Identifying and Alleviating Concept Drift in Streaming Tensor Decomposition
Identifying and Alleviating Concept Drift in Streaming Tensor Decomposition
复制标题
DOI：
10.1007/978-3-030-10928-8_20
复制
发表时间：
2018-04
期刊：
影响因子：
--
通讯作者：
Ravdeep Pasricha;Ekta Gujral;E. Papalexakis
中科院分区：
文献类型：
--
作者：
Ravdeep Pasricha;Ekta Gujral;E. Papalexakis
张量分解被用于从社交网络到医疗应用的各种数据挖掘应用中，并且在发现潜在结构或概念化数据方面非常有用。许多现实世界的应用程序本质上是动态的，它们的数据也是如此。为了处理数据的这种动态性质，存在各种在线张量分解算法。所有这些算法的一个核心假设是，潜在概念的数量在整个流中保持固定。然而，情况不必如此。流中的每个传入批次可能具有不同数量的潜在概念，并且从一个张量批次到另一个张量批次的潜在概念的差异可以提供对我们在特定应用中的发现如何随着时间的推移而表现和偏离的见解。在本文中，我们定义“概念”和“概念漂移”的上下文中的流张量分解，作为整个流的潜在概念的可变性的表现。此外，我们还介绍了SeekAndDestroy（方法名称是在Metallica第一张专辑中的歌曲之后（也是对Metallica的致敬）），这是一种检测流张量分解中概念漂移的算法，并且能够产生对该漂移鲁棒的结果。据我们所知，这是第一个研究流张量分解中概念漂移的工作。我们广泛评估SeekAndDestroyon合成数据集，这些数据集表现出各种各样的现实漂移。我们的实验证明了SeekAndDestroy在检测概念漂移和减轻其影响方面的有效性，产生的结果与一次性分解整个张量的质量相似。此外，在真实的数据集中，SeekAndDestroy的性能优于其他流基线，同时发现了新的有用组件。与本文相关的代码可在以下网址获得： https://github.com/ravdeep003/conceptDrift .
Tensor decompositions are used in various data mining applications from social network to medical applications and are extremely useful in discovering latent structures orconceptsin the data. Many real-world applications are dynamic in nature and so are their data. To deal with this dynamic nature of data, there exist a variety of online tensor decomposition algorithms. A central assumption in all those algorithms is that the number of latent concepts remains fixed throughout the entire stream. However, this need not be the case. Every incoming batch in the stream may have a different number of latent concepts, and the difference in latent concepts from one tensor batch to another can provide insights into how our findings in a particular application behave and deviate over time. In this paper, we define “concept” and “concept drift” in the context of streaming tensor decomposition, as the manifestation of the variability of latent concepts throughout the stream. Furthermore, we introduceSeekAndDestroy(The method name is after (and a tribute to) Metallica’s song from their first album (who also owns the copyright for the name)), an algorithm that detects concept drift in streaming tensor decomposition and is able to produce results robust to that drift. To the best of our knowledge, this is the first work that investigates concept drift in streaming tensor decomposition. We extensively evaluateSeekAndDestroyon synthetic datasets, which exhibit a wide variety of realistic drift. Our experiments demonstrate the effectiveness ofSeekAndDestroy, both in the detection of concept drift and in the alleviation of its effects, producing results with similar quality to decomposing the entire tensor in one shot. Additionally, in real datasets,SeekAndDestroyoutperforms other streaming baselines, while discovering novel useful components. Code related to this paper is available at: https://github.com/ravdeep003/conceptDrift .