Margin-based active learning for structured predictions

Keyphrases

• Active Learning 100%
• Active Learning Methods 40%
• Annotated Dataset 20%
• Application Domain 20%
• Binary Prediction 20%
• Complexity Prediction 40%
• Data Requirements 20%
• Domain Complexity 20%
• Early Prediction 20%
• Extraction System 20%
• Labeling Problem 20%
• Learning Paradigm 20%
• Machine Learning Techniques 20%
• Multi-class Prediction 20%
• Multiple Instances 20%
• Named Entity Extraction 20%
• Natural Language Processing Applications 20%
• Pipeline Model 20%
• Prediction Problems 60%
• Relation Extraction 20%
• Semantic Role Labeling 20%
• Structured Output Spaces 20%
• Structured Prediction 100%
• Synthetic Data 20%

Computer Science

• Active Learning 100%
• Active Learning Technique 66%
• Application Domain 33%
• Based Active Learning 100%
• Data Requirement 33%
• Machine Learning Technique 33%
• Multiple Instance 33%
• Natural Language Processing 33%
• Semantic Role Labeling 33%
• Synthetic Data 33%