WO2021189977A1

WO2021189977A1 - Address coding method and apparatus, and computer device and computer-readable storage medium

Info

Publication number: WO2021189977A1
Application number: PCT/CN2020/136330
Authority: WO
Inventors: 李硕
Original assignee: 平安科技（深圳）有限公司
Priority date: 2020-08-31
Filing date: 2020-12-15
Publication date: 2021-09-30
Also published as: CN112069276A; CN112069276B

Abstract

An address coding method and apparatus, and a computer device and a computer-readable storage medium, which belong to the technical field of intelligent decision-making. The method comprises: when an address coding request is received, performing a word segmentation operation on address text to be coded that is carried by the address coding request, so as to obtain a territory word group sequence (S101); extracting a territory word of a lowest-level administrative region from the territory word group sequence, and matching the extracted territory word with a pre-stored territory code dictionary so as to determine a territory code corresponding to the extracted territory word (S102); determining, according to the determined territory code, a target trie tree corresponding to the address text to be coded (S103); determining, from the target trie tree, POI information corresponding to the address text to be coded, and taking same as a coding result of the address text to be coded (S104); and detecting a credibility type of the coding result by means of a trained credibility detection model (S105). By means of the method, a requirement for coding massive pieces of address text can be met, and the reliability of a coding result can be ensured.

Description

Address coding method, device, computer equipment and computer readable storage medium

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office with the application number CN 202010899558.3 and the name "Address coding method, device, computer equipment and computer-readable storage medium" on August 31, 2020, and its entire contents Incorporated in this application by reference.

Technical field

This application relates to the technical field of intelligent decision-making, and in particular to an address encoding method, device, computer equipment, and computer-readable storage medium.

Background technique

Address coding refers to finding the corresponding location (latitude and longitude) on the earth according to the address text. Address coding technology is used in many fields, such as logistics, map search and other fields. Taking the map search field as an example, the map application needs to encode the user's search address to display the search result on the map.

technical problem

The inventor realizes that as the number of users grows, the demand for address coding is increasing, reaching tens of millions or even hundreds of millions. The existing method of using the interface to call the server for coding, the coding speed is too slow, and it is difficult to meet the demand. , And it is difficult to guarantee reliability.

Technical solutions

This application provides an address encoding method, the method includes:

When an address encoding request is received, perform a word segmentation operation on the address text to be encoded carried in the address encoding request to obtain a geographical phrase sequence;

Extracting the regional words of the lowest-level administrative region from the regional phrase sequence, and matching the regional words of the lowest-level administrative region with a pre-stored regional code dictionary to determine the region code corresponding to the regional words of the lowest-level administrative region;

Determine the target trie tree corresponding to the address text to be encoded according to the determined region code;

Determining the POI information corresponding to the address text to be encoded from the target trie tree as the encoding result of the address text to be encoded;

Through the trained credibility detection model, the credibility type of the coding result is detected.

This application also provides an address encoding device, which includes:

The word segmentation module is used to perform word segmentation operations on the address text to be encoded carried in the address encoding request when an address encoding request is received, to obtain a regional phrase sequence;

The matching module is used to extract the regional words of the lowest-level administrative region from the regional phrase sequence, and match the regional words of the lowest-level administrative region with a pre-stored regional coding dictionary to determine the regional words of the lowest-level administrative region Corresponding geographic code;

The first determining module is configured to determine the target trie tree corresponding to the address text to be encoded according to the determined geographic code;

A second determining module, configured to determine the POI information corresponding to the address text to be encoded from the target trie tree, as the encoding result of the address text to be encoded;

The detection module is used to detect the credibility type of the coding result through the trained credibility detection model.

The present application also provides a computer device including a processor, a memory, and a computer program stored on the memory and executable by the processor, wherein the computer program is executed by the processor When, implement the following steps:

The present application also provides a computer-readable storage medium with a computer program stored on the computer-readable storage medium, and when the computer program is executed by a processor, the following steps are implemented:

Description of the drawings

In order to explain the technical solutions of the embodiments of the present application more clearly, the following will briefly introduce the drawings used in the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present application. Ordinary technicians can obtain other drawings based on these drawings without creative work.

FIG. 1 is a schematic flowchart of an address encoding method provided by an embodiment of this application;

2 is a schematic flowchart of another address encoding method provided by an embodiment of this application;

FIG. 3 is a schematic flowchart of yet another address encoding method provided by an embodiment of this application;

FIG. 4 is a schematic block diagram of an address encoding device provided by an embodiment of this application;

FIG. 5 is a schematic block diagram of the structure of a computer device related to an embodiment of the application.

The following specific embodiments will further illustrate this application in conjunction with the above-mentioned drawings.

