CN115080847A - Address library supplementing method and system, electronic equipment and readable storage medium - Google Patents

Abstract

The embodiments of this specification provide a method, system, electronic device, and readable storage medium for supplementing an address database, wherein the method includes: acquiring address crowdsourcing collection data; acquiring address database data and map POI data; combining the address database data and the map POI data, obtain address data that does not exist in the address database in the address crowdsourcing collection data, to obtain supplementary address data; and output the supplementary address data to the address database. The above solution can greatly reduce the expansion cost of the address library, improve the update speed, and have a high accuracy rate.

Description

Address library supplement method, system, electronic device and readable storage medium

technical field

The embodiments of this specification relate to the technical field of data processing, and in particular, to a method, system, electronic device, and readable storage medium for supplementing an address library.

Background technique

The address library is a knowledge base that stores real address information. Various address services such as address search, address completion, and address error correction rely heavily on the address library. The data accuracy, completeness and coverage of the address database will greatly affect the performance of various address services.

The traditional address database expansion method based on manual supplementation requires a large number of acquisition equipment and manual field inspections to supplement the new addresses and verify the existing addresses, which requires a lot of manpower and time, and the update real-time performance is poor. The high cost further limits the expansion of the address library.

The contents in the Background section are merely technologies known to the disclosed person, and do not of course represent the prior art in the field.

SUMMARY OF THE INVENTION

In view of this, the embodiments of this specification provide an address library supplement method, system, electronic device and readable storage medium, which can greatly reduce the expansion cost of the address library, improve the update speed, and have a higher accuracy rate.

First, an embodiment of this specification provides a method for supplementing an address library, including:

Obtain address crowdsourcing to collect data;

Obtain address database data and map POI data;

combining the address database data and the map POI data, acquiring address data that does not exist in the address database in the address crowdsourcing collection data to obtain supplementary address data;

The supplemental address data is output to the address library.

Optionally, before acquiring the address data that does not exist in the address database in the address crowdsourcing collection data, the method further includes:

The address crowdsourcing collection data is preprocessed to obtain initial address data with unified format and complete information, which is used to obtain the supplementary address data.

Optionally, performing preprocessing on the address crowdsourcing collection data to obtain initial address data with complete unified format information, including:

Divide the address text in the address crowdsourcing collection data into blocks, identify the address attribute type, and supplement the administrative division information to obtain the initial address data, where the initial address data includes: coordinate data and corresponding address text data, the address text data includes: address attribute type and administrative division information.

Optionally, combining the address database data and the map POI data to obtain address data that does not exist in the address database in the address crowdsourcing collection data to obtain supplementary address data, including:

Coordinate matching is performed between the initial address data and the map POI data, and the initial address data corresponding to the POI blocks that match the coordinates and have a preset address attribute type are reserved as candidate address data;

Combined with the address database data and the map interest point data, duplicate data and noise data in the candidate address data are eliminated to obtain the supplementary address data.

Optionally, performing coordinate matching between the initial address data and the map POI data, and retaining the initial address data of the POI blocks that match the coordinates and have a preset address attribute type, as candidate address data, including:

According to the coordinate data in the initial address data, query the POI blocks whose coordinates in the map POI data match;

Coordinate matching is performed between the POI sub-blocks with the building attribute type contained in the POI block and the coordinate data, and the initial address data whose coordinates are matched is retained as the candidate address data.

Optionally, by combining the address database data and the map POI data, eliminating duplicate data and noise data in the candidate address data to obtain the supplementary address data, including at least one of the following:

Combining the address database data and the POI information in the map POI data, and matching the candidate address data, it is determined that it belongs to a duplicate address, or the address attribute type belongs to a newly added building, or belongs to a classification of a newly added mount address , then remove the initial address data with duplicate addresses;

According to the POI information in the map POI data, rough clustering is performed on the candidate address data, noise data in the candidate address data is removed, and the supplementary address data is obtained.

