Affiliation:
1. Systems and Computing Department Federal University of Campina Grande Paraíba Brazil
Abstract
AbstractTo alleviate challenges related to mobility, travel planning, and time management, individuals commonly encounter the necessity of locating points of interest (POIs) that either share the same physical building or are situated within interconnected buildings. However, existing search engines face difficulties in retrieving groups of places based on keywords and topological relations among their respective regions, for instance, when a user wants to find a residential building that is connected to a green area. This is primarily due to their limited consideration of POIs as mere points in space, rather than recognizing the polygonal geometries of their boundaries. In this work, we present a spatial search solution based on textual and topological query parameters for efficiently retrieving groups of POIs whose neighborhoods and boundaries satisfy specific restrictions. To perform this type of search efficiently, we propose the Topo‐MSJ algorithm, building upon the well‐established “Multi‐Star‐Join” (MSJ) algorithm, by including an efficient approach to handle queries with topological requirements. To assess the effectiveness of our solution, we conduct a performance evaluation by comparing the execution time of Topo‐MSJ with equivalent spatial SQL queries. The experimental analysis, performed on real datasets, reveals that Topo‐MSJ exhibits a faster execution time compared with equivalent SQL queries, additionally providing a simplified spatial pattern query notation.