Embodiments of the present invention

The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

The flowchart shown in the drawings is only an example, and does not necessarily include all contents and operations/steps, nor does it have to be executed in the described order. For example, some operations/steps can also be decomposed, combined or partially combined, so the actual execution order may be changed according to actual conditions.

It should be understood that the terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in the specification of this application and the appended claims, unless the context clearly indicates other circumstances, the singular forms "a", "an" and "the" are intended to include plural forms.

It should also be understood that the term "and/or" used in the specification and appended claims of this application refers to any combination of one or more of the items listed in the associated and all possible combinations, and includes these combinations .

The embodiments of the present application provide an address encoding method, device, equipment, and computer-readable storage medium. Among them, the address encoding method is mainly applied to an address encoding device, which is a distributed server cluster composed of multiple servers. Among them, the address encoding device is configured with the Spark framework.

Hereinafter, some embodiments of the present application will be described in detail with reference to the accompanying drawings. In the case of no conflict, the following embodiments and features in the embodiments can be combined with each other.

Please refer to FIG. 1. FIG. 1 is a schematic flowchart of an address encoding method according to an embodiment of the application.

As shown in FIG. 1, the address encoding method includes steps S101 to S105.

Step S101: When an address encoding request is received, a word segmentation operation is performed on the address text to be encoded carried in the address encoding request to obtain a geographical phrase sequence.

When the address encoding device receives an address encoding request, it extracts the address text to be encoded from the address encoding request, and then uses NLP (Natural Language Processing) technology to segment the address text to be encoded, that is, the address text to be encoded represents the administrative Regions, roads, and other characters are segmented. When segmenting administrative regions, the characters of the lowest administrative region contained in the address text to be encoded are used as the boundary to obtain a sequence of administrative regions, roads and/or other phrases. For example, "xx" "70m next to xx garden, xx street, xx street, xx district, city", the word segmentation result is "xx street, xx district, xx city, xx road/xx garden/70m next to xx garden".

Step S102, extracting the regional words of the lowest-level administrative region from the regional phrase sequence, and matching the regional words of the lowest-level administrative region with a pre-stored regional coding dictionary to determine the corresponding regional words of the lowest-level administrative region Geographical code.

Wherein, the address coding device stores a pre-built regional coding dictionary, and the regional coding dictionary contains the regional names of the national administrative divisions, the regional codes of the national administrative divisions, and the mapping relationship between the two.

After obtaining the regional phrase sequence, extract the regional words of the lowest administrative region from the phrase sequence, and then match the regional words of the lowest administrative region with the pre-built regional coding dictionary to determine the corresponding regional words of the lowest administrative region Geographical code.

In an embodiment, the matching the regional words of the lowest-level administrative region with a pre-stored regional code dictionary to determine the region codes corresponding to the regional words of the lowest-level administrative region is specifically: The regional words of the administrative region are compared with the pre-stored regional coding dictionary to determine the pre-stored regional names in the pre-stored regional coding dictionary that match the regional words of the lowest-level administrative region; based on the pre-stored regional coding dictionary, the pre-stored regional names and the pre-stored regions The mapping relationship of codes determines the pre-stored area code corresponding to the pre-stored area name matching the area word of the lowest-level administrative area; the determined pre-stored area code is used as the area code corresponding to the area word of the lowest-level administrative area.

That is, in the pre-built regional coding dictionary, find the pre-stored regional names matching the regional words of the lowest level administrative region, and then based on the pre-built regional coding dictionary, the mapping relationship between the pre-stored regional names and the pre-stored regional codes , Determine the pre-stored area code corresponding to the pre-stored area name matching the area word of the lowest-level administrative area, and the determined pre-stored area code is the area code corresponding to the area word of the lowest-level administrative area.

It is understandable that if the lowest-level administrative area in the address text to be coded is a street (town), then the geographic code corresponding to the regional word of the lowest-level administrative area is the street (town) code; if the lowest level in the address text to be coded If the administrative region is a district, the region code corresponding to the region word of the lowest administrative region is the district code; if the lowest administrative region in the address text to be coded is a city, the region code corresponding to the region word of the lowest administrative region is the city code .

Step S103: Determine the target trie tree corresponding to the address text to be encoded according to the determined geographic code;

Among them, the address coding device stores two pre-built Trie trees corresponding to each street (town) in the country. The two Trie trees use the geographic code of the corresponding street (town) as index information, and one of the Trie trees contains the corresponding All road information under the street (town) (defined as the first trie tree), the road information includes the road name, and the other Trie tree contains all POI (Point of Interest) information under the corresponding street (town) (definition It is the second trie tree), POI information includes POI name, address and latitude and longitude.

