%0 Conference Proceedings %T On the Volume of Geo-referenced Tweets and Their Relationship to Events Relevant for Migration Tracking %+ Austrian Institute of Technology [Vienna] (AIT) %+ Federal Ministry of Interior [Austrian Federal Government] (BM.I) %A Neubauer, Georg %A Huber, Hermann %A Vogl, Armin %A Jager, Bettina %A Preinerstorfer, Alexander %A Schirnhofer, Stefan %A Schimak, Gerald %A Havlik, Denis %Z Part 7: Analytics and Visualization %< avec comité de lecture %( IFIP Advances in Information and Communication Technology %B 11th International Symposium on Environmental Software Systems (ISESS) %C Melbourne, Australia %Y Ralf Denzer %Y Robert M. Argent %Y Gerald Schimak %Y Jiří Hřebíček %I Springer %3 Environmental Software Systems. Infrastructures, Services and Applications %V AICT-448 %P 520-530 %8 2015-03-25 %D 2015 %R 10.1007/978-3-319-15994-2_53 %K migration %K tweets %K geolocation %K early indicator %K push factor %Z Computer Science [cs]Conference papers %X Migration is a major challenge for the European Union, resulting in early preparedness being an imperative for target states and their stakeholders such as border police forces. This preparedness is necessary for multiple reasons, including the provision of adequate search and rescue measures. To support preparedness, there is a need for early indicators for detection of developing migratory push-factors related to imminent migration flows. To address this need, we have investigated the daily number of geo-referenced Tweets in three regions of Ukraine and the whole of Japan from August 2014 until October 2014. This analysis was done by using the data handling tool Ubicity. Additionally, we have identified days when relevant natural, civil or political events took place in order to identify possible event triggered changes of the daily number of Tweets. In all the examined Ukrainian regions a considerable increase in the number of daily Tweets was observed for the election day of a new parliament. Furthermore, we identified a significant decrease in the number of daily Tweets for the Crimea for the whole examined period which could be related to the political changes that took place. The natural disasters identified in Japan do not show a clear relationship with the changes in the degree of use of the social media tool Twitter. The results are a good basis to use communication patterns as future key indicator for migration analysis. %G English %Z TC 5 %Z WG 5.11 %2 https://inria.hal.science/hal-01328602/document %2 https://inria.hal.science/hal-01328602/file/978-3-319-15994-2_53_Chapter.pdf %L hal-01328602 %U https://inria.hal.science/hal-01328602 %~ IFIP %~ IFIP-AICT %~ IFIP-TC %~ IFIP-TC5 %~ IFIP-WG %~ IFIP-AICT-448 %~ IFIP-WG5-11 %~ IFIP-ISESS