Filters








320 Hits in 6.2 sec

Pruning Strategies Based on the Upper Bound of Information Gain for Discriminative Subgraph Mining [chapter]

Kouzou Ohara, Masahiro Hara, Kiyoto Takabayashi, Hiroshi Motoda, Takashi Washio
2009 Lecture Notes in Computer Science  
Thus, to improve its efficiency, we propose pruning methods based on the upper-bound of information gain that is used as a criterion for discriminability of subgraphs in Cl-GBI.  ...  The upper-bound of information gain of a subgraph is the maximal one that its super graph can achieve.  ...  This work was partly supported by a grant-in-aid for Young Scientists (B) No. 19700145 from MEXT, Japan.  ... 
doi:10.1007/978-3-642-01715-5_5 fatcat:casamhyr5va3rk524nm3gl2nqu

LTS: Discriminative subgraph mining by learning from search history

Ning Jin, Wei Wang
2011 2011 IEEE 27th International Conference on Data Engineering  
We discover that search history of discriminative subgraph mining is very useful in computing empirical upper-bounds of discrimination scores of subgraphs.  ...  The search space for discriminative subgraphs is usually prohibitively large.  ...  ACKNOWLEDGMENT We thank Xifeng Yan and Sayan Ranu for their kindly providing implementation of LEAP and graphSig.  ... 
doi:10.1109/icde.2011.5767922 dblp:conf/icde/JinW11 fatcat:eym7suj6bbejfixq3nee4ok3k4

Mining significant graph patterns by leap search

Xifeng Yan, Hong Cheng, Jiawei Han, Philip S. Yu
2008 Proceedings of the 2008 ACM SIGMOD international conference on Management of data - SIGMOD '08  
Our new mining method revealed that the widely adopted branch-and-bound search in data mining literature is indeed not the best, thus sketching a new picture on scalable graph pattern discovery.  ...  Furthermore, graph classifiers built on mined patterns outperform the up-to-date graph kernel method in terms of efficiency and accuracy, demonstrating the high promise of such patterns.  ...  [5] derived a frequency upper bound of discriminative measures such as information gain and Fisher score, showing a relationship between frequency and discriminative measures.  ... 
doi:10.1145/1376616.1376662 dblp:conf/sigmod/YanCHY08 fatcat:jpmfkdtzsnfj7g2ozmcupbqcnu

Iterative Graph Feature Mining for Graph Indexing

Dayu Yuan, Prasenjit Mitra, Huiwen Yu, C. Lee Giles
2012 2012 IEEE 28th International Conference on Data Engineering  
Next, we propose a basic branch and bound algorithm to mine the features.  ...  Most of the "mine-at-once" algorithms involve frequent subgraph or subtree mining over the whole graph database.  ...  Fig. 2 . 2 One counterexample for the anti-monotonic of minSup(p, Q) Statement 4 (Branch Upper Bound): For a feature p, a branch upper bound exists such that ∀p ⊃ p, gain(p ) Fig. 3 . 3 Graph features  ... 
doi:10.1109/icde.2012.11 dblp:conf/icde/YuanMYG12 fatcat:rqnckyclkffxbjrhhr7yvlo4ii

Discriminative frequent subgraph mining with optimality guarantees

Marisa Thoma, Hong Cheng, Arthur Gretton, Jiawei Han, Hans-Peter Kriegel, Alex Smola, Le Song, Philip S. Yu, Xifeng Yan, Karsten M. Borgwardt
2010 Statistical analysis and data mining  
Second, our submodular quality function criterion can be integrated into gSpan, the state-of-the-art tool for frequent subgraph mining, and help to prune the search space for discriminative frequent subgraphs  ...  The goal of frequent subgraph mining is to detect subgraphs that frequently occur in a dataset of graphs.  ...  The responsibility for this publication lies with the authors. The authors would like to thank Siegfried Nijssen for fruitful discussions.  ... 
doi:10.1002/sam.10084 fatcat:zodsnci3azcypaag7d7coh4xeu

