[go: up one dir, main page]

CN116257703B - Address automatic cleaning method based on weight analysis model of geographic information system - Google Patents

Address automatic cleaning method based on weight analysis model of geographic information system

Info

Publication number
CN116257703B
CN116257703B CN202310214957.5A CN202310214957A CN116257703B CN 116257703 B CN116257703 B CN 116257703B CN 202310214957 A CN202310214957 A CN 202310214957A CN 116257703 B CN116257703 B CN 116257703B
Authority
CN
China
Prior art keywords
coordinate
address
longitude
coordinate set
latitude
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310214957.5A
Other languages
Chinese (zh)
Other versions
CN116257703A (en
Inventor
彭彦
戴峻
李宇鹏
孙爱超
陈康
陈月婷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Changyuan Information Technology Co ltd
Jiangsu Changyuan Information Technology Co ltd
Original Assignee
Nanjing Changyuan Information Technology Co ltd
Jiangsu Changyuan Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing Changyuan Information Technology Co ltd, Jiangsu Changyuan Information Technology Co ltd filed Critical Nanjing Changyuan Information Technology Co ltd
Priority to CN202310214957.5A priority Critical patent/CN116257703B/en
Publication of CN116257703A publication Critical patent/CN116257703A/en
Application granted granted Critical
Publication of CN116257703B publication Critical patent/CN116257703B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Remote Sensing (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本发明公开了一种基于地理信息系统权重分析模型的地址自动清洗方法，基于标准坐标系对地址信息中的大要素切分比对，与主体的名称、地址单向查询以及大要素联合查询的结果构建权重分析模型，权重值可动态完善，从而判断地址信息的准确性、进行纠偏获得准确的地址信息，提高对地址信息匹配的准确性。

The present invention discloses an automatic address cleaning method based on a geographic information system weight analysis model. The method segments and compares large elements in address information based on a standard coordinate system, and constructs a weight analysis model based on the results of a one-way query of the subject's name and address and a joint query of large elements. The weight value can be dynamically improved to judge the accuracy of the address information, perform corrections to obtain accurate address information, and improve the accuracy of address information matching.

