- 310 views
- 514 downloads
An Interaction-driven Approach for Inferring the Polarity of Collaborations in Wikipedia and Political Preferences on Twitter
-
- Author / Creator
- Makazhanov, Aibek
-
In this thesis we explore interactions of users of two major information sources, namely Wikipedia and Twitter. In particular, we show that revision histories of Wikipedia articles contain interaction patterns which can be used to build collaboration profiles of editors. Such profiles can be classified as positive or negative, corresponding to productive or counter-productive collaboration. Additionally, profiles can be applied to a number of related tasks, such as predicting votes in Wikipedia administrator elections or detecting controversial articles. We further extend the ideas behind collaboration profiles and adapt the approach to the problem of predicting political preference of Twitter users. By considering tweeting on a party-specific topic as a form of user-party interaction, we show that a record of such interactions, produced by a user during an election campaign, can be used to predict her political preference. Additionally, we analyze how does predicted political preference of different groups of users change over time. Our results suggest that politically active users are less likely to change their preference during the course of an election campaign.
-
- Subjects / Keywords
-
- Graduation date
- Spring 2013
-
- Type of Item
- Thesis
-
- Degree
- Master of Science
-
- License
- This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.