An Approach to Handling Spatio-Temporal Contexts on Linked Open Data for Points of Interest

Abstract

Many people and organizations, including governments, for example, publicly disseminate data in various ways and for various reasons. For publishing data on the Semantic Web, the Linked Data principles have become the norm. However, numerous difficult jobs and problems must be resolved while linking the vast amount of data by utilizing Linked Open Data (LOD) sources. This thesis outlines our contributions to understanding spatiotemporal contexts to address those issues. Creating various housing options and opportunities is one of the core ideas behind smart growth. As a result, geographical elements like Points of Interest (POIs) affected the real estate market and buyers' choices. Using data from diverse sources, we provide a domain-specific strategy in this work. When estimating instance value for POIs, this work attempts to handle spatiotemporal contexts. We refer to the POI evaluation events that include spatiotemporal notions of the target real estate site as spatio-temporal contexts. We gather the data and format it into a common format to forecast the instance value of an object. Then, we construct a domain-specific ontology to evaluate the application of control in city planning. The instances are then filled using Federated SPARQL Query using data from Endpoints. The design of our prototype system to manage spatio-temporal contexts on LOD for POIs has now been prepared and put into practice.