If the lowest level administrative area in the address text to be coded is a street (town), compare the street (town) code with the index information of the pre-stored Trie tree to find the corresponding street (town) in the address text to be coded Define the first trie tree and the second trie tree found corresponding to the first trie tree and the second trie tree as the target trie tree.

If the lowest-level administrative area in the address text to be coded is a district, then according to the district code, the codes of all streets (towns) under the district can be found from the pre-stored geographic code dictionary, and then each street (town) under the district can be further coded. ) The code is compared with the index information of the pre-stored Trie tree, and the target trie tree corresponding to each town in the district in the address text to be coded can be found.

If the lowest administrative area in the address text to be coded is a city, you can find the codes of all the districts under the city from the pre-stored area code dictionary according to the city code, and further find all the streets under each district under the city (Town) code, and then compare the code of each street (town) under each district with the index information of the pre-stored Trie tree, you can find each town in each district under the city in the address text to be coded The corresponding target trie tree.

Step S104: Determine the POI information corresponding to the address text to be encoded from the target trie tree, as the encoding result of the address text to be encoded.

If the lowest-level administrative area in the address text to be encoded is a street (town), match by road, that is, the road name in the address text to be encoded is the first trie in the target trie tree corresponding to the street (town) The tree is matched, and the maximum forward matching algorithm is used to calculate the text length that the road name in the address text to be encoded can match in the first trie tree. When the road name in the address text to be encoded can be matched in the first trie tree If the length of the text reaches the preset first threshold, it is considered that the two are consistent, and it is confirmed that the road name is matched in the first trie tree.

Further, according to the road name matched in the first trie tree, the second trie tree in the target trie tree is matched by number, that is, the number in the address text to be encoded is matched with the second trie tree Perform matching, so as to find from the second trie tree the number that matches the number in the address text to be encoded under the matched road name, so that the address text to be encoded can be determined from the second trie tree Corresponding POI information.

In addition, when matching by number, if the second trie tree does not have a number that is exactly the same as the number in the address text to be encoded, the number whose matching degree with the number in the address text to be encoded reaches the preset second threshold is selected. , As the number that matches the number in the address text to be encoded. Wherein, both the preset first threshold and the preset second threshold can be flexibly set according to actual needs, and are not limited here.

If the lowest-level administrative region in the address text to be encoded is a district or city, then the road name in the address text to be encoded is matched with the first trie tree corresponding to each town in the district, or the address text in the address to be encoded The road name is matched with the first trie tree corresponding to each town in the city.

In one embodiment, after determining the POI information corresponding to the address text to be encoded from the target trie tree, the method further includes: determining whether there are fuzzy words and/or numbers in the address text to be encoded; If there is a fuzzy word in the address text to be encoded, the fuzzy word is added after the determined POI information as the encoding result; if there is a number in the address text to be encoded, the number is normalized, and the After the determined POI information, a normalized number is added as the encoding result; if there are fuzzy words and numbers in the address text to be encoded, the numbers are normalized, and the determined POI After the information, fuzzy words and normalized numbers are sequentially added as the encoding result.

That is, after matching the POI information corresponding to the address text to be encoded, it is also determined whether there are fuzzy words and/or numbers in the address text to be encoded, such as side, opposite, southeast direction, and numbers such as 200m. If there are fuzzy words in the address text to be encoded, add fuzzy words after the matched POI information as the encoding result; if there are numbers in the address text to be encoded, normalize the numbers to 1-100 For example, the value of 200m is 100m, if it is 70m, the value is 70m, and the normalized number is added after the matched POI information as the encoding result. If there is a number after the fuzzy word, the number is normalized, and the normalized number is added to the fuzzy word as the encoding result.

Step S104: Detect the credibility type of the coding result through the trained credibility detection model.

After the encoding result is obtained, the credibility evaluation of the encoding result is also performed. Specifically, the encoding result is input to the pre-trained credibility detection model to obtain the credibility type of the encoding result output by the credibility detection model. Reliability types include complete accuracy, relatively accurate, basic preparation, and inaccuracy.

In an embodiment, as shown in FIG. 2, before step S101, step S106 to step S107 are included.

Step S106: Obtain national administrative division data, and construct a regional coding dictionary based on the national administrative division data.