Optionally, before outputting the supplementary address data to the address library, the method further includes:

Perform spatial clustering on the supplementary address data to obtain cluster center coordinates and corresponding address description texts as supplementary address data output to the address library.

Optionally, performing spatial clustering on the supplementary address data to obtain cluster center coordinates and corresponding address description text, including:

Use the anti-noise clustering algorithm to cluster the data describing the same location to obtain the coordinates of the cluster center;

Clustering and supplementing the address text data describing the same location in the supplementary address data to obtain the corresponding address description text.

The embodiments of this specification also provide an address library supplement system, including:

a first data acquisition unit, adapted to acquire address crowdsourcing collection data;

a second data acquisition unit, adapted to acquire map point of interest data;

a third data acquisition unit, adapted to acquire address database data;

a supplementary data generating unit, adapted to acquire, in combination with the map POI data, address data that does not exist in the address database in the address crowdsourcing collection data, as supplementary address data;

An output unit, adapted to output the supplementary address data to the address library.

Optionally, the system further includes: a preprocessing unit, adapted to preprocess the address crowdsourcing collection data to obtain initial address data with complete unified format information, so as to obtain the supplementary address data.

Optionally, the system further includes: a spatial clustering unit, adapted to perform spatial clustering on the supplementary address data, to obtain the coordinates of the cluster center and the corresponding address description text, as supplementary addresses output to the address library data.

Embodiments of the present specification further provide an electronic device, including a memory and a processor, where the memory stores a computer program that can run on the processor, and the processor executes any of the foregoing when the computer program runs. The steps of the method described in the examples.

The embodiments of the present specification further provide a computer-readable storage medium on which a computer program is stored, and when the computer program runs, the steps of the method described in any of the foregoing embodiments are executed.

The address database supplement method of the embodiment of this specification is adopted, and the address data that does not exist in the address database in the address crowdsourcing collection data is obtained in combination with the map interest point data, as supplementary address data, wherein the map Point-of-interest data, which can cross-validate the address crowdsourcing collection data. Since the data volume of the address crowdsourcing collection data is huge, and the map POI data is accurate, the address database can be supplemented efficiently and accurately, so it can be extremely It greatly reduces the expansion cost of the address library, improves the update speed of the address library, and has a higher accuracy rate.

Further, by preprocessing the address crowdsourcing collection data to obtain initial address data with complete unified format information, which is used to obtain the supplementary address data, the accuracy and information of the supplementary address database data can be further improved. completeness.

Further, the initial address data is obtained by dividing the address text in the address crowdsourcing collection data into blocks, identifying the address attribute type, and supplementing the administrative division information, so that the address text in the initial address data is obtained. The data format is unified and complete, including information such as address attribute types and administrative division information, which facilitates matching with the map interest points, improves the accuracy and completeness of the obtained supplementary address data, and further improves the supplementary address data. The accuracy and completeness of the address database data.

Further, by performing coordinate matching between the initial address data and the map POI data, the initial address data corresponding to the POI blocks whose coordinates are matched and have a preset address attribute type are reserved as candidate address data, and then combined with all the original address data. The address database data and the map POI data, the duplicate data and noise data in the candidate address data are eliminated, and the supplementary address data is obtained, which can improve the accuracy of the address database data and save the address database data. storage.

Further, according to the coordinate data in the initial address data, query the POI blocks whose coordinates are matched in the POI data on the map, and then compare the POI sub-blocks with building attribute types contained in the POI blocks with Coordinate matching is performed on the coordinate data, and the initial address data of the coordinate matching is retained as the candidate address data, which can further improve the accuracy of the supplementary address data and save the storage space of the address database data.

