Research On Real Estate Big Data System

As a basic production element,land plays an important role in people's social and economic life.In order to better satisfy the requirements of social development and meet public needs,it is required that the precision management ability in land management should be incessantly improved together with the quality of land management.With the establishment of a unified registration system for real estate,data on people,land and housing is for the first time integrated during unified registration of real-estate data.However,real estate data registration lacks necessary social and economic attributes.The advent of the era of big data enables an effective complement of the dynamic acquisition of social and economic attribute data,and also creates important opportunities for the refinement of land management,thus making the big data decision analysis by land management department possible on the basis of real estate data registration.This paper constructs a real estate big data system by taking real estate data as the core,adopting big data technology,and integrating social and economic attributes,which further creates big data sets conducive to decision-making support for land management.Firstly,based on the land management theory,the paper deeply analyzes the interaction among three factors,namely people,land and property,and the influence of interaction on land management,clarifies that big data of real estate shall take registration data of real estate as the basis,collect information of three facets,namely "people,land and property," use data with enough technology and integrated with social and economic attributes such as web spider,geographical coding,etc.,and constructs structured big data set on real estate with time-space characteristics.Based on this,the paper sorts out content on people,land and property in real estate big data in detail from the aspects of data on natural person,legal representative,land basic information,land transaction,land management and utilization,property,building,property transaction,etc.,and then comes up with the framework of real estate big data analysis system.Secondly,with focus on the construction of real estate big data databases,the construction of big data systems is realized in this paper from four aspects,namely data collection,data fusion,data warehousing and data updating.The collection of the registration data of real state concerning people,housing,and land is analyzed specifically in this paper,and the application of web crawling framework Scrapy when collecting data with social and economic attributes is discussed,and two different data fusion schemes,namely spatial and attribute fusion,are proposed.Through the establishment of E-R models of people,land and housing,data table structures of real estate big data databases is formed,thus realizing the dynamic update of real estate big data according to the characteristics of data sources.Finally,this paper analyzes and deeply explores the relationship between people and property in the real estate big data set in Chengdu City.In terms of population structure,slight discrepancy exists in gender distribution among the population of the study area while the distribution varies considerably among different age groups,with the largest proportion of the group aged between 30 and 39 years.In terms of property spatial distribution and within spatial scale of the sub-district office,the great discrepancy exists in the number of registration and the distribution of the average registration price for one house,two houses or more than two different types of houses in the study area.The study demonstrates that real estate big data system can acquire the distinct characteristic of population structure change and the spatial distribution difference of property as well as the interaction between people and house,so it is of great significance in research and application in land management.