Publications

Detailed Information

An Efficient Algorithm for Subgraph Isomorphism using Dynamic Programming on Directed Acyclic Graphs : 비순환 방향 그래프에서의 동적 프로그래밍을 이용한 효율적인 부분그래프 동형 알고리즘

DC Field Value Language
dc.contributor.advisor박근수-
dc.contributor.author한명지-
dc.date.accessioned2018-05-28T16:19:26Z-
dc.date.available2018-05-28T16:19:26Z-
dc.date.issued2018-02-
dc.identifier.other000000149660-
dc.identifier.urihttps://hdl.handle.net/10371/140658-
dc.description학위논문 (박사)-- 서울대학교 대학원 : 공과대학 전기·컴퓨터공학부, 2018. 2. 박근수.-
dc.description.abstractSubgraph isomorphism (or subgraph matching) is one of the fundamental problems in graph analysis.
It is an NP-hard problem, and extensive research has been done to develop practical solutions for subgraph isomorphism.
Although a great deal of progress has been made, the existing solutions still show a limited scalability in handling large real graphs. The state-of-the-art algorithms are based on the general framework of backtracking which recursively finds all embeddings of a query graph in a data graph.
In this thesis we introduce three new techniques: dynamic programming on directed acyclic graphs,
the adaptive matching order with DAG-ordering, and pruning by failing sets,
which together lead to a much faster and more scalable algorithm DP-iso for subgraph isomorphism.
Extensive experiments with real datasets show that DP-iso outperforms the fastest existing solution by up to 4 orders of magnitude with respect to the running time and
up to 6 orders of magnitude with respect to the number of recursions.
-
dc.description.tableofcontentsChapter 1 Introduction 1
1.1 Background 1
1.2 Challenges 3
1.3 Our Ideas 4
1.4 Organization 6
Chapter 2 Preliminaries 7
2.1 Notations 7
2.2 Problem Definition 7
2.3 Directed Acyclic Graph 8
2.4 Related Work 8
2.4.1 Subgraph Matching 8
2.4.2 Other Work on Subgraph Matching 10
2.4.3 Dynamic Programming 10
Chapter 3 Overview of DP-iso 12
3.1 Algorithm Outline 12
3.2 Leaf Decomposition 13
Chapter 4 Dynamic Programming on DAG 14
4.1 CS Structure 14
4.1.1 Definition of CS Structure 14
4.1.2 Properties of CS Structure 15
4.2 DP on DAG 15
4.2.1 Weak Embedding 15
4.2.2 CS Refinement using DP on DAG 16
4.2.3 Time Complexity 18
4.3 Optimizing CS 19
Chapter 5 DAG-Ordering and Adaptive Matching Order 21
5.1 Backtracking Based on DAG-Ordering 21
5.1.1 DAG-Ordering and Extendable Candidates 21
5.1.2 Backtracking Framework 23
5.2 Adaptive Matching Order 24
5.2.1 Infrequent-Path-First Strategy and Tree-like Paths 24
5.2.2 Weight Array 25
5.2.3 Adaptive Matching Order using Weight Array 26
5.3 Backtracking Process 27
Chapter 6 Pruning by Failing Sets 29
6.1 Search Tree Node 29
6.2 General Idea 30
6.3 Failing Set 31
6.3.1 Definition of Failing Set 31
6.3.2 Computing Failing Sets 32
6.3.3 Correctness of Failing Set Computation 33
6.4 Refined Backtracking Process 34
Chapter 7 Performance Evaluation 37
7.1 Experimental Settings 37
7.1.1 Environment 37
7.1.2 Datasets 38
7.1.3 Query Graphs 39
7.1.4 Performance Measurement 39
7.2 Experimental Results 41
7.2.1 Comparing with CFL-Match 41
7.2.2 Comparing with TurboISO 43
7.2.3 Evaluating Boosting Technique 45
7.2.4 Evaluating Pruning Technique by Failing Sets 45
7.2.5 Additional Experiments 48
Chapter 8 Conclusion 52
Bibliography 53
-
dc.formatapplication/pdf-
dc.format.extent3317249 bytes-
dc.format.mediumapplication/pdf-
dc.language.isoen-
dc.publisher서울대학교 대학원-
dc.subjectSubgraph Isomorphism-
dc.subjectSubgraph Matching-
dc.subjectDynamic Programming-
dc.subjectDirected Acyclic Graph-
dc.subjectDAG-
dc.subjectNP-hard-
dc.subject.ddc621.3-
dc.titleAn Efficient Algorithm for Subgraph Isomorphism using Dynamic Programming on Directed Acyclic Graphs-
dc.title.alternative비순환 방향 그래프에서의 동적 프로그래밍을 이용한 효율적인 부분그래프 동형 알고리즘-
dc.typeThesis-
dc.contributor.AlternativeAuthorMyoungji Han-
dc.description.degreeDoctor-
dc.contributor.affiliation공과대학 전기·컴퓨터공학부-
dc.date.awarded2018-02-
Appears in Collections:
Files in This Item:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share