Further, by combining the address database data and the POI information in the map POI data, and matching the candidate address data, it is determined that it belongs to a duplicate address, or the address attribute type belongs to a newly added building, or belongs to a newly added hanging. According to the classification of the carrier address, the initial address data with repeated addresses is eliminated; and, according to the point of interest information in the map interest point, the candidate address data is roughly clustered, and the noise data in the candidate address data is removed, The supplementary address data is obtained. Using the above method to analyze, sort and cluster data can effectively identify and eliminate duplicate address data and noise data, thereby improving the accuracy and usability of the supplementary address data and saving the storage space of address database data.

Further, by spatially clustering the supplementary address data, the coordinates of the cluster center and the corresponding address description text are obtained as supplementary address data of the address database, which can further improve the accuracy of the supplementary address data of the address database. Spend.

Description of drawings

In order to explain the technical solutions of the embodiments of the present specification more clearly, the following briefly introduces the drawings that are required to be used in the embodiments of the present specification or the description of the prior art. Obviously, the drawings described below are only for the purposes of the present specification. In some embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

FIG. 1 shows a flowchart of a method for supplementing an address library in an embodiment of the present specification.

FIG. 2 shows a flowchart of another method for supplementing an address library in an embodiment of the present specification.

FIG. 3 is a schematic diagram showing a specific manner of obtaining candidate address data through cross-matching of map POI data in a specific scenario in the embodiment of the present specification.

FIG. 4 shows a schematic structural diagram of an address library supplementing system in an embodiment of the present specification.

FIG. 5 shows a schematic structural diagram of an electronic device in an embodiment of the present specification.

Detailed ways

As described in the background art, the traditional address database expansion method based on manual supplementation requires a large number of collection equipment and manual field inspections to supplement the newly added addresses and verify the existing addresses, which requires a lot of manpower and time. , the update has poor real-time performance and high cost, which further limits the expansion of the address library.

In response to the above problems, the embodiments of this specification provide a corresponding supplementary solution to the address database. Specifically, based on the address crowdsourcing collection data with a huge amount of data, combined with the map point of interest data, the address crowdsourcing collection data that does not exist in the collection data is obtained. The address data in the address library is used as supplementary address data. Wherein, through the map POI data, cross-validation can be performed on the address crowdsourcing collection data. Since the data volume of the address crowdsourcing collection data is huge, and the map POI data is accurate, it is possible to efficiently and accurately check the address database. Therefore, the expansion cost of the address library can be greatly reduced, the update speed of the address library can be improved, and the accuracy rate can be higher.

In order for those skilled in the art to better understand the technical concepts, working principles and advantages of the embodiments of this specification, and to implement the technical solutions of the embodiments of this specification, hereinafter, referring to the accompanying drawings, and in combination with specific application scenarios, the embodiments of this specification are described through some specific examples. The optional examples are described in detail.

Referring to the flowchart of the address database supplement method shown in FIG. 1, in some embodiments of this specification, the data supplement of the address database can be implemented by the following steps:

S11 , obtaining address crowdsourcing collection data.

In a specific implementation, map data may be collected by a large number of manually-driven or automatically-driven vehicles driving on the road to obtain the address crowdsourcing collection data.

In order to further improve the update speed of the address database, each vehicle can upload the collected address crowdsourcing data in real time, or upload it when there is a network, or upload it as soon as possible according to a preset time interval, for example, it can be uploaded every day, or every week. upload etc.

In order to save network transmission resources, in other embodiments, sensors can be set on the vehicle to monitor environmental changes in real time, and compared with a map (specifically, a high-precision map), and when a road change is found, the data is uploaded for Address library additions.

As an alternative example, the vehicle may form a map based on a vision algorithm or based on a lidar. In some embodiments, a camera can be used to collect video data, and through deep learning algorithm or image recognition, a large amount of data is fed to the deep learning algorithm, so that the data processing device has a stronger recognition ability, thereby improving data processing. Ability.

S12, acquire address database data and map POI data.

Point of Interest (POI) data refers to meaningful points on the map, such as shops, bars, gas stations, hospitals, stations, etc.

