scholarly journals An approach framework of transfer learning, adversarial training and hierarchical multi-task learning - a case study of disinformation detection with offensive text

2022 ◽  
Vol 2161 (1) ◽  
pp. 012049
Author(s):  
Shravan Chandra ◽  
Bhaskarjyoti Das
Keyword(s):  
Transfer Learning ◽  
Error Feedback ◽  
Task Learning ◽  
Learning Tasks ◽  
Linguistic Patterns ◽  
Social Media Platforms ◽  
Adversarial Training ◽  
Hierarchical Nature ◽  
Definition Of

Abstract With society going online and disinformation getting accepted as a phenomena that we have to live with, there is a growing need to automatically detect offensive text on modern social media platforms. But the lack of enough balanced labeled data, constantly evolving socio-linguistic patterns and ever-changing definition of offensive text make it a challenging task. This is a common pattern witnessed in all disinformation detection tasks such as detection of propaganda, rumour, fake news, hate etc. The work described in this paper improves upon the existing body of techniques by bringing in an approach framework that can surpass the existing benchmarks. Firstly, it addresses the imbalanced and insufficient nature of available labeled dataset. Secondly, learning using relates tasks through multi-task learning has been proved to be an effective approach in this domain but it has the unrealistic requirement of labeled data for all related tasks. The framework presented here suitably uses transfer learning in lieu of multi-task learning to address this issue. Thirdly, it builds a model explicitly addressing the hierarchical nature in the taxonomy of disinformation being detected as that delivers a stronger error feedback to the learning tasks. Finally, the model is made more robust by adversarial training. The work presented in this paper uses offensive text detection as a case study and shows convincing results for the chosen approach. The framework adopted can be easily replicated in other similar learning tasks facing a similar set of challenges.

Mediated Public Diplomacy Redefined: Foreign Stakeholder Engagement via Paid, Earned, Shared, and Owned Media

2019 ◽  
Vol 63 (12) ◽  
pp. 1665-1683 ◽  
Author(s):  
Guy J. Golan ◽  
Ilan Manor ◽  
Phillip Arceneaux
Keyword(s):  
Social Media ◽  
News Media ◽  
Stakeholder Engagement ◽  
Media Coverage ◽  
Public Diplomacy ◽  
The Social ◽  
Social Media Platforms ◽  
Media Platform ◽  
Definition Of

Mediated public diplomacy literature examines the engagement of foreign audiences by governments via mediated channels. To date, scholars have examined the competitive contest between global rivals in promoting and contesting one another’s frames as reflected in global news media coverage. Recognizing the meaningful impact of social media platforms, along with the global rise of government-sponsored media organizations, the current study builds on previous mediated public diplomacy scholarship by expanding the scope of the literature beyond the earned media perspective to also include paid, shared, and owned media. The article presents a revised definition of the term mediated public diplomacy along with a case study of government to foreign stakeholder engagement via the social media platform, Twitter.

scholarly journals An overview of multi-task learning

2017 ◽  
Vol 5 (1) ◽  
pp. 30-43 ◽  
Author(s):  
Yu Zhang ◽  
Qiang Yang
Keyword(s):  
Supervised Learning ◽  
Language Processing ◽  
Health Informatics ◽  
Web Applications ◽  
Task Learning ◽  
Learning Tasks ◽  
Promising Area ◽  
Speed Up ◽  
Definition Of ◽  
Theoretical Analyses

Abstract As a promising area in machine learning, multi-task learning (MTL) aims to improve the performance of multiple related learning tasks by leveraging useful information among them. In this paper, we give an overview of MTL by first giving a definition of MTL. Then several different settings of MTL are introduced, including multi-task supervised learning, multi-task unsupervised learning, multi-task semi-supervised learning, multi-task active learning, multi-task reinforcement learning, multi-task online learning and multi-task multi-view learning. For each setting, representative MTL models are presented. In order to speed up the learning process, parallel and distributed MTL models are introduced. Many areas, including computer vision, bioinformatics, health informatics, speech, natural language processing, web applications and ubiquitous computing, use MTL to improve the performance of the applications involved and some representative works are reviewed. Finally, recent theoretical analyses for MTL are presented.

