Out-of-bag discriminative graph mining
Pages 109 - 114
Abstract
In class-labeled graph databases, each graph is associated with one from a finite set of classes, which induces associations between classes and subgraphs occurring in the database graphs. The subgraphs with strong class associations are called discriminative subgraphs. In this work, discriminative subgraphs are repeatedly mined on bootstrap samples of a graph database in order to improve on estimation of subgraph associations. The number of times a subgraph occurs in a graph associated with each class (support values) is recorded over the out-of-bag instances of the bootstrap process. We investigate sample mean and maximum likelihood estimation for the approximation of the true underlying support from these empirical values. It is shown that both significantly improve on the process, compared to single runs of discriminative graph mining, by applying the methods to publicly available toxicological databases, and validating support values, class bias, and class significance. In toxicology, the detection of subgraphs (fragments of chemical structure) that induce toxicity is a major goal. Apart from the subgraph associations being statistically validated, the number of subgraphs created by the proposed methods are much lower than for ordinary discriminative graph mining, which is often a bottleneck in the application of computational models to such databases, and hinders interpretation of the results.
References
[1]
Leo Breiman. Out-Of-Bag Estimation. Technical Report, Statistics Department, University of California, 1996.
[2]
Björn Bringmann, Siegfried Nijssen, and Albrecht Zimmermann. From Local Patterns to Classification Models. In Saso Džeroski, Bart Goethals, and Pance Panov, editors, Inductive Databases and Constraint-Based Data Mining, pages 127--154. Springer New York, 2010.
[3]
Tom Bylander. Estimating Generalization Error on Two-Class Datasets Using Out-of-Bag Estimates. Machine Learning, 48(1-3):287--297, 2002.
[4]
Yun Chi, Richard R. Muntz, Siegfried Nijssen, and Joost N. Kok. Frequent Subtree Mining - An Overview. Fundamenta Informaticae, 66(1--2):161--198, 2004.
[5]
Jun Huan, Wei Wang, Jan Prins, and Jiong Yang. SPIN: Mining Maximal Frequent Subgraphs From Graph Databases. In Proceedings of the Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '04, pages 581--586, New York, NY, USA, 2004. ACM.
[6]
Jeroen Kazius, Ross McGuire, and Roberta Bursi. Derivation and Validation of Toxicophores for Mutagenicity Prediction. Journal of Medicinal Chemistry, 48(1):312--320, 2005.
[7]
Stefan Kramer, Luc De Raedt, and Christoph Helma. Molecular Feature Mining in HIV Data. In KDD '01: Proceedings of the Seventh ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 136--143, New York, NY, USA, 2001. ACM.
[8]
Andreas Maunz, Christoph Helma, and Stefan Kramer. Efficient Mining for Structurally Diverse Subgraph Patterns in Large Molecular Databases. Machine Learning, 83:193--218, 2011.
[9]
Hiroto Saigo, Sebastian Nowozin, Tadashi Kadowaki, Taku Kudo, and Koji Tsuda. gBoost: A Mathematical Programming Approach to Graph Classification and Regression. Machine Learning, 75(1):69--89, 2009.
[10]
Claudia Suenderhauf, Felix Hammann, Andreas Maunz, Christoph Helma, and Jörg Huwyler. Combinatorial QSAR Modeling of Human Intestinal Absorption. Molecular Pharmaceutics, 8(1):213--224, 2011.
- Out-of-bag discriminative graph mining
Recommendations
Learning block-preserving graph patterns and its application to data mining
Recently, due to the rapid growth of electronic data having graph structures such as HTML and XML texts and chemical compounds, many researchers have been interested in data mining and machine learning techniques for finding useful patterns from graph-...
Mining Frequent Subgraph Patterns from Uncertain Graph Data
In many real applications, graph data is subject to uncertainties due to incompleteness and imprecision of data. Mining such uncertain graph data is semantically different from and computationally more challenging than mining conventional exact graph ...
Mining of Frequent Externally Extensible Outerplanar Graph Patterns
ICMLA '08: Proceedings of the 2008 Seventh International Conference on Machine Learning and ApplicationsAn outerplanar graph is a planar graph which can be embedded in the plane in such a way that all of vertices lie on the outer boundary. Many chemical compounds are known to be expressed by outerplanar graphs. In this paper, firstly, we introduce an ...
Comments
Information & Contributors
Information
Published In
March 2013
2124 pages
Copyright © 2013 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]
Sponsors
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 18 March 2013
Check for updates
Author Tags
Qualifiers
- Research-article
Conference
SAC '13
Sponsor:
Acceptance Rates
SAC '13 Paper Acceptance Rate 255 of 1,063 submissions, 24%;
Overall Acceptance Rate 1,650 of 6,669 submissions, 25%
Upcoming Conference
SAC '25
- Sponsor:
- sigapp
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 83Total Downloads
- Downloads (Last 12 months)1
- Downloads (Last 6 weeks)0
Reflects downloads up to 09 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in