2026-05-02T13:25:57Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/4801 2021-12-13T02:29:37Z 1582963302567:1597824273898 user-ucsy

Content Driven Tweets Annotation during Natural Disasters Win, Si Si Mar Aung, Than Nwe Nowadays, Twitter, Social Networking Site,becomes most popular microblogging service andpeople have started publishing data on the use of it innatural disasters. Twitter has also created theopportunities for first responders to know the criticalinformation and work effective reactions to impactedcommunities. This paper presents the automatedannotation system that can detect the tweets whichcontain critical information or not. Annotation is doneat tweet level with three labels by using the publiclyavailable annotated datasets. LibLinear classifier isused to build a model for automatic tweets annotation.The annotation system also creates disaster relatedcorpus with new tweets collected from Twitter API andannotated on real time manner. The performance ofthis model is evaluated based on different disasterrelated datasets and new Myanmar_Earthquake_2016dataset derived from Twitter. The experiments show ahigh agreement rate between the annotation of thissystem and the annotators. 2017-02-16 http://hdl.handle.net/20.500.12678/0000004801 https://meral.edu.mm/records/4801