#### DMCA

## Pattern Discovery

### Citations

650 | gspan: Graph-based substructure pattern mining.
- Yan, Han
- 2002
(Show Context)
Citation Context ...A. Copyright 2013 ACM 978-1-4503-2037-5/13/06 ...$15.00. 25], graph clustering and classification [3], and graph query processing [2]. A wealth of mining algorithms have been proposed along the years =-=[1, 9, 11, 12, 17, 20, 23]-=-, among which many of them handle various constraints on top of the frequency requirement [26]. However, the constrained graph pattern mining problem so far still poses a huge challenge to the existin... |

310 | An Apriori-based algorithm for mining frequent substructures from graph data, in:
- Inokuchi, Washio, et al.
- 2000
(Show Context)
Citation Context ...frequent 2 http://www.weibo.com Figure 23: A conversation example in Sina Weibo Figure 24: Weibo: An interaction pattern in conversations patterns in graph transactions, e.g., AGM by Inokuchi et al., =-=[10, 11]-=-, FSG by Kuramochi and Karypis, [12], Borgelt and Berthold, [1], gSpan by Yan and Han, [23] and FFSM by Huan et al., [9]. These algorithms aim to find the complete frequent pattern set, but suffer fro... |

201 | Graph indexing: A frequent structure-based approach,” in - Yan, Yu, et al. - 2004 |

194 | Efficient mining of frequent subgraphs in the presence of isomorphism.
- Huan, Wang, et al.
- 2003
(Show Context)
Citation Context ...A. Copyright 2013 ACM 978-1-4503-2037-5/13/06 ...$15.00. 25], graph clustering and classification [3], and graph query processing [2]. A wealth of mining algorithms have been proposed along the years =-=[1, 9, 11, 12, 17, 20, 23]-=-, among which many of them handle various constraints on top of the frequency requirement [26]. However, the constrained graph pattern mining problem so far still poses a huge challenge to the existin... |

165 | Mining Molecular Fragments: Finding Relevant Substructures of Molecules,"
- Borgelt, Berthold
- 2002
(Show Context)
Citation Context ...A. Copyright 2013 ACM 978-1-4503-2037-5/13/06 ...$15.00. 25], graph clustering and classification [3], and graph query processing [2]. A wealth of mining algorithms have been proposed along the years =-=[1, 9, 11, 12, 17, 20, 23]-=-, among which many of them handle various constraints on top of the frequency requirement [26]. However, the constrained graph pattern mining problem so far still poses a huge challenge to the existin... |

140 | Frequent Substructure-Based Approaches for Classifying Chemical Compounds,”
- Deshpande, Kuramochi, et al.
- 2005
(Show Context)
Citation Context ..., requires prior specific permission and/or a fee. SIGMOD’13, June 22–27, 2013, New York, New York, USA. Copyright 2013 ACM 978-1-4503-2037-5/13/06 ...$15.00. 25], graph clustering and classification =-=[3]-=-, and graph query processing [2]. A wealth of mining algorithms have been proposed along the years [1, 9, 11, 12, 17, 20, 23], among which many of them handle various constraints on top of the frequen... |

120 | An efficient algorithm for discovering frequent subgraphs.
- Kuramochi, Karypis
- 2004
(Show Context)
Citation Context |

118 | Mining Frequent Itemsets with Convertible Constrants,
- Peo, Han, et al.
- 2001
(Show Context)
Citation Context ...ts into the mining process. Recent work has highlighted the importance of constraint-based mining in the context of mining frequent itemsets, sequential patterns, associations, and graphs. Pei et al. =-=[16]-=- identify a class of hard constraints called convertible constraints and develop its pushing method in itemset setting. The study [15] overviews the principles of pattern-growth methods for constraine... |

100 | Complete mining of frequent patterns from graphs: Mining graph data.
- Inokuchi, Washio, et al.
- 2003
(Show Context)
Citation Context |