That is, before step S101, a regional coding dictionary needs to be constructed in advance. Specifically, the national administrative division data is collected. The national administrative division data contains the 8-bit code of province-city-district-street (town) across the country. The 8-bit code, from left to right, the first two digits represent the province code, The four digits represent the city code, the first six digits represent the district code, and the last two digits represent the street (town) code, for example:

Extract the first two digits plus 6 0s from the 8-digit code of each province-city-district-street (town) to get the province code, associate the province code with the province name; then extract the first four digits plus 4 One 0, get the city code, associate the city code with the city name; then extract the first six digits and add 2 zeros to get the district code, and associate the district code with the district name; the 8-bit code is the street (town) code, and Street (town) name association; according to the associated province code and province name, city code and city name, district code and district name, street (town) code and street (town) name, a regional code dictionary can be obtained.

Step S107: Obtain national POI data, construct a Trie tree corresponding to each street or town according to the national POI data, and store the constructed Trie tree in a distributed manner.

That is, before step S101, the Trie tree corresponding to each street (town) needs to be constructed in advance. Specifically, collect national POI data from geographic information providers or city open data platforms. Each POI contains information such as category, name, address, latitude and longitude, and then construct the Trie tree corresponding to each street or town based on the national POI data, and then distribute it. Storing the constructed Trie tree can avoid the memory overflow problem of the Chinese Trie tree stored on a single server, and greatly speed up the calculation speed of text matching.

In an embodiment, the construction of the Trie tree corresponding to each street or town according to the national POI data is specifically: cleaning the national POI data; based on the pre-configured Hadoop framework, using the Hive of the Hadoop framework Component, store the cleaned national POI data in the Hive table in a preset format to obtain a standard POI library; based on the pre-configured Spark framework, create a component task for the standard POI library; execute the component task to obtain each street Or the Trie tree corresponding to the town.

Considering that the national POI data is not only for address coding, the collected national POI information point data may include redundant data. For this, first clean the national POI information point data to filter out unnecessary redundancy data. After that, the Hadoop framework based on the address encoding device (the address encoding device is equipped with the Hadoop framework), using the Hive component of Hadoop, in accordance with "province-city-district-street (town)-road-number- In the format of "POI name", the cleaned national POI data is stored in the Hive table to obtain the standard POI library.

Further, based on the Spark framework of the address encoding device, a component task is created for the standard POI library to construct a trie tree based on the standard POI library. Specifically, when creating a component task for the standard POI library, configure corresponding task parameters on the Spark component page, and the task parameters include execution time, Spark code, and so on. Among them, the Spark code defines the processing procedure when constructing the trie tree according to the standard POI library. The processing procedure includes:

a. Initialize the address and POI in the standard POI library;

b. Based on the street (town) code in the regional coding dictionary, split the standard POI library. Specifically, take the street (town) code as the split benchmark for each street (town) in the standard POI library and its subordinates The information of all roads constructs two trie trees. One trie tree includes all road information under the corresponding street (town) and is defined as the first trie tree, and the other Trie tree contains all the POI information under the corresponding street (town). (POI name and latitude and longitude), defined as the second trie tree.

By executing this component task, you can perform the above processing on the standard POI library in the Hive table to obtain two Trie trees corresponding to each street (town). The two Trie trees corresponding to each street (town) are stored in a distributed manner, and the street (town) code is used as the index information of the corresponding Trie tree.

As a result, by using a distributed server cluster, combined with a distributed stored Trie tree, when the address text is subsequently encoded, the encoding speed can be significantly improved.

In one embodiment, as shown in FIG. 3, before step S101, step S108 is included.

Step S108, training a credibility detection model to obtain a trained credibility detection model.

That is, before step S101, a credibility detection model for detecting the credibility of the encoding result needs to be pre-trained.

In one embodiment, the step S108 specifically includes: collecting address text with accurate latitude and longitude as training data; encoding the training data to obtain an encoding result, and extracting features of the training data during the encoding process; The SVM model is trained according to the coding result and the feature of the training data, and a trained SVM model is obtained as a trained credibility detection model.

The credibility detection model may be a support vector machine SVM model. Specifically, first collect the address text with accurate latitude and longitude as training data, first encode the training data, and extract the characteristics of the training data during the encoding process, such as whether there are administrative districts, corresponding levels of administrative districts, road matching ratios, and similar house numbers Rate, POI similarity rate, whether there is fuzzy value, fuzzy value distance ratio, and then train the SVM model according to the coding result of the training data and the characteristics of the training data.

The type of credibility that the SVM model needs to distinguish is divided into four situations: completely accurate, relatively accurate, basic preparation, and inaccurate. The task of the SVM model is to summarize the rules of the coding results of the training data. These four situations are divided into four categories , Gradually form an autonomous judgment logic curve, and divide the coding result of training data whose spherical distance between the actual latitude and longitude of the training data is less than 20m into completely accurate, and divide the spherical distance from the actual longitude and latitude of the training data within the range of 20-100m. The training data coding result is divided into more accurate ones. The training data coding result with the spherical distance between the actual latitude and longitude of the training data in the range of 100-1000m is divided into the basic preparation, and the spherical distance between the actual longitude and latitude of the training data is greater than 1000m. The coding result of the training data is divided into inaccurate, and the standard for judging the credibility type is obtained, and the trained SVM model is obtained, which is used as the trained credibility detection model to evaluate the credibility of the subsequent coding results. Lay the foundation.

