Disaster affected communities are increasingly turning to social media for communication and coordination. This includes reports on needs (demands) offers (supplies) of resources required during emergency situations. Identifying matching such requests with potential responders can substantially accelerate relief efforts. Current work disaster management agencies is labor intensive, there substantial interest in automated tools.We present machine–learning methods automatically identify match...

This work contributes to the study of retweet behavior on Twitter surrounding real-world events. We analyze over a million tweets pertaining three events, present general tweet properties in such topical datasets and qualitatively most tweeted/viral content pieces. Findings include clear relationship between sparse/dense patterns type itself; suggesting need link-based diffusion models.

Citizen sensing, with a billion plus active users and tweets/week, is complemented by shared information from contextually relevant Web of Data (blogs, news, media objects) background knowledge. How can these enable us in informing, understanding managing broad variety activities events locally around the world? Twitris, currently version 3, scalable interactive platform which continuously collects, aggregates, integrates, analyzes above forms data knowledge to give deeper insights, as well...

Social media platforms facilitate the emergence of citizen communities that discuss real-world events. Their content reflects a variety intent ranging from social good (e.g., volunteering to help) commercial interest criticizing product features). Hence, mining data can aid in filtering support organizations, such as an emergency management unit for resource planning. However, effective is inherently challenging due ambiguity interpretation, and sparsity relevant behaviors data. In this...

Recent decades have seen a significant increase in the frequency, intensity, and impact of natural disasters other emergencies, forcing governments around world to make emergency response disaster management national priorities. The growth extremely large complex datasets—commonly referred as <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">big data</i> —and various advances information communications technology computing now support more...

Social media platforms provide a valuable source of public health information, as one-third US adults seek specific information online. Many antitobacco campaigns have recognized such trends among youth and shifted their advertising time effort toward digital platforms. Timely evidence is needed to inform the adaptation changing social platforms.In this study, we conducted content analysis major on Facebook using machine learning natural language processing (NLP) methods, well traditional...

The work of Emergency Management (EM) agencies requires timely collection relevant data to inform decision-making for operations and public communication before, during, after a disaster. However, the limited human resources available deploy field is persistent problem EM agencies. Thus, many these have started leveraging social media as supplemental source new venue engage with public. While prior research has analyzed potential benefits attitudes practitioners when during disasters, gap...

Background Preventing youth exposure to cigarette smoking is a public health priority. One of the most effective ways reduce tobacco use increase prices products. Minimum floor price laws (MFPLs) are relatively new but more feasible strategy that sets below which product cannot be sold. We aim examine effects minimum (MFPs) on among in Virginia. Methods An agent-based modelling (ABM) was developed from bottom-up evaluate influence increasing MFPs middle and high school students’ behaviour...

Predicting user intent and detecting the corresponding slots from text are two key problems in Natural Language Understanding (NLU). Since annotated datasets only available for a handful of languages, our work focuses particularly on zero-shot scenario where target language is unseen during training. In context learning, this task typically approached using representations pre-trained multilingual models such as mBERT or by fine-tuning data automatically translated into language. We propose...

Abstract Each year, significant investment of time and resources is made to improve diversity within engineering across a range federal state agencies, private/not-for-profit organizations, foundations. In spite decades investments, efforts have not yielded desired returns - participation by minorities continues lag at when STEM workforce requirements are increasing. recent years new stream data has emerged online social networks, including Twitter, Facebook, LinkedIn that acts as key sensor...

The public expects a prompt response from emergency services to address requests for help posted on social media. However, the information overload of media experienced by these organizations, coupled with their limited human resources, challenges them timely identify and prioritize critical requests. This is particularly acute in crisis situations where any delay may have severe impact effectiveness response. While has been extensively studied during crises, there work formally...

Existing approaches to cyber defense have been inadequate at defending the targets from advanced persistent threats (APTs). APTs are stealthy and orchestrated attacks, which target both corporations governments exfiltrate important data. In this paper, we present a novel comprehensibility manipulation framework (CMF) generate haystack of hard comprehend fake documents, can be used for deceiving attackers increasing cost data exfiltration by wasting their time resources. CMF requires an...

This analysis examines the literature on gendered media coverage of women candidates for higher office, and considers how biases in treatment based gender may be evident or exacerbated by promulgation fake news. Using 2016 Presidential election cycle United States as a case study, two news stories are investigated, which, like most at time, exhibited favor candidacy Donald Trump demonized denigrated his opponent, Hillary Clinton. Findings suggest that Pizzagate Health Scare evince narratives...

Detecting malicious intent behavior such as sharing hate speech has become an important challenge for social networking platforms. The method of automated detection media posts is often challenged by the complexity capturing context user expression with potential intent. We hypothesize that semantic features can help enrich representation word senses in a post machine learning algorithms. This paper presents novel empirical study diverse classification task on posts. Specifically, we present...

Transliteration is very common on social media, but transliterated text not adequately handled by modern neural models for various NLP tasks. In this work, we combine data augmentation approaches with a Teacher-Student training scheme to address issue in cross-lingual transfer setting fine-tuning state-of-the-art pre-trained multilingual language such as mBERT and XLM-R. We evaluate our method Hindi Malayalam, also introducing new datasets benchmarking real-world scenarios: one sentiment...

We study online social group dynamics based on how members diverge in their discussions. Previous studies mostly focused the link structure to characterize dynamics, whereas behavior of content generation discussions is not well understood. Particularly, we use Jensen-Shannon (JS) divergence measure topics user-generated contents, and it progresses over time. Twitter messages (tweets) multiple real-world events (natural disasters activism) with different times demographics. also model...

We present a study that examines how social media activism campaign aimed at improving gender diversity within engineering gained and maintained momentum in its early period. examined over 50,000 Tweets posted the first ~75 days of #ILookLikeAnEngineer found diverse participation - types users increased activity crucial moments. categorize these triggers into four types: 1) Event-Driven: Alignment with offline events related to issue (Diversity SFO, Disrupt, etc.); 2) Media-Driven: News...

Timely access to information of critical resources is the fundamental requirement for disaster management domain. Existing systems support services lack interoperability and have several challenges efficiently heterogeneous information, especially open data. Such include heterogeneity data sources formats (e.g., a hospital resource entity in Data. Gov versus OpenStreetMap), inconsistency vocabularies an type Data.Gov schema OpenStreetMap schema), incompleteness single source all relevant...

Abstract Social media has become an alternative communication mechanism for the public to reach out emergency services during time-sensitive events. However, information overload of social experienced by these services, coupled with their limited human resources, challenges them timely identify, prioritize, and organize critical requests help. In this paper, we first present a formal model serviceability called Social-EOC , which describes elements serviceable message posted in expressing...