Enhancing the interface for ontology-supported homonym search

Tian Tian, James Geller, Soon Ae Chun

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

Keyword-based search engines face a complication when a search term is a homonym, that is, a word with multiple meanings. In a pervious study, we have built an Ontology-Supported Web Search system to categorize the homonymous terms and display their suggested completions separately. The suggestions of the homonyms are extracted from an ontology developed in this research. In this paper, we present new ways of displaying homonymous search results. To help the user better understand and differentiate the homonymous terms, we display snippets of search results belonging to the suggested completions separately as follows. For every homonymous sense, the snippets are in a separate pane, and panes are tiled horizontally, i.e., are next to each other. In addition, we present an improvement of Google's instant feature, in that snippets are refreshed as soon as a user hovers with the mouse on a suggested completion for more than a certain period, e.g. two seconds.

Original languageEnglish (US)
Title of host publicationAdvanced Information Systems Engineering Workshops - CAiSE 2011 International Workshops, Proceedings
PublisherSpringer Verlag
Pages544-553
Number of pages10
ISBN (Print)9783642220555
DOIs
StatePublished - 2011

Publication series

NameLecture Notes in Business Information Processing
Volume83 LNBIP
ISSN (Print)1865-1348

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Management Information Systems
  • Business and International Management
  • Information Systems
  • Modeling and Simulation
  • Information Systems and Management

Keywords

  • Homonyms
  • Improvement of search experience
  • Ontology
  • Ontology-supported Web search
  • Suggested completions
  • Web search with homonyms

Fingerprint

Dive into the research topics of 'Enhancing the interface for ontology-supported homonym search'. Together they form a unique fingerprint.

Cite this