论文部分内容阅读
实体识别在数据质量管理中起着重要的作用，然而，当前的实体识别方法仅局限于数据库中的单个关系，而没有考虑多关系数据库中的实体识别，而多个关系中的信息可以有消息用于增强实体识别结果，基于这个考虑，本文提出了基于识别层次的实体识别算法，按照重要性组织关系，按照层次逐层进行识别，实验结果本文提出的方法可以在多关系的数据库中快速有效地实现实体识别且优于现有算法。“,”Entity resolution plays an important role in data quality management. However, current entity resolution focuses on single relation but not considers entity resolution on multi-relation databases. Clearly, in a multi-relation database, the information from multiple relations could improve the results of entity resolution. According to this consideration, this paper proposes entity resolution algorithms based on hierarchy. This method organizes relations according to the importance and performs entity resolution along the hierarchy. Experimental results show that the method proposed in this paper could perform the entity resolution efficiently and effectively. It outperforms current methods.