282 Hits in 9.8 sec

WHOSe Heritage: Classification of UNESCO World Heritage "Outstanding Universal Value" Documents with Soft Labels [article]

Nan Bai, Renqian Luo, Pirouz Nourian, Ana Pereira Roders
2021 arXiv   pre-print
Furthermore, manual annotation of heritage values and attributes from multi-source textual data, which is currently dominant in heritage studies, is knowledge-demanding and time-consuming, impeding systematic  ...  The study shows that the best models fine-tuned from BERT and ULMFiT can reach 94.3% top-3 accuracy.  ...  The authors are grateful for all the constructive comments from the anonymous reviewers.  ... 
arXiv:2104.05547v2 fatcat:yi7chwjrenbjnczyysjr7pukvy

An Ensemble Deep Learning Model for Drug Abuse Detection in Sparse Twitter-Sphere [article]

Han Hu and NhatHai Phan and James Geller and Stephen Iezzi and Huy Vo and Dejing Dou and Soon Ae Chun
2019 arXiv   pre-print
., many studies that primarily utilize social media data, such as postings on Twitter, to study drug abuse-related activities use machine learning as a powerful tool for text classification and filtering  ...  In this study, we approach this problem by designing an ensemble deep learning model that leverages both word-level and character-level features to classify abuse-related tweets.  ...  Acknowledgements The authors gratefully acknowledge the support from the National Science Foundation (NSF) grants CNS-1650587, CNS-1747798, CNS-1624503, and CNS-1850094.  ... 
arXiv:1904.02062v1 fatcat:vdmfs4rctnde5dbixgbna2i3xy

Supporting ESL Writing by Prompting Crowdsourced Structural Feedback

Yi-Ching Huang, Jiunn-Chia Huang, Hao-Chuan Wang, Jane Yung-jen Hsu
2017 AAAI Conference on Human Computation & Crowdsourcing  
First, we compared our crowd-based method with three naïve machine learning (ML) methods and got the best performance on the identification of topic sentence and irrelevant sentence in the article.  ...  Writing is challenging, especially for non-native speakers.  ...  Electronics, and Advantech.  ... 
dblp:conf/hcomp/HuangHWH17 fatcat:fy3rwt7l6zaljldeuj36mzvieq

Cyberbullies in Twitter: A focused review

Nicolas Tsapatsoulis, Vasiliki Anastasopoulou
2019 Zenodo  
The purpose of these studies is to find smart and sophisticated ways and methods to detect these cyberbullying incidents and in case that it is not possible to fully eliminate them to provide the means  ...  We present the theoretical roots of the term cyberbullying and we discuss thoroughly some influential studies to motivate new researchers that work in the general area of cybersecurity and privacy.  ...  ACKNOWLEDGMENT The authors acknowledge research funding from the European Union's Horizon 2020 research and innovation programme under the Marie Skodowska -Curie ENCASE project, Grant Agreement No. 691025  ... 
doi:10.5281/zenodo.2657415 fatcat:unp5mr5kwvaknn73uhua5iv7lu

Crowdsourcing Twitter annotations to identify first-hand experiences of prescription drug use

Nestor Alvaro, Mike Conway, Son Doan, Christoph Lofi, John Overington, Nigel Collier
2015 Journal of Biomedical Informatics  
Self-reported patient data has been shown to be a valuable knowledge source for post-market pharmacovigilance.  ...  In order to achieve this goal we explore machine learning with data crowdsourced from laymen annotators.  ...  Social media has been shown to be a promising data source for pharmacovigilance data due to its real-time nature and utility in providing insights into off-label consumer habits [6, 7] .  ... 
doi:10.1016/j.jbi.2015.11.004 pmid:26556646 fatcat:hyfrygomzbeqnaj32pjd2bkymq

Seeing Should Probably not be Believing: The Role of Deceptive Support in COVID-19 Misinformation on Twitter

Chaoyuan Zuo, Ritwik Banerjee, Hossein Shirazi, Fateme Hashemi Chaleshtori, Indrakshi Ray
2022 ACM Journal of Data and Information Quality  
Social media posts often leverage the trust readers have in prestigious news agencies and cite news articles as a way of gaining credibility.  ...  Nevertheless, it is not always the case that the cited article supports the claim made in the social media post.  ...  , since we only consider posts that contain links to reputable news agencies, and discard content derived from other kinds of user-generated content (e.g., blogs or other social media platforms). ( 8  ... 
doi:10.1145/3546914 fatcat:sbosl7grkba7vl6p6bsqvagdwy

Intentional Control of Type I Error over Unconscious Data Distortion: a Neyman-Pearson Approach to Text Classification [article]

Lucy Xia, Richard Zhao, Yanhui Wu, Xin Tong
2018 arXiv   pre-print
Digital texts have become an increasingly important source of data for social studies.  ...  However, textual data from open platforms are vulnerable to manipulation (e.g., censorship and information inflation), often leading to bias in subsequent empirical analysis.  ...  We did not use the labels got from this crowdsourcing method in our analysis due to their subpar label quality.  ... 
arXiv:1802.02558v2 fatcat:7sgxgqciszfgdopyrygwd2af7a

Contextual Multi-View Query Learning for Short Text Classification in User-Generated Data [article]

