Efficient answering of why-not questions in similar graph matching

Md. Saiful Islam, Chengfei Liu, Jianxin Li

Research output: Chapter in Book/Conference paperConference paper

3 Citations (Scopus)

Abstract

© 2016 IEEE.Graph data management and matching similar graphs are very important for many applications including bioinformatics, computer vision, VLSI design, bug localization, road networks, social and communication networking. Many graph indexing and similarity matching techniques have already been proposed for managing and querying graph data. In similar graph matching, a user is returned with the database graphs whose distances with the query graph are below a threshold. In such query settings, a user may not receive certain database graphs that are very similar to the query graph if the initial query graph is inappropriate/imperfect for the expected answer set. To exemplify this, consider a drug designer who is looking for chemical compounds that could be the target of her hypothetical drug before realizing it. In response to her query, the traditional search system may return the structures from the database that are most similar to the query graph. However, she may get surprised if some of the expected targets are missing in the answer set. She may then seek assistance from the system by asking Is there other query graph that can match my expected answer set?. The system may then modify her initial query graph to include the missing answers in the new answer set. Here, we study this kind of problem of answering why-not questions in similar graph matching for graph databases.
Original languageEnglish
Title of host publication2016 IEEE 32nd International Conference on Data Engineering, ICDE 2016
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages1476-1477
Number of pages2
DOIs
Publication statusPublished - 2016
Externally publishedYes
Event2016 IEEE 32nd International Conference on Data Engineering, ICDE 2016 - Helsinki, Finland
Duration: 16 May 201620 May 2016

Conference

Conference2016 IEEE 32nd International Conference on Data Engineering, ICDE 2016
CountryFinland
CityHelsinki
Period16/05/1620/05/16

Fingerprint Dive into the research topics of 'Efficient answering of why-not questions in similar graph matching'. Together they form a unique fingerprint.

Cite this