Near-optimal supervised feature selection among frequent subgraphs [chapter]

Marisa Thoma, Hong Cheng, Arthur Gretton, Jiawei Han, Hans-Peter Kriegel, Alex Smola, Le Song, Philip S. Yu, Xifeng Yan, Karsten Borgwardt
2009 Proceedings of the 2009 SIAM International Conference on Data Mining  
Second, our submodular quality function criterion can be integrated into gSpan, the state-of-the-art tool for frequent subgraph mining, and help to prune the search space for discriminative frequent subgraphs  ...  On large graphs, however, one faces the enormous problem that the number of these frequent subgraphs may grow exponentially with the size of the graphs, but only few of them possess enough discriminative  ...  (2.8) to provide an upper bound for the CORK values of supergraphs of a given subgraph S and exploit this information for pruning the search space in a branch-and-bound fashion.  ... 
doi:10.1137/1.9781611972795.92 dblp:conf/sdm/ThomaCGHKSSYYB09 fatcat:b5duyrszofgexo6djtvgxnzpje

Semi-supervised feature selection for graph classification

Xiangnan Kong, Philip S. Yu
2010 Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '10  
Then we propose a branch-and-bound algorithm to efficiently search for optimal subgraph features by judiciously pruning the subgraph search space.  ...  the subgraph feature mining process.  ...  However, when there are not enough labeled graphs, the pruning ability of the upper-bound based on labeled graphs can be poor, thus making it infeasible to find discriminative subgraph features within  ... 
doi:10.1145/1835804.1835905 dblp:conf/kdd/KongY10 fatcat:2kiysijawrcehjkokvjalz3kzi

Mining Top-K Graph Patterns that Jointly Maximize Some Significance Measure

Yong Liu, Jianzhong Li, Jinghua Zhu, Hong Gao
2010 Journal of Computers  
The pruning condition for the patterns T is not empty based on Theorem 6 and 7. In information gain measure is shown in Theorem 7.  ...  For later iterations of greedy selection, we can define a In Line 1 of Function MiningNextPattern, p≠min(p) similar upper bound for pruning.  ... 
doi:10.4304/jcp.5.4.565-572 fatcat:jyxhtwfbivdtxflqllmutb2cwi

On the Usefulness of Weight-Based Constraints in Frequent Subgraph Mining [chapter]

Frank Eichinger, Matthias Huber, Klemens Böhm
2010 Research and Development in Intelligent Systems XXVII  
In particular, we study frequent subgraph mining in the presence of weight-based constraints and explain how to integrate them into mining algorithms.  ...  A lower bound predicate c l for a pattern p is a predicate with the following structure: An upper bound predicate c u in turn is as follows: c u (p) := ( e 2 ∈ E(p) : measure(e 2 ) > t u ) ∨ (|p| < size  ...  Acknowledgments We thank Zahir Balaporia (Schneider National, Inc.) and Chris Clifton (Purdue University) for providing us with the logistics dataset [9] .  ... 
doi:10.1007/978-0-85729-130-1_5 dblp:conf/sgai/EichingerHB10 fatcat:etdvl2r3hnfzblll3hoabjsuzi

Semi-supervised Clustering of Graph Objects: A Subgraph Mining Approach [chapter]

Xin Huang, Hong Cheng, Jiong Yang, Jeffery Xu Yu, Hongliang Fei, Jun Huan
2012 Lecture Notes in Computer Science  
We derive an upper bound of the objective function based on which, a branch-and-bound algorithm is proposed to speedup subgraph mining.  ...  As there is no predefined feature set for the graph objects, we propose to use discriminative subgraph patterns as the features.  ...  With the derived upper boundq, we can develop a branch-and-bound subgraph mining algorithm on top of gSpan for mining the optimal subgraph feature set T * .  ... 
doi:10.1007/978-3-642-29038-1_16 fatcat:6bahaz6rbfcjnmtyyfdo2pcga4

