A Spatio-Temporal Geocoding Model for Vector Data Integration

Vector data integration is an important function in Urban Public Participation GIS Platform (UPPGP). Most current researches drill down the issue without considering two points: (1) the inner connections among different urban elements. (2) The temporal meaning of each object. The neglect of these points causes redundant storage and inefficient retrieval problems in smart city applications. In view of that, a spatio-temporal geocoding model for vector data integration is proposed in this paper. The model regards the urban entity element as the bridge between economic element and event element, so the task turns to find a way to uniquely identify the urban entities to avoid ambiguity and redundancy when entity objects connect with other type of objects during integration. Based on the object-oriented spatio-temporal data model, the entity object is constructed by type, space and time codes using concept lattice and regional GeoHash technologies. The method computes code similarity for each entity object to decide whether to put the object into storage. Experiments on the real UPPGP of Sino-Singapore Tianjin Eco-city show that it can avoid data redundancy and ambiguity effectively.