Matches in ScholarlyData for { <https://w3id.org/scholarlydata/inproceedings/www2012/paper/1261> ?p ?o. }
Showing items 1 to 16 of
16
with 100 items per page.
- 1261 creator alex-smola.
- 1261 creator amr-ahmed.
- 1261 creator kostas-tsioutsiouliklis.
- 1261 creator liangjie-hong.
- 1261 creator siva-gurumurthy.
- 1261 type InProceedings.
- 1261 label "Discovering Geographical Topics from Twitter Streams".
- 1261 sameAs 1261.
- 1261 abstract "Micro-blogging services have become indispensable communication tools for online users for disseminating breaking news and eyewitness accounts, and even for organizing flash mobs and protest groups. For instance, Twitter was heavily used in a number of events and emergencies, ranging from elections, earthquakes and tsunamis, as well as playing as an instrumental role in facilitating political upheavals in the Middle East. Recently, micro-blogging services like Twitter, along with other location sharing services such as Foursquare, Gowalla and Facebook Places, have started to support users to specify their location in messages, either explicitly, by letting users to choose their place, or implicitly, by enabling geo-tagging functionality. Such functionalities present an important opportunity to answer several questions: 1) how information is created and shared across geographical locations, 2) how spatial and linguistic characteristics of people vary across regions, and 3) how to model human mobility. It is a challenge to discover topics and identify users’ interests from these geo-tagged messages due to the sheer amount of data and a diversity of language variations used on Twitter. Although some prior work has investigated on how location information can be used to better understand patterns in social photo sharing services, models developed for such data are usually limited and cannot easily be applied to content-rich social media. In addition, user preferences are not completely considered in previous models. In this paper, we address the problem of modeling geographical topical patterns on Twitter by introducing a novel sparse topic model, which utilizes both statistical topic models and sparse coding techniques to provide a principled method for uncovering different language patterns and common interests shared across the world. Our approach is vital for applications such as user profiling, content recommendation and topic tracking and the method can be easily extended in a number of ways. We show interesting topics identified by the model and demonstrate its effectiveness on the task of predicting locations of new messages and outperform non-trivial baselines.".
- 1261 hasAuthorList authorList.
- 1261 isPartOf proceedings.
- 1261 keyword "Geographical Modeling".
- 1261 keyword "Sparse Modeling".
- 1261 keyword "Topic Models".
- 1261 keyword "Twitter".
- 1261 title "Discovering Geographical Topics from Twitter Streams".