An online matching method of place names and addresses based on spatiotemporal big data

Abstract

The invention discloses an online matching method for place names and addresses based on spatiotemporal big data, comprising: A. extracting geographic entities from large-scale topographic map data, matching geographic entities with industrial and commercial enterprise point information, and performing matching on successfully matched address information. Sort out the word segmentation, split the matching data into buildings, units and households, and integrate with the civil affairs and public security place names and addresses to form an initial place name and address matching database; B. Based on the initial place name and address matching database, collect industry The place-name address data is updated into the initial place-name-address matching database according to the place-name-address update mechanism, and a place-name history database is established based on the relationship between historical place-names and current place-names to realize multi-dimensional and fine-grained matching of place-name addresses. The invention updates and maintains the place-name-address matching database through multiple channels, ensures the dynamic and current situation of the data, and improves the matching accuracy and accuracy by combining the multi-dimensional and fine-grained place-name-address matching mechanism.

Description

technical field [0001] The invention relates to the field of geographic information systems, in particular to an online matching method for place names and addresses based on spatiotemporal big data. Background technique [0002] In recent years, with the construction and development of smart cities, urban informatization has changed from digital cities to smart cities, and spatiotemporal data plays an important role in the construction of smart cities. Place-name address data is a geospatial expression, an important part of spatio-temporal data, and an important reference for location information in urban management, which puts forward new requirements for the maintenance and retrieval and matching technology of place-name address database. In the process of urban development, multiple departments have accumulated a lot of place name and address data based on their own business, which often leads to various and ununiform descriptions of the same location. The current urban ...

AI Technical Summary

Problems solved by technology

[0005] (1) The traditional field surveying and mapping collection method has a long data production cycle and poor data performance

[0006] The traditional method of collecting place names and addresses in field surveying and mapping takes the block as the unit, and collects in a group of about 2 people. Before collecting, you need to prepare the map of the survey area, field work certificates, folders, pens and other tools, and carry out according to the job requirements. Collect and record, and mark the points on the drawings. After the field survey and mapping is completed, the office needs to spatialize the surveyed drawings and enter the relevant attribute data into the database to finally form a place name address matching database. , the data collected in this way has a long cycle and consumes a lot of labor. It is only suitable for use when there is no data related to place names and addresses. The update frequency of this method is low, and the current situation of the place names and addresses database cannot be guaranteed.

[0007] (2) The existing place name and address matching database lacks a good update and maintenance mechanism

There are certain limitations in the traditional way of updating place name and address data, such as the POI data update method, the use of POI for address matching update is greatly affected by the current situation of the data; while the method of collecting place name and address based on the mobile terminal has a high investment cost, which is difficult for foreign businesses. The professional quality requirements of the collection personnel are relatively high

If only the place name and address matching database is built, without updating and maintenance in the later stage, the current situation of the matching database cannot be guaranteed, which will affect the accuracy of the final place name and address matching. Therefore, it is necessary to establish a low-cost, short-period data dynamic update method

[0009] (3) The current place name address matching method is not compatible with the multi-dimensional and fine-grained nature of the address

Most of the existing place name and address matching libraries are organized and matched according to the tree structure, but basically do not reach the split granularity of buildings and households, and cannot match more accurately

Most of the currently established place name and address databases are matched with point-shaped element information, which cannot match lines (roads) and planes (administrative districts), without considering the multi-dimensionality of place names and addresses, and cannot perform effective matching

Images