A heterogeneous information network (HIN) is a ubiquitous data model, consisting of multiple types of entities and relations. Names of entities in HINs are inherently ambiguous, making it difficult to fully disambiguate a HIN. In this paper, we introduce the task of exploratory entity linking for HINs. Given a partially disambiguated HIN, we aim at linking ambiguous names to disambiguated entities in the HIN if their referent entities are present. We also try to “explore” other alternatives by discovering new entities and adding them to the HIN. A partial classification EM-based approach is proposed to address this task. We present a constrained probability propagation model to link surface names to entities in the HIN. New entity detection process is modeled as a maximum edge weight clique problem. Experiments illustrate that our method outperforms state-of-the-art methods for entity linking with HINs and author name disambiguation.

中文翻译：

---

HEEL：异构信息网络的探索性实体链接

异构信息网络（HIN）是一种普遍存在的数据模型，由多种类型的实体和关系组成。HIN中的实体名称本质上是模棱两可的，因此很难完全消除HIN的歧义。在本文中，我们介绍了HIN探索性实体链接的任务。考虑到部分歧义的HIN，我们旨在将歧义名称链接到HIN中歧义的实体（如果存在其引用实体）。我们还尝试通过发现新实体并将其添加到HIN中来“探索”其他选择。提出了一种基于EM的部分分类的方法来解决此任务。我们提出了一种约束概率传播模型，以将表面名称链接到HIN中的实体。新的实体检测过程被建模为最大边缘权重集团问题。