In the address encoding method provided by the foregoing embodiment, when an address encoding request is received, the word segmentation operation is performed on the address text to be encoded carried in the address encoding request to obtain a geographical phrase sequence; and then the geographical word of the lowest administrative region is extracted from the geographical phrase sequence , Match the region word of the lowest level administrative region with the pre-stored region code dictionary to determine the region code corresponding to the region word of the lowest level administrative region; then determine the target trie tree corresponding to the address text to be encoded according to the determined region code; then , Determine the POI information corresponding to the address text to be encoded from the target trie tree, as the encoding result of the address text to be encoded; finally, detect the credibility type of the encoding result through the trained credibility detection model. Since the implementation of the above method is based on a distributed server cluster and combined with a trie tree, the encoding speed can be significantly improved when the address text is encoded, so that the encoding of a large amount of address text can be completed in a short time to meet address encoding As required, the trained credibility detection model is finally used to evaluate the credibility of the coding result, which can ensure the reliability of the subsequent use of the coding result.

Please refer to FIG. 4, which is a schematic block diagram of an address encoding device according to an embodiment of the application.

As shown in FIG. 4, the address encoding device 400 includes: a word segmentation module 401, a matching module 402, a first determination module 403, a second determination module 404, and a detection module 405.

The word segmentation module 401 is configured to, when an address encoding request is received, perform word segmentation operations on the address text to be encoded carried in the address encoding request to obtain a geographic phrase sequence;

The matching module 402 is configured to extract the regional words of the lowest-level administrative region from the regional phrase sequence, and match the regional words of the lowest-level administrative region with a pre-stored region coding dictionary to determine the region of the lowest-level administrative region The geographical code corresponding to the word;

The first determining module 403 is configured to determine the target trie tree corresponding to the address text to be encoded according to the determined geographic code;

The second determining module 404 is configured to determine the POI information corresponding to the address text to be encoded from the target trie tree, as the encoding result of the address text to be encoded;

The detection module 405 is configured to detect the reliability type of the encoding result through the trained reliability detection model.

It should be noted that those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working process of the above described device and each module and unit can refer to the corresponding process in the foregoing address encoding method embodiment. I won't repeat them here.

The apparatus provided in the foregoing embodiment may be implemented in the form of a computer program, and the computer program may run on the computer device shown in FIG. 5.

Please refer to FIG. 5, which is a schematic block diagram of the structure of a computer device according to an embodiment of the application. The computer device may be a personal computer (personal computer, PC), a server, and other devices with data processing functions.

As shown in FIG. 5, the computer device includes a processor, a memory, and a network interface connected through a system bus, where the memory may be volatile or non-volatile.

The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any address encoding method.

The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

The internal memory provides an environment for the running of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any address encoding method.

The network interface is used for network communication, such as sending assigned tasks. Those skilled in the art can understand that the structure shown in FIG. 5 is only a block diagram of part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may Including more or fewer parts than shown in the figure, or combining some parts, or having a different arrangement of parts.

It should be understood that the processor may be a central processing unit (Central Processing Unit, CPU), and the processor may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSP), and application specific integrated circuits (Application Specific Integrated Circuits). Circuit, ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor.

Wherein, in an embodiment, the processor is used to run a computer program stored in a memory to implement the following steps:

When an address encoding request is received, the word segmentation operation is performed on the address text to be encoded carried in the address encoding request to obtain a geographical phrase sequence; the geographical word of the lowest administrative region is extracted from the geographical phrase sequence, and the lowest level Matching the regional words of the administrative area with the pre-stored regional code dictionary to determine the regional code corresponding to the regional words of the lowest-level administrative area; determine the target trie tree corresponding to the address text to be coded according to the determined regional code; The POI information corresponding to the address text to be encoded is determined from the target trie tree as the encoding result of the address text to be encoded; the credibility type of the encoding result is detected through the trained credibility detection model .

In some embodiments, the processor implementing the matching of the regional words of the lowest-level administrative region with a pre-stored region code dictionary to determine the region codes corresponding to the regional words of the lowest-level administrative region includes:

Comparing the regional words of the lowest-level administrative region with a pre-stored regional coding dictionary to determine the pre-stored regional names in the pre-stored regional coding dictionary that match the regional words of the lowest-level administrative region;

Based on the mapping relationship between the pre-stored area name and the pre-stored area code in the pre-stored area code dictionary, determine the pre-stored area code corresponding to the pre-stored area name that matches the area word of the lowest-level administrative area;

The determined pre-stored area code is used as the area code corresponding to the area word of the lowest-level administrative region.

