Crowdsourcing Malware Family Annotation: Joint Class-Determined Tag Extraction and Weakly-Tagged Sample Inference

Keyphrases

• Crowdsourcing 100%
• Malware Family 100%
• Tag Extraction 100%
• Locator 75%
• Incremental Parsing 75%
• Malware 50%
• Parsing 50%
• Expectation-maximization Algorithm 25%
• Maximum Likelihood 25%
• Maximum Likelihood Estimation 25%
• Family Activities 25%
• Benchmark Dataset 25%
• Confusion Matrix 25%
• Anti-malware 25%
• Extract Class 25%
• Estimation Scheme 25%
• Security Community 25%
• Annotation pipeline 25%
• Malice 25%
• Unknown Truths 25%

Computer Science

• Annotation 100%
• Parsing 100%
• Malware Family 100%
• Malware 60%
• maximum-likelihood 40%
• Security Community 20%
• Likelihood Estimation 20%
• Estimation Scheme 20%
• Confusion Matrix 20%
• Family Behavior 20%