ENT Rank: Retrieving Entities for Topical Information Needs through Entity-Neighbor-Text Relations
ENT Rank: Retrieving Entities for Topical Information Needs through Entity-Neighbor-Text Relations
复制标题
DOI：
10.1145/3331184.3331257
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Laura Dietz
中科院分区：
文献类型：
--
作者：
Laura Dietz
相关工作已经证明了在文本检索中利用实体信息的有用性；在这里，我们探讨相反的情况：在实体检索中利用有关文本的信息。我们对实体-邻居-文本（ENT）关系的相关性进行建模，以导出学习对实体进行排名的模型。我们专注于检索（多个）相关实体的任务，以响应“寨卡热”等主题信息需求。 ENT Rank 模型旨在利用半结构化知识资源（例如维基百科）进行实体检索。 ENT Rank 模型结合了 (1) 实体相关性的既定特征，(2) 来自相邻实体（共同提及或页面提及）的信息，通过 (3) 通过传统检索模型（如 BM25 和 RM3）获得的文本上下文的相关性分数。
Related work has demonstrated the helpfulness of utilizing information about entities in text retrieval; here we explore the converse: Utilizing information about text in entity retrieval. We model the relevance of Entity-Neighbor-Text (ENT) relations to derive a learning-to-rank-entities model. We focus on the task of retrieving (multiple) relevant entities in response to a topical information need such as "Zika fever". The ENT Rank model is designed to exploit semi-structured knowledge resources such as Wikipedia for entity retrieval. The ENT Rank model combines (1) established features of entity-relevance, with (2) information from neighboring entities (co-mentioned or mentioned-on-page) through (3) relevance scores of textual contexts through traditional retrieval models such as BM25 and RM3.