Performance of active learning models for screening prioritization in systematic reviews: a simulation study into the Average Time to Discover relevant records

Keyphrases

• Simulation Study 100%
• Active Learning 100%
• Learning Model 100%
• Screening Prioritization 100%
• Time to Discovery 85%
• Over Sampling 42%
• Term Frequency-inverse Document Frequency (TF-IDF) 28%
• Naïve Bayes 28%
• Logistic Regression 14%
• Range of Values 14%
• Random Forest 14%
• Performance Measures 14%
• Feature Extraction Methods 14%
• Different Datasets 14%
• Classification Methods 14%
• Number of Publications 14%
• Screening Process 14%
• Machine Learning Software 14%
• Doc2Vec 14%
• Regression Support Vector Machine 14%
• Sampled Values 14%
• New Statistics 14%
• Review Dataset 14%

Computer Science

• Active Learning 100%
• Simulation Study 100%
• Support Vector Machine 14%
• Performance Measure 14%
• Random Decision Forest 14%
• Classification Technique 14%
• Logistic Regression 14%
• Screening Process 14%
• Machine Learning 14%
• Learning System 14%
• Feature Extraction 14%

Chemical Engineering

• Learning System 100%
• Support Vector Machine 100%