Discovering geographical topics in the twitter stream

By Liangjie Hong, Amr Ahmed, Siva Gurumurthy, Alex Smola, Kostas Tsioutsiouliklis. 2012

410 citations recorded by the source corpus.

Micro-blogging services have become indispensable communication tools for online users for disseminating breaking news, eyewitness accounts, individual expression, and protest groups. Recently, Twitter, along with other online social networking services such as Foursquare, Gowalla, Facebook and Yelp, have started supporting location services in their messages, either explicitly, by letting users choose their places, or implicitly, by enabling geo-tagging, which is to associate messages with latitudes and longitudes. This functionality allows researchers to address an exciting set of questions: 1) How is information created and shared across geographical locations, 2) How do spatial and linguistic characteristics of people vary across regions, and 3) How to model human mobility. Although many attempts have been made for tackling these problems, previous methods are either complicated to be implemented or oversimplified that cannot yield reasonable performance. It is a challenge task to discover topics and identify users' interests from these geo-tagged messages due to the sheer amount of data and diversity of language variations used on these location sharing services. In this paper we focus on Twitter and present an algorithm by modeling diversity in tweets based on topical diversity, geographical diversity, and an interest distribution of the user. Furthermore, we take the Markovian nature of a user's location into account. Our model exploits sparse factorial coding of the attributes, thus allowing us to deal with a large and diverse set of covariates efficiently. Our approach is vital for applications such as user profiling, content recommendation and topic tracking. We show high accuracy in location estimation based on our model. Moreover, the algorithm identifies interesting topics based on location and language.

Subfields: Epidemiology, Statistical and Nonlinear Physics, Transportation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Epidemiology

In Statistical and Nonlinear Physics

In Transportation

Citations in the corpus

Is the Sample Good Enough? Comparing Data from Twitter's Streaming API with Twitter's Firehose (2013, background): “Finally, (Yin et al. 2011; Hong et al. 2012; Pozdnoukhov and Kaiser 2011) all study the problem of identifying topics in geographical Twitter datasets, proposing models to extract topics relevant to different geographical areas in the data.”
Geolocation Prediction in Social Media Data by Finding Location Indicative Words (2012, methodology): “…been applied to the study of geospatially-related tasks including user dialect (Eisenstein et al., 2010), topic discovery (Yin et al., 2011), object matching (Dalvi et al., 2012), factorisation of different geospatial features (Hong et al., 2012), and spatial-temporal analysis (Bauer et al., 2012).”
Jointly modeling aspects, ratings and sentiments for movie recommendation (JMARS) (2014, methodology): “Our approach borrows from [7] by designing an exponential additive model in terms of θu and θm.”
Collaborative Filtering beyond the User-Item Matrix (2014, background): “Work by Hong et al. [2012] has proposed a new algorithm to discover geographical topics from for location recommendation.”
Text-Based Twitter User Geolocation Prediction (2014, neutral): “, 2011) or topic regions (Eisenstein et al., 2010; Hong et al., 2012).”
Regions, Periods, Activities: Uncovering Urban Dynamics via Cross-Modal Representation Learning (2017, background): “Geographical topic discovery [31, 19, 35, 15, 36, 24, 34] aims at modeling the topics in different regions.”
SAR: A sentiment-aspect-region model for user preference analysis in geo-tagged reviews (2015, methodology): “There also exist proposals [5], [27], [24] that model user preferences to geographical topics based on geo-tagged tweets.”
A location-based news article recommendation with explicit localized semantic analysis (2013, background): “Hong et al. proposed a generative model to discover geographical topics in the tweet stream [9].”
Fine-grained location extraction from tweets with temporal awareness (2014, background): “considered the user’s interest to model users’ geographic behavior [6].”
Learning geographical preferences for point-of-interest recommendation (2013, methodology): “Also, people have used gaussian geographical distribution to model region over locations [20, 25, 10].”
Geolocation for Twitter: Timing Matters (2016, background): “Geolocation – the task of identifying a social media message’s location – can support a variety of downstream applications, such as advertising, personalization, event discovery, trend analysis and disease tracking (Watanabe et al., 2011; Hong et al., 2012; Kulshrestha et al., 2012; Broniatowski et ”
Semi-supervised User Geolocation via Graph Convolutional Networks (2018, background): “…2010); (2) geographical topic models that learn region-speciﬁc topics, but don’t scale to the magnitude of social media (Eisen-stein et al., 2010; Hong et al., 2012; Ahmed et al., 2013); and (3) supervised models which are often framed as text classiﬁcation (Serdyukov et al., 2009; Wing and…”