作者: Jean-Claude Thill , Mikhail Mozolin
DOI: 10.1007/978-3-642-59787-9_17
关键词: Heuristics 、 Spatial Interaction Model 、 Feedforward neural network 、 Trustworthiness 、 Artificial intelligence 、 Artificial neural network 、 Underpinning 、 Machine learning 、 Regression 、 Spatial interaction
摘要: Though it has often been criticized for providing too crude a rendition of processes underpinning revealed patterns interaction between geo-referenced entities, spatial modelling persisted as one the methodological pillars several sciences, including regional science, geography and transportation (Fotheringham O’Kelly 1989; Ortuzar Willumsen 1994; Sen Smith 1995; Isard et al. 1998). Traditionally, model is calibrated by well known fitting optimization techniques, leastsquares regression, maximum likelihood, or numerical heuristics.