层以及应用服务层等。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储于一计算机可读取存储介质中，该计算机可读指令在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和易失性存储器中的至少一种。非易失性存储器可包括只读存储器(Read-Only Memory，ROM)、磁带、软盘、闪存或光存储器等。易失性存储器可包括随机存取存储器(Random Access Memory，RAM)或外部高速缓冲存储器。作为说明而非局限，RAM可以是多种形式，比如静态随机存取存储器(Static Random Access Memory，SRAM)或动态随机存取存储器(Dynamic Random Access Memory，DRAM)等。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种基于人工智能的地址标准化方法，包括：
    获取待处理地址对应的层级关键词列表；
    基于所述层级关键词列表中的层级字段值对所述待处理地址进行切分处理得到切分字段；
    从所述层级关键词列表中查找与所述切分字段对应的区域层级，并将未能查找到区域层级的所述切分字段作为待处理切分字段；
    获取与所述待处理地址对应的地址配置库，从所述地址配置库中获取与所述待处理切分字段对应的区域层级；及
    根据所述待处理地址对应的区域层级得到标准化的地址。
  2. 根据权利要求1所述的方法，其中，所述获取待处理地址对应的层级关键词列表，包括：
    获取历史地址数据，对所述历史地址数据中的区域层级进行标注得到层级标注值；
    从所述历史地址数据中获取与所述层级标注值对应的层级字段值、所述层级字段值对应的层级关键词以及所述层级关键词的层级位置信息；及
    根据每一个所述区域层级对应的所述层级标注值、所述层级字段值、所述层级关键词以及所述层级关键词的层级位置信息，建立每一个区域层级对应的层级关键词列表，所述层级关键词列表存储于区块链中。
  3. 根据权利要求2所述的方法，其中，所述基于所述层级关键词列表中的层级字段值对所述待处理地址进行切分处理得到切分字段，包括：
    将所述待处理地址中的地址字段与所述层级关键词列表中的所述层级字段值进行匹配，将匹配成功的所述层级字段值提取为切分字段值，将匹配成功的所述地址字段提取为待切分字段；
    从所述层级关键词列表中获取与所述切分字段值对应的层级关键词，以及所述层级关键词对应的层级位置信息；
    获取所述层级位置信息对应的层级位置替换符，根据所述层级位置替换符对所述待切分字段进行替换处理得到待替换地址；及
    根据所述待替换地址中的所述位置替换符对所述待替换地址进行切分处理，得到多个切分字段。
  4. 根据权利要求3所述的方法，其中，所述基于所述层级关键词列表中的层级字段值对所述待处理地址进行切分处理得到切分字段之后，所述方法还包括：
    查找所述切分字段中对应的层级关键词的个数；
    当所述层级关键词的个位为一个时，从所述层级关键词列表中查找与所述切分字段对应的区域层级；及
    当所述层级关键词的个数为多于一个时，继续对所述待处理地址进行切分处理，直至 得到包括一个层级关键词的切分字段。
  5. 根据权利要求1所述的方法，其中，所述并将未能查找到区域层级的所述切分字段作为待处理切分字段之后，所述方法还包括：
    在所述待处理地址中获取与所述待处理切分字段相邻的相邻切分字段；
    获取所述相邻切分字段对应的相邻区域层级，以根据所述相邻区域层级确定所述待处理切分字段的区域层级；及
    根据各所述切分字段对应的区域层级得到所述待处理地址对应的标准化地址。
  6. 根据权利要求1所述的方法，其中，所述获取与所述待处理地址对应的地址配置库，包括：
    获取地址标识对应的网页数据，所述地址标识是根据最大级别的层级深度对应的行政区域确定；
    从所述网页数据中查找与所述地址标识对应的子行政区域，以及所述地址标识对应的区域层级、各所述子行政区域对应的子区域层级；及
    根据所述地址标识、所述子行政区域、所述区域层级以及所述子区域层级构建所述地址标识对应的地址配置库。