In some embodiments, after the processor realizes the determination of the POI information corresponding to the address text to be encoded from the target trie tree, the method further includes:

Determine whether there are fuzzy words and/or numbers in the address text to be encoded;

If there are fuzzy words in the address text to be encoded, add fuzzy words after the determined POI information as the encoding result;

If there are numbers in the address text to be encoded, normalize the numbers, and add the normalized numbers after the determined POI information as the encoding result;

If there are fuzzy words and numbers in the address text to be encoded, the numbers are normalized, and fuzzy words and normalized numbers are sequentially added after the determined POI information as the encoding result.

In some embodiments, before the processor realizes that when an address encoding request is received, the word segmentation operation is performed on the address text to be encoded carried in the address encoding request to obtain a sequence of regional phrases, including:

Acquiring national administrative division data, and constructing a regional coding dictionary based on the national administrative division data;

Obtain the national POI data, construct the Trie tree corresponding to each street or town according to the national POI data, and store the constructed Trie tree in a distributed manner.

In some embodiments, the processor implementing the construction of the Trie tree corresponding to each street or town according to the national POI data includes:

Clean the national POI data;

Based on the pre-configured Hadoop framework, using the Hive component of the Hadoop framework, the cleaned national POI data is stored in the Hive table in a preset format to obtain the standard POI library;

Create component tasks for the standard POI library based on the pre-configured Spark framework;

Perform the component task to obtain the Trie tree corresponding to each street or town.

Train the credibility detection model to obtain the trained credibility detection model.

In some embodiments, the processor implements the trained credibility detection model to obtain a trained credibility detection model, including:

Collect address text with accurate latitude and longitude as training data;

Encoding the training data to obtain an encoding result, and extracting features of the training data during the encoding process;

The SVM model is trained according to the coding result and the feature of the training data, and a trained SVM model is obtained as a trained credibility detection model.

The embodiments of the present application also provide a computer-readable storage medium, and the computer-readable storage medium may be volatile or non-volatile. A computer program is stored on the computer-readable storage medium, and the computer program includes program instructions. The method implemented when the program instructions are executed can refer to the various embodiments of the address encoding method of the present application.

The computer-readable storage medium may be the internal storage unit of the computer device described in the foregoing embodiment, for example, the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), and a secure digital (Secure Digital, SD) equipped on the computer device. ) Card, Flash Card, etc.

Further, the computer-readable storage medium may mainly include a storage program area and a storage data area, where the storage program area may store an operating system, an application program required by at least one function, etc.; the storage data area may store Data created by the use of nodes, etc.

The blockchain referred to in the present invention is a new application mode of computer technology 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 for verification. The validity of the information (anti-counterfeiting) and the 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.

It should be noted that in this article, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, It also includes other elements that are not explicitly listed, or elements inherent to the process, method, article, or system. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, article, or system that includes the element.

The serial numbers of the foregoing embodiments of the present application are for description only, and do not represent the superiority or inferiority of the embodiments. The above are only specific implementations of this application, but the protection scope of this application is not limited to this. Anyone familiar with the technical field can easily think of various equivalents within the technical scope disclosed in this application. Modifications or replacements, these modifications or replacements shall be covered within the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

Claims

An address encoding method, wherein the address encoding method includes the following steps:

When an address encoding request is received, perform a word segmentation operation on the address text to be encoded carried in the address encoding request to obtain a geographical phrase sequence;

Extracting the regional words of the lowest-level administrative region from the regional phrase sequence, and matching the regional words of the lowest-level administrative region with a pre-stored regional code dictionary to determine the region code corresponding to the regional words of the lowest-level administrative region;

Determine the target dictionary tree trie tree corresponding to the address text to be encoded according to the determined geographic code;

Determine the POI information of the information point corresponding to the address text to be encoded from the target trie tree, as the encoding result of the address text to be encoded;

Through the trained credibility detection model, the credibility type of the coding result is detected.
The address encoding method according to claim 1, wherein when an address encoding request is received, performing word segmentation operation on the address text to be encoded carried in the address encoding request to obtain a sequence of geographic phrases includes:

Acquiring national administrative division data, and constructing a regional coding dictionary based on the national administrative division data;

Obtain the national POI data, construct the Trie tree corresponding to each street or town according to the national POI data, and store the constructed Trie tree in a distributed manner.
The address encoding method according to claim 2, wherein the construction of the Trie tree corresponding to each street or town according to the national POI data comprises:

