Empirical assessment of bias in machine learning diagnostic test accuracy studies.
Ryan J. CrowleyYuan Jin TanJohn P. A. IoannidisPublished in: J. Am. Medical Informatics Assoc. (2020)
Keyphrases
- text mining
- machine learning
- information extraction
- bias variance
- natural language processing
- error rate
- data mining
- knowledge discovery
- empirical studies
- high accuracy
- medical diagnosis
- data analysis
- trade off
- fold cross validation
- decision making
- artificial intelligence
- empirical data
- statistical significance
- meta analysis
- machine learning approaches
- test data
- decision trees
- variance reduction
- supervised learning
- expert systems
- statistically significant
- explanation based learning
- experimental design
- measurement error
- diagnostic tests
- false positive and false negative
- bias variance analysis
- highly accurate
- theoretical analysis
- bayesian networks
- computational intelligence
- reinforcement learning
- pattern recognition
- natural language
- semi supervised
- knowledge representation
- classification accuracy
- computational cost
- active learning