Entity Retrieval

Generalizing recent attention to retrieving entities and not just documents, we introduce two entity retrieval tasks: list completion and entity ranking. For each task, we propose and evaluate several algorithms. One of the core challenges is to overcome the very limited amount of information that serves as input—to address this challenge we explore different representations of list descriptions and/or example entities, where entities are represented not just by a textual description but also by the description of related entities. For evaluation purposes we make use of the lists and categories available in Wikipedia. Experimental results show that cluster-based contexts improve retrieval results for both tasks.