  7. 根据权利要求1所述的方法，其中，还包括：
    获取地址配置库，从所述地址配置库中提取个性化配置信息；及
    按照预设规则对所述个性化配置信息进行通用化处理，以将所述地址配置库中的个性化配置信息转换为通用配置信息。
  8. 一种基于人工智能的地址标准化装置，包括：
    第一获取模块，用于获取待处理地址对应的层级关键词列表；
    切分模块，用于基于所述层级关键词列表中的层级字段值对所述待处理地址进行切分处理得到切分字段；
    查找模块，用于从所述层级关键词列表中查找与所述切分字段对应的区域层级，并将未能查找到区域层级的所述切分字段作为待处理切分字段；
    第二获取模块，用于获取与所述待处理地址对应的地址配置库，从所述地址配置库中获取与所述待处理切分字段对应的区域层级；及
    标准化模块，用于根据所述待处理地址对应的区域层级得到标准化的地址。
  9. 一种计算机设备，包括存储器和一个或者多个处理器，所述存储器存储有计算机可读指令，所述计算机可读指令被所述一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取待处理地址对应的层级关键词列表；
    基于所述层级关键词列表中的层级字段值对所述待处理地址进行切分处理得到切分字段；
    从所述层级关键词列表中查找与所述切分字段对应的区域层级，并将未能查找到区域 层级的所述切分字段作为待处理切分字段；
    获取与所述待处理地址对应的地址配置库，从所述地址配置库中获取与所述待处理切分字段对应的区域层级；及
    根据所述待处理地址对应的区域层级得到标准化的地址。
  10. 根据权利要求9所述的计算机设备，其中，所述处理器执行所述计算机可读指令时所实现的所述获取待处理地址对应的层级关键词列表，包括：
    获取历史地址数据，对所述历史地址数据中的区域层级进行标注得到层级标注值；
    从所述历史地址数据中获取与所述层级标注值对应的层级字段值、所述层级字段值对应的层级关键词以及所述层级关键词的层级位置信息；及
    根据每一个所述区域层级对应的所述层级标注值、所述层级字段值、所述层级关键词以及所述层级关键词的层级位置信息，建立每一个区域层级对应的层级关键词列表，所述层级关键词列表存储于区块链中。
  11. 根据权利要求10所述的计算机设备，其中，所述处理器执行所述计算机可读指令时所实现的所述基于所述层级关键词列表中的层级字段值对所述待处理地址进行切分处理得到切分字段，包括：
    将所述待处理地址中的地址字段与所述层级关键词列表中的所述层级字段值进行匹配，将匹配成功的所述层级字段值提取为切分字段值，将匹配成功的所述地址字段提取为待切分字段；
    从所述层级关键词列表中获取与所述切分字段值对应的层级关键词，以及所述层级关键词对应的层级位置信息；
    获取所述层级位置信息对应的层级位置替换符，根据所述层级位置替换符对所述待切分字段进行替换处理得到待替换地址；及
    根据所述待替换地址中的所述位置替换符对所述待替换地址进行切分处理，得到多个切分字段。
  12. 根据权利要求11所述的计算机设备，其中，所述处理器执行所述计算机可读指令时还实现以下步骤：
    查找所述切分字段中对应的层级关键词的个数；
    当所述层级关键词的个位为一个时，从所述层级关键词列表中查找与所述切分字段对应的区域层级；及
    当所述层级关键词的个数为多于一个时，继续对所述待处理地址进行切分处理，直至得到包括一个层级关键词的切分字段。
  13. 根据权利要求9所述的计算机设备，其中，所述处理器执行所述计算机可读指令时还实现以下步骤：
    在所述待处理地址中获取与所述待处理切分字段相邻的相邻切分字段；
    获取所述相邻切分字段对应的相邻区域层级，以根据所述相邻区域层级确定所述待处 理切分字段的区域层级；及
    根据各所述切分字段对应的区域层级得到所述待处理地址对应的标准化地址。
  14. 根据权利要求9所述的计算机设备，其中，所述处理器执行所述计算机可读指令时所实现的所述获取与所述待处理地址对应的地址配置库，包括：
    获取地址标识对应的网页数据，所述地址标识是根据最大级别的层级深度对应的行政区域确定；
    从所述网页数据中查找与所述地址标识对应的子行政区域，以及所述地址标识对应的区域层级、各所述子行政区域对应的子区域层级；及
