Soft Querying Features in GeoJSON Documents: The GeoSoft Proposal

Abstract

Since the advent of textit{JSON} as a popular format for exchanging large amounts of data, a novel category of NoSQL database systems, named textit{JSON}document stores, has emerged for storing textit{JSON} data sets; in fact, these novel databases are able to natively manage collections of textit{JSON} documents. To help analysts and data engineers query and integrate textit{JSON} data sets persistently saved in textit{JSON} document stores, the J-CO Framework has been developed (at the University of Bergamo, Italy): it is built around a novel query language, named J-CO-QL^{+}, that provides sophisticated features, including soft-querying capabilities. However, J-CO-QL^{+} (as the other languages for querying textit{JSON} data sets) is designed to be general purpose; consequently, it can be cumbersome for users to apply it on specific data formats. This is the case of GeoJSON, a specific and popular textit{JSON} data format that is designed to represent geographical information layers. This paper presents the latest evolution of GeoSoft, a novel high-level “domain-specific language” that is specifically designed to express complex queries on the GeoJSON documents, including soft-queries. GeoSoft is inspired to the classical SQL language, so as to reduce the learning curve of potential users. GeoSoft queries are translated into J-CO-QL^{+} scripts, to be actually executed.