Abstract
It is difficult to directly apply conventional significance tests to compare the performance of network classification models because network data instances are not independent and identically distributed. Recent work [6] has shown that paired t-tests applied to overlapping network samples will result in unacceptably high levels (e.g., up to 50%) of Type I error (i.e., the tests lead to incorrect conclusions that models are different, when they are not). Thus, we need new strategies to accurately evaluate network classifiers. In this paper, we analyze the sources of bias (e.g. dependencies among network data instances) theoretically and propose analytical corrections to standard significance tests to reduce the Type I error rate to more acceptable levels, while maintaining reasonable levels of statistical power to detect true performance differences. We validate the effectiveness of the proposed corrections empirically on both synthetic and real networks.
Chapter PDF
Similar content being viewed by others
References
Bengio, Y., Grandvalet, Y.: No unbiased estimator of the variance of k-fold cross-validation. Journal of Machine Learning Research 5, 1089â1105 (2004)
Dietterich, T.: Approximate statistical tests for comparing supervised classification learning algorithms. Neural Computation 10, 1895â1923 (1998)
Franklin, J.N.: Matrix Theory. Dover Publications, Mineola (1993)
Macskassy, S., Provost, F.: Classification in networked data: A toolkit and a univariate case study. Journal of Machine Learning Research 8, 935â983 (2007)
Nadeau, C., Bengio, Y.: Inference for the generalization error. Machine Learning Journal 52(3), 239â281 (2003)
Neville, J., Gallagher, B., Eliassi-Rad, T., Wang, T.: Correcting evaluation bias of relational classifiers with network cross validation. Knowledge and Information Systems, 1â25 (2011)
Owen, A.B.: Variance of the number of false discoveries. Journal of the Royal Statistical Society: Series B (Statistical Methodology)Â 67, 411â426 (2005)
Sen, P., Namata, G., Bilgic, M., Getoor, L., Gallagher, B., Eliassi-Rad, T.: Collective classification in network data. AI Magazine 29(3), 93â106 (2008)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
Âİ 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wang, T., Neville, J., Gallagher, B., Eliassi-Rad, T. (2011). Correcting Bias in Statistical Tests for Network Classifier Evaluation. In: Gunopulos, D., Hofmann, T., Malerba, D., Vazirgiannis, M. (eds) Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2011. Lecture Notes in Computer Science(), vol 6913. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-23808-6_33
Download citation
DOI: https://doi.org/10.1007/978-3-642-23808-6_33
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-23807-9
Online ISBN: 978-3-642-23808-6
eBook Packages: Computer ScienceComputer Science (R0)