1990
DOI: 10.1109/49.46879
|View full text |Cite
|
Sign up to set email alerts
|

Predicting software development errors using software complexity metrics

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
68
0

Year Published

1996
1996
2012
2012

Publication Types

Select...
6
4

Relationship

1
9

Authors

Journals

citations
Cited by 195 publications
(68 citation statements)
references
References 10 publications
0
68
0
Order By: Relevance
“…In a similar study, Agresti and Evanco [9] presented a model to predict defect density based on the product and process characteristics for Ada program. Moreover, there are many papers advocating statistical models and software metrics [10,11]. Most of them are based on size and complexity metrics.…”
Section: Related Workmentioning
confidence: 99%
“…In a similar study, Agresti and Evanco [9] presented a model to predict defect density based on the product and process characteristics for Ada program. Moreover, there are many papers advocating statistical models and software metrics [10,11]. Most of them are based on size and complexity metrics.…”
Section: Related Workmentioning
confidence: 99%
“…A prediction scheme is said proactive if it predicts FP software modules merely based on some static characteristics of software code itself. For example, program code complexity [5], code size [15], [16], object-oriented metrics [8], and revision history [11] are the metrics most commonly studied. The faulty patterns are also studied by Mizuno and Kikuno [13] to discriminate FP modules and non-fault-prone (NFP) modules using a spam filtering tool.…”
Section: Introductionmentioning
confidence: 99%
“…There are two categories for estimating software faults using software metrics, using statistical techniques and using machine learning. The models using statistical classification techniques include Discriminant Analysis [2] and Factor Analysis [3] and those using machine learning classification techniques include decision trees [4], artificial neural networks [5], support vector machines [6] [7], etc. Many of these studies have dealt with finding the subsets of the software metrics that are most likely to predict the existence of faults.…”
Section: Introductionmentioning
confidence: 99%