Description

Address automatic cleaning method based on weight analysis model of geographic information system
Technical Field
The invention relates to an address automatic cleaning method, in particular to an address automatic cleaning method based on a geographic information system weight analysis model.
Background
The application scenario of the intellectual property data information is more and more required, for example, the address information can be used for monitoring the maintenance condition of the intellectual property in the target area and the migration real-time change condition, but the reality and the accuracy of the address information are depended.
In the links of application and the like, address information related to the main bodies of the applicant and the like is in the conditions of irregular filling, instability and incompleteness, and when data information is put in storage, the address information cannot be accurately matched with a real address.
The method for judging the division of the regional categories of the patent applicant is that keyword word segmentation is carried out on the address of the applicant, provincial county matching is carried out on the address of the applicant to achieve regional matching, for example, "new regional double-sub-floor B base 11 building of Zhengsu province, zhenjiang city" is formed through word segmentation, "Jiangsu province", "Zhenjiang city" and "new regional double-sub-floor B base 11 building" are word libraries of [ small elements ], probability calculation is carried out, and effective regional information is formed, and the regional matching method is commonly used for national regional sorting and pre-screening in the express industry. However, the blind statistical method based on probability analysis cannot solve the problem of determining the right of a plurality of address information, and meanwhile, judging errors easily occur under the address information of complex interference, for example, serious interference exists when word segmentation and matching are performed on 'Shanghai city Tibetan, lunan Beijing street 118'.
Disclosure of Invention
The invention aims to provide an address automatic cleaning method, which is used for judging the accuracy of address information and correcting deviation through a weight analysis model constructed based on a geographic information system.
The technical scheme is that the address automatic cleaning method based on the geographic information system weight analysis model comprises the following steps:
Step 100, inquiring longitude and latitude coordinates of a warehouse-in name of a main body through a standard coordinate system to obtain a coordinate set A { A 1,…,An }, wherein n is more than or equal to 1, inquiring longitude and latitude coordinates of a warehouse-in address of the main body through the standard coordinate system to obtain a coordinate set B { B 1,…,Bm }, wherein m is more than or equal to 1, judging whether the coordinate set A is completely consistent with the coordinate set B, and judging that warehouse-in address information deviation exists if the coordinate set A is inconsistent with the coordinate set B;
step 200 of selecting said set of coordinates a and/or said set of coordinates B,
Step 201, carrying out reverse analysis and verification by using a map tool of the standard coordinate system to obtain a verification address, extracting a large element from the verification address, obtaining a matching address by using the large element and the warehouse-in name, inquiring the longitude and latitude coordinates of the matching address through the standard coordinate system to obtain a coordinate set C { C 1,…,Cp }, wherein p is more than or equal to 1,
Step 202, respectively executing step 201 according to each longitude and latitude coordinate of the selected coordinate set, and obtaining i groups of coordinate sets C, wherein i=n or m or not more than n+m;
step 300, establishing a linear regression equation according to the ranges of all longitude and latitude coordinates of the coordinate set A and the coordinate set B, i and the coordinate set C;
and 400, taking the coordinate set A and the coordinate set B, i as three weight items and giving weight values, constructing a weighted model, calculating the signal value of each longitude and latitude coordinate and the regression line, and taking the longitude and latitude coordinate corresponding to the minimum signal value as an effective address.
Specifically, the subject is a non-individual.
Specifically, the standard coordinate system comprises a GCJ-02 coordinate system, a Mars coordinate system and a geodetic coordinate system.
Further, the signal value is w×|k|, the|k| is an absolute value of a discrete value of the longitude and latitude coordinate and the regression line, and W is a weight value of a weight item corresponding to the longitude and latitude coordinate.
Further, when the weighting model is built, the initial assignment of the weight value of each weight item is 1/weight item number, and then the model precision training is realized on the weight value by manually comparing the consistency of regression lines of voting and/or warehousing address minor element reverse comparison, verifying the calculation result to perfect the weighting model.
Further, the manual comparison voting refers to manual accuracy judgment of longitude and latitude coordinates of the warehouse-in name, the real address of the main body, the major element and the warehouse-in name.
The address automatic cleaning method based on the geographic information system weight analysis model has the advantages that the weight analysis model is built based on the large element segmentation ratio pairs in the address information, the name of the main body, the one-way address query and the large element joint query results, and the weight value can be dynamically perfected, so that the accuracy of the address information is judged, the correction is carried out to obtain the accurate address information, and the accuracy of the address information matching is improved.