Empowering Women Entrepreneurs Through Microfinance - A Case of Rural Bangladesh

2017 ◽  
Vol 3 (2) ◽  
pp. 7
Author(s):  
Saida Parvin
Keyword(s):  
Women's Empowerment ◽  
Assessment Tools ◽  
Women’S Empowerment ◽  
Case Study Method ◽  
Policy Makers ◽  
Theoretical Contribution ◽  
Donor Agencies ◽  
Definition Of ◽  
The Impact

Women’s empowerment has been at the centre of research focus for many decades. Extant literature examined the process, outcome and various challenges. Some claimed substantial success, while others contradicted with evidence of failure. But the success remains a matter of debate due to lack of empirical evidence of actual empowerment of women around the world. The current study aimed to address this gap by taking a case study method. The study critically evaluates 20 cases carefully sampled to include representatives from the entire country of Bangladesh. The study demonstrates popular beliefs about microfinance often misguide even the borrowers and they start living in a fabricated feeling of empowerment, facing real challenges to achieve true empowerment in their lives. The impact of this finding is twofold; firstly there is a theoretical contribution, where the definition of women’s empowerment is proposed to be revisited considering findings from these cases. And lastly, the policy makers at governmental and non-governmental organisations, and multinational donor agencies need to revise their assessment tools for funding.

The Rapid Iterative Test and Evaluation Method (RITE)

2018 ◽  
Author(s):  
Michael C. Medlock
Keyword(s):  
Evaluation Method ◽  
Test And Evaluation ◽  
Original Case ◽  
Definition Of

This chapter begins with a discussion of the philosophy and then definition of the RITE method. It then delves into the benefits of this method and provides practical notes on running RITE tests effectively. The chapter concludes with an overview of the original case study behind the 2002 article documenting this method.

Poe and His Global Advocates

2018 ◽  
pp. 596-617
Author(s):  
Emron Esplin
Keyword(s):  
Twentieth Century ◽  
South America ◽  
Rio De La Plata ◽  
Río De La Plata ◽  
Literary Traditions ◽  
La Plata ◽  
Original Works ◽  
Definition Of ◽  
Global Reputation

This essay explores Edgar Allan Poe’s extraordinary relationships with various literary traditions across the globe, posits that Poe is the most influential US writer on the global literary scene, and argues that Poe’s current global reputation relies at least as much on the radiance of the work of Poe’s literary advocates—many of whom are literary stars in their own right—as it does on the brilliance of Poe’s original works. The article briefly examines Poe’s most famous French advocates (Baudelaire, Mallarmé, Valéry); glosses the work of his advocates throughout Europe, Asia, and the Americas; and offers a concise case study of Poe’s influence on and advocacy from three twentieth-century writers from the Río de la Plata region of South America (Quiroga, Borges, and Cortázar). The essay concludes by reading the relationships between Poe and his advocates through the ancient definition of astral or stellar influence.

scholarly journals Sustainability and Stakeholder Awareness: A Case Study of a Scottish University

2021 ◽  
Vol 13 (8) ◽  
pp. 4186
Author(s):  
Abdulhakeem Raji ◽  
Abeer Hassan
Keyword(s):  
Signaling Theory ◽  
Relevant Information ◽  
Voluntary Exercise ◽  
Communication Tool ◽  
Individual Interviews ◽  
Study Approach ◽  
Research Findings ◽  
Definition Of ◽  
The University