In a specific implementation, the map POI data may include various basic information such as POI name, address, coordinates, and address attribute type, wherein the attribute type may include, for example, building attribute information. In addition, it can also include some detailed information, which may include business attribute information, etc., where the business attribute information can include, for example, the vertical industry to which it belongs, business hours, and other information that may be of interest to users, such as: price, rating, comment, menu, etc. , the specific details type is related to the POI type. Among them, a point of interest of a commercial attribute type may include a point of interest block of multiple building attributes.

It can be understood that, the embodiment of the present specification does not make any limitation on the business attribute type of the POI.

S13, in combination with the address database data and the map point of interest data, acquire address data that does not exist in the address database in the address crowdsourcing collection data, as supplementary address data.

In a specific implementation, in order to improve the accuracy of the obtained supplementary address data and save the data storage resources of the address database, the address crowdsourcing data can be coordinately matched with the map POI data, and the coordinate matching and the preset address attribute can be reserved. The initial address data corresponding to the type of POI block is used as candidate address data. After that, the address database data and the map POI data can be combined to eliminate duplicate data and noise data in the candidate address data to obtain the desired address. The supplementary address data is described.

S14, outputting the supplementary address data to the address library.

Although the data collected by the address crowdsourcing is not accurate enough and the data quality is not high, the data volume is huge, and the map POI data is accurate. Therefore, through step S13, through the map POI data, the address crowdsourcing data can be collected. Cross-validation is performed. Therefore, by using the above embodiment, the address database can be supplemented efficiently and accurately, so the expansion cost of the address database can be greatly reduced, the update speed of the address database can be improved, and the accuracy rate can be higher.

In a specific implementation, in order to further improve the quality of the supplementary data of the address library, the above-mentioned embodiment may be further expanded and optimized. The following is a detailed description of some specific examples combined with specific application scenarios.

Referring to the flowchart of another address library supplement method shown in FIG. 2, in other embodiments of this specification, the address library may be supplemented by the following steps:

S21 , obtaining address crowdsourcing collection data.

For how to obtain the address crowdsourcing collection data, please refer to the detailed introduction of step S11 in the foregoing embodiment.

S22, preprocessing the address crowdsourcing collection data to obtain initial address data with complete unified format information, so as to obtain the supplementary address data.

In the specific implementation, the format of the address crowdsourcing collection data obtained initially may not be uniform, or the information may be missing. In response to this situation, the address text in the address crowdsourcing collection data may be divided into blocks Segment, identify the address attribute type, and supplement the administrative division information to obtain the initial address data, and the generated initial address data with complete format unified information may include: coordinate data and corresponding address text data, where the address text data includes : Address attribute type and administrative division information.

A specific example is given below. An address crowdsourcing collection data is: Champagne International Building 3 Building 1402 has an elevator, coordinates 120.110081, 30.308145, including coordinate data and corresponding address text data. After step S22 preprocessing, the address crowdsourcing collection The address text in the data is divided into blocks, the address attribute information is identified, and the administrative division information is supplemented. The obtained initial address data of the corresponding entry can be:

prov=Zhejiang Province city=Hangzhou city district=Gongshu District poi=Champagne International houseno=3 buildings roomno=1402other=Elevator.

Among them, the address information is divided into blocks, the POI of the point of interest information is identified as "Champagne International", the building identification is 3, the house number is 1402, other information includes "there is an elevator", and the supplementary administrative division information includes: "prov = City of Zhejiang Province = District of Hangzhou City = Gongshu District”, that is, the administrative division information is located in Gongshu District, Hangzhou City, Zhejiang Province.

S23, acquire address database data and map POI data.

For details on how to obtain the map POI data, reference may be made to the foregoing step S22, which will not be described here.

S24, combining the address database data and the map POI data, acquire address data in the initial address data that does not exist in the address database to obtain supplementary address data.

