作者: Elaine J. Weyuker , Thomas J. Ostrand , Robert M. Bell
关键词:
摘要: We have been investigating different prediction models to identify which files of a large multi-release industrial software system are most likely contain the largest numbers faults in next release. To make predictions we considered number file characteristics and change information about files, built fully- automatable that do not require user any statistical expertise. now consider effect adding developer as factor assess extent this affects quality predictions.