Summary

Fehrman, Elaine; Egan, Vincent; Gorban, Alexander N.; Levesley, Jeremy; Mirkes, Evgeny M.; Muhammad, Awaz K.

doi:10.1007/978-3-030-10442-9_5

Cited by 1 publication

(8 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The drug consumption data set was collected and made publicly available by Fehrmann et al (2017) on the UCI ML Repository (Dua & Graff, 2019). The data were gathered between March 2011 and March 2012 and comprise drug usage records of N = 1, 885 individuals ( n female = 942, n male = 943).…”

Section: Methodsmentioning

confidence: 99%

“…Despite the name and subscales, the ImpSS is regarded as a tool to assess general sensation-seeking (Zuckerman, 1994). Unfortunately, Fehrmann et al (2017) only provided participants’ aggregated scores on these scales, preventing us from including all items as separate predictors (for an example of using a large number of individual questionnaire items as predictors, see Pargent & Albert-von der Gönna, 2018).…”

Section: Methodsmentioning

confidence: 99%

“…Conveniently, Fehrmann et al (2017) transformed all predictors into continuous variables. That is, they quantified all ordinal (e.g., the answers on the Likert scales) and all nominal (e.g., gender or ethnicity) predictors so that ML algorithms can be easily applied.…”

Section: Methodsmentioning

confidence: 99%

“…We want to demonstrate a selection of the most commonly used CSL methods on a psychological data set to propagate the usefulness of CSL in psychological classification problems 1 . For our empirical demonstration, we use the openly available drug consumption data set (Fehrmann et al, 2017) from the University of California Irvine (UCI) ML Repository (Dua & Graff, 2019). This data set contains records of the usage of 18 legal and illegal drugs for 1,885 individuals.…”

Section: In Psychological Sciencementioning

confidence: 99%

“…Thanks to Fehrmann et al (2017) making their data set on drug consumption and personality publicly available on the UCI ML Repository (Dua & Graff, 2019), another body of research has emerged over the last few years: Many studies have been conducted with the goal of predicting drug consumption on the basis of the Big Five as well as impulsivity and sensation seeking (Kumari et al, 2018; Liu, 2019; Nath et al, 2017; Qiao et al, 2019). In these studies, different kinds of ML algorithms (e.g., random forests or artificial neural networks) were used to predict the usage of various licit or illicit drugs.…”

Section: In Psychological Sciencementioning

confidence: 99%

See 4 more Smart Citations

Everything has its price: Foundations of cost-sensitive machine learning and its application in psychology.

Sterner

Goretzko

Pargent

2023

Psychological Methods

View full text Add to dashboard Cite

Psychology has seen an increase in the use of machine learning (ML) methods. In many applications, observations are classified into one of two groups (binary classification). Off-the-shelf classification algorithms assume that the costs of a misclassification (false positive or false negative) are equal. Because this is often not reasonable (e.g., in clinical psychology), cost-sensitive machine learning (CSL) methods can take different cost ratios into account. We present the mathematical foundations and introduce a taxonomy of the most commonly used CSL methods, before demonstrating their application and usefulness on psychological data, that is, the drug consumption data set (N = 1, 885) from the University of California Irvine ML Repository. In our example, all demonstrated CSL methods noticeably reduced mean misclassification costs compared to regular ML algorithms. We discuss the necessity for researchers to perform small benchmarks of CSL methods for their own practical application. Thus, our open materials provide R code, demonstrating how CSL methods can be applied within the mlr3 framework (https://osf.io/cvks7/).

show abstract

Section: Methodsmentioning

confidence: 99%