In a specific implementation, the initial address data may be coordinate-matched with the map POI data, and the initial address data corresponding to the POI blocks that match the coordinates and have a preset address attribute type may be reserved as candidate address data; then , as an optional step, the address database data and the map POI data can be combined to eliminate duplicate data and noise data in the candidate address data to obtain the supplementary address data.

The specific example in step S22 is still used for description: first, for the initial address data prov=Zhejiang Province city=Hangzhou city district=Gongshu District poi=Champagne International houseno=3 buildings roomno=1402other=there is an elevator, according to the coordinate data therein , query the map point of interest data, perform coordinate matching, and obtain a commercial attribute AOI area "Champagne International", that is, refer to the map interest point 30 area shown in Figure 3, and which contains 6 AOI blocks of architectural attributes, Referring to the 6 dotted frame areas in Fig. 3, the coordinate data in the initial address is matched with the coordinates of the 6 AOI blocks, and it is found that some of the data in the initial address data do not hit the building (as shown by the dots in Fig. 3 ) ), it can be directly discarded as noise data, and other data hits the building (indicated by the X point in Figure 3), then they are retained as candidate address data.

Next, in order to further improve the accuracy and usability of the supplementary address data and save the storage space of the address database data, duplicate data and noise data in the candidate address data may be eliminated. Some examples of alternatives are shown below. Optional way 1, combine the address database data and the POI information in the map POI data, and match the candidate address data to determine that it is a duplicate address, or the address attribute type belongs to a newly added building, or belongs to a newly added building. The classification of mount addresses can eliminate the initial address data with duplicate addresses.

Still taking the above specific example as an example, for the convenience of description, the address in the above example prov=Zhejiang province city=Hangzhou city district=Gongshu district poi=Champagne international houseno=3 roomno=1402 is recorded as address A, combined with the address database Data and map POI data, that is, the AOI area "Champagne International" obtained by querying the map POI data is recorded as address B, and address B is matched with address A according to the length of the common substring and the length of the common substring shared by the two. In the proportion of the length of address A, the proportion of the length of the common substring in the length of address B, and the type of different string fragments between addresses A and B, whether there are data with the same name near address A and address B, etc. A linear classifier may be used to perform cross-matching between the map POI data and the initial address data, and it is obtained that the address type corresponding to the initial address data is a newly added building.

Optional way 2, perform rough clustering on the candidate address data according to the point of interest information in the map interest point data, remove noise data in the candidate address data, and obtain the supplementary address data.

Continuing with the description of the above example, if the new building "3" has multiple initial address data (dot data), most of the data may be in a certain building, and a small part is in other buildings. Retain the data on a building with the largest number of dots, and remove other data as noise data.

In a specific implementation, the above-mentioned optional mode 1 and optional mode 2 may be implemented in combination.

In a specific implementation, the data processed in step S24 can be directly used as supplementary address data, and then step S26 is directly executed to output to the address library, or the data processed in step S25 can be used only as rough selection supplementary address data, Step S25 is further executed.

S25: Perform spatial clustering on the rough selected supplementary address data to obtain cluster center coordinates and corresponding address description texts as supplementary address data output to the address library.

In a specific implementation, on the one hand, an anti-noise clustering algorithm can be used to cluster the data describing the same place to obtain the coordinates of the cluster center; on the other hand, the address text describing the same place in the supplementary address data The data is clustered and supplemented to obtain the corresponding address description text.

For example, an anti-noise clustering algorithm such as Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is used to cluster a batch of data describing the same location to obtain the coordinates of the cluster center. Among them, the DBSCAN algorithm is a density-based spatial clustering algorithm, which divides regions with sufficient density into clusters, and finds clusters of arbitrary shapes in a spatial database with noise, which defines clusters as density-connected points the largest collection of . It can be understood that other anti-noise clustering algorithms can also be used, such as other types of density-based spatial clustering algorithms, or other distance-based spatial clustering algorithms, which are not limited in the embodiments of this specification. The specific type of spatial clustering algorithm used.

