A new swarm-based framework for handwritten authorship identification in forensic document analysis

Keyphrases

• Feature Selection 100%
• Writer Identification 100%
• Forensic Analysis 100%
• Authorship Attribution 100%
• Feature Selection Methods 42%
• Identification Framework 28%
• Class Distribution 28%
• Classification Accuracy 28%
• Selection Framework 28%
• Paper Documents 14%
• Class-specific 14%
• Feature Extraction 14%
• Pattern Recognition 14%
• Word Shape 14%
• Computationally Expensive 14%
• Attribute Selection 14%
• Handwriting Analysis 14%
• Near-optimal 14%
• Computational Cost 14%
• Individual Writing 14%
• Machine Learning Tasks 14%
• Handwriting Style 14%
• Feature Class 14%
• Identification Task 14%
• Good Characteristics 14%
• Biometric Applications 14%
• Framework Selection 14%
• Identification Feature 14%
• Biometrics 14%
• Sequential Selection 14%
• Forensic Application 14%
• High-dimensional Data 14%
• Writing Style 14%
• IAM Dataset 14%

Computer Science

• Feature Selection 100%
• Feature Extraction 100%
• Documentary Analysis 100%
• Class Distribution 18%
• Classification Accuracy 18%
• Machine Learning 9%
• Good Performance 9%
• High Dimensional Data 9%
• Selection Phase 9%
• Learning System 9%
• Biometrics 9%
• Pattern Recognition 9%
• Extracted Feature 9%
• Computational Cost 9%