77 | Fg-index: towards verification-free query processing on graph databases,” in
- Cheng, Ke, et al.
- 2007
(Show Context)
Citation Context ...sion and/or a fee. SIGMOD’13, June 22–27, 2013, New York, New York, USA. Copyright 2013 ACM 978-1-4503-2037-5/13/06 ...$15.00. 25], graph clustering and classification [3], and graph query processing =-=[2]-=-. A wealth of mining algorithms have been proposed along the years [1, 9, 11, 12, 17, 20, 23], among which many of them handle various constraints on top of the frequency requirement [26]. However, th... |

77 | Substructure discovery in the SUBDUE system.
- Holder, Cook, et al.
- 1994
(Show Context)
Citation Context ...ound ɛ. Due to its constraint on diameter bound and spider-based merging, it is hard for SpiderMine to find skinny patterns as we defined. Other important works in single graph setting include SUBDUE =-=[8]-=-, SEuS was proposed by Ghazizadeh and Chawathe [6], and GREW [13], which have been thoroughly discussed in [26]. MoSS [5] is proposed for mining complete patterns in single graphs, which, as any other... |

69 | P.S.: Mining significant graph patterns by leap search.
- Yan, Cheng, et al.
- 2008
(Show Context)
Citation Context ...oo large to handle. ORIGAMI as proposed in [7] is an algorithm to find a representative pattern set based on output space sampling. Other works include structural leap search introduced by Yan et al. =-=[22]-=-, which adopts structural similarity to mine significant graph patterns efficiently and directly from two graph datasets. Mining frequent pattern in single-graph setting is harder than mining in graph... |

68 | Pathsim: meta pathbased top-k similarity search in heterogeneous information networks,”
- Sun, Han, et al.
- 2011
(Show Context)
Citation Context ...rticular type of constrained frequent patterns — the “skinny” patterns. Recently, the importance of these skinny patterns has been increasingly recognized in heterogenous information network analysis =-=[19]-=- and social network studies. A “skinny” pattern, formally called a l-long δ-skinny pattern, is essentially a graph pattern with a long backbone of length l from which short twigs no longer than δ bran... |

53 | Graph indexing: tree + delta >= graph. - Zhao, Yu, et al. - 2007 |

40 |
Constrained frequent pattern mining: a pattern-growth view,
- Pei, Han
- 2002
(Show Context)
Citation Context ...emsets, sequential patterns, associations, and graphs. Pei et al. [16] identify a class of hard constraints called convertible constraints and develop its pushing method in itemset setting. The study =-=[15]-=- overviews the principles of pattern-growth methods for constrained frequent patternmining according to the constraints in itemset settings and sequential pattern mining. Another work for itemset set... |

26 | Support computation for mining frequent subgraphs in a single graph
- Fiedler, Borgelt
- 2007
(Show Context)
Citation Context ...ns as we defined. Other important works in single graph setting include SUBDUE [8], SEuS was proposed by Ghazizadeh and Chawathe [6], and GREW [13], which have been thoroughly discussed in [26]. MoSS =-=[5]-=- is proposed for mining complete patterns in single graphs, which, as any other algorithm mining for the complete pattern set, suffers from the same scalability issue as the input graph size grows. Al... |

25 | Direct mining of discriminative and essential frequent patterns via model-based search tree
- Fan, Zhang, et al.
- 2008
(Show Context)
Citation Context ...y examine locally those relevant candidates. Direct mining has been studied not on constraints on patterns but on certain utility of patterns such as discriminativeness measured on the mining results =-=[4]-=-, which is a different problem from ours. Figure 2: An architectural view of the direct mining framework An architectural overview comparing our direct mining approach and traditional mining approach ... |

23 | Margin:Maximal frequent subgraph mining.
- Thomas, Valluri, et al.
- 2006
(Show Context)
Citation Context |

22 | ORIGAMI: Mining representative orthogonal graph patterns
- Hasan, Chaoji, et al.
- 2007
(Show Context)
Citation Context ...to find patterns larger than a threshold due to the exponential number of pattern candidates to enumerate and examine. Probabilistic algorithms returning a sample of the pattern output space, ORIGAMI =-=[7]-=- as a representative, would miss all but a few patterns of interest. The most relevant work is SpiderMine [26] designed for mining large patterns. However, due to the diameter bound constraint and the... |

