Research And Application On Data Model Of The Real Estate Big Data

Spatial big data management has always been one of the research focuses in GIS field.Real estate registration data is a typical spatial big data,and real estate registration data model is the core of the basic platform of real estate registration information management.This paper takes the data of China's national real estate registration information management basic platform as the research object,aiming to propose a kind of high-order data model that adopts the coexistence of relational and non-relational data storage modes,and improve the query efficiency of big data of the national real estate registration information management basic platform.First of all,this paper introduces the research background and significance,and investigates the research status of real estate registration and data model at home and abroad.By analyzing the current situation,the existing problems are sorted out.The research content and technical route are clarified.Data is a symbolic representation of objective things,and models are abstractions of the real world and data characteristics.Therefore,we analyze the characteristics of real estate data before the model design,mainly from the data source,data type,data quantity estimation and data characteristics.Secondly,the data model is the foundation of the database.The development of spatial database technology has experienced five periods,including file system,file and relational database hybrid management system,spatial database engine,object-relational spatial database management system,and big data management technology.Big data management technology is applied to the design of real estate data model,especially big data processing technology,big data storage technology and big data query technology.Then,based on two design ideas to design.One is organic integration of relational and non-relational,the other is to use ETL and ElasticSearch to improve query efficiency.The design process is to clarify problems,design ideas and principles,conceptual model research,logical model design and optimization,physical model design,model testing and verification.Based on the existing conceptual model,the data logic model is divided into summary data area and Shared data area.Combining the characteristics of summary data area and Shared data area,the logical organization structure is optimized according to the application needs.Physical model is designed for network and user needs.Through the optimization of organizational logic structure and the application of big data technology,the performance of business system is greatly improved.The validity of the new model is verified by the prototype system.Finally,the real estate big data model meets the application needs of the national real estate registration information management basic platform,and the establishment of the model provides reference value for the construction of real estate big data platform in various Provinces and cities.