At the same time, the text can be clustered and supplemented to obtain the address description text. For example, there are two addresses: Zhejiang Province->Hangzhou City->Gongshu District->[Road]->Poly Champagne International->3 Buildings and Zhejiang Province- >Hangzhou City->Gongshu District->Human Road->Poly Champagne International->Building 3->Runyuan Pharmacy, you can supplement the first address and get, Zhejiang Province->Hangzhou City->Gongshu District ->Human Road->Poly Champagne International->3 Buildings.

S26, outputting the supplementary address data to the address library.

Finally, the cluster center coordinates and address description text generated in step S25 can be used as supplementary address data to be supplemented into the warehouse, or the data obtained in step 24 can be directly supplemented into the warehouse as supplementary address data.

By adopting the above embodiment, in step S22, the address crowdsourcing collection data is preprocessed to obtain initial address data with complete format unified information, which is used to obtain the supplementary address data, which can further improve the address database obtained by supplementation. Data accuracy and completeness of information.

More specifically, the initial address data is obtained by dividing the address text in the address crowdsourcing collection data into blocks, identifying the address attribute type, and supplementing the administrative division information, so that the address in the initial address data is The text data format is unified and complete, including information such as address attribute type and administrative division information, which is conducive to matching with the map interest points, and improves the accuracy and completeness of the obtained supplementary address data, thereby further improving the supplementary data. The accuracy and completeness of the address database data.

In addition, in step S24, by performing coordinate matching between the initial address data and the map POI data, the initial address data corresponding to the POI blocks whose coordinates match and have a preset address attribute type are reserved as candidate address data , and then combine the address database data and the map interest point data to eliminate duplicate data and noise data in the candidate address data to obtain the supplementary address data, which can improve the accuracy of the address database data and save money. Storage space for address database data.

Wherein, according to the coordinate data in the initial address data, query the POI blocks whose coordinates match in the POI data on the map, and then compare the POI sub-blocks with the building attribute type contained in the POI block with all the POI blocks. Coordinate matching is performed on the coordinate data, and the initial address data of the coordinate matching is reserved as the candidate address data, which can further improve the accuracy of the supplementary address data and save the storage space of the address database data.

By combining the address database data and the POI information in the map POI data, and matching the candidate address data, it is determined that it belongs to a duplicate address, or the address attribute type belongs to a newly added building, or belongs to a newly added mount address and, according to the POI information in the map POI, perform rough clustering on the candidate address data, remove the noise data in the candidate address data, and obtain the The supplementary address data is described. Using the above method to analyze, sort and cluster data can effectively identify and eliminate duplicate address data and noise data, thereby improving the accuracy and usability of the supplementary address data and saving the storage space of address database data.

In addition, through step S25, spatial clustering is performed on the supplementary address data to obtain the coordinates of the cluster center and the corresponding address description text, which are used as supplementary address data of the address database, which can further improve the supplementary address data of the address database. accuracy.

The address library supplement method according to the embodiment of the present invention has been described in detail above in conjunction with specific application examples. The embodiment of the present invention also provides a corresponding address library supplement system. Corresponding descriptions are given in conjunction with specific application examples.

First, with reference to the schematic structural diagram of the address database complementing system shown in FIG. 4 , in some embodiments of the present invention, as shown in FIG. 4 , the address database complementing system 40 may include: a first data acquisition unit 41 and a second data acquisition unit 42 and third data acquisition unit 43, supplementary data generation unit 44 and output unit 45, wherein:

The first data acquisition unit 41 is adapted to acquire address crowdsourcing collection data;

The second data obtaining unit 42 is adapted to obtain map point of interest data;

The third data acquisition unit 43 is adapted to acquire address database data;

The supplementary data generating unit 44 is adapted to combine the address database data and the map POI data to obtain address data that does not exist in the address database in the address crowdsourcing collection data, as supplementary address data;

The output unit 45 is adapted to output the supplementary address data to the address library.