Payam Karisani, Negin Karisani, Li Xiong
2021 arXiv   pre-print
., for the early detection of outbreaks or for extracting personal observations--often suffers from the lack of enough training data, short document length, and informal language model.  ...  The experiments testify that our model is applicable to multiple important representative Twitter tasks and also significantly outperforms the existing baselines.  ...  Towards identifying drug side effects from of social media tasks that are not based on queries social media using active learning and crowd sourc-  ... 
arXiv:2112.02611v1 fatcat:t4c63auyqndwrpvx6xs3afeqoq

Enhancing Mobile App User Understanding and Marketing with Heterogeneous Crowdsourced Data: A Review

Bin Guo, Yi Ouyang, Tong Guo, Longbing Cao, Zhiwen Yu
2019 IEEE Access  
This paper reviews CrowdApp, a research field that leverages heterogeneous crowdsourced data for mobile app user understanding and marketing.  ...  It has some key differentiating characteristics which make it different from traditional markets.  ...  App developers also publish and disseminate posts about their marketing activities in the social media. B.  ... 
doi:10.1109/access.2019.2918325 fatcat:de763kc4qbdy5ijo55jxyhzgt4

A Reliable Weighting Scheme for the Aggregation of Crowd Intelligence to Detect Fake News

Franklin Tchakounté, Ahmadou Faissal, Marcellin Atemkeng, Achille Ntyam
2020 Information  
Several authors have proposed systems to detect fake news in social networks using crowd signals through the process of crowdsourcing.  ...  This fake news can cause enormous difficulties for users and institutions.  ...  post as fake or not fake Known at start (Tacchini et al. 2017) One type (Simple) Use algorithms derived from crowdsourcing (BLC) for making decision Logistic regression and crowdsourcing  ... 
doi:10.3390/info11060319 fatcat:gd2tuadpsrazzh2twfvc5orx3i

Detecting Cyberbullying and Cyberaggression in Social Media [article]

Despoina Chatzakou, Ilias Leontiadis, Jeremy Blackburn, Emiliano De Cristofaro, Gianluca Stringhini, Athena Vakali, Nicolas Kourtellis
2019 arXiv   pre-print
Using various state-of-the-art machine learning algorithms, we classify these accounts with over 90% accuracy and AUC.  ...  More than half of young social media users worldwide have been exposed to such prolonged and/or coordinated digital harassment.  ...  The paper reflects only the authors' views and the Agency and the Commission are not responsible for any use that may be made of the information it contains.  ... 
arXiv:1907.08873v1 fatcat:4tkcdria2rhntg5lawcafeixxi

Social Media Relevance Filtering Using Perplexity-Based Positive-Unlabelled Learning

Sunghwan Mac Kim, Stephen Wan, Cécile Paris, Andreas Duenser
2020 International Conference on Web and Social Media  
Our PPUL method generally outperforms strong PU Learning baselines, which we demonstrate on five different data sets: the Hazardous Product Review data set, two well known social media data sets, and two  ...  In this paper, we introduce our Perplexity variant of Positive-Unlabelled Learning (PPUL) framework as a means to perform social media relevance filtering.  ...  Acknowledgments We would like to thank the project's software engineers, Brian Jin and James McHugh, for supporting this research. We also thank the reviewers for their insightful feedback.  ... 
dblp:conf/icwsm/Kim0PD20 fatcat:azycnmyr5jcdjp6qvugyzsoyji

A Survey of COVID-19 Misinformation: Datasets, Detection Techniques and Open Issues [article]

A.R. Sana Ullah, Anupam Das, Anik Das, Muhammad Ashad Kabir, Kai Shu
2021 arXiv   pre-print
Misinformation during pandemic situations like COVID-19 is growing rapidly on social media and other platforms.  ...  Researchers are trying their best to mitigate this problem using different approaches based on Machine Learning (ML), Deep Learning (DL), and Natural Language Processing (NLP).  ...  finally the social media posts were collected from Facebook, Twitter, Instagram, Youtube and TikTok.  ... 
arXiv:2110.00737v2 fatcat:vhafkc3i6bcaxpmjmrt7pm2mxe

Beyond Social Media Analytics: Understanding Human Behaviour and Deep Emotion using Self Structuring Incremental Machine Learning [article]

Tharindu Bandaragoda
2020 arXiv   pre-print
An event detection technique was developed to automatically monitor those identified topic pathways for significant fluctuations in social behaviours using multiple indicators such as volume and sentiment  ...  For fast-paced, a self-structuring and incremental learning technique was developed to automatically capture salient topics and corresponding dynamics over time.  ...  Learning from generalisation Topic separation with IKASL The features vectors derived from batches of social media messages are used to self structure the adaptive structure that consists of chains of  ... 
arXiv:2009.09078v1 fatcat:izo3eyjc4vdo3jnttt7omvsv5q

Characterizing Abhorrent, Misinformative, and Mistargeted Content on YouTube [article]

Kostantinos Papadamou
2021 arXiv   pre-print
Finally, when studying pseudoscientific misinformation, we find that YouTube suggests more pseudoscientific content regarding traditional pseudoscientific topics (e.g., flat earth) than for emerging ones  ...  Following a data-driven quantitative approach, we analyze thousands of videos on YouTube, to shed light on: 1) the risks of YouTube media consumption by young children; 2) the role of the recommendation  ...  For many users, YouTube has also become one of the most important information sources for news, world events, and various other topics [1, 2] .  ... 
arXiv:2105.09819v1 fatcat:k2mdclt4orgunnlwb6hnnrka4i
« Previous Showing results 1 — 15 out of 282 results