EECS500 Fall 2016 Department Colloquium

Shih-Feng Sun
Statistical Fault Localization
White 411
1:00 - 2:00 PM
November 11, 2016

Software fault localization is one of the most expensive, tedious and time consuming activities in program debugging.  Coverage-based statistical fault localization (CBSFL) techniques have been proposed for characterizing the “suspiciousness” of program elements based on code-coverage profiles and PASS/FAIL labels for a set of test or operational executions.  The resulting suspiciousness scores are typically intended to be used to rank program elements for inspection by developers, on the assumption that program elements which cause labeled failures will tend to receive high ranks.  Many CBSFL metrics are proposed.  This talk will introduce CBSFL techniques and also discuss the common structure of several effective metrics.  This commonality can be used to better understand the metrics’ strengths and weaknesses, which have been little explored in the literature.


Shih-Feng Sun is a PhD candidate in the Department of Electrical Engineering and Computer Science at Case Western Reserve University.  His research involves statistical fault localization and causal inference theory in software engineering.