Classification of Software Behaviors for Failure Detection: A Discriminative Pattern Mining Approach
Software is a ubiquitous component of our daily life. We of-
ten depend on the correct working of software systems. Due
to the di±culty and complexity of software systems, bugs
and anomalies are prevalent. Bugs have caused billions of
dollars loss, in addition to privacy and security threats. In
this work, we address software reliability issues by proposing
a novel method to classify software behaviors based on past
history or runs. With the technique, it is possible to gener-
alize past known errors and mistakes to capture failures and
anomalies. Our technique ¯rst mines a set of discriminative
features capturing repetitive series of events from program
execution traces. It then performs feature selection to select
the best features for classi¯cation. These features are then
used to train a classi¯er to detect failures. Experiments and
case studies on traces of several benchmark software systems
and a real-life concurrency bug from MySQL server show the
utility of the technique in capturing failures and anomalies.
On average, our pattern-based classi¯cation technique out-
performs the baseline approach by 24.68% in accuracy1.
Date: August 30, 2009
Book Title: Proc. 2009 ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining (KDD'09)
Type: InProceedings
Address: Paris, France
Downloads: 84
Has 1 soft copy
remote link
Bibtex
@InProceedings{Classification_of_Software_Behaviors_for,
author = "David Lo and Hong Cheng and Jiawei Han and Siau-Cheng Khoo and ChengNian Sun",
title = "{Classification of Software Behaviors for Failure Detection: A Discriminative Pattern Mining Approach}",
month = "August",
year = "2009",
address = ", Paris, France",
booktitle = "Proc. 2009 ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining (KDD'09)",
}