MEGA: Machine Learning-Enhanced Graph Analytics for Infodemic Risk Management

Hang, Ching Nam; Yu, Pei-Duo; Ling, Lin; Tan, Chee Wei

doi:10.1101/2020.10.24.20215061

Cited by 2 publications

(2 citation statements)

References 77 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Only by continuously comparing and calibrating data can the accuracy and credibility of data be improved and the presence of erroneous information be minimized. However, this is a time-consuming process, and Hang et al had proposed that the machine learning-enhanced graph analytics (MEGA) model used automatic feature-based vertex embedding to process the data to calculate accurate Infodemic risk scores [ 40 ]. In short, we have to choose the appropriate method to validate the data.…”

Section: Resultsmentioning

confidence: 99%

Infodemic: Challenges and solutions in topic discovery and data process

Zhang,

Pan,

Lin

et al. 2023

Arch Public Health

View full text Add to dashboard Cite

Background The Coronavirus Disease 2019 (COVID-19) pandemic was a huge shock to society, and the ensuing information problems had a huge impact on society at the same time. The urgent need to understand the Infodemic, i.e., the importance of the spread of false information related to the epidemic, has been highlighted. However, while there is a growing interest in this phenomenon, studies on the topic discovery, data collection, and data preparation phases of the information analysis process have been lacking. Objective Since the epidemic is unprecedented and has not ended to this day, we aimed to examine the existing Infodemic-related literature from January 2019 to December 2022. Methods We have systematically searched ScienceDirect and IEEE Xplore databases with some search limitations. From the searched literature we selected titles, abstracts and keywords, and limitations sections. We conducted an extensive structured literature search and analysis by filtering the literature and sorting out the available information. Results A total of 47 papers ended up meeting the requirements of this review. Researchers in all of these literatures encountered different challenges, most of which were focused on the data collection step, with few challenges encountered in the data preparation phase and almost none in the topic discovery section. The challenges were mainly divided into the points of how to collect data quickly, how to get the required data samples, how to filter the data, what to do if the data set is too small, how to pick the right classifier and how to deal with topic drift and diversity. In addition, researchers have proposed partial solutions to the challenges, and we have also proposed possible solutions. Conclusions This review found that Infodemic is a rapidly growing research area that attracts the interest of researchers from different disciplines. The number of studies in this field has increased significantly in recent years, with researchers from different countries, including the United States, India, and China. Infodemic topic discovery, data collection, and data preparation are not easy, and each step faces different challenges. While there is some research in this emerging field, there are still many challenges that need to be addressed. These findings highlight the need for more articles to address these issues and fill these gaps.

show abstract

Section: Resultsmentioning

confidence: 99%

Infodemic: Challenges and solutions in topic discovery and data process

Zhang,

Pan,

Lin

et al. 2023

Arch Public Health

View full text Add to dashboard Cite

show abstract

“…They use epidemiology models to characterize the basic reproduction number R 0 and provide platform-dependent numerical estimates of rumors' amplification. Hang et al [38] study graph-based framework to infodemiology using joint hierarchical clustering and cloud computing, which is a key to designing scalable data analytics for infodemic control. In addition, they use statistical machine learning to exploit the statistics of data to accelerate computation.…”

Section: Data Analytics Computation and Applicationsmentioning

confidence: 99%

Data-driven analytics of COVID-19 ‘infodemic’

Wan

Xiang

et al. 2022

Int J Data Sci Anal

View full text Add to dashboard Cite

The rampant of COVID-19 infodemic has almost been simultaneous with the outbreak of the pandemic. Many concerted efforts are made to mitigate its negative effect to information credibility and data legitimacy. Existing work mainly focuses on fact-checking algorithms or multi-class labeling models that are less aware of the intrinsic characteristics of the language. Nor is it discussed how such representations can account for the common psycho-socio-behavior of the information consumers. This work takes a data-driven analytical approach to (1) describe the prominent lexical and grammatical features of COVID-19 misinformation; (2) interpret the underlying (psycho-)linguistic triggers in terms of sentiment, power and activity based on the affective control theory; (3) study the feature indexing for anti-infodemic modeling. The results show distinct language generalization patterns of misinformation of favoring evaluative terms and multimedia devices in delivering a negative sentiment. Such appeals are effective to arouse people's sympathy toward the vulnerable community and foment their spreading behavior.

show abstract

MEGA: Machine Learning-Enhanced Graph Analytics for Infodemic Risk Management

Cited by 2 publications

References 77 publications

Infodemic: Challenges and solutions in topic discovery and data process

Infodemic: Challenges and solutions in topic discovery and data process

Data-driven analytics of COVID-19 ‘infodemic’

Contact Info

Product

Resources

About