Evolutionary Acquisition of Multiple TTSP Graph Patterns with Wildcards by Clustering TTSP Graphs
Evolutionary Acquisition of Multiple TTSP Graph Patterns with Wildcards by Clustering TTSP Graphs
复制标题
通过对 TTSP 图进行聚类来进化获取带有通配符的多个 TTSP 图模式
DOI：
10.1109/iwcia52852.2021.9626029
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Uchida Tomoyuki
中科院分区：
文献类型：
--
作者：
Kawasaki Yuma;Miyahara Tetsuhiro;Kuboyama Tetsuji;Suzuki Yusuke;Uchida Tomoyuki
从图结构化数据中获取知识是机器学习和数据挖掘中的一项重要任务。 TTSP（双端串联并联）图用作电网和调度的数据模型。我们提出了一种多重 TTSP 图模式，它是带有通配符的 TTSP 图模式的有限集，并给出了 TTSP 图的聚类过程。然后，我们提出了一种进化学习方法，通过对正 TTSP 图进行聚类，从正和负 TTSP 图数据中获取具有通配符的特征多个 TTSP 图模式。实验结果表明，我们提出的进化学习方法获得了具有通配符的特征多个 TTSP 图模式。
Knowledge acquisition from graph structured data is an important task in machine learning and data mining. TTSP (Two-Terminal Series Parallel) graphs are used as data models for electric networks and scheduling. We propose a multiple TTSP graph pattern, which is a finite set of TTSP graph patterns, with wildcards and give a clustering procedure of TTSP graphs. Then we propose an evolutionary learning method for obtaining characteristic multiple TTSP graph patterns with wildcards, from positive and negative TTSP graph data by clustering positive TTSP graphs. Experimental results show that our proposed evolutionary learning method obtains characteristic multiple TTSP graph patterns with wildcards.