11. Geospatial Data Mining and Analytics for Real-Estate Applications

Market information on housing is relevant for good decision making of households as well as for real estate economics and is also of systemic interest. The official statistics often provide only annual indices on national level based on transactions of previous year. The stakeholders of the real estate markets however request analysis of higher spatial resolution on local level based recent transactions. This paper focuses on methodology of automated data acquisition, analysis and visualization of data from the housing market. Data retrieved from observing real estate markets (to rent/to buy) are used for statistical modelling of value-descriptive parameters, for estimating and forecasting real estate market fundamentals, which also can reveal market risks. This paper elaborates methods of automated data acquisition based on web mining, reviews methods of econometric spatial modelling with impact from proximity in order to deduct efficiency parameters within different categories. The analysis and the resulting visualization at different granularity of spatial, temporal and typological effects provides relevant information for decision making.