Research On The Validity Of Attributes For Code-Based Software Ouality Assessment

Nowadays,the evaluation of software quality is becoming more and more important.People lack quantitative experimental evaluation of the practicality of the quality model.For example,in the prediction of code quality,it is impossible to prove that the accuracy and usability of the quality evaluation result is better,and there is no actual experiment to prove that the quality attribute can better describe the code.Aiming at the problem that the performance of the quality model in the prediction of code quality cannot be quantitatively measured,based on the sensitivity of the quality evaluation model to code defects,a machine learning-based quality evaluation attribute validity verification method is proposed.This method conducts comparative experiments by controlling variables.First,extract the basic metric elements;then,convert them into quality attributes of the software;finally,to verify the quality evaluation model and the effectiveness of medium quality attributes,this paper compares machine learning methods based on quality attributes with which based on text features,and conducts experimental evaluation in two data sets.The machine learning evaluation method based on quality attributes is also designed and verified in this paper,which can optimize the traditional evaluation method for accurate and efficient evaluation of software quality.The result shows that the effectiveness of quality attributes under control variables is better,and lead by 15%in AdaBoostClassifier;when the text feature extraction method is increased to 50-150 dimensions,the performance of the text feature in the four machine learning algorithms overtakes the quality attributes;but when the peak is reached,quality attributes are more stable.This also provides a direction for the optimization of the quality model and the use of quality assessment in different situations.The usability and significance of the method are proved by experiments.