Submissions:2018/Discovering the Hidden Semantics of 'See Also' via Wikipedia-based Explicit Semantic Analysis
- Discovering the Hidden Semantics of 'see Also' via Wikipedia-based Explicit Semantic Analysis
- Theme (optional)
- Tech & Tools
- Type of submission
- Jacob Fong
- E-mail address
- Wikimedia username
- Affiliation(s) (optional)
Stanford Online High School, Class of '18
The Wikipedia Manual of Style describes the 'see also' as reflecting "links that would be present in a comprehensive article on the topic...", and explains the importance of common sense as the utmost principle in deciding the relevance of an article in ‘see also’. This presentation explores the extent to which 'see also' articles are thematically related to their parent articles as opposed to minor subjects present in them. I implement Wikipedia-based Explicit Semantic Analysis (Gabrilovich, Markovitch, 2007) and Minor-subject Relatedness as measures of the above opposed, respectively. The implications of my results are for a proposed automatic extension of 'see also' sections across Wikipedia based on the semantic analysis of related page populations. Further, I'll discuss a heuristic for the aforementioned automation.
Examples of Interesting 'See Also' Phenomena:
- One of the nine ‘see also’ under page San Francisco is a link to UCERF3, the California Earthquake Forecast, to which three paragraphs discussing the 1906 earthquake in San Francisco pertain vaguely. Does this make for a thematic relation or a Minor-subject relation, and should a heuristic include this kind of page?
- The Magnitsky Act was not a part of Sergei Magnitsky's 'see also'. Does this make for a thematic relation or a Minor-subject relation, and should a heuristic include this kind of page?
- Length of presentation
- 15-30 min
- Special requests
- Thunderbolt-compatible projector
- Preferred room size
- Have you presented on this topic previously? If yes, where/when?
- If you will be incorporating a slidedeck during your presentation, do you agree to upload it to Commons before your session, with a CC-BY-SA 4.0 license, including suitable attribution in the slidedeck for any images used?
- Will you attend WikiConference North America if your submission is not accepted?
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 four tildes. (~~~~).
- Add your username here.