No abstract

Nowadays, text annotation plays an important role within real-time social media mining. Social media analysis provides actionable information to its users in times of natural disasters. This paper presents an approach to a real-time two layer text annotation system for social media stream to the domain of natural disasters. The proposed system annotates raw tweets from Twitter into two types such as Informative or Not Informative as first layer. And then it annotates again five information types based on Informative tweets only as second layer. Based on the first and second layer annotation results, this system provides the tweets with user desired informative type in real time. In this system, annotation is done at tweet level by using word and phrase level features with LibLinear classifier. All features are in the form of Ngram nature based on part of speech (POS) tag, Sentiment Lexicon and especially created Disaster Lexicon. The validation of this system is performed based on different disaster related datasets and new Myanmar_Earthquake_2016 dataset derived from Twitter. The annotated datasets generated from this work can also be used by interested research communities to study the social media natural disaster related research.

Twitter, Social Networking Site, becomes most popular microblogging service and people have started publishing data on the use of it in natural disasters. Twitter has also created the opportunities for first responders to know the critical information and work effective reactions for impacted communities. This paper introduces the tweet monitoring system to identify the messages that people updated during natural disasters into a set of information categories and provide user desired target information type automatically. In this system, classification is done at tweet level with three labels by using LibLinear classifier. This system is intended to extract the small number of informational and actionable tweets from large amounts of raw tweets on Twitter using machine learning and natural language processing (NLP). Feature extraction of this work exploited only linguistic features, sentiment lexicon based features and especially disaster lexicon based features. The monitoring system also creates disaster related corpus with new tweets collected from Twitter API and annotation is done on real time manner. The performance of this system is evaluated based on four publicly available annotated datasets. The experiments showed the classification accuracy on the proposed features set is higher than the classifier based on neural word embeddings and standard bag-of-words (BOW) models. This system automatically annotated the Myanmar_Earthquake_2016 dataset at 75% accuracy on average.

No abstract

No abstract