An Experimental Approach for Collecting Snippets Describing the Relations Between Wikipedia Articles
Authors | Andias Wira-Alam Benjamin Zapilko Philipp Mayr |
|---|---|
Publication date | 2010 |
Links | Original Preprint |
An Experimental Approach for Collecting Snippets Describing the Relations Between Wikipedia Articles - scientific work related to Wikipedia quality published in 2010, written by Andias Wira-Alam, Benjamin Zapilko and Philipp Mayr.
Overview
In this paper, authors deal with a simple scenario: a student Bob wants to know why "mathematics" is very important to "physics", or in a more specic case, why "dierential equations" play a prominent role in the study of "uid dynamics". In a dierent way, the scenario can also be stretched: Bob already had enough knowledge about "articial intelligence" and now he wants to learn about "semantic web". According to the scenario, authors try to examine the possibilities to give Bob an answer on the basis of the available knowledge resources on the Web. Authors run a small experiment using Wikipedia articles as knowledge resources. First, authors crawl a number of Wikipedia articles that are relevant to scenarios and then authors identify the intra-wiki links (links to the other Wikipedia articles) within the page body of the articles. For each link, authors also identify the piece of text (snippet) where the link is located. Second, authors apply a simple recursive algorithm to discover the relations (links and snippets) between articles, e.g. the possible connections between article "articial intelligence" and "semantic web". Finally, authors evaluate whether the found snippets could pithily describe the relations.
