Home Location Identification of Twitter Users

Abstract

We present a new algorithm for inferring the home location of Twitter users at different granularities, including city, state, time zone or geographic region, using the content of users' tweets and their tweeting behavior. Unlike existing approaches, our algorithm uses an ensemble of statistical and heuristic classifiers to predict locations and makes use of a geographic gazetteer dictionary to identify place-name entities. We find that a hierarchical classification approach, where time zone, state or geographic region is predicted first and city is predicted next, can improve prediction accuracy. We have also analyzed movement variations of Twitter users, built a classifier to predict whether a user was travelling in a certain period of time and use that to further improve the location detection accuracy. Experimental evidence suggests that our algorithm works well in practice and outperforms the best existing algorithms for predicting the home location of Twitter users.

Exploring Social Media for Event Attendance

“…Many papers tackle the problem of estimating the current location of users or their home from non geo-located tweets [2], [3], [4], [5], [6]. Compared to these proposals, we have a different objective as we do not want to estimate the user's exact location at the time of the post, but classify the single posts on the basis of the user's future, current and past attendance to a given event.…”

Modeling Twitter Engagement in Real-World Events

“…First, we infer the location directly from the user's event-related tweets if the present location is mentioned/attached. Otherwise, we infer the home location from the user's profile and tweets using the methods mentioned in [32] to infer the geolocations of Twitter users. We then verify the extracted location information with the diurnal patterns of the user's tweets [33].…”

FamilyID: A Hybrid Approach to Identify Family Information from Microblogs

“…Knowledge discovery in social networks is a hot research area. For instance, methods have been proposed to identify user attributes, such as gender, age, location [12], location type [9], activities [6], personalities [15], etc. There are also proposals to make predictions based on information and activities in social networks, e.g., to predict stock rates based on user tweets [3].…”