Clean the national POI data;

Based on the pre-configured distributed system Hadoop framework, using the Hive component of the data warehouse tool of the Hadoop framework, the cleaned national POI data is stored in the Hive table in a preset format to obtain the standard POI library;

Create component tasks for the standard POI library based on the pre-configured computing engine Spark framework;

Perform the component task to obtain the Trie tree corresponding to each street or town.
The address coding method according to claim 1, wherein the matching the regional words of the lowest-level administrative region with a pre-stored region code dictionary to determine the region codes corresponding to the regional words of the lowest-level administrative region comprises :

Comparing the regional words of the lowest-level administrative region with a pre-stored regional coding dictionary to determine the pre-stored regional names in the pre-stored regional coding dictionary that match the regional words of the lowest-level administrative region;

Based on the mapping relationship between the pre-stored area name and the pre-stored area code in the pre-stored area code dictionary, determine the pre-stored area code corresponding to the pre-stored area name that matches the area word of the lowest-level administrative area;

The determined pre-stored area code is used as the area code corresponding to the area word of the lowest-level administrative region.
The address encoding method according to claim 1, wherein after determining the POI information corresponding to the address text to be encoded from the target trie tree, the method further comprises:

Determine whether there are fuzzy words and/or numbers in the address text to be encoded;

If there are fuzzy words in the address text to be encoded, add fuzzy words after the determined POI information as the encoding result;

If there are numbers in the address text to be encoded, normalize the numbers, and add the normalized numbers after the determined POI information as the encoding result;

If there are fuzzy words and numbers in the address text to be encoded, the numbers are normalized, and fuzzy words and normalized numbers are sequentially added after the determined POI information as the encoding result.
The address encoding method according to claim 1, wherein when an address encoding request is received, performing word segmentation operation on the address text to be encoded carried in the address encoding request to obtain a sequence of geographic phrases includes:

Train the credibility detection model to obtain the trained credibility detection model.
The address encoding method according to claim 6, wherein said training a credibility detection model to obtain a trained credibility detection model comprises:

Collect address text with accurate latitude and longitude as training data;

Encoding the training data to obtain an encoding result, and extracting features of the training data during the encoding process;

A support vector machine SVM model is trained according to the encoding result of the training data and the feature, and a trained SVM model is obtained as a trained credibility detection model.
An address encoding device, wherein the address encoding device includes:

The word segmentation module is used to perform word segmentation operations on the address text to be encoded carried in the address encoding request when an address encoding request is received, to obtain a regional phrase sequence;

The matching module is used to extract the regional words of the lowest-level administrative region from the regional phrase sequence, and match the regional words of the lowest-level administrative region with a pre-stored regional coding dictionary to determine the regional words of the lowest-level administrative region Corresponding geographic code;

The first determining module is configured to determine the target trie tree corresponding to the address text to be encoded according to the determined geographic code;

A second determining module, configured to determine the POI information corresponding to the address text to be encoded from the target trie tree, as the encoding result of the address text to be encoded;

The detection module is used to detect the credibility type of the coding result through the trained credibility detection model.
A computer device, wherein the computer device includes a processor, a memory, and a computer program stored on the memory and executable by the processor, wherein when the computer program is executed by the processor, the The following steps:

When an address encoding request is received, perform a word segmentation operation on the address text to be encoded carried in the address encoding request to obtain a geographical phrase sequence;

Extracting the regional words of the lowest-level administrative region from the regional phrase sequence, and matching the regional words of the lowest-level administrative region with a pre-stored regional code dictionary to determine the region code corresponding to the regional words of the lowest-level administrative region;

Determine the target dictionary tree trie tree corresponding to the address text to be encoded according to the determined geographic code;

Determine the POI information of the information point corresponding to the address text to be encoded from the target trie tree, as the encoding result of the address text to be encoded;

Through the trained credibility detection model, the credibility type of the coding result is detected.
The computer device according to claim 9, when the computer program is executed by the processor, the following steps are further implemented:

Acquiring national administrative division data, and constructing a regional coding dictionary based on the national administrative division data;

Obtain the national POI data, construct the Trie tree corresponding to each street or town according to the national POI data, and store the constructed Trie tree in a distributed manner.
The computer device according to claim 10, wherein said constructing the Trie tree corresponding to each street or town according to the national POI data comprises:

Clean the national POI data;

Based on the pre-configured distributed system Hadoop framework, using the Hive component of the data warehouse tool of the Hadoop framework, the cleaned national POI data is stored in the Hive table in a preset format to obtain the standard POI library;

