Automatic Generalization of Residential Areas Based on “Paradigm” Theory and Big Data

“Paradigm” theory is an important ideological and practical tool for scientific research. The research means methods of Geographic Information Science follow the laws four paradigms. Automatic cartographic generalization not only key link map making, but also a recognized difficult hot issue. Based on large-scale data deep learning technology, automatic problem-solving model proposed in this paper. According to problems faced by residential area selection simplification, models simplification based big are constructed respectively, which provides new ideas schemes solve simplification.