Abstract
With the proliferation of the mobile use of the web, spatial keyword query (SKQ) services are gaining in importance. However, state-of-the-art SKQ systems do not provide systematic functionality that allows users to ask why some known object is unexpectedly missing from a query result and do not provide an explanation for such missing objects. In this demonstration, we present a system called YASK, a whY-not question Answering engine for Spatial Keyword query services, that is capable of answering whynot questions posed in response to answers to spatial keyword topk queries. Two explanation and query refinement models, namely preference adjustment and keyword adaption, are implemented in YASK. The system provides users not only with the reasons why desired objects are missing from query results, but provides also relevant refined queries that revive the expected but missing objects. This demonstration gives attendees hands-on experience with YASK through a map-based GUI interface in which attendees can issue spatial keyword queries, pose why-not questions, and visualize the results.
Original language | English |
---|---|
Pages (from-to) | 1501-1504 |
Number of pages | 4 |
Journal | Proceedings of the VLDB Endowment |
Volume | 9 |
Issue number | 13 |
DOIs | |
Publication status | Published - Sept 2016 |
Event | 42nd International Conference on Very Large Data Bases, VLDB 2016 - New Delhi, India Duration: 5 Sept 2016 → 9 Sept 2016 |
Scopus Subject Areas
- Computer Science (miscellaneous)
- General Computer Science