Qi Song-Answering Why-Questions in Multi-Attributed Graphs
Qi Song received the BS and MS degrees in computer science from Beihang University, in China. He is currently working toward the PhD degree in computer science at Washington State University under supervision of Prof. Yinghui Wu. His dissertation topic is “Towards user-friendly graph exploration” and his boarder research interests include graph data mining, graph query models and languages, knowledge graph query and completion, and deep learning on graph data. For more information, please visithttp://eecs.wsu.edu/~qsong/.
Subgraph queries have been routinely used to search graphs e.g., social networks and knowledge bases. With little knowledge of underlying data, users often need to rewrite queries multiple times to reach desirable answers. Why-questions are studied to explain missing (as “Why-not” questions) or unexpected answers (as “Why” questions). We makes a first step to answer why-questions for subgraph queries in attributed graphs. We approach query rewriting and construct query rewrites, which modify original subgraph queries to identify desired entities that are specified by Why-questions. Several algorithms are developed to dynamically select “picky” operators that ensure to change (estimated) answers closer to desired ones, and incur cost determined by the size of query results and questions only.
Furthermore, in real life applications, users may give a set of examples that descri bes desired answers instead of explicitly give missing or unexpected answers. In this talk, I will discuss our recent progress on answering why-questions in attributed graphs [1-3].
 Qi Song, Mohammad Hossein Namaki, Yinghui Wu. Answering Why-Questions for Subgraph Queries in Multi-Attributed Graphs. In ICDE 2019.
 Mohammad Hossein Namaki, Qi Song, Yinghui Wu, Shengqi Yang. Answering Why-questions by Exemplars in Attributed Graphs. In SIGMOD 2019.
 Mohammad Hossein Namaki, Qi Song, Yinghui Wu, Jiaxing Pi. NAVIGATE: Explainable Visual Graph Exploration by Examples. In SIGMOD 2019 (demo).