Using the above-mentioned address database complementing system 40, through the map POI data, it is possible to perform cross-validation on the address crowdsourcing collection data. Since the data volume of the address crowdsourcing collection data is huge, and the map POI data is accurate, it can efficiently To supplement the address database accurately, the whole process can be completed automatically, without the need for manual on-site inspection and verification, which can greatly reduce the expansion cost of the address database, improve the update speed of the address database, and have a high accuracy rate.

In a specific implementation, with continued reference to FIG. 4 , the address database supplementation system may further include: a preprocessing unit 46 , adapted to preprocess the address crowdsourcing collection data to obtain initial address data with complete unified format information, so as to for obtaining the supplemental address data.

In a specific implementation, with continued reference to FIG. 4 , the address library supplementation system may further include: a spatial clustering unit 47, adapted to perform spatial clustering on the supplementary address data to obtain cluster center coordinates and corresponding address description texts , as supplementary address data output to the address library.

For the specific implementation, working principle, and advantages of each functional unit of the address database supplement system, reference may be made to the detailed introduction in the foregoing method embodiments, which will not be described in detail here.

Referring to the schematic structural diagram of the electronic device shown in FIG. 5 , an embodiment of the present specification further provides an electronic device 50 , which includes a memory 51 and a processor 52 , and the memory 51 stores a memory 51 that can run on the processor 52 . A computer program, wherein when the processor 52 runs the computer program, the steps of the address library supplement method described in any of the foregoing embodiments can be executed. For specific steps, reference may be made to the detailed description of the foregoing method embodiments.

The electronic device 50 may be a computer device that can process data locally, such as a personal computer, a tablet computer, etc., or a computer cluster, or a cloud server, and the memory 51 may be a distributed storage device, except for storage In addition to the computer program, one or more of the address crowdsourcing collection data, address database data, and map interest point data may also be stored.

The processor 52 may be a single-core processor, a multi-core processor, a general-purpose processor, or a special-purpose processor capable of processing a large amount of data. The specific structure and implementation of the processor are described here. The method is not limited in any way.

In addition, in a specific implementation, referring to FIG. 5 , the electronic device 50 may further include a display 53, which is suitable for outputting a map corresponding to the supplementary address data, and may also output and display the intermediate execution process, for example, display and address library Cross-match results of data and map POI data.

In a specific implementation, with continued reference to FIG. 5 , the electronic device may further include an input interface 54, through which the user interacts with the user, so that the user can select crowdsourced address collection data, map interest point data, etc., and process the results of the intermediate data. Filter, and make some necessary or personalized settings, etc.

In other embodiments, the acquisition crowdsourcing address collection data, map interest point data, etc. may be acquired through the communication interface 55 .

In a specific implementation, the memory 51 , the processor 52 , the display 53 , the input interface 54 and the communication interface 55 can communicate through the bus 56 .

The embodiments of the present specification further provide a computer-readable storage medium on which a computer program is stored, wherein, when the computer program runs, the steps of the methods described in any of the foregoing embodiments may be performed. For specific steps, reference may be made to the foregoing embodiments. , and will not be repeated here.

In a specific implementation, the computer-readable storage medium may be various suitable readable storage mediums such as an optical disc, a mechanical hard disk, and a solid-state hard disk.

Although the embodiments of the present specification are disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be based on the scope defined by the claims.

Claims (13)

1. An address bank supplementing method, comprising:

acquiring address crowdsourcing acquisition data;

acquiring address database data and map interest point data;

combining the address database data and the map interest point data to obtain address data which does not exist in the address database in the address crowdsourcing acquisition data so as to obtain supplementary address data;

outputting the supplemental address data to the address repository.