21 | GREW–A Scalable Frequent Subgraph Discovery Algorithm,
- Kuramochi, Karypis
- 2004
(Show Context)
Citation Context ...tterns efficiently and directly from two graph datasets. Mining frequent pattern in single-graph setting is harder than mining in graphtransaction setting due to the complexity of support computation =-=[13, 26]-=-. One of the most recent work in this category is SpiderMine [26] which is designed to find the top-K largest patterns with a high probability of 1−ɛ for any user-specified error bound ɛ. Due to its c... |

21 | gPrune: A constraint pushing framework for graph pattern mining,
- Zhu, Yan, et al.
- 2007
(Show Context)
Citation Context ...tonicity and succinctness and develops efficient constraint-based frequent itemset mining algorithms for a single constraint. Constraints in graph settings are more complicated [18]. The study gPrune =-=[27]-=- discusses pruning properties and techniques in both pattern space and data space. SkinnyMine, on the other hand, examines the properties of constraints that would admit our proposed direct mining fra... |

21 | Mining colossal frequent patterns by core pattern fusion
- Zhu, Yan, et al.
- 2007
(Show Context)
Citation Context ...ramework such that all target patterns can be reached fast (making it direct) with minimum visits to irrelevant ones (making it precise). As illustrated in Figure 1 from a pattern lattice perspective =-=[28]-=-, the target mining result is “carved” by the constraint into pattern clusters indicated by the shaded areas. Probabilistic methods cannot accomplish the mining task because most of them are only capa... |

15 | SEuS: Structure extraction using summaries
- Ghazizadeh, Chawathe
- 2002
(Show Context)
Citation Context ...d spider-based merging, it is hard for SpiderMine to find skinny patterns as we defined. Other important works in single graph setting include SUBDUE [8], SEuS was proposed by Ghazizadeh and Chawathe =-=[6]-=-, and GREW [13], which have been thoroughly discussed in [26]. MoSS [5] is proposed for mining complete patterns in single graphs, which, as any other algorithm mining for the complete pattern set, su... |

9 |
melting, large graphs by edge manipulation
- Gelling
- 2012
(Show Context)
Citation Context ...nsightful information diffusion patterns and different roles of users in the dissemination process. These patterns are key to network manipulation to either facilitate or hinder information diffusion =-=[21]-=-. Unfortunately, existing algorithms are not able to mine these skinny patterns efficiently. Essentially, algorithms exhausting all frequent patterns, gSpan [23] as a representative, are not able to f... |

9 | Mining top-k large structural patterns in a massive network.
- Zhu, Qu, et al.
- 2011
(Show Context)
Citation Context ...ry processing [2]. A wealth of mining algorithms have been proposed along the years [1, 9, 11, 12, 17, 20, 23], among which many of them handle various constraints on top of the frequency requirement =-=[26]-=-. However, the constrained graph pattern mining problem so far still poses a huge challenge to the existing algorithms mainly due to the inherent bottleneck in the mining paradigm. In essence, mining ... |

5 |
Mining frequent itemsets with convertible constraints
- J, Lakshmanan
- 2001
(Show Context)
Citation Context ...nts in itemset settings and sequential pattern mining. Another work for itemset settings is proposed by Zhu et al. [28], which mines large patterns probabilistically by core pattern fusion. Ng et al. =-=[14]-=- study three important classes of constraints: monotonicity, antimonotonicity and succinctness and develops efficient constraint-based frequent itemset mining algorithms for a single constraint. Const... |

5 |
SPIN: mining maximal frequent subgraphs from graph databases
- Prins, Yang
- 2004
(Show Context)
Citation Context |

3 | Efficient topological olap on information network
- Qu, Zhu, et al.
- 2011
(Show Context)
Citation Context ... monotonicity, antimonotonicity and succinctness and develops efficient constraint-based frequent itemset mining algorithms for a single constraint. Constraints in graph settings are more complicated =-=[18]-=-. The study gPrune [27] discusses pruning properties and techniques in both pattern space and data space. SkinnyMine, on the other hand, examines the properties of constraints that would admit our pro... |