    根据所述地址标识、所述子行政区域、所述区域层级以及所述子区域层级构建所述地址标识对应的地址配置库。
  15. 一个或多个存储有计算机可读指令的计算机可读存储介质，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取待处理地址对应的层级关键词列表；
    基于所述层级关键词列表中的层级字段值对所述待处理地址进行切分处理得到切分字段；
    从所述层级关键词列表中查找与所述切分字段对应的区域层级，并将未能查找到区域层级的所述切分字段作为待处理切分字段；
    获取与所述待处理地址对应的地址配置库，从所述地址配置库中获取与所述待处理切分字段对应的区域层级；及
    根据所述待处理地址对应的区域层级得到标准化的地址。
  16. 根据权利要求15所述的存储介质，其中，所述计算机可读指令被所述处理器执行时所实现的所述获取待处理地址对应的层级关键词列表，包括：
    获取历史地址数据，对所述历史地址数据中的区域层级进行标注得到层级标注值；
    从所述历史地址数据中获取与所述层级标注值对应的层级字段值、所述层级字段值对应的层级关键词以及所述层级关键词的层级位置信息；及
    根据每一个所述区域层级对应的所述层级标注值、所述层级字段值、所述层级关键词以及所述层级关键词的层级位置信息，建立每一个区域层级对应的层级关键词列表，所述层级关键词列表存储于区块链中。
  17. 根据权利要求16所述的存储介质，其中，所述计算机可读指令被所述处理器执行时所实现的所述基于所述层级关键词列表中的层级字段值对所述待处理地址进行切分处理得到切分字段，包括：
    将所述待处理地址中的地址字段与所述层级关键词列表中的所述层级字段值进行匹配，将匹配成功的所述层级字段值提取为切分字段值，将匹配成功的所述地址字段提取为待切分字段；
    从所述层级关键词列表中获取与所述切分字段值对应的层级关键词，以及所述层级关 键词对应的层级位置信息；
    获取所述层级位置信息对应的层级位置替换符，根据所述层级位置替换符对所述待切分字段进行替换处理得到待替换地址；及
    根据所述待替换地址中的所述位置替换符对所述待替换地址进行切分处理，得到多个切分字段。
  18. 根据权利要求17所述的存储介质，其中，所述计算机可读指令被所述处理器执行时还实现以下步骤：
    查找所述切分字段中对应的层级关键词的个数；
    当所述层级关键词的个位为一个时，从所述层级关键词列表中查找与所述切分字段对应的区域层级；及
    当所述层级关键词的个数为多于一个时，继续对所述待处理地址进行切分处理，直至得到包括一个层级关键词的切分字段。
  19. 根据权利要求15所述的存储介质，其中，所述计算机可读指令被所述处理器执行时还实现以下步骤：
    在所述待处理地址中获取与所述待处理切分字段相邻的相邻切分字段；
    获取所述相邻切分字段对应的相邻区域层级，以根据所述相邻区域层级确定所述待处理切分字段的区域层级；及
    根据各所述切分字段对应的区域层级得到所述待处理地址对应的标准化地址。
  20. 根据权利要求15所述的存储介质，其中，所述计算机可读指令被所述处理器执行时所实现的所述获取与所述待处理地址对应的地址配置库，包括：
    获取地址标识对应的网页数据，所述地址标识是根据最大级别的层级深度对应的行政区域确定；
    从所述网页数据中查找与所述地址标识对应的子行政区域，以及所述地址标识对应的区域层级、各所述子行政区域对应的子区域层级；及
    根据所述地址标识、所述子行政区域、所述区域层级以及所述子区域层级构建所述地址标识对应的地址配置库。
PCT/CN2021/109461 2020-11-16 2021-07-30 基于人工智能的地址标准化方法、装置、设备和存储介质 Ceased WO2022100154A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011278707.0A CN112364114A (zh) 2020-11-16 2020-11-16 地址标准化方法、装置、计算机设备和存储介质
CN202011278707.0 2020-11-16

