Strategy for processing and analyzing social media data streams in emergencies

Abstract

People are using social media to a greater extent, particularly in emergency situations. However, approaches for processing and analyzing the vast quantities of data produced currently lag far behind. In this paper we discuss important steps, and the associated challenges, for processing and analyzing social media in emergencies. In our research project EmerGent, a huge volume of low-quality messages will be continuously gathered from a variety of social media services such as Facebook or Twitter. Our aim is to design a software system that will process and analyze social media data, transforming the high volume of noisy data into a low volume of rich content that is useful to emergency personnel. Therefore, suitable techniques are needed to extract and condense key information from raw social media data, allowing detection of relevant events and generation of alerts pertinent to emergency personnel.

From social networks to emergency operation centers: A semantic visualization approach

“…To allow this, it is crucial to define solutions that facilitate the collection and identification of reliable and useful data from the real-time stream of social messages in order to avoid issues such as lack of trust, overwhelming volume of data, and low quality of data [25,26,8]. Algorithms have been developed to meet the challenges involved in evaluating the quality of citizen-generated content [9] or removing noisy redundant data [27]. Since the focus of this research was on information visualization, this section reviews examples of visual analytics tools built to analyze and visualize information collected from social networks.…”

Social Media in Crisis Management: An Evaluation and Analysis of Crisis Informatics Research

“…Different algorithmic approaches have been applied to study and include citizen-generated content (C2A) from social media (Imran et al, 2015). On the one hand, they are supposed to identify or predict critical events and to convert the high volume of big and noisy data, which cannot be managed by emergency managers in a short time before or during large-scale emergencies, to a low volume of rich and thick content (Moi et al, 2015). On the contrary, algorithms are supposed to identify underlying patterns (e.g., mood or geospatial correlations) applying statistical approaches or visual analytics (Brynielsson, Johansson, Jonsson, & Westling, 2014;Fuchs, Andrienko, Andrienko, Bothe, & Stange, 2013).…”

“…Additionally, De Albuquerque, Herfort, Brenning, and Zipf (2015) prove that geographical approaches for quantitatively assessing social media messages might be helpful to enhance important content. Moi et al (2015) suggest a system to process and investigate social media data, transforming the high volume of noisy data into a low volume of rich content that emergency personnel can use. To succeed, they categorize the steps of information gathering and data preparation, information mining, data enrichment, alert detection, information visualization, semantic data modeling with ontologies, and information quality assessment.…”

Preprocessing Arabic text on social media

“…Development and application or computerized tools to extract knowledge and information from Arabic text on social media platforms is a complex task. To this end, several methods including text cleaning, natural language processing (NLP), normalization, learning algorithms, and application design tools must be developed [ 1 , 2 , 3 ].…”