Drawings
FIG. 1 is a position diagram of longitude and latitude coordinates A 1 in example 1;
Fig. 2 is a position diagram of longitude and latitude coordinates B 1 in example 1;
Fig. 3 is a position diagram of longitude and latitude coordinates C 1 in example 1;
FIG. 4 is a linear regression diagram of example 1;
fig. 5 is a position diagram of longitude and latitude coordinates a 1 in example 2;
FIG. 6 is a position diagram of longitude and latitude coordinates B 1 in example 2;
fig. 7 is a position diagram of longitude and latitude coordinates C 1 in example 2;
FIG. 8 is a linear regression diagram of example 2;
fig. 9 is a position diagram of longitude and latitude coordinates a 1 in example 3;
fig. 10 is a position diagram of longitude and latitude coordinates B 1 in example 3;
fig. 11 is a position diagram of longitude and latitude coordinates C 1 in example 3;
FIG. 12 is a linear regression diagram in example 3;
fig. 13 is a position diagram of latitude and longitude coordinates a 1 in example 4;
fig. 14 is a position diagram of latitude and longitude coordinates a 2 in example 4;
Fig. 15 is a position diagram of latitude and longitude coordinates C 1 of the 1 st set of coordinates C in example 4;
fig. 16 is a position diagram of latitude and longitude coordinates C 1 of the 3 rd set of coordinates C in example 4;
fig. 17 is a linear regression diagram in example 4.
Detailed Description
The invention is further elucidated below in connection with the drawings and the specific embodiments.
An address automatic cleaning method based on a weight analysis model of a geographic information system specifically comprises the following steps:
step 100, inquiring longitude and latitude coordinates of a warehouse-in name of a main body through a standard coordinate system to obtain a coordinate set A { A 1,…,An }, wherein n is more than or equal to 1, inquiring longitude and latitude coordinates of a warehouse-in address of the main body through the standard coordinate system to obtain a coordinate set B { B 1,…,Bm }, and m is more than or equal to 1, judging whether the coordinate set A is completely consistent with the coordinate set B, and judging that warehouse-in address information deviation exists if the coordinate set A is inconsistent with the coordinate set B.
Because the names of the personal subjects, namely the names, do not have the attribute of the addresses of the subjects, in the technical scheme of the application, the subjects refer to enterprises and institutions and the like, and the names of the subjects have one or more address attributes when being indexed by GIS. When the data information is put in storage, the data of which the main body is an individual is filtered first.
The standard coordinate system includes a GCJ-02 coordinate system, a Mars coordinate system, and a geodetic coordinate system.
Step 200 of selecting one of the coordinate set a and/or the coordinate set B,
Step 201, carrying out reverse analysis and verification by using a map tool of a standard coordinate system with one longitude and latitude coordinate of a selected coordinate set to obtain a verification address, extracting a large element from a verification address cutting word, obtaining a matching address by using the large element and a warehouse-in name, inquiring the longitude and latitude coordinate of the matching address by using the standard coordinate system to obtain a coordinate set C { C 1,…,Cp }, p is more than or equal to 1,
Step 202, executing step 201 with each longitude and latitude coordinate of the selected coordinate set, and obtaining i groups of coordinate sets C, i=n or m or not more than n+m.
Map tools using a standard coordinate system, such as a Goldmap using a GCJ-02 coordinate system, and map tools using a Mars coordinate system, such as a hundred degree map.
And 300, establishing a linear regression equation according to the ranges of all longitude and latitude coordinates of the coordinate set A and the coordinate set B, i and the coordinate set C.
And 400, taking the coordinate set A and the coordinate set B, i as three weight items and giving weight values, constructing a weighted model, calculating the signal value of each longitude and latitude coordinate and the regression line, and taking the longitude and latitude coordinate corresponding to the minimum signal value as an effective address.
When the weighting model is built, the weight value W of each weight item is initially assigned to be 1/weight item number, and then the model precision training is realized on the weight value by manually comparing regression line consistency of voting and/or warehousing address minor element reverse comparison, verifying the calculation result to perfect the weighting model. The manual comparison voting refers to manual accuracy judgment of longitude and latitude coordinates of a warehouse-in name, a real address of a main body, a major element and the warehouse-in name.
The signal value is W, K is the absolute value of the discrete value of the longitude and latitude coordinates and the regression line, and W is the weight value of the weight item corresponding to the longitude and latitude coordinates. For example, the signal value of the longitude and latitude coordinate a 2 is calculated, and W is a dynamic parameter of the weighting item, i.e. the coordinate set a, which can be weighted.
The following examples illustrate the application of the address auto-cleaning method of the present invention from several cases of bias in the information of the binned data.
Example 1
Taking the case that the warehouse-in address of the patent applicant (namely the main body) is false address information as an example, the automatic address cleaning method is adopted to rectify deviation.
Warehouse name of patent applicant Nanjing university
The warehouse-in address of the patent applicant is that Puzhu south road 30 No. 8020 mail box 32 boxes in Pu kou area of Nanjing, jiangsu province
