Improving the detection accuracy of evolutionary coupling

Simple item page
dc.contributor.authorMondal, M.
dc.contributor.authorRoy, C.K.
dc.contributor.authorSchneider, K.A.
dc.date.accessioned2015-07-29T19:18:50Z
dc.date.available2015-07-29T19:18:50Z
dc.date.issued2013
dc.description.abstractIf two or more program entities (e.g., files, classes, methods) co-change frequently during software evolution, these entities are said to have evolutionary coupling. The entities that frequently co-change (i.e., exhibit evolutionary coupling) are likely to have logical coupling (or dependencies) among them. Association rules and two related measurements, Support and Confidence, have been used to predict whether two or more co-changing entities are logically coupled. In this paper, we propose and investigate a new measurement, Significance, that has the potential to improve the detection accuracy of association rule mining techniques. Our preliminary investigation on four open-source subject systems implies that our proposed measurement is capable of extracting coupling relationships even from infrequently co-changed entity sets that might seem insignificant while considering only Support and Confidence. Our proposed measurement, Significance (in association with Support and Confidence), has the potential to predict logical coupling with higher precision and recall.en_US
dc.description.refereedYesen_US
dc.identifier.doi10.1109/ICPC.2013.6613853
dc.identifier.urihttp://hdl.handle.net/1880/50694
dc.publisherIEEEen_US
dc.publisher.urlhttp://dx.doi.org/10.1109/ICPC.2013.6613853en_US
dc.rightsAn error occurred on the license name.*
dc.rights.uriAn error occurred getting the license - uri.*
dc.titleImproving the detection accuracy of evolutionary couplingen_US
dc.typeunknown
Files
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.84 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Surfnet