- Social Media and Politics
- Sentiment Analysis and Opinion Mining
- Misinformation and Its Impacts
- Hate Speech and Cyberbullying Detection
- Text and Document Classification Technologies
- Social Media in Health Education
- Complex Network Analysis Techniques
- Sexual Assault and Victimization Studies
- Digital Marketing and Social Media
- Advanced Text Analysis Techniques
- Topic Modeling
- Spam and Phishing Detection
- Mental Health via Writing
- Computational and Text Analysis Methods
- Media Influence and Politics
- Biomedical Text Mining and Ontologies
- Web and Library Services
- Public Relations and Crisis Communication
- Health Literacy and Information Accessibility
- Outsourcing and Supply Chain Management
- Face and Expression Recognition
- LGBTQ Health, Identity, and Policy
- Multi-Criteria Decision Making
- Disaster Management and Resilience
- Media Studies and Communication
University of Alabama at Birmingham
2022-2024
University of South Carolina
2017-2022
Kent State University
2020
University of Alabama
2020
Indiana University Bloomington
2020
University of Cape Town
2020
Conference Board
2020
University of North Carolina at Charlotte
2019
Kermanshah University of Medical Sciences
2013-2018
Hamedan University of Medical Sciences
2014-2017
Researchers have collected Twitter data to study a wide range of topics. This growing body literature, however, has not yet been reviewed systematically synthesize Twitter-related papers. The existing literature review papers limited by constraints traditional methods manually select and analyze samples topically related goals this retrospective are identify dominant topics Twitter-based research, summarize the temporal trend topics, interpret evolution withing last ten years. mines large...
In recent years, we have been faced with a series of natural disasters causing tremendous amount financial, environmental and human losses. The unpredictable nature behaviour makes it hard to comprehensive situational awareness (SA) support disaster management. Using opinion surveys is traditional approach analyse public concerns during disasters; however, this limited, expensive time-consuming. Luckily, the advent social media has provided scholars an alternative means analysing concerns....
The understanding of the public response to COVID-19 vaccines is key success factor control pandemic. To understand response, there a need explore opinion. Traditional surveys are expensive and time-consuming, address limited health topics, obtain small-scale data. Twitter can provide great opportunity opinion regarding vaccines. current study proposes an approach using computational human coding methods collect analyze large number tweets wider perspective on vaccine. This identifies...
Opinion polls have been the bridge between public opinion and politicians in elections. However, developing surveys to disclose people's feedback with respect economic issues is limited, expensive, time-consuming. In recent years, social media such as Twitter has enabled people share their opinions regarding Social provided a platform for collecting large amount of data. This article proposes computational mining approach explore discussion during an election. Current related studies use...
Oil refinery process releases toxic pollutants into aqueous environment.Phenol and its derivations as the most important pose severe environmental concern.In this study, rectangle anaerobic stabilization pond (ASP) consisting of feed tank with workload 60 Lit (1 × 0.2 1) meter phenol was used.This study evaluated interactive effect concentration (200-400 mg/l), temperature (8-24°C) Hydraulic retention time (HRT) (2-5 d) on efficiency for oil wastewater treatment.In experiments were carried...
Sexual harassment has been the topic of thousands research articles in 20th and 21st centuries. Several review papers have developed to synthesize literature about sexual harassment. While traditional studies provide valuable insights, these some limitations including analyzing a limited number papers, being time-consuming labor-intensive, focusing on few topics, lacking temporal trend analysis. To address limitations, this paper employs both computational qualitative approaches identify...
One of the challenges for text analysis in medical domains is analyzing large-scale documents. As a consequence, finding relevant documents has become more difficult. popular methods to retrieve information based on discovering themes topic modeling. The help same with and without query. In this paper, we present novel approach modeling using fuzzy clustering. To evaluate our model, experiment two datasets evaluation metrics carried out through document classification show that model...
We analyze a set of Twitter hashtags to ascertain how contemporary parlance in social media can illuminate the rich cultural intersections between modern forms work, use technology, and physical mobility. network word co-occurrence analysis topic modeling reveal several thematic areas discourse present Twitter, each with its own affiliated terms distinctive emphases. The first theme centers on worker identity is currently dominated by experiences digital nomads. second focuses practicalities...
Twitter’s APIs are now the main data source for social media researchers. A large number of studies have utilized Twitter diverse research interests. users can share their precise real-time location, and provide this information as longitude latitude. These geotagged help to study human activities movements different applications. Compared mostly small-scale samples in domains, such science, collecting offers samples. There is a fundamental question whether represent non-geotagged users....
ABSTRACT Social media based digital epidemiology has the potential to support faster response and deeper understanding of public health related threats. This study proposes a new framework analyze unstructured textual data via Twitter users' post (tweets) characterize negative sentiments nonhealth concerns in relations corpus regarding diet, diabetes, exercise obesity (DDEO). Through collection six million Tweets for one month, this identified prominent topics users as it relates...
Objective:The goal of this study is to understand how people experience sexism and sexual harassment in the workplace by discovering themes 2,362 experiences posted on Everyday Sexism Project's website everydaysexism.com.Method: This used both quantitative qualitative methods.The method was a computational framework collect analyze large number experiences.The analysis topics generated text mining method.Results: Twenty-three were coded then grouped into three overarching from sex...
ABSTRACT Although there are millions of transgender people in the world, a lack information exists about their health issues. This issue has consequences for medical field, which only nascent understanding how to identify and meet this population's health‐related needs. Social media sites like Twitter provide new opportunities overcome these barriers by sharing personal experiences. Our research employs computational framework collect tweets from self‐identified users, detect those that...
Using big data has been a prevailing research trend in various academic fields. However, no studies have explored the scope and structure of across disciplines. In this article, we applied topic modeling word co-occurrence analysis methods to identify key topics from more than 36,000 publications all disciplines between 2012 2017. The results revealed several associated with storage, collection large datasets; were predominantly published computational Other identified show influence...
Abstract Social media has become a mainstream channel of communication during the COVID‐19 pandemic. While some studies have been developed on investigating public opinion social data regarding pandemic, there is no study analyzing anti‐quarantine comments media. This collected and analyzed near 80,000 tweets to understand comments. Using text mining, we found 11 topics representing different issues such as comparing flu health side effects quarantine. We believe that this shines light...
To combat health disinformation shared online, there is a need to identify and characterize the prevalence of topics by trolls managed individuals promote discord. The current literature limited few dominated vaccination. goal this study analyze breadth discussed left (liberal) right (conservative) Russian on Twitter. We introduce an automated framework based mixed methods including both computational qualitative techniques. Results suggest that 48 health-related topics, ranging from diet...
Given the rapidly unfolding nature of COVID-19 pandemic, there is an urgent need to streamline literature synthesis growing scientific research elucidate targeted solutions. Traditional systematic review studies have restrictions, including analyzing a limited number papers, having various biases, being time-consuming and labor-intensive, focusing on few topics, lack data-driven tools. This has collected 9298 papers representing published through May 5, 2020. We used frequency analysis find...
Selecting IT service providers in information systems outsourcing involves both qualitative and quantitative evaluations. This paper proposes an integrated multi-criteria decision-making (MCDM) framework to effectively handle uncertainty subjectivity the vendor selection process. The proposed methods apply fuzzy logic approach integrate survey data into traditional decision models such as envelope analysis (DEA), analytical hierarchy process (AHP) methods, TOPSIS. Based on case studies from...