Step 100, inquiring the warehouse-in name through a GCJ-02 coordinate system to obtain only one longitude and latitude coordinate (see figure 1), inquiring the warehouse-in address through the GCJ-02 coordinate system to obtain only one longitude and latitude coordinate (see figure 2), and judging that the warehouse-in address information deviation exists by manually judging that B 1 is a transceiver station near Nanjing industrial university through figure 3 due to inconsistent A 1、B1.
Step 200, selecting the coordinate set A in the coordinate set A and the coordinate set B,
Step 201, carrying out reverse analysis and verification through a Goldmap by using only longitude and latitude coordinates A 1 of a coordinate set A to obtain a verification address, namely, a Puzhu southward No. 30 of Nanjing, jiangsu province, cutting words of the verification address, extracting a large element, namely, nanjing, jiangsu province, obtaining a matching address by using a large element and a warehouse-in name, namely, the Nanjing industrial university of Nanjing, jiangsu province, inquiring the matching address through a GCJ-02 coordinate system, and obtaining only one longitude and latitude coordinate (see figure 3), wherein the coordinate set C { C 1 (118.640081,32.082496) }.
Step 300, a linear regression equation is established by using the ranges of all longitude and latitude coordinates of the coordinate set A, the coordinate set B and the coordinate set C of the 1 group (see figure 4).
And 400, constructing an equal weight weighting model by taking a coordinate set A, a coordinate set B and a coordinate set C as three weight items, wherein the initial assignment of weight values is 1/3, then calculating the signal value of each longitude and latitude coordinate and a regression line, and judging that the signal value of the longitude and latitude coordinate A 1、C1 is equal and minimum and is an effective address if the longitude and latitude coordinate (118.640081,32.082496) is nearest to the regression line.
And through actual manual judgment and verification, the results of the method are consistent.
Example 2
Taking the case that the warehouse-in address of the patent applicant (namely the main body) is the wrong address information as an example, the automatic address cleaning method is adopted to correct deviation.
Warehouse-in name of patent applicant Jiangsu Minyue optical glasses Co., ltd
The warehouse-in address of the patent applicant is Jiangsu Danyang province economic development area Ji Lianglu No. 200
Step 100, inquiring the warehouse-in name through a GCJ-02 coordinate system to obtain only one longitude and latitude coordinate (see figure 5), wherein the positioning information is in Beijing, the coordinate set A { A 1 (116.462689,39.878249) }, inquiring the warehouse-in address through the GCJ-02 coordinate system to obtain only one longitude and latitude coordinate (see figure 6), and the coordinate set B { B 1 (119.617603,32.036547) } is in Jiangsu Zhenjiang Danyang, and judging that the warehouse-in address information deviation exists due to the inconsistency of A 1、B1.
Step 200, selecting a coordinate set A and a coordinate set B in the coordinate set B,
Step 201, carrying out reverse analysis and verification through a Gaode map by using only longitude and latitude coordinates B 1 of a coordinate set B to obtain a verification address, namely, no. Ji Lianglu of Danyang city of Zhenjiang, jiangsu province, cutting words of the verification address, extracting a large element, namely, danyang city of Zhenjiang province, jiangsu province, obtaining a matching address by using a large element and a warehouse-in name, namely, a Jiangsu ming's optical glasses limited company of Danyang city of Zhenjiang province, jiangsu province, inquiring the matching address through a GCJ-02 coordinate system, and obtaining only one longitude and latitude coordinate (see figure 7), wherein the positioning information is in Zhenjiang Danyang, and the coordinate set C { C 1 (119.617603,32.036547) }.
Step 300, a linear regression equation is established by using the ranges of all longitude and latitude coordinates of the coordinate set A, the coordinate set B and the coordinate set C of the 1 group (see figure 8).
And 400, constructing an equal weight weighting model by taking a coordinate set A, a coordinate set B and a coordinate set C as three weight items, wherein the initial assignment of weight values is 1/3, then calculating the signal value of each longitude and latitude coordinate and a regression line, and judging that the signal value of the longitude and latitude coordinate B 1、C1 is equal and minimum and is an effective address if the longitude and latitude coordinate (119.617603,32.036547) is nearest to the regression line.
In this embodiment, "dan Yang Sheng" in the warehouse-in address is an error message, and the error written as "dan yang province" in dan yang city can be rectified by the invention.
And through actual manual judgment and verification, the results of the method are consistent.
Example 3
Taking the case that the warehouse-in address of the patent applicant (namely the main body) is the missing address information as an example, the automatic address cleaning method is adopted to correct deviation.
Warehouse name of the patent applicant Nanjing Rasen Fuel Co., ltd
The warehouse-in address of the patent applicant is saddle town Feng Cun in Liuhe region of Nanjing, jiangsu province
Step 100, inquiring the warehouse-in name through a GCJ-02 coordinate system to obtain only one longitude and latitude coordinate (see figure 9), wherein the positioning information is in Jiangsu Nanjing, then the coordinate set A { A 1 (118.802459,32.414463) }, inquiring the warehouse-in address through the GCJ-02 coordinate system to obtain only one longitude and latitude coordinate (see figure 10), and the positioning information is in Jiangsu Nanjing, then the coordinate set B { B 1 (118.804505,32.40086) }, and judging that the warehouse-in address information deviation exists due to the inconsistency of A 1、B1.
Step 200, selecting a coordinate set A and a coordinate set B in the coordinate set B,
