Enhancing Wikipedia with semantic technologies

From AcaWiki
Jump to: navigation, search

Citation: Lian Hoy Lee, Christof Lutteroth, Gerald Weber (2011) Enhancing Wikipedia with semantic technologies. iUBICOM ’11: The 6th International Workshop on Ubiquitous and Collaborative Computing (RSS)

Download: http://www.bcs.org/upload/pdf/ewic_ubi11_paper3.pdf

Tagged: Computer Science (RSS) Wikipedia (RSS), semantic web (RSS), semantic technology (RSS)


This paper looks at user interfaces for querying, with an emphasis on ease of use. It presents their "platform for semantic interface enhancements" and shows three examples of enhancements they have made:

  • a range query, which is limited by the performance of the underlying database (triplestore); they suggest that histograms and graphs based on range queries would be particularly useful
  • an article recommender
  • a type-ahead interface like Google Instant

These were evaluated by having 10 people use them over a number of days; the range slider was found to be useful by 9 of the users who wanted to be able to do range search across the Web.

In the introduction and background for the paper, they also explore previous work using two main ideas for searching semantic networks:

  • graph pattern builders lik SPARQLExplorer and Disco
    • a subtype are relationship finders like Query Builder, Relationship Finder, and iSPARQL
  • faceted browser (e.g. Ontogator, OpenLink Software, Neofonie)

and introduce interface evaluation criteria from Lei 2006, who says interfaces should have:

  • a gradual learning curve
  • expressiveness
  • intuitive presentation of results
  • low response time

Faceted browsing, they say, is recommended by Hahn et al. 2010 as "the best way of introducing semantic querying capability to the user." They suggest complementing keyword searching, for instance where semantic data is stronger (e.g. formats of multimedia data). Other ideas are to use semantic data to "detect missing or broken links between articles in Wikipedia, and suggest to contributors on how they could be fixed"

They also note various novel applications that could be enabled by semantic technologies, including existing ones such as DBPedia Mobile (Bizer et al. 2009).

Selected references

  • C. Bizer, J. Lehmann, G. Kobilarov, S. Auer, C. Becker, R. Cyganiak, and S. Hellmann. Dbpediaa crystallization point for the web of data. Web Semantics: Science, Services and Agents on the

World Wide Web, 2009. ISSN 1570-8268.

Theoretical and practical relevance:

This paper is a useful source for an overview of existing interfaces for making semantic queries feasible for endusers.

Rangefinding deserves further exploration.