REGlobe is a virtual globe based real estate web app which provides an informational environment for house buyers to understand, compare and contrast the investment value of real estate. It mainly includes two models: one is predicting the community house price through public infrastructure distribution, the other is predicting the specific house price through VGI data.

The first model evaluates the community house price using the surrounding POIs (school, hospital, park, market, entertainment, subway), roads, business districts etc. Through machine learning, we get the geographical relevance of house prices and predict community house prices according to the surrounding public infrastructure distribution.

The second model evaluates the specific house price using VGI data. These data are collected by house sellers, real estate agents, and house buyers, such as house floor, house orientation, house inner panorama, house decoration etc. We calculate a house score through these data and revise the first model of house price assessment for the specific house in a community.