Step 201, carrying out reverse analysis and verification through a Goldmap by using only longitude and latitude coordinates B 1 of a coordinate set B to obtain a verification address, namely, a Jiangsu Nanjing city, feng Cun village commission, cutting words of the verification address, extracting a large element, namely, jiangsu Nanjing city, and obtaining a matching address by using the large element and a warehouse entry name, namely, nanjing Ralssen Fuel Limited company, jiangsu Nanjing city, wherein the matching address is queried through a GCJ-02 coordinate system, only one longitude and latitude coordinate is obtained (see figure 11), and positioning information is in Jiangsu Nanjing, so that the coordinate set C { C 1 (119.617603,32.036547) }.
Step 300, a linear regression equation is established by using the ranges of all longitude and latitude coordinates of the coordinate set A, the coordinate set B and the coordinate set C of the 1 group (see figure 12).
And 400, constructing an equal weight weighting model by taking a coordinate set A, a coordinate set B and a coordinate set C as three weight items, wherein the initial assignment of weight values is 1/3, then calculating the signal value of each longitude and latitude coordinate and a regression line, and judging that the longitude and latitude coordinate C 1 is an effective address if the signal value is the smallest and the longitude and latitude coordinate (119.617603,32.036547) is nearest to the regression line.
In the embodiment, the error correction can be performed by the invention because of inaccurate positioning range caused by missing detailed element information in the warehouse-in address.
And through actual manual judgment and verification, the results of the method are consistent.
Example 4
Taking the case that the warehouse-in address of the contact person (namely the main body) is the wrong address information as an example, the automatic address cleaning method is adopted to correct the deviation.
Warehouse-in name of contact person, jiangsu university of science and technology economic management school
The warehouse-in address of the contact person is Zhenjiang city Dantu area long glory path 666 number
Step 100, inquiring the warehouse-in name through a GCJ-02 coordinate system to obtain two longitude and latitude coordinates, A 1 (119.468078,32.197314), positioning information in Zhenjiang city (see figure 13), A 2 (116.357342,39.993032), positioning information in sea lake area (see figure 14) of Beijing city, inquiring the warehouse-in address through the GCJ-02 coordinate system to obtain only one longitude and latitude coordinate, and judging that the warehouse-in address information deviation exists due to inconsistency of A 1、A2、B1, wherein the coordinate set A { A 1(119.468078,32.197314),A2 (116.357342,39.993032) }, the coordinate set B { B 1 (119.360779,32.109739) } is obtained.
Step 200. Selecting a coordinate set a and a coordinate set B,
Step 201, carrying out reverse analysis and verification through a Goldmap by using only longitude and latitude coordinates B 1 of a coordinate set B to obtain a verification address, namely a long light path 666 number in Dantu region of Zhenjiang city in Jiangsu province, extracting a large element from the verification address, namely Zhensu province, obtaining a matching address by using a large element and a warehouse-in name, namely Jiangsu province's university of Jiangsu science and technology economic management institute, inquiring the matching address through a GCJ-02 coordinate system to obtain only one longitude and latitude coordinate (see figure 15), and obtaining a1 st set of coordinate set C { C 1 (119.468078,32.197307) } when positioning information is in Jiangsu Zhenjiang.
Step 202, carrying out reverse analysis and verification through a Goldmap by using one longitude and latitude coordinate A 1 of a coordinate set A to obtain a verification address, extracting a word from the verification address to obtain a major element, namely Jiangsu Zhenjiang city, obtaining a matching address by using a major element and a warehouse-in name, namely Jiangsu Zhenjiang city Jiangsu technical university economic management institute, inquiring the matching address through a GCJ-02 coordinate system to obtain only one longitude and latitude coordinate, wherein positioning information is in Jiangsu Zhenjiang city, and a2 nd set of coordinate sets C { C 1 (119.468078,32.197307) };
And (3) carrying out reverse analysis and verification through a Goldmap by using the other longitude and latitude coordinate A 2 of the coordinate set A to obtain a verification address, and extracting a large element of the verification address by cutting words, wherein the matched address is queried through a GCJ-02 coordinate system to obtain only one longitude and latitude coordinate (see figure 16), and the 3 rd coordinate set C { C 1 (116.357342,39.993028) } is positioned in the sea lake area of Beijing city.
Step 300, a linear regression equation is established by using the ranges of all longitude and latitude coordinates of the coordinate set A, the coordinate set B and the 3 groups of coordinate set C (see figure 17).
And 400, taking the coordinate set A, the coordinate set B and the 3 groups of coordinate sets C as three weight items, constructing an equal weight weighting model by initially assigning 1/3 weight values, calculating the signal value of each longitude and latitude coordinate and the regression line, wherein the longitude and latitude coordinate (119.468078,32.197314) is nearest to the regression line, and judging the longitude and latitude coordinate A 1 as an effective address if the signal value is minimum.
In this embodiment, the error writes the address of the division area without "economic management university" as the address of the "Jiangsu university of science and technology" department, which results in inaccurate positioning range, and the correction can be performed by the present invention.
And through actual manual judgment and verification, the results of the method are consistent.

