Refining fine-tuned transformers with hand-crafted features for gender screening on question-answering communities

Keyphrases

• Artificial Intelligence Architecture 50%
• Automatic Gender Recognition 50%
• Bayes 50%
• Community Question Answering 50%
• Community Question Answering Sites 50%
• Complex Goals 50%
• Demographic Analysis 50%
• Dependency Parsing 50%
• Discriminant Function 50%
• Downstream Task 50%
• Encoder 50%
• Encrypted Information 50%
• Gender Classification 50%
• Gender Disparity 50%
• Generative Transformer 50%
• Handcrafted Features 100%
• Linguistic Features 50%
• Linguistic Properties 50%
• Machine Learning Analysis 50%
• Malicious Behavior 50%
• Maximum Entropy 50%
• Model Fine-tuning 50%
• Multiple Disadvantage 50%
• New Subjects 50%
• Overfitting 50%
• Pre-trained Encoder 50%
• Prediction Rate 50%
• Question Answering Communities 100%
• Small Dataset 50%
• Statistical Classifiers 50%
• Target Domain 50%
• Traditional Classifiers 50%
• Transference 50%
• Transformer 100%
• Wrapper 50%
• XLNet 50%

Computer Science

• Artificial Intelligence 100%
• dependency parse 100%
• Discriminant Function 100%
• Field Research 100%
• Gender Recognition 100%
• Interpretability 100%
• Linguistic Property 100%
• Machine Learning 100%
• Malicious Activity 100%

Psychology

• Artificial Intelligence 100%