PINA: Leveraging Side Information in eXtreme Multi-label Classification via Predicted Instance Neighborhood Aggregation

Keyphrases

• Neighborhood Aggregation 100%
• Extreme multi-label Classification 100%
• State-of-the-art Techniques 14%
• Extracting Information 14%
• Real-world Application 14%
• Enhancement Mechanism 14%
• Semantic Features 14%
• Publicly Available 14%
• Benchmark Dataset 14%
• Electronic Commerce 14%
• Transformer 14%
• Product Search 14%
• Recommendation System 14%
• Label Space 14%
• Data Enhancement 14%
• Consistent Gain 14%
• Label Metadata 14%

Computer Science

• Side Information 100%
• Large Data Set 33%
• Experimental Result 33%
• World Application 33%
• Semantic Feature 33%
• Training Instance 33%
• Extract Information 33%
• Data Enhancement 33%