Unsupervised Context Retrieval for Long-tail Entities
This addresses the challenge of monitoring long-tail entities in media streams, which is incremental as it builds on existing entity representation methods.
The paper tackles the problem of retrieving textual contexts for monitoring long-tail entities, which lack rich representations in knowledge bases, by proposing an unsupervised method that leverages established entities as support, and evaluation shows its suitability and robustness for out-of-KB entities.
Monitoring entities in media streams often relies on rich entity representations, like structured information available in a knowledge base (KB). For long-tail entities, such monitoring is highly challenging, due to their limited, if not entirely missing, representation in the reference KB. In this paper, we address the problem of retrieving textual contexts for monitoring long-tail entities. We propose an unsupervised method to overcome the limited representation of long-tail entities by leveraging established entities and their contexts as support information. Evaluation on a purpose-built test collection shows the suitability of our approach and its robustness for out-of-KB entities.