Miao Zhengjie, Zeng Qitian, Glavic Boris, Roy Sudeepa
Duke University.
IIT.
Proc ACM SIGMOD Int Conf Manag Data. 2019 Jun;2019:485-502. doi: 10.1145/3299869.3300066.
Provenance and intervention-based techniques have been used to explain surprisingly high or low outcomes of aggregation queries. However, such techniques may miss interesting explanations emerging from data that is in the provenance. For instance, an unusually low number of publications of a prolific researcher in a certain venue and year can be explained by an increased number of publications in another venue in the same year. We present a novel approach for explaining outliers in aggregation queries through . That is, explanations are outliers in the opposite direction of the outlier of interest. Outliers are defined w.r.t. patterns that hold over the data in aggregate. We present efficient methods for mining such (), discuss how to use ARPs to generate and rank explanations, and experimentally demonstrate the efficiency and effectiveness of our approach.
基于来源和干预的技术已被用于解释聚合查询中令人惊讶的高或低结果。然而，这些技术可能会遗漏来源数据中出现的有趣解释。例如，一位多产的研究人员在某一场所和年份的出版物数量异常低，可以通过同年在另一场所的出版物数量增加来解释。我们提出了一种通过 来解释聚合查询中的异常值的新颖方法。也就是说，解释是与感兴趣的异常值方向相反的异常值。异常值是根据聚合数据中存在的模式来定义的。我们提出了挖掘此类 （）的有效方法，讨论了如何使用ARP来生成解释并对其进行排名，并通过实验证明了我们方法的效率和有效性。