Claims (3)

1.一种基于地理信息系统权重分析模型的地址自动清洗方法，其特征在于包括以下步骤：1. An automatic address cleaning method based on a geographic information system weight analysis model, characterized by comprising the following steps: 步骤100：对主体的入库名称通过标准坐标系查询经纬度坐标，得到坐标集A{A1,…,An}，n≥1，对主体的入库地址通过所述标准坐标系查询经纬度坐标，得到坐标集B{B1,…,Bm}，m≥1，判断所述坐标集A与所述坐标集B是否完全一致，如不一致则判断存在入库地址信息偏差；Step 100: Query the longitude and latitude coordinates of the entry name of the subject using a standard coordinate system to obtain a coordinate set A{A 1 ,…,A n }, where n≥1; query the longitude and latitude coordinates of the entry address of the subject using the standard coordinate system to obtain a coordinate set B{B 1 ,…,B m }, where m≥1; determine whether the coordinate set A is completely consistent with the coordinate set B; if not, determine that there is a deviation in the entry address information; 步骤200：选择所述坐标集A和/或所述坐标集B，Step 200: Select the coordinate set A and/or the coordinate set B, 步骤201：以选择的坐标集的其中一个经纬度坐标，通过使用所述标准坐标系的地图工具进行反向解析验证得到验证地址，对所述验证地址切词提取得到大要素，以所述大要素+所述入库名称得到匹配地址，对所述匹配地址通过所述标准坐标系查询经纬度坐标，得到坐标集C{C1,…,Cp}，p≥1，Step 201: Using one of the longitude and latitude coordinates in the selected coordinate set, reverse parsing and verification are performed using the map tool of the standard coordinate system to obtain a verification address. The verification address is segmented and extracted to obtain large elements. The matching address is obtained by combining the large elements with the stored name. The longitude and latitude coordinates of the matching address are searched using the standard coordinate system to obtain a coordinate set C{C 1 ,…,C p }, where p≥1. 步骤202：以选择的坐标集的其它每个经纬度坐标，分别执行步骤201；共得到i组坐标集C，i＝n或m或不大于n+m；Step 202: Execute step 201 for each other latitude and longitude coordinate of the selected coordinate set; obtain a total of i coordinate sets C, where i = n or m or not greater than n + m; 步骤300：以坐标集A、坐标集B、i组坐标集C的所有经纬度坐标的范围，建立线性回归方程；Step 300: Establish a linear regression equation based on the range of all longitude and latitude coordinates of coordinate set A, coordinate set B, and i groups of coordinate set C; 步骤400：以坐标集A、坐标集B、i组坐标集C为三项权重项并赋权重值，构建加权模型，再计算每个经纬度坐标与回归线的信值，以最小信值对应的经纬度坐标为有效地址；Step 400: Using coordinate set A, coordinate set B, and i sets of coordinate sets C as three weight items and assigning weight values, a weighted model is constructed. The confidence value between each longitude and latitude coordinate and the regression line is calculated, and the longitude and latitude coordinate corresponding to the minimum confidence value is regarded as the valid address. 所述信值为W*|K|，|K|为经纬度坐标与回归线离散值的绝对值，W为该经纬度坐标对应的权重项的权重值；构建加权模型时，各权重项的权重值初始赋值均为1/权重项项数，而后通过人工比对投票和/或入库地址小要素反向对比的回归线一致性，验证计算结果完善加权模型，对权重值实现模型精度训练；所述人工比对投票是指对所述入库名称、所述主体的真实地址、所述大要素+所述入库名称，各自经纬度坐标的人工准确性判断。The confidence value is W*|K|, |K| is the absolute value of the longitude and latitude coordinates and the discrete value of the regression line, and W is the weight value of the weight item corresponding to the longitude and latitude coordinates; when constructing the weighted model, the initial weight value assignment of each weight item is 1/the number of weight items, and then the calculation results are verified and improved through manual comparison voting and/or reverse comparison of the regression line consistency of the small elements of the storage address, and the model accuracy training of the weight value is achieved; the manual comparison voting refers to the manual accuracy judgment of the longitude and latitude coordinates of the storage name, the real address of the subject, the large element + the storage name. 2.根据权利要求1所述的基于地理信息系统权重分析模型的地址自动清洗方法，其特征在于：所述主体为非个人。2. The automatic address cleaning method based on the geographic information system weight analysis model according to claim 1 is characterized in that the subject is non-individual. 3.根据权利要求1所述的基于地理信息系统权重分析模型的地址自动清洗方法，其特征在于：所述标准坐标系包括GCJ-02坐标系、火星坐标系、大地坐标系。3. The automatic address cleaning method based on the geographic information system weight analysis model according to claim 1 is characterized in that the standard coordinate system includes the GCJ-02 coordinate system, the Martian coordinate system, and the earth coordinate system.
CN202310214957.5A 2023-03-08 2023-03-08 Address automatic cleaning method based on weight analysis model of geographic information system Active CN116257703B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310214957.5A CN116257703B (en) 2023-03-08 2023-03-08 Address automatic cleaning method based on weight analysis model of geographic information system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310214957.5A CN116257703B (en) 2023-03-08 2023-03-08 Address automatic cleaning method based on weight analysis model of geographic information system

