Geocoding of traffic-related events from Twitter

Juan C. Salazar, Miguel Torres-Ruiz, Clodoveu A. Davis, Marco Moreno-Ibarra

Research output: Contribution to journalConference articlepeer-review

2 Scopus citations

Abstract

Nowadays social networks provide information with high correlation with events that are occurring in the worldwide. Twitter is a microblogging network of real time posts in which people know different classes of events such as concerts, festivals, demonstrations, etc. Other relevant topic is traffic congestions; user-generated content is useful to assist drivers in avoiding crowded areas. Therefore, this work is oriented towards following specific steps focused on improving the geocoding of traffic-related events that are associated with a number of geographic elements. Preliminary results have increased the precision and recall of locating geographic elements, achieving 85% and 83%, from a baseline of 36% and 30% respectively.

Original languageEnglish
Pages (from-to)14-25
Number of pages12
JournalProceedings of the Brazilian Symposium on GeoInformatics
Volume2015-November
StatePublished - 2015
Event16th Brazilian Symposium on GeoInformatics, GEOINFO 2015 - Campos do Jordao, Brazil
Duration: 29 Nov 20152 Dec 2015

Fingerprint

Dive into the research topics of 'Geocoding of traffic-related events from Twitter'. Together they form a unique fingerprint.

Cite this