Create component tasks for the standard POI library based on the pre-configured computing engine Spark framework;

Perform the component task to obtain the Trie tree corresponding to each street or town.
The computer device according to claim 9, wherein the matching the regional words of the lowest-level administrative region with a pre-stored region code dictionary to determine the region codes corresponding to the regional words of the lowest-level administrative region comprises:

Comparing the regional words of the lowest-level administrative region with a pre-stored regional coding dictionary to determine the pre-stored regional names in the pre-stored regional coding dictionary that match the regional words of the lowest-level administrative region;

Based on the mapping relationship between the pre-stored area name and the pre-stored area code in the pre-stored area code dictionary, determine the pre-stored area code corresponding to the pre-stored area name that matches the area word of the lowest-level administrative area;

The determined pre-stored area code is used as the area code corresponding to the area word of the lowest-level administrative region.
The computer device according to claim 9, when the computer program is executed by the processor, the following steps are implemented:

Determine whether there are fuzzy words and/or numbers in the address text to be encoded;

If there are fuzzy words in the address text to be encoded, add fuzzy words after the determined POI information as the encoding result;

If there are numbers in the address text to be encoded, normalize the numbers, and add the normalized numbers after the determined POI information as the encoding result;

If there are fuzzy words and numbers in the address text to be encoded, the numbers are normalized, and fuzzy words and normalized numbers are sequentially added after the determined POI information as the encoding result.
The computer device according to claim 9, wherein when the computer program is executed by the processor, the following steps are implemented:

Train the credibility detection model to obtain the trained credibility detection model.
The computer device according to claim 14, wherein the training a credibility detection model to obtain a trained credibility detection model comprises:

Collect address text with accurate latitude and longitude as training data;

Encoding the training data to obtain an encoding result, and extracting features of the training data during the encoding process;

A support vector machine SVM model is trained according to the encoding result of the training data and the feature, and a trained SVM model is obtained as a trained credibility detection model.
A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the following steps are implemented:

When an address encoding request is received, perform a word segmentation operation on the address text to be encoded carried in the address encoding request to obtain a geographical phrase sequence;

Extracting the regional words of the lowest-level administrative region from the regional phrase sequence, and matching the regional words of the lowest-level administrative region with a pre-stored regional code dictionary to determine the region code corresponding to the regional words of the lowest-level administrative region;

Determine the target dictionary tree trie tree corresponding to the address text to be encoded according to the determined geographic code;

Determine the POI information of the information point corresponding to the address text to be encoded from the target trie tree, as the encoding result of the address text to be encoded;

Through the trained credibility detection model, the credibility type of the coding result is detected.
The computer device according to claim 16, when the computer program is executed by the processor, the following steps are further implemented:

Acquiring national administrative division data, and constructing a regional coding dictionary based on the national administrative division data;

Obtain the national POI data, construct the Trie tree corresponding to each street or town according to the national POI data, and store the constructed Trie tree in a distributed manner.
The computer device according to claim 17, wherein said constructing the Trie tree corresponding to each street or town according to the national POI data comprises:

Clean the national POI data;

Based on the pre-configured distributed system Hadoop framework, using the Hive component of the data warehouse tool of the Hadoop framework, the cleaned national POI data is stored in the Hive table in a preset format to obtain the standard POI library;

Create component tasks for the standard POI library based on the pre-configured computing engine Spark framework;

Perform the component task to obtain the Trie tree corresponding to each street or town.
The computer device according to claim 16, wherein the matching the regional words of the lowest-level administrative region with a pre-stored region code dictionary to determine the region codes corresponding to the regional words of the lowest-level administrative region comprises:

Comparing the regional words of the lowest-level administrative region with a pre-stored regional coding dictionary to determine the pre-stored regional names in the pre-stored regional coding dictionary that match the regional words of the lowest-level administrative region;

Based on the mapping relationship between the pre-stored area name and the pre-stored area code in the pre-stored area code dictionary, determine the pre-stored area code corresponding to the pre-stored area name that matches the area word of the lowest-level administrative area;

The determined pre-stored area code is used as the area code corresponding to the area word of the lowest-level administrative region.
The computer device according to claim 16, when the computer program is executed by the processor, the following steps are implemented:

Determine whether there are fuzzy words and/or numbers in the address text to be encoded;

If there are fuzzy words in the address text to be encoded, add fuzzy words after the determined POI information as the encoding result;

If there are numbers in the address text to be encoded, normalize the numbers, and add the normalized numbers after the determined POI information as the encoding result;

If there are fuzzy words and numbers in the address text to be encoded, the numbers are normalized, and fuzzy words and normalized numbers are sequentially added after the determined POI information as the encoding result.