Predicting online lecture ratings based on gesturing and vocal behavior

Cheng, Dong; Salamin, Hugues; Salvagnini, Pietro; Cristani, Marco; Vinciarelli, Alessandro; Murino, Vittorio

doi:10.1007/s12193-013-0142-z

Cited by 15 publications

(6 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This accuracy is obtained by training the classifier without the unfairness measure in the loss function. The mean accuracy obtained from an unfair classifier is comparable to accuracy reported in recent studies on TED talk data (Cheng et al 2014;Tanveer et al 2019). However, note that our language model (Le and Mikolov 2014) is much simpler when compared to methods used in the cited studies.…”

Section: Prediction Accuracysupporting

confidence: 83%

FairyTED: A Fair Rating Predictor for TED Talk Data

Acharyya

Das

Chattoraj

et al. 2020

AAAI

View full text Add to dashboard Cite

With the recent trend of applying machine learning in every aspect of human life, it is important to incorporate fairness into the core of the predictive algorithms. We address the problem of predicting the quality of public speeches while being fair with respect to sensitive attributes of the speakers, e.g. gender and race. We use the TED talks as an input repository of public speeches because it consists of speakers from a diverse community and has a wide outreach. Utilizing the theories of Causal Models, Counterfactual Fairness and state-of-the-art neural language models, we propose a mathematical framework for fair prediction of the public speaking quality. We employ grounded assumptions to construct a causal model capturing how different attributes affect public speaking quality. This causal model contributes in generating counterfactual data to train a fair predictive model. Our framework is general enough to utilize any assumption within the causal model. Experimental results show that while prediction accuracy is comparable to recent work on this dataset, our predictions are counterfactually fair with respect to a novel metric when compared to true data labels. The FairyTED setup not only allows organizers to make informed and diverse selection of speakers from the unobserved counterfactual possibilities but it also ensures that viewers and new users are not influenced by unfair and unbalanced ratings from arbitrary visitors to the ted.com website when deciding to view a talk.

show abstract

Section: Prediction Accuracysupporting

confidence: 83%

FairyTED: A Fair Rating Predictor for TED Talk Data

Acharyya

Das

Chattoraj

et al. 2020

AAAI

View full text Add to dashboard Cite

show abstract

“…This accuracy is obtained by training the classifier without the unfairness measure in the loss function. The mean accuracy obtained from an unfair classifier is comparable to accuracy reported in recent studies on TED talk data (Cheng et al 2014;. However, note that our language model (Le and Mikolov 2014) is much simpler when compared to methods used in the cited studies.…”

Section: Prediction Accuracysupporting

confidence: 79%

FairyTED: A Fair Rating Predictor for TED Talk Data

Acharyya¹,

Das²,

Chattoraj³

et al. 2019

Preprint

View full text Add to dashboard Cite

With the recent trend of applying machine learning in every aspect of human life, it is important to incorporate fairness into the core of the predictive algorithms. We address the problem of predicting the quality of public speeches while being fair with respect to sensitive attributes of the speakers, e.g. gender and race. We use the TED talks as an input repository of public speeches because it consists of speakers from a diverse community and has a wide outreach. Utilizing the theories of Causal Models, Counterfactual Fairness and stateof-the-art neural language models, we propose a mathematical framework for fair prediction of the public speaking quality. We employ grounded assumptions to construct a causal model capturing how different attributes affect public speaking quality. This causal model contributes in generating counterfactual data to train a fair predictive model. Our framework is general enough to utilize any assumption within the causal model. Experimental results show that while prediction accuracy is comparable to recent work on this dataset, our predictions are counterfactually fair with respect to a novel metric when compared to true data labels. The FairyTED setup not only allows organizers to make informed and diverse selection of speakers from the unobserved counterfactual possibilities but it also ensures that viewers and new users are not influenced by unfair and unbalanced ratings from arbitrary visitors to the ted.com website when deciding to view a talk.

show abstract

“…Thus, our results were similar to the reports of many other educational scientists. 19 , 27 , 28 Our respondents also agree that facilitators were using audiovisual aids effectively, such as slides. Therefore, the current study supports earlier research findings.…”

Section: Discussionmentioning

confidence: 84%

Evaluation of doctors' performance as facilitators in basic medical science lecture classes in a new Malaysian medical school

Ismail¹,

Salam²,

Alattraqchi³

et al. 2015

AMEP

View full text Add to dashboard Cite

BackgroundDidactic lecture is the oldest and most commonly used method of teaching. In addition, it is considered one of the most efficient ways to disseminate theories, ideas, and facts. Many critics feel that lectures are an obsolete method to use when students need to perform hands-on activities, which is an everyday need in the study of medicine. This study evaluates students’ perceptions regarding lecture quality in a new medical school.MethodsThis was a cross-sectional study conducted of the medical students of Universiti Sultan Zainal Abidin. The study population was 468 preclinical medical students from years 1 and 2 of academic year 2012–2013. Data were collected using a validated instrument. There were six different sections of questions using a 5-point Likert scale. The data were then compiled and analyzed, using SPSS version 20.ResultsThe response rate was 73%. Among 341 respondents, 30% were male and 70% were female. Eighty-five percent of respondents agree or strongly agree that the lectures had met the criteria with regard to organization of lecture materials. Similarly, 97% of students agree or strongly agree that lecturers maintained adequate voices and gestures.ConclusionMedical students are quite satisfied with the lecture classes and the lectures. However, further research is required to identify student-centered teaching and learning methods to promote active learning.

show abstract

Predicting online lecture ratings based on gesturing and vocal behavior

Cited by 15 publications

References 42 publications

FairyTED: A Fair Rating Predictor for TED Talk Data

FairyTED: A Fair Rating Predictor for TED Talk Data

FairyTED: A Fair Rating Predictor for TED Talk Data

Evaluation of doctors' performance as facilitators in basic medical science lecture classes in a new Malaysian medical school

Contact Info

Product

Resources

About

Predicting online lecture ratings based on gesturing and vocal behavior

Cited by 15 publications

References 42 publications

FairyTED: A Fair Rating Predictor for TED Talk Data

FairyTED: A Fair Rating Predictor for TED Talk Data

FairyTED: A Fair Rating Predictor for TED Talk Data

Evaluation of doctors&#39; performance as facilitators in basic medical science lecture classes in a new Malaysian medical school

Contact Info

Product

Resources

About

Evaluation of doctors' performance as facilitators in basic medical science lecture classes in a new Malaysian medical school