This paper adopted a case study approach to investigate the sustainability practices of a Scottish university in order to understand if sustainability forms part of its central policy agenda. As such, the paper focuses on the levels of awareness and disclosure of their sustainable practices, measuring the impacts and effectiveness of those initiatives. This paper introduces signaling theory to explore the idea that appropriate communication via integrated thinking can close the gap between the organization and its stakeholders. We believe that the provision of this relevant information will lead to better communication between the organization and its stakeholders, supporting a signaling theory interpretation. Therefore, we are suggesting that integrated thinking is an internal process that organizations can follow to increase the level of disclosure as a communication tool with stakeholders. From the literature reviewed, four themes were identified (definition of university sustainability, sustainability awareness, disclosure framework within universities, and level of accountability). The research adopted a pragmatic view and conducted individual interviews with participants belonging to three stakeholder groups (members of the university’s senior management, the governing council, and the student union executive). Although this study focused on just one Scottish university, it should still provide some insight for the better understanding of the underpinning issues surrounding the sustainability accountability practices of Scottish universities in general. The research findings indicated that the university prioritized only two sustainability dimensions—economic and environmental—and that the university still perceived sustainability as a voluntary exercise. Additionally, it is evident that the university had no framework in place for measuring its sustainability delivery—and therefore had no established medium of communicating these activities to its stakeholders. Moreover, research findings showed that the social and educational context of sustainability was lacking at the university. The university has done little or nothing to educate its stakeholders on sustainability.

scholarly journals A Tale of Two Disasters

2020 ◽  
Vol 4 (Supplement_1) ◽  
pp. 694-694
Author(s):  
Tammy Mermelstein
Keyword(s):  
Older Adults ◽  
Social Media ◽  
Interest Group ◽  
Media Analysis ◽  
Ice Storm ◽  
Social Media Analysis ◽  
Individual Survival ◽  
Social Media Platforms

Abstract Preparing for or experiencing a disaster is never easy, but how leaders communicate with older adults can ease a situation or make it exponentially worse. This case study describes two disasters in the same city: Hurricane Harvey and the 2018 Houston Texas Ice Storm and the variation in messaging provided to and regarding older adults. For example, during Hurricane Harvey, the primary pre-disaster message was self-preparedness. During the storm, messages were also about individual survival. Statements such as “do not [climb into your attic] unless you have an ax or means to break through,” generated additional fear for older adults and loved ones. Yet, when an ice storm paralyzed Houston a few months later, public messaging had a strong “check on your elderly neighbors” component. This talk will explore how messaging for these events impacted older adults through traditional and social media analysis, and describe how social media platforms assisted people with rescue and recovery. Part of a symposium sponsored by Disasters and Older Adults Interest Group.

scholarly journals Augmenting Crop Detection for Precision Agriculture with Deep Visual Transfer Learning—A Case Study of Bale Detection

2020 ◽  
Vol 13 (1) ◽  
pp. 23
Author(s):  
Wei Zhao ◽  
William Yamada ◽  
Tianxin Li ◽  
Matthew Digman ◽  
Troy Runge
Keyword(s):  
Object Detection ◽  
Transfer Learning ◽  
Precision Agriculture ◽  
Crop Production ◽  
Domain Adaptation ◽  
Training Data ◽  
Detection Accuracy ◽  
Detection Model ◽  
Agriculture Products

In recent years, precision agriculture has been researched to increase crop production with less inputs, as a promising means to meet the growing demand of agriculture products. Computer vision-based crop detection with unmanned aerial vehicle (UAV)-acquired images is a critical tool for precision agriculture. However, object detection using deep learning algorithms rely on a significant amount of manually prelabeled training datasets as ground truths. Field object detection, such as bales, is especially difficult because of (1) long-period image acquisitions under different illumination conditions and seasons; (2) limited existing prelabeled data; and (3) few pretrained models and research as references. This work increases the bale detection accuracy based on limited data collection and labeling, by building an innovative algorithms pipeline. First, an object detection model is trained using 243 images captured with good illimitation conditions in fall from the crop lands. In addition, domain adaptation (DA), a kind of transfer learning, is applied for synthesizing the training data under diverse environmental conditions with automatic labels. Finally, the object detection model is optimized with the synthesized datasets. The case study shows the proposed method improves the bale detecting performance, including the recall, mean average precision (mAP), and F measure (F1 score), from averages of 0.59, 0.7, and 0.7 (the object detection) to averages of 0.93, 0.94, and 0.89 (the object detection + DA), respectively. This approach could be easily scaled to many other crop field objects and will significantly contribute to precision agriculture.

