SUDMAD: Sequential and unsupervised decomposition of a multi-author document based on a hidden markov model

Keyphrases

• Hidden Markov Model 100%
• Document-oriented 100%
• Multi-author 100%
• Classification Results 40%
• Increased Demand 20%
• Superior Performance 20%
• Training Data 20%
• Benchmark Dataset 20%
• Document Analysis 20%
• Authorship Analysis 20%
• Learning Procedure 20%
• Model Learning 20%
• Plagiarism Detection 20%
• Writing Style 20%
• Forensic Analysis 20%
• Civil Law 20%
• Sentence-based 20%
• Specific Language 20%
• Document Context 20%
• Contextual Correlation 20%
• Contextual Relations 20%
• Copyright Issues 20%
• Document Decomposition 20%

Computer Science

• Document Author 100%
• classification result 66%
• Written Document 66%
• Superior Performance 33%
• Training Data 33%
• Statistical Approach 33%
• Plagiarism Detection 33%
• Document Context 33%
• Language Specific 33%
• Authorship Analysis 33%
• Markov Learning Model 33%
• Initial Value 33%