Preference-based top-k spatial keyword queries

  • Authors:

  • Jinzeng Zhang;Dongqi Liu;Xiaofeng Meng

  • Affiliations:

  • Renmin University of China, Beijing, China;Renmin University of China, Beijing, China;Renmin University of China, Beijing, China

  • Venue:
  • Proceedings of the 1st international workshop on Mobile location-based service
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

With proliferation of geo-positioning and geo-tagging, spatial keyword query has been an attractive and challenging topic that blooms various interesting applications in spatial databases. However, spatial keyword queries in those fields do not always satisfy users requirements. One reason for those queries contain fuzzy linguistic labels that often embody users preference, e.g., finding hotels near Zhongguancun with moderate price. In this paper, we define Preference-based Top-k Spatial Keyword (PTkSK) query for coping with fuzzy labels. Existing works only consider on spatial locations, textual descriptions and overlook users preference. In order to answer (PTkSK) queries efficiently, we propose a hybrid index structure called AIR-tree(Attribute-Inverted file R-tree), which combines location proximity with preference similarity and textual relevance. Given a query with fuzzy labels, we propose an efficient strategy of computing users preference to transform these labels into the [0,1] interval, and design two effective search methods to find relevant objects for satisfying users requirements. Our experiments on synthetic and real datasets demonstrate that our proposed methods are scalable and efficient.