Publications (2)

Publication Number Publication Date
CN116257703A CN116257703A (en) 2023-06-13
CN116257703B true CN116257703B (en) 2025-09-16

Family

ID=86687685

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310214957.5A Active CN116257703B (en) 2023-03-08 2023-03-08 Address automatic cleaning method based on weight analysis model of geographic information system

Country Status (1)

Country Link
CN (1) CN116257703B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106056625A (en) * 2016-05-25 2016-10-26 中国民航大学 Airborne infrared moving target detection method based on geographical homologous point registration
CN106991185A (en) * 2017-04-10 2017-07-28 携程计算机技术（上海）有限公司 The hotel's latitude and longitude information maintaining method and system of OTA websites

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004107219A1 (en) * 2003-05-29 2004-12-09 Locateplus Holdings Corporation Current mailing address identification and verification
CN110263022B (en) * 2019-05-08 2023-03-14 深圳丝路天地电子商务有限公司 Hotel data matching method and device

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106056625A (en) * 2016-05-25 2016-10-26 中国民航大学 Airborne infrared moving target detection method based on geographical homologous point registration
CN106991185A (en) * 2017-04-10 2017-07-28 携程计算机技术（上海）有限公司 The hotel's latitude and longitude information maintaining method and system of OTA websites

Also Published As

Publication number Publication date
CN116257703A (en) 2023-06-13

Similar Documents

Publication Publication Date Title
CN104572645B (en) Interest point data association method and device
US7668651B2 (en) Reverse geocoding system using combined street segment and point datasets
CN110146789B (en) Intelligent operation and inspection reporting method and device
CN101826274B (en) Correction Method of Vector Traffic Digital Map Based on Floating Car Data
CN114692236B (en) Big data-oriented territorial space planning base map base number processing method
CN110895878A (en) A Generating Method of Traffic State Virtual Detector Based on GE-GAN
CN107368480B (en) A method and device for locating and recognizing error types of points of interest data
CN113434708A (en) Address information detection method and device, electronic equipment and storage medium
CN112131752A (en) A robust estimation algorithm for super-strong collapse pollution rate based on quasi-quasi-calibration
CN117196918A (en) A method, device, equipment and storage medium for determining building carbon emissions
CN103064953B (en) Position-based is registered the POI update method of data
WO2023123929A1 (en) Abnormal application recognition method and device
CN115438061B (en) A method and system for updating customer standard address database
CN116257703B (en) Address automatic cleaning method based on weight analysis model of geographic information system
CN106897683B (en) Ground object detection method and system of remote sensing image
CN110688439A (en) Method for automatically identifying and analyzing enterprise information based on regional geocoding
CN117197730B (en) Repair evaluation method for urban space distortion image
CN110619134A (en) Integrated detection and restoration method for solving problem of flying spot and spot density of road network data
CN115292962B (en) Path similarity matching method and device based on track rarefaction and storage medium
CN116303870A (en) Gas flow value mapping method and system for gas pipe network data management
CN116991806A (en) Automatic soil data superposition analysis method based on spatial data
CN116662048A (en) Survey data warehouse entry checking and correcting method and device, electronic equipment and storage medium
CN120873101B (en) Two-dimensional vector geographic database defect detection method and computer-readable storage medium
CN112381941A (en) Aviation flight image coordinate correction method
CN107273380B (en) A method and device for determining the position of a target point and checking data

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant