Estimating the Number of Faults in Code

Authors:
John E. Gaffney
Affiliations:
National Weather Service, U.S. Department of Commerce, Silver Spring, MD, 20910./ Federal System Division, IBM Corporation, Gaithersburg, MD 20879.
Venue:
IEEE Transactions on Software Engineering
Year:
1984

Citing 0
Cited 2

Regression via Classification applied on software defect estimation

Expert Systems with Applications: An International Journal
Review: On the application of genetic programming for software engineering predictive modeling: A systematic review

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.01

Visualization

Abstract

This paper provides formulas relating the number of faults or ``bugs'' to the number of lines of code and to the number of conditional jumps. A result is that there are, on the average, about 21 bugs per KSLOC discoverable after successful compilation. A major motivation for the work presented here was to determine if some published data suggest any relationship between the level of the coding language employed (e.g., Jovial is a ``high'' level language) and the number of bugs found. It is shown that the number of bugs appears not to be a function of the ``level'' of the coding language employed. In addition, knowledge of items such as the size of the vocabulary (operator and operand) used appears to be of little consequence to the estimate of bug content beyond that based on SLOC count.