A behavior analytics approach to identifying tweets from crisis regions

Kumar, Shamanth; Hu, Xia; Liu, Huan

doi:10.1145/2631775.2631814

Cited by 21 publications

(11 citation statements)

References 13 publications

(6 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One branch of studies has focused on analyzing the geotagging behaviors of people in disasters [23], [24]. For example, Kumar et al proposed an approach to identify whether a tweet is generated from crisis regions based on the historical geotags [25]. Another branch of the studies has examined the utility of geotagged social media posts for disaster response and recovery [26].…”

Section: Related Workmentioning

confidence: 99%

A Hybrid Machine Learning Pipeline for Automated Mapping of Events and Locations From Social Media in Disasters

Fan

Mostafavi

2020

IEEE Access

View full text Add to dashboard Cite

The objective of this study is to propose and test a hybrid machine learning pipeline to uncover the unfolding of disaster events corresponding to different locations from social media posts during disasters. Effective disaster response and recovery require a comprehensive understanding of disaster situations, i.e., unfolding of disaster events and geographic distribution of the disruptions. Existing studies have employed machine learning methods to conduct coarse-grained event detection and analyze the geographical location information from geotagged social media data. However, only a very small fraction of the entire set of social media data includes geotagged information, which may not directly correspond to events described in the content of posts. In addition, the coarse-grained information detected by existing approaches is tokenbased, which does not provide sufficient information for situation awareness. Hence, the detection of location and finer-grained event information could significantly improve the utility, credibility, and interpretability of social media data for situation awareness. To address these limitations, this study proposed a hybrid machine learning pipeline that makes use of all relevant tweets to uncover the evolution of disaster events across different locations. The pipeline integrates Named Entity Recognition for detecting locations mentioned in the posts, location fusion approach to extract coordinates of the locations and remove noise information, fine-tuned BERT model for classifying posts with humanitarian categories, and graph-based clustering to identify credible situational information. The application of the study is demonstrated using the data set collected from Twitter during the 2017 Hurricane Harvey in Houston. The results show the capability of the proposed hybrid pipeline for automated mapping of events across time and space from social media posts with considerable accuracy. The findings also suggest that the potential for forensic analysis of disasters using mapped events and their evolution, and based on the variation of social media attention to different locations in disasters. Hence, this method could provide a useful tool to support emergency managers, public officials, residents, first responders, and other stakeholders in rapid situation awareness across time and space. INDEX TERMS Machine learning pipeline, social media, disasters, automated mapping.

show abstract

Section: Related Workmentioning

confidence: 99%

A Hybrid Machine Learning Pipeline for Automated Mapping of Events and Locations From Social Media in Disasters

Fan

Mostafavi

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Unbalanced classes are a fundamental challenge for machine learning with techniques to mitigate their influence evolving [37]. For example Starbird, Grace and Leysia [17] adopt an algorithmic approach, but a sampling approach adopted by Morstatter, Lubold, Pon-Barry, Pfeffer and Liu [16] and Kumar et al [38] is more common for Twitter case studies. An advantage of considering linked images in comparison to the text content of micro-blogs, is the degree of imbalance is significantly less [1].…”

Section: The Needle In the Haystackmentioning

confidence: 99%

Identifying Witness Accounts from Social Media Using Imagery

Truelove

Khoshelham

Mclean

et al. 2017

IJGI

View full text Add to dashboard Cite

This research investigates the use of image category classification to distinguish images posted to social media that are Witness Accounts of an event. Only images depicting observations of the event, captured by micro-bloggers at the event, are considered Witness Accounts. Identifying Witness Accounts from social media is important for services such as news, marketing and emergency response. Automated image category classification is essential due to the large number of images on social media and interest in identifying witnesses in near real time. This paper begins research of this emerging problem with an established procedure, using a bag-of-words method to create a vocabulary of visual words and classifier trained to categorize the encoded images. In order to test the procedure, a set of images were collected for case study events, Australian Football League matches, from Twitter. Evaluation shows an overall accuracy of 90% and precision and recall for both classes exceeding 83%.

show abstract

“…At abroad, existing researches on the use of Twitter in natural hazard is more manifold (Kryvasheyeu et al, 2016). Some researchers study its contribution to situational awareness (Vieweg et la., 2010;Power et al, 2014), and some pay attention to practical field of classifying disaster message, detecting events and identifying risking areas (Earle et al, 2011;Sakaki et al, 2010;Kumar et al, 2014;Lmran et al, 2013;Caragea et al, 2011). Especially for tropical cyclone disaster, Kryvasheyeu and others presented a multiscale analysis of Twitter activity and Hurricane Sandy and found a strong relationship between Sandy's path and topic related Twitter activity (Kryvasheyeu et al, 2016).…”

Section: Introductionmentioning

confidence: 99%

The Typhoon Disaster Analysis Based on Weibo Topic Heat

Yuan

Gong

Wang

et al. 2018

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Could social media data be utilized in hazard evaluation? Typhoon disaster as one of the costly disaster has become devastating threats for human. Moreover, social media change the communication way of human and citizens can turn to this platform to express disasterrelated information at real time. Therefore, social media improves situational awareness and widens the method of hazard information acquiring. With more and more studies investigating in relationship between social media response and degree of damage, the strong correlation has been proved. Weibo as one of the most popular social media in China can provide data with posted text, location, user identification and other additional information. Combining with 10 tropical cyclones and Weibo data in 2013, We perform a quantitative analysis between the grade of hazard situation and Weibo related topic heat in province scale. We provide a new model of Weibo topic heat to evaluate the Weibo activity in study area. Also we demonstrate the hazard assessing formula is H&thinsp;=&thinsp;1.8845&thinsp;ln(&alpha;)&thinsp;+&thinsp;15.636 in tropical cyclone disaster. High level goodness of curve fitting also suggest that this equation can be used for rapid assessment of hazard caused by tropical cyclones.

show abstract

A behavior analytics approach to identifying tweets from crisis regions

Cited by 21 publications

References 13 publications

A Hybrid Machine Learning Pipeline for Automated Mapping of Events and Locations From Social Media in Disasters

A Hybrid Machine Learning Pipeline for Automated Mapping of Events and Locations From Social Media in Disasters

Identifying Witness Accounts from Social Media Using Imagery

The Typhoon Disaster Analysis Based on Weibo Topic Heat

Contact Info

Product

Resources

About