Using Siamese Graph Neural Networks for Similarity-Based Retrieval in Process-Oriented Case-Based Reasoning

Using Siamese Graph Neural Networks for Similarity-Based Retrieval in Process-Oriented Case-Based Reasoning
复制标题

在面向过程的基于案例的推理中使用连体图神经网络进行基于相似性的检索

DOI:
10.1007/978-3-030-58342-2_15
复制
发表时间:
2020
期刊:
影响因子:
--
通讯作者:
Ralph Bergmann
Ralph Bergmann
中科院分区:
--
文献类型:
--
作者:
Maximilian Hoffmann;Lukas Malburg;Patrick Klein;Ralph Bergmann

文献摘要

参考文献

被引文献

相似文献

基于相似度的语义图检索广泛应用于实际场景,例如业务工作流领域。为了解决检索过程中复杂且耗时的图相似度计算问题,在面向过程的基于案例的推理(POCBR)中使用了mac / facc方法,其中从预选的候选图集中提取相似图。这些图是使用计算成本低廉的相似度量进行相似计算得出的。本文的贡献是一种新的相似度度量,其中使用两个siameseGraph神经网络(gnn)生成的向量空间嵌入来近似精确但因此计算复杂的图相似度度量的相似性。我们的方法包括语义图的特定编码方案,使其能够在神经网络中使用。评估检查了这些模型在预选检索候选者和近似两个工作流域的图相似度度量的真实相似度方面的质量和性能。结果显示了该方法在MAC/FAC场景中的巨大潜力,无论是作为预选模型还是作为图相似性度量的近似值。
Similarity-based retrieval of semantic graphs is widely used in real-world scenarios, e. g., in the domain of business workflows. To tackle the problem of complex and time-consuming graph similarity computations during retrieval, theMAC/FACapproach is used inProcess-Oriented Case-Based Reasoning(POCBR), where similar graphs are extracted from a preselected set of candidate graphs. These graphs result from a similarity computation with a computationally inexpensive similarity measure. The contribution of this paper is a novel similarity measure where vector space embeddings generated by two siameseGraph Neural Networks(GNNs) are used to approximate the similarities of a precise but therefore computationally complex graph similarity measure. Our approach includes a specific encoding scheme for semantic graphs that enables their usage in neural networks. The evaluation examines the quality and performance of these models in preselecting retrieval candidates and in approximating the ground-truth similarities of the graph similarity measure for two workflow domains. The results show great potential of the approach for being used in a MAC/FAC scenario, either as a preselection model or as an approximation of the graph similarity measure.
通过优化方法进行广义案例的相似性评估
DOI: --
发表时间: 2002
期刊: European Conference on Case-Based Reasoning
影响因子: --
作者:
Babak Mougouie;R. Bergmann
通讯作者: R. Bergmann
相似性和实用性的基础
DOI: --
发表时间: 2007
期刊: The Florida AI Research Society
影响因子: --
作者:
M. Richter
通讯作者: M. Richter
通过面向过程的基于案例的推理来适应科学工作流程
DOI: 10.1007/978-3-030-29249-2_26
发表时间: 2019
期刊:
影响因子: --
作者:
Christian Zeyen;Lukas Malburg;Ralph Bergmann
通讯作者: Ralph Bergmann
面向过程的基于案例的推理
DOI: 10.1016/j.is.2013.06.004
发表时间: 2014
期刊: Inf. Syst.
影响因子: --
作者:
Mirjam Minor;S. Montani;Juan A. Recio
通讯作者: Juan A. Recio
学习工作流嵌入以提高面向过程的基于案例的推理的基于相似性的检索的性能
DOI: --
发表时间: 2019
期刊: International Conference on Case-Based Reasoning
影响因子: --
作者:
P. Klein;Lukas Malburg;R. Bergmann
通讯作者: R. Bergmann