2. The method of claim 1, wherein prior to obtaining address data in the address crowd-sourced collection data that is not present in the address repository, further comprising:

preprocessing the address crowdsourcing acquisition data to obtain initial address data with a uniform format and complete information, so as to obtain the supplementary address data.

3. The method of claim 2, wherein the preprocessing the address crowdsourcing acquisition data to obtain initial address data with complete format unification information comprises:

performing block segmentation on an address text in the address crowdsourcing acquisition data, identifying an address attribute type, and supplementing administrative division information to obtain initial address data, wherein the initial address data comprises: coordinate data and corresponding address text data, the address text data comprising: address attribute type and administrative division information.

4. The method of claim 3, wherein the obtaining, in combination with the address library data and the map point of interest data, address data that is not present in the address library in the address crowd-sourced collection data to obtain supplemental address data comprises:

carrying out coordinate matching on the initial address data and the map interest point data, and keeping the initial address data which is matched with the coordinates and corresponds to the interest point block with a preset address attribute type as candidate address data;

and removing repeated data and noise data in the candidate address data by combining the address database data and the map interest point data to obtain the supplementary address data.

5. The method of claim 4, wherein the coordinate matching the initial address data with the map point of interest data, and the retaining initial address data of the point of interest block with coordinate matching and a preset address attribute type as candidate address data comprises:

inquiring an interest point block matched with the coordinates in the map interest point data according to the coordinate data in the initial address data;

and carrying out coordinate matching on the interest point subblocks with the building attribute types contained in the interest point block and the coordinate data, and keeping initial address data of the coordinate matching as the candidate address data.

6. The method of claim 4, wherein said culling of duplicate data and noise data from said candidate address data to obtain said supplemental address data in combination with said address repository data and said map point of interest data comprises at least one of:

matching the candidate address data with the interest point information in the map interest point data by combining address database data and the interest point information in the map interest point data, determining that the candidate address data belongs to a repeated address, or the address attribute type belongs to a newly added building, or the candidate address data belongs to the classification of a newly added mounting address, and then rejecting initial address data with repeated addresses;

and according to the interest point information in the map interest point data, performing rough clustering on the candidate address data, and removing noise data in the candidate address data to obtain the supplementary address data.

7. The method of any of claims 1-6, wherein prior to outputting the supplemental address data to the address bank, further comprising:

and carrying out spatial clustering on the supplementary address data to obtain a clustering center coordinate and a corresponding address description text which are used as supplementary address data output to the address library.

8. The method of claim 7, wherein the spatially clustering the supplemental address data to obtain cluster center coordinates and corresponding address description text comprises:

clustering data describing the same place by using an anti-noise clustering algorithm to obtain a clustering center coordinate;

and clustering and supplementing the address text data describing the same place in the supplemented address data to obtain a corresponding address description text.

9. An address repository replenishment system, comprising:

the first data acquisition unit is suitable for acquiring the address crowdsourcing acquisition data;

the second data acquisition unit is suitable for acquiring the point data of interest of the map;

the supplementary data generating unit is suitable for acquiring address data which is not in the address database in the address crowdsourcing acquisition data by combining the map interest point data, and the address data is used as supplementary address data;

an output unit adapted to output the supplemental address data to the address bank.

10. The system of claim 9, further comprising: and the preprocessing unit is suitable for preprocessing the address crowdsourcing acquisition data to obtain initial address data with complete format unified information so as to obtain the supplementary address data.

11. The system of claim 9 or 10, further comprising: and the spatial clustering unit is suitable for performing spatial clustering on the supplementary address data to obtain a clustering center coordinate and a corresponding address description text as supplementary address data output to the address library.

12. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program being executable on the processor, wherein the processor, when executing the computer program, performs the steps of the method of any of claims 1 to 8.

13. A computer-readable storage medium, on which a computer program is stored, wherein the computer program performs the steps of the method of any one of claims 1 to 8 when executed.

Images