SWIFT WARM ACQUAINTANCE EXPLORATION WITH KEYWORDS

Abstract

conventional spatial queries, including range seek and nearest neighbor retrieval, contain only conditions on gadgets’ geometric properties. Today, many modern applications name for novel types of queries that intention to locate items enjoyable each a spatial predicate, and a predicate on their related texts. As an example, in preference to considering all of the eating places, a nearest neighbor query might as an alternative ask for the eating place this is the closest amongst the ones whose menus comprise “steak, spaghetti, brandy” all on the equal time. Currently, the first-class approach to such queries is based totally on the ir²-tree, which, as shown on this paper, has some deficiencies that severely impact its efficiency. Inspired by this, we increase a brand new access technique referred to as the spatial inverted index that extends the conventional inverted index to deal with multidimensional information, and comes with algorithms that may answer nearest neighbor queries with keywords in real time. As confirmed by means of experiments, the proposed techniques outperform the ir²-tree in query reaction time extensively, often with the aid of a thing of orders of significance.