Submissions/Improving Link Coverage in Wikipedia
After careful consideration, the Programme Committee has decided not to accept the below submission at this time. Thank you to the author(s) for participating in the Wikimania 2015 programme submission, we hope to still see you at Wikimania this July.
- Submission no.
- Title of the submission
- Improving Link Coverage in Wikipedia
- Type of submission (discussion, hot seat, panel, presentation, tutorial, workshop)
- E-mail address
- Country of origin
- Affiliation, if any (organisation, company etc.)
- Stanford Univeristy
- Wikimedia Foundation
- Personal homepage or blog
- Abstract (at least 300 words to describe your proposal)
With more than 33 million articles, Wikipedia is an immense body of knowledge that can be accessed through its articles. Wikipedia users who wish to learn about a concept usually start from one article and explore other articles as needed to gain more context about the topic of interest. One of the most common ways users explore other articles is through article links. However, it is not clear how much the current structure of article links supports users in finding material that they need to understand the context of a topic. This research proposes one way to measure how good is the current link structure and recommends a methodology for improving it.
Currently, the task of adding links to articles is left to the editors with different level of expertise in editing and knowledge about a topic. This results in underlinking and overlinking problems that can interfer with readers’ experience when accessing knowledge. Several methods have been proposed for adding missing links to Wikipedia, but they are all based on the static structure of the existing links. However, the ultimate purpose of a wiki-link is to aid navigation. Therefore we propose a way to use session logs in order to optimize for the article links considering how readers currently access Wikipedia pages. We present the result of an experiment in which we added links to articles and we show how by doing so we improve reader experience. We close by discussing the challenges the link prediction model faces moving forward and future research directions.
- Technology, Interface & Infrastructure
- Length of session (if other than 30 minutes, specify how long)
- 30 minutes
- Will you attend Wikimania if your submission is not accepted?
- Slides or further information (optional)
- Special requests
If you are interested in attending this session, please sign with your username below. This will help reviewers to decide which sessions are of high interest. Sign with a hash and four tildes. (# ~~~~).