Face Value: Remote facial expression analysis adds predictive power to perceived effectiveness for selecting anti-tobacco PSAs

Parvanta, Claudia; Hammond, Robert W; He, Weiwei; Zemen, Rahel; Boddupalli, Samantha; Walker, Kim; Chen, H; Harner, R N

doi:10.1080/10810730.2022.2100016

Cited by 4 publications

(4 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We chose to move forward by collaborating with the lab's technology partner (Pedersen, 2021) to develop the capability to collect facial expression and eye tracking data online. The data and studies suggest online biometric data collection is a valuable tool for connecting with uncommon population segments (Parvanta et al, 2022) and can be used to collect data quickly. Our experience is also consistent with prior research describing challenges with crowdsourced populations.…”

Section: Volume 7 Number 12mentioning

confidence: 99%

“…Our experience is also consistent with prior research describing challenges with crowdsourced populations. We have encountered significant loss of usable data with field online data collection (Hammond et al, 2022;Parvanta et al, 2022). Moreover, eye tracking (ET) visualizations from online field data seem to be more diffused and to include more spurious focus points when compared to lab-based equipment.…”

Section: Volume 7 Number 12mentioning

confidence: 99%

See 1 more Smart Citation

Crowdsourced Online Biometric Studies: Is the juice worth the squeeze?

W Hammond,

Wang

2023

MBR

View full text Add to dashboard Cite

While promising, online biometric studies with crowdsourced participants have vagaries that can limit success. This paper shares researcher experience with biometric online data collection and a study that compared the differences of online and lab-based eye tracking (ET) and facial expression (FE) data. Students and the public were separated into two groups to place orders from a well-known restaurant online menu application. Group 1 accessed the menu web application routed through an online biometric data collection tool while Group 2 interacted with the menu web application through a desktop application that collected the biometric data. The study found little difference in the user experience metrics, the facial expression data, or the number of gaze points between the two data collection methods. Interestingly and consistent with field experience a significant difference in the ET data was observed. The vagaries of the field present significant hazards for researchers and even more so if study participants are compensated. Recommendations for researchers to optimize research effort are provided regarding recruitment, participant compensation, analysis effort and expertise, maximizing the usable data, and employing eye tracking tools.

show abstract

Section: Volume 7 Number 12mentioning

confidence: 99%

Section: Volume 7 Number 12mentioning

confidence: 99%

Crowdsourced Online Biometric Studies: Is the juice worth the squeeze?

W Hammond,

Wang

2023

MBR

View full text Add to dashboard Cite

show abstract

“…This term encompasses ‘a wide array of services, policies, and practices within an equally wide range of institutional and governmental entities’ (Green et al, 2022, p. 1). These are ‘concerned with promoting the conditions that give everyone a chance to have a healthy life’ (Parvanta et al, 2017, n.p.). Public health places a considerable emphasis on health promotion as a way to promote good health in the broader community.…”

Section: Social Media Influencers and Public Health Pedagogymentioning

confidence: 99%

Public health pedagogy and digital misinformation: Health professional influencers and the politics of expertise

Thompson

2022

Journal of Sociology

View full text Add to dashboard Cite

This article asks: ‘To what extent can health professional influencers function as health pedagogues, educating their audiences and protecting public health in an era of digital misinformation?’ The article teases out that question by applying Content Analysis and Framing Analysis to a selection of TikTok and Instagram posts by Dr Michael Mrozinski, a Scottish general practitioner who is based in Australia. The posts seek to debunk online misinformation and provide facts regarding COVID-19. Mrozinski's social media content exemplifies what the article terms ‘public health pedagogy’ (PHP) – pedagogy that is informed by public health principles and that is undertaken outside traditional educational institutions. The article also asks: ‘How exactly does Mrozinski respond to misinformation actors and to what extent does this diminish the effectiveness of his PHP?’ The article investigates whether Mrozinski's hostility towards these actors actually invokes stereotypes of medical experts as elitist and uncaring. Those stereotypes are commonly expounded by misinformation and conspiracy actors.

show abstract

“…In human-computer interactions, by identifying the facial expressions and further analyze the expressions, researchers could find out the feelings of the users to merchandises. In social network analysis, it could seek and extract relations among people, by analysing the interactions based on facial expressions [6]. Besides conventional machine learning solutions, with the deepening of research, neutral expressions have also been added to basic facial expressions by researchers, forming seven basic facial expressions in today's facial expression recognition research.…”

Section: Introductionmentioning

confidence: 99%

Recognizing facial emotion leveraging convolutional neural network

Yang,

Zhang,

Zheng

2023

ACE

View full text Add to dashboard Cite

Facial expression recognition technology is a technology based on the combination of artificial intelligence technology and biology, and it is interdisciplinary research. The appearance of this technology shows the diversity of the development direction and application fields of computer technology, but it also means that the development of facial expression recognition technology needs not only the support of computer technology but also the exploration and progress of biology. At present, to recognize facial expressions a system usually contains three stages, namely, face detection stage, feature extraction stage, and face expression recognition stage. The facial expression recognition algorithm is transformed into a classification problem, and a convolutional neural network (CNN) is adopted, where features of the facial expressions could be automatically extracted. It takes as input a preprocessed image and can directly produce encoded features and predictions, leveraging the end-to-end strategy. It could discard the complicated intermediate modeling process of traditional machine learning. In this work a CNN is implemented for the recognition of facial expression on Fer2013 dataset. Moreover, the effectiveness of different numbers of CNN layers are testified on this problem. It could be concluded that with deeper architectures, the CNN tends to perform better.

show abstract

Face Value: Remote facial expression analysis adds predictive power to perceived effectiveness for selecting anti-tobacco PSAs

Cited by 4 publications

References 66 publications

Crowdsourced Online Biometric Studies: Is the juice worth the squeeze?

Crowdsourced Online Biometric Studies: Is the juice worth the squeeze?

Public health pedagogy and digital misinformation: Health professional influencers and the politics of expertise

Recognizing facial emotion leveraging convolutional neural network

Contact Info

Product

Resources

About