Publications (1)

Publication Number Publication Date
WO2022100154A1 true WO2022100154A1 (zh) 2022-05-19

Family

ID=74514957

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/109461 Ceased WO2022100154A1 (zh) 2020-11-16 2021-07-30 基于人工智能的地址标准化方法、装置、设备和存储介质

Country Status (2)

Country Link
CN (1) CN112364114A (zh)
WO (1) WO2022100154A1 (zh)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115168546A (zh) * 2022-07-26 2022-10-11 思必驰科技股份有限公司 地址识别方法及电子设备和存储介质
CN116737705A (zh) * 2023-06-12 2023-09-12 深圳房讯通信息技术有限公司 一种押品信息标准化方法、装置、设备以及存储介质
CN117251517A (zh) * 2023-09-12 2023-12-19 河南省农业科学院农业经济与信息研究所 大数据视野下的年鉴行政区划信息匹配方法和模型
CN117992440A (zh) * 2024-02-06 2024-05-07 浪潮卓数大数据产业发展有限公司 一种地址标准化拆分方法及装置
CN118193664A (zh) * 2024-05-13 2024-06-14 云南省标准化研究院 一种统一社会信用代码行政区划数据补全方法与系统
CN120223612A (zh) * 2025-05-26 2025-06-27 中国民用航空局信息中心 一种民航物流数据交换方法

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112364114A (zh) * 2020-11-16 2021-02-12 深圳壹账通智能科技有限公司 地址标准化方法、装置、计算机设备和存储介质
CN113704574B (zh) * 2021-08-27 2024-02-09 北京市律典通科技有限公司 地址标准化的方法及装置
CN114003812A (zh) * 2021-10-29 2022-02-01 深圳壹账通智能科技有限公司 地址匹配方法、系统、设备及存储介质
CN114638217B (zh) * 2022-03-14 2025-10-17 支付宝(杭州)信息技术有限公司 地址文本处理方法及装置
CN115204167A (zh) * 2022-05-30 2022-10-18 连通(杭州)技术服务有限公司 一种基于地址信息确定行政区域的方法与设备
CN115062862B (zh) * 2022-07-08 2026-01-09 连通(杭州)技术服务有限公司 一种基于多个地址信息的行政区域预测方法与设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090182975A1 (en) * 2008-01-11 2009-07-16 International Business Machines Corporation Dynamic address translation with load page table entry address
CN106970903A (zh) * 2016-01-13 2017-07-21 阿里巴巴集团控股有限公司 物流系统中地址信息的处理方法及装置
CN109033086A (zh) * 2018-08-03 2018-12-18 银联数据服务有限公司 一种地址解析、匹配的方法及装置
CN109670122A (zh) * 2018-09-25 2019-04-23 平安科技（深圳）有限公司 地址信息回填方法、装置、设备及计算机可读存储介质
CN111914557A (zh) * 2020-07-31 2020-11-10 上海燕汐软件信息科技有限公司 地址解析方法、装置、设备及计算机可读存储介质
CN112364114A (zh) * 2020-11-16 2021-02-12 深圳壹账通智能科技有限公司 地址标准化方法、装置、计算机设备和存储介质

