Predicting Risk Perception: New Insights from Data Science

Bhatia, Sudeep

doi:10.1287/mnsc.2018.3121

Cited by 66 publications

(97 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given that we have fewer judgment targets (<=200) than independent variables (300), we chose three regularized regression techniques: lasso, ridge, and support vector machines. We also included k-nearest neighbors (KNN) regression since it was used in work on predicting risk perceptions with word embeddings [19]. We describe our secondary models (lasso, SVR, KNN) more in the SI, and instead focus here on our primary model, ridge regression.…”

Section: Methodsmentioning

confidence: 99%

“…When a new entity is presented to be judged, it applies its learned mapping to the semantic vector for the new entity to predict participant judgments for that entity. Some prior work has used this approach to generate word norms in psycholinguistic studies [18], and to study risk perception [19]. In principle, this approach can be applied to any type of human judgment, as long as judgment targets are in the word embeddings vocabulary.…”

Section: From Representation To Judgmentmentioning

confidence: 99%

See 1 more Smart Citation

Semantic representations extracted from large language corpora predict high-level human judgment in seven diverse behavioral domains

Richie¹,

Zou²,

Bhatia³

2018

Preprint

Self Cite

View full text Add to dashboard Cite

Recent advances in machine learning, combined with the increased availability of large natural language datasets, have made it possible to uncover semantic representations that characterize what people know about and associate with a wide range of objects and concepts. In this paper, we examine the power of word embeddings, a popular approach for uncovering semantic representations, for studying high-level human judgment. Word embeddings are typically applied to linguistic and semantic tasks, however we show that word embeddings can be used to predict complex theoretically- and practically-relevant human perceptions and evaluations in domains as diverse as social cognition, health behavior, risk perception, organizational behavior, and marketing. By learning mappings from word embeddings directly onto judgment ratings, we outperform a similarity-based baseline as well as common metrics of human inter-rater reliability. Word embeddings are also able to identify the concepts that are most associated with observed perceptions and evaluations, and can thus shed light on the psychological substrates of judgment. Overall, we provide new methods and insights for predicting and understanding high-level human judgment, with important applications across the social and behavioral sciences.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: From Representation To Judgmentmentioning

confidence: 99%

Semantic representations extracted from large language corpora predict high-level human judgment in seven diverse behavioral domains

Richie¹,

Zou²,

Bhatia³

2018

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“… 2019 ). In another instance, the same author conducted experiments with 300 participants and predicted the perceived risk of several risk sources using a vector-space representation of natural language, concluding that the word distribution of language successfully captures human perception of risk (Bhatia 2019 ). Similar work has been conducted by Jaidka et al.…”

Section: Introductionmentioning

confidence: 99%

Public risk perception and emotion on Twitter during the Covid-19 pandemic

Dyer

Kolic

2020

Appl Netw Sci

View full text Add to dashboard Cite

Successful navigation of the Covid-19 pandemic is predicated on public cooperation with safety measures and appropriate perception of risk, in which emotion and attention play important roles. Signatures of public emotion and attention are present in social media data, thus natural language analysis of this text enables near-to-real-time monitoring of indicators of public risk perception. We compare key epidemiological indicators of the progression of the pandemic with indicators of the public perception of the pandemic constructed from $$\sim 20$$ ∼ 20 million unique Covid-19-related tweets from 12 countries posted between 10th March and 14th June 2020. We find evidence of psychophysical numbing: Twitter users increasingly fixate on mortality, but in a decreasingly emotional and increasingly analytic tone. Semantic network analysis based on word co-occurrences reveals changes in the emotional framing of Covid-19 casualties that are consistent with this hypothesis. We also find that the average attention afforded to national Covid-19 mortality rates is modelled accurately with the Weber–Fechner and power law functions of sensory perception. Our parameter estimates for these models are consistent with estimates from psychological experiments, and indicate that users in this dataset exhibit differential sensitivity by country to the national Covid-19 death rates. Our work illustrates the potential utility of social media for monitoring public risk perception and guiding public communication during crisis scenarios.

show abstract

“…Given that we have fewer judgment targets (< = 200) than independent variables (300), we chose three regularized regression techniques: lasso, ridge, and support vector machines. We also included k-nearest neighbors (KNN) regression since it was used in work on predicting risk perceptions with word embeddings (Bhatia, 2019). We describe our secondary models (lasso, SVR, KNN) more in the SI, and instead focus here on our primary model, ridge regression.…”

Section: Methodsmentioning

confidence: 99%

“…When a new entity is presented to be judged, it applies its learned mapping to the semantic vector for the new entity to predict participant judgments for that entity. Some prior work has used this approach to generate word norms in psycholinguistic studies (Hollis, Westbury, & Lefsrud, 2017), and to study risk perception (Bhatia, 2019) and brand judgments (Bhatia & Olivola, 2018). In principle, this approach can be applied to any type of human judgment, as long as judgment targets are in the word embeddings vocabulary.…”

Section: From Representation To Judgmentmentioning

confidence: 99%

Predicting High-Level Human Judgment Across Diverse Behavioral Domains

Richie

Zou

Bhatia

2019

Collabra: Psychology

Self Cite

View full text Add to dashboard Cite

Recent advances in machine learning, combined with the increased availability of large natural language datasets, have made it possible to uncover semantic representations that characterize what people know about and associate with a wide range of objects and concepts. In this paper, we examine the power of word embeddings, a popular approach for uncovering semantic representations, for studying high-level human judgment. Word embeddings are typically applied to linguistic and semantic tasks, however we show that word embeddings can be used to predict complex theoretically-and practically-relevant human perceptions and evaluations in domains as diverse as social cognition, health behavior, risk perception, organizational behavior, and marketing. By learning mappings from word embeddings directly onto judgment ratings, we outperform a similarity-based baseline and perform favorably compared to common metrics of human inter-rater reliability. Word embeddings are also able to identify the concepts that are most associated with observed perceptions and evaluations, and can thus shed light on the psychological substrates of judgment. Overall, we provide new methods and insights for predicting and understanding high-level human judgment, with important applications across the social and behavioral sciences.

show abstract

Predicting Risk Perception: New Insights from Data Science

Cited by 66 publications

References 60 publications

Semantic representations extracted from large language corpora predict high-level human judgment in seven diverse behavioral domains

Semantic representations extracted from large language corpora predict high-level human judgment in seven diverse behavioral domains

Public risk perception and emotion on Twitter during the Covid-19 pandemic

Predicting High-Level Human Judgment Across Diverse Behavioral Domains

Contact Info

Product

Resources

About