scholarly journals Multi-objective parametrization of interatomic potentials for large deformation pathways and fracture of two-dimensional materials

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Xu Zhang ◽  
Hoang Nguyen ◽  
Jeffrey T. Paci ◽  
Subramanian K. R. S. Sankaranarayanan ◽  
Jose L. Mendoza-Cortes ◽  
...  
Keyword(s):  
Large Deformation ◽  
Functional Form ◽  
Principal Component ◽  
Interatomic Potentials ◽  
Two Dimensional ◽  
Multi Objective ◽  
Multi Objective Genetic Algorithm ◽  
Two Dimensional Materials ◽  
Definition Of

AbstractThis investigation presents a generally applicable framework for parameterizing interatomic potentials to accurately capture large deformation pathways. It incorporates a multi-objective genetic algorithm, training and screening property sets, and correlation and principal component analyses. The framework enables iterative definition of properties in the training and screening sets, guided by correlation relationships between properties, aiming to achieve optimal parametrizations for properties of interest. Specifically, the performance of increasingly complex potentials, Buckingham, Stillinger-Weber, Tersoff, and modified reactive empirical bond-order potentials are compared. Using MoSe2 as a case study, we demonstrate good reproducibility of training/screening properties and superior transferability. For MoSe2, the best performance is achieved using the Tersoff potential, which is ascribed to its apparent higher flexibility embedded in its functional form. These results should facilitate the selection and parametrization of interatomic potentials for exploring mechanical and phononic properties of a large library of two-dimensional and bulk materials.

Innovative promotional strategies and diversification of flavoured mass merchandise cigar products: a case study of Swedish match

2021 ◽  
pp. tobaccocontrol-2020-056145 ◽  
Author(s):  
Ollie Ganz ◽  
Mary Hrywna ◽  
Kevin R J Schroth ◽  
Cristine D Delnevo
Keyword(s):  
Regulatory Authority ◽  
Business Practice ◽  
Smoking Prevention ◽  
Tobacco Products ◽  
The Family ◽  
Promotional Efforts ◽  
Promotional Strategies ◽  
Definition Of ◽  
The U.S

In 2009, the Family Smoking Prevention and Tobacco Control Act (TCA) granted the U.S. Food and Drug Administration (FDA) regulatory authority over tobacco products, although initially this only included cigarettes, smokeless tobacco and roll-your-own tobacco. In 2016, the deeming rule extended regulatory authority to include all tobacco products, including cigars. The deeming rule prohibited the introduction of new tobacco products into the marketplace without proper marketing authorisation and laid out pathways for tobacco companies to follow. The deeming rule should have frozen the cigar marketplace in 2016. In this paper, we describe how the cigarillo marketplace, nevertheless, continues to diversify with new brands, flavors, styles and packaging sizes entering the market regularly. As an example, we highlight recent promotional efforts by Swedish Match North America (Swedish Match) for their popular cigarillo brands, including White Owl, Night Owl and Garcia y Vega’s Game brand. We argue that ambiguities in the TCA make it unclear whether Swedish Match’s seemingly new cigarillos fit the definition of new tobacco products and, if so, whether they are on the market legally. Swedish Match and other cigarillo companies may be taking advantage of these ambiguities to promote a variety of cigarillo flavors and styles in innovative ways. Given that cigars are combustible tobacco products that pose many of the same risks as cigarettes, this business practice raises significant concerns regarding the protection of public health, particularly among young people.

Sign in / Sign up

Export Citation Format

Share Document