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106557896A (zh) * 2015-09-25 2017-04-05 阿里巴巴集团控股有限公司 网络数据处理方法、装置及系统
CN111724110A (zh) * 2020-06-16 2020-09-29 苏宁云计算有限公司 地址信息处理方法、装置、计算机设备和存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090182975A1 (en) * 2008-01-11 2009-07-16 International Business Machines Corporation Dynamic address translation with load page table entry address
CN106970903A (zh) * 2016-01-13 2017-07-21 阿里巴巴集团控股有限公司 物流系统中地址信息的处理方法及装置
CN109033086A (zh) * 2018-08-03 2018-12-18 银联数据服务有限公司 一种地址解析、匹配的方法及装置
CN109670122A (zh) * 2018-09-25 2019-04-23 平安科技（深圳）有限公司 地址信息回填方法、装置、设备及计算机可读存储介质
CN111914557A (zh) * 2020-07-31 2020-11-10 上海燕汐软件信息科技有限公司 地址解析方法、装置、设备及计算机可读存储介质
CN112364114A (zh) * 2020-11-16 2021-02-12 深圳壹账通智能科技有限公司 地址标准化方法、装置、计算机设备和存储介质

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115168546A (zh) * 2022-07-26 2022-10-11 思必驰科技股份有限公司 地址识别方法及电子设备和存储介质
CN116737705A (zh) * 2023-06-12 2023-09-12 深圳房讯通信息技术有限公司 一种押品信息标准化方法、装置、设备以及存储介质
CN117251517A (zh) * 2023-09-12 2023-12-19 河南省农业科学院农业经济与信息研究所 大数据视野下的年鉴行政区划信息匹配方法和模型
CN117251517B (zh) * 2023-09-12 2024-05-17 河南省农业科学院农业经济与信息研究所 大数据视野下的年鉴行政区划信息匹配方法和模型
CN117992440A (zh) * 2024-02-06 2024-05-07 浪潮卓数大数据产业发展有限公司 一种地址标准化拆分方法及装置
CN118193664A (zh) * 2024-05-13 2024-06-14 云南省标准化研究院 一种统一社会信用代码行政区划数据补全方法与系统
CN120223612A (zh) * 2025-05-26 2025-06-27 中国民用航空局信息中心 一种民航物流数据交换方法

Also Published As

Publication number Publication date
CN112364114A (zh) 2021-02-12

Similar Documents

Publication Publication Date Title
WO2022100154A1 (zh) 基于人工智能的地址标准化方法、装置、设备和存储介质
CN108959244B (zh) 地址分词的方法和装置
WO2021189977A1 (zh) 地址编码方法、装置、计算机设备及计算机可读存储介质
WO2018177316A1 (zh) 信息识别方法、计算设备及存储介质
WO2021147222A1 (zh) 基于ocr的表格版式恢复方法、装置、电子设备及存储介质
WO2021147252A1 (zh) 基于ocr的表格版式恢复方法、装置、电子设备及存储介质
WO2016165538A1 (zh) 一种地址数据的管理方法和装置
CN112528174A (zh) 基于知识图谱和多重匹配的地址修整补全方法及应用
CN113850081B (zh) 基于人工智能的文本处理方法、装置、设备及介质
CN112699775A (zh) 基于深度学习的证件识别方法、装置、设备及存储介质
CN111291099B (zh) 一种地址模糊匹配方法、系统及计算机设备
CN111814664A (zh) 识别图纸中标注的方法、装置、计算机设备及存储介质
CN113868351B (zh) 一种地址聚类方法、装置、电子设备及存储介质
CN116612276B (zh) 一种图像处理方法、装置、设备及存储介质
CN111444368B (zh) 构建用户画像的方法、装置、计算机设备及存储介质
CN108228657B (zh) 一种关键字检索的实现方法及装置
CN116431625A (zh) 一种地理实体的定位分析方法、装置及计算机设备
CN116414823A (zh) 一种基于分词模型的地址定位方法和装置
CN113360789B (zh) 兴趣点数据处理方法、装置、电子设备及存储介质
CN107748778A (zh) 一种提取地址的方法及装置
CN113553458A (zh) 图数据库中的数据导出方法及装置
CN110134920A (zh) 绘文字兼容显示方法、装置、终端及计算机可读存储介质
CN110688995B (zh) 地图查询的处理方法，计算机可读存储介质和移动终端
CN103559177A (zh) 一种地名识别方法及装置
CN114510935B (zh) 双地址文本识别方法、装置、计算机设备和存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21890685

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 23.08.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 21890685

Country of ref document: EP

Kind code of ref document: A1