Change search
ReferencesLink to record
Permanent link

Direct link
Troll hunting in the information age
Umeå University, Faculty of Science and Technology, Department of Computing Science.
2015 (Swedish)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

This thesis presents an attempt to use linear regression to predict the number of authors who contributed to a text, in this case a forum. To do this a set of features which is usually used in authorship attribution has been applied. After a functionality for feature extraction was implemented a model was trained on how the number of authors affects the richness of a text. This model is then used to predict how many people contributed to a there-to unseen text. The result of this work shows that it works rather well to guess the number of authors who contributed to a text if the number of authors are fairly small, but as the number of authors grow, it gets harder to guess the number of authors. Possibly because the variance does not grow enough linear for large number of authoers.

Place, publisher, year, edition, pages
2015. , 29 p.
, UMNAD, 1027
National Category
Engineering and Technology
URN: urn:nbn:se:umu:diva-108375OAI: diva2:852731
Educational program
Bachelor of Science Programme in Computing Science
Available from: 2015-09-10 Created: 2015-09-10 Last updated: 2015-09-10Bibliographically approved

Open Access in DiVA

fulltext(269 kB)49 downloads
File information
File name FULLTEXT01.pdfFile size 269 kBChecksum SHA-512
Type fulltextMimetype application/pdf

By organisation
Department of Computing Science
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 49 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 269 hits
ReferencesLink to record
Permanent link

Direct link