| Since the provisional Regulations on The Registration of Real Estate was officially implemented on March 1,2015,the unified registration of real estate in China has been carried out in an all-round way.Before this,land registration and house registration were registered by different competent departments,and there was a backward state of decentralized management of registration data and imbalanced informatization level.Such decentralized registration brought difficulties to the development of real estate registration.Therefore,it is imperative to establish a unified standard real estate information management platform.At the same time,the real estate information management platform database migration and integration technology is the key link.Based on the real estate registration stock data integration and migration project in Panyu District,Guangzhou(Panyu Real Estate database migration and integration),this paper sorted out the migration and integration content of real estate database and studied the technology of database migration and integration.The main contents are as follows:(1)The theory of real estate registration and database construction are sorted out.Firstly,the data structure to be migrated is analyzed,the database standard is formulated,and the technical process of database migration is designed according to the migration rules.Secondly,the types of real estate registration data after migration are sorted out,and the ownership data,registration data and archive data are integrated respectively,and the association relationship is established.In terms of database construction,Oracle is used for registration database,SQL is used as database language,and the spatial data platform is developed based on Arc GIS.(2)In view of the low efficiency of manual matching in the genus association of data integration graph,two address similarity matching algorithms,namely "Difflib" and "cosine similarity",are proposed by using the existing multi-source spatial data of registration,and the effective matching between the spatial data of registration and the "address" based algorithm is realized in Python language.Single matching and batch matching are realized,and the accuracy of the two algorithms and manual matching is evaluated by the accuracy rate.Through the research of this technology,the workload of manual matching is reduced and the efficiency of data integration is improved.(3)In view of the problem that the addresses of some registered data are missing and fuzzy and difficult to be located in data integration,which makes it difficult to land on the map,the "address" matching method is used to complete the land on the map.A set of technical route of address matching is designed,which uses four-way reasoning,personal visit,GIS spatial positioning,social data mining and Internet search to analyze the complex situation of the current situation and match the exact address of the desired land.In the later stage,the work of settlement is finally completed through field survey and visit.Through the application of this technology to solve the address missing,fuzzy data difficult to settle the problem.Stock of this topic by guangzhou panyu district real estate registration data integration migration project as the background,multi-source data for this article research figure is related to match the "address" technology and the "address" matches the fall of the practice,to explore the key technology of real estate database migration and integration,solved the difficult problems in integration process,improve the efficiency of integration,It reduces the workload of operators and achieves the goal of county and district real estate database migration and integration better.There are 22 figures,8 tables and 62 references in this thesis. |
PDF Full Text Request