Multi-graph-view subgraph mining for graph classification

Jia Wu, Zhibin Hong, Shirui Pan, Xingquan Zhu, Zhihua Cai, Chengqi Zhang
2015 Knowledge and Information Systems  
Specifically, we derive an evaluation criterion to estimate the discriminative power and redundancy of subgraph features across all views, with a branch-and-bound algorithm being proposed to prune subgraph  ...  To solve the problem, we propose a Cross Graph-View Subgraph Feature based Learning (gCGVFL) algorithm that explores an optimal set of subgraphs, across multiple graph-views, as features to represent graphs  ...  For example, on multi-graph-view Image dataset as shown in Figure 10 (B), UgCGVFL needs about 9000ms to mine the discriminative subgraphs, whereas by using the upper bound pruning gCGVFL only takes about  ... 
doi:10.1007/s10115-015-0872-1 fatcat:fjfwjnhmsbfrhddxyprgwoarlu

Multi-label Feature Selection for Graph Classification

Xiangnan Kong, Philip S. Yu
2010 2010 IEEE International Conference on Data Mining  
subgraph feature mining process.  ...  Then a branch-andbound algorithm is proposed to efficiently search for optimal subgraph features by judiciously pruning the subgraph search space using multiple labels.  ...  We further observe that in all tasks and evaluation criteria, our multi-label feature selection algorithm with multilabel classification (gMLC+BOOSTEXTER) outperforms the  ... 
doi:10.1109/icdm.2010.58 dblp:conf/icdm/KongY10 fatcat:wtqcozabinebzcg76bclstqx7e

Multi-graph-view Learning for Graph Classification

Jia Wu, Zhibin Hong, Shirui Pan, Xingquan Zhu, Zhihua Cai, Chengqi Zhang
2014 2014 IEEE International Conference on Data Mining  
In addition, a bag based multi-graph model is further used to relax labeling by only requiring one label for a bag of graphs, which represent one object.  ...  Experiments on real-world learning tasks demonstrate the performance of MGVBL for complicated object classification.  ...  Acknowledgments The work was supported by Australian Research Council (ARC) Discovery Projects under Grant No. DP140100545 and DP140102206.  ... 
doi:10.1109/icdm.2014.97 dblp:conf/icdm/WuHPZCZ14 fatcat:2ax3yvc4mjg3nhw4uh6kwuj63i

Dual active feature and sample selection for graph classification

Xiangnan Kong, Wei Fan, Philip S. Yu
2011 Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '11  
Current research on graph classification focuses on mining discriminative subgraph features under supervised settings. The basic assumption is that a large number of labeled graphs are available.  ...  To address this challenge, we demonstrate how one can simultaneously estimate the usefulness of a query graph and a set of subgraph features.  ...  ACKNOWLEDGEMENTS The work is supported in part by NSF through grants IIS-0905215, DBI-0960443, OISE-0968341 and OIA-0963278.  ... 
doi:10.1145/2020408.2020511 dblp:conf/kdd/KongFY11 fatcat:qcsfmkyz5fbldbvz5kbo6nmcdm

Efficient mining for structurally diverse subgraph patterns in large molecular databases

Andreas Maunz, Christoph Helma, Stefan Kramer
2010 Machine Learning  
This is confirmed experimentally by running times reduced by more than 60% compared to ordinary (static) upper bound pruning.  ...  We present a new approach to large-scale graph mining based on so-called backbone refinement classes.  ...  This upper bound has the antimonotonic property and can thus be used for pruning the refinements of a structure, if their upper bound falls below a given threshold.  ... 
doi:10.1007/s10994-010-5187-6 fatcat:7h2zyhn2gfasdlobkg5re4nbsq
« Previous Showing results 1 — 15 out of 320 results