Skip to content

LanguageMachines/LamaEvents

Repository files navigation

About Lama Events

Lama Events is a calendar application listing events in the near future. The events are detected and selected by a fully automatic procedure in the Dutch Twitter stream (courtesy of Twiqs.nl). Tweets referring to the same future events are clustered based on the frequent co-occurrence of words (names, phrases) and temporal expressions that characterize the event. The date and time of the event is automatically determined based on direct and indirect time references in the texts of the tweets in a cluster. The demo shows a day-by-day ranked list of automatically detected events in the Dutch language area (Netherlands and Flanders).

Lama Events is different from other future calendar applications because of its fully automatic operation, and because of its broad focus on any social events. It does not only list big events such as sports games, concerts, or iPhone releases, but it will also catch events that no other calendar lists: smaller-scale regional events.

Lama Events makes mistakes! It occasionally projects a past event into the future, and sometimes groups together tweets that do not refer to the same event. The estimated date and the estimated time in hours:minutes may differ. Lama event is part of ongoing research at the Language and Speech Technology group of the Centre for Language Studies at Radboud University Nijmegen.

Lama Events was developed by Mustafa Erkan Başar (@me_basar)(programming, design) on the basis of research by Florian Kunneman (@Flowian7) and Ali Hürriyetoğlu (@hurrial) as part of the ADNEXT (@Adnext_Commit) work package of the COMMIT/ Infiniti-project (@COMMIT_nl) project, with contributions from Nelleke Oostdijk, Antal van den Bosch (@antalvdb), Maarten van Gompel (@proycon), and Erik Tjong Kim Sang (@tkserik). Erkan Başar's work was made possible by the Erasmus programme.

Link: http://lamaevents.cls.ru.nl/

About

Lama Events is a calendar application listing events in the near future. The events are detected and selected by a fully automatic procedure in the Dutch Twitter stream.

Topics

Resources

Stars

Watchers

Forks

Packages

No packages published