[1]陆子豪,荆晓远.基于改进 SMOTE 的半监督极限学习机缺陷预测[J].计算机技术与发展,2021,31(12):21-25.[doi:10. 3969 / j. issn. 1673-629X. 2021. 12. 004]
　LU Zi-hao,JING Xiao-yuan.Semi-supervised Extreme Learning Machine Based on Improved SMOTE for Software Defect Prediction[J].,2021,31(12):21-25.[doi:10. 3969 / j. issn. 1673-629X. 2021. 12. 004]
点击复制

基于改进 SMOTE 的半监督极限学习机缺陷预测()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年12期
页码:
21-25
栏目:
人工智能
出版日期:
2021-12-10

文章信息/Info

Title:
Semi-supervised Extreme Learning Machine Based on Improved SMOTE for Software Defect Prediction
文章编号:
1673-629X(2021)12-0021-05
作者:
陆子豪1 ; 荆晓远2
1. 南京邮电大学 计算机学院,江苏 南京 210003;
2. 南京邮电大学 自动化学院,江苏 南京 210003
Author(s):
LU Zi-hao1 ; JING Xiao-yuan2
1. School of Computer,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;
2. School of Automation,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
关键词:
软件缺陷预测; 半监督学习; 极限学习机; 机器学习; 栈式自编码器
Keywords:
software defect prediction; semi-supervised learning; extreme learning machine; machine learning; stacked denoising auto en鄄coder
分类号:
TP31
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 12. 004
摘要:
软件缺陷预测研究中,研究人员通常使用有标记的数据进行预测模型的构建。 但是在实际应用中往往会存在有标记样本不足的情况,为了应对这种状况,专家学者引入了半监督学习。 尽管近年来不断有学者提出项目内的半监督缺陷预测方法,这些方法的预测准确度还有很大的提升空间。 该文提出了一种新的半监督极限学习机软件缺陷预测方法,即基于改进 SMOTE 的半监督极限学习机方法。 首先提出了一个改进的 SMOTE 采样方法来缓解类不平衡问题,其次使用了栈式去噪自动编码器来保留和获得更好的特征表示,最后为了提高模型的学习速率,获得更好的泛化性能,引入了极限学习机。 通过在 NASA 和 AEEEM 数据集上的大量实验,结果表明该方法与基线方法相比获得了更好的预测性能。
Abstract:
In software defect prediction research,researchers usually use labeled data to build predictive models. However,in practical applications,? there are often insufficient labeled samples. In order to cope with this situation, semi-supervised learning is introduced.Although some semi-supervised defect prediction methods have been proposed in recent years,there is still much room for improvement in the accuracy of these methods. We propose a semi-supervised software defect prediction method based on improved SMOTE extreme learning machine. Firstly,? an improved SMOTE sampling method is proposed to alleviate the problem of class imbalance. Secondly,as tacked denoising auto encoder? ?is used to retain and obtain better feature representations. Finally, in order to improve the learning efficiency of the model and obtain better generalization performance,an extreme learning machine is introduced. A large number of experiments on NASA and AEEEM data sets show that the proposed method has better prediction performance than the baseline methods.

相似文献/References:

[1]陈锦禾 沈洁.基于信息熵的主动学习半监督分类研究[J].计算机技术与发展,2010,(02):110.
　CHEN Jin-he,SHEN Jie.Active Learning Based on Information Entropy for Semi- supervised Classification[J].,2010,(12):110.
[2]陈东泳 钟尚平.基于半监督学习的JPEG图像通用隐写检测方法[J].计算机技术与发展,2009,(02):169.
　CHEN Dong-yong,ZHONG Shang-ping.A Universal Steganalysis Method for JPEG Images Based on Semi - supervised Learning[J].,2009,(12):169.
[3]徐庆伶 汪西莉.一种基于支持向量机的半监督分类方法[J].计算机技术与发展,2010,(10):115.
　XU Qing-ling,WANG Xi-li.A Novel Semi-Supervised Classification Method Based on SVM[J].,2010,(12):115.
[4]戴林 姜梅.基于半监督学习的入侵检测系统[J].计算机技术与发展,2011,(01):162.
　DAI Lin JIANG Mei.Semi-Supervised Learning-Based Network Intrusion Detection System[J].,2011,(12):162.
[5]李亚娥,汪西莉.一种自适应的半监督图像分类算法[J].计算机技术与发展,2013,(02):112.
　LI Ya-e,WANG Xi-li.An Adaptive Semi-supervised Image Classification Algorithm[J].,2013,(12):112.
[6]朱乔亚,陈可佳,方彪. 采用位置信息的半监督链接预测方法[J].计算机技术与发展,2015,25(07):63.
　ZHU Qiao-ya,CHEN Ke-jia,FANG Biao. A Semi-supervised Link Prediction Method Using Place Features[J].,2015,25(12):63.
[7]陆海洋[],荆晓远[],董西伟[],等. 基于代价敏感学习的软件缺陷预测方法[J].计算机技术与发展,2015,25(11):58.
　LU Hai-yang[],JING Xiao-yuan[],DONG Xi-wei[],et al. Software Defect Prediction Based on Cost-sensitive Learning[J].,2015,25(12):58.
[8]史作婷,吴 迪,荆晓远,等.类不平衡稀疏重构度量学习软件缺陷预测[J].计算机技术与发展,2018,28(06):125.[doi:10.3969/ j. issn.1673-629X.2018.06.028]
　SHI Zuo-ting,WU Di,JING Xiao-yuan,et al.Prediction of Defect of Class-imbalance Sparse Reconstruction Metric Learning Software[J].,2018,28(12):125.[doi:10.3969/ j. issn.1673-629X.2018.06.028]
[9]王晴[],荆晓远[][],朱阳平[],等. 基于局部稀疏重构度量学习的软件缺陷预测[J].计算机技术与发展,2016,26(11):54.
　WANG Qing[],JING Xiao-yuan[][],ZHU Yang-ping[],et al. Software Defect Prediction of Metric Learning Based on Local Sparse Reconstruction[J].,2016,26(12):54.
[10]郑文静,李雷. 基于聚类核的半监督情感分类算法研究[J].计算机技术与发展,2016,26(12):87.
　ZHENG Wen-jing,LI Lei. Research on Semi-supervised Sentiment Classification Based on Cluster Kernel[J].,2016,26(12):87.
[11]张志武[],荆晓远[][],吴飞[]. 基于非负稀疏图的协同训练软件缺陷预测[J].计算机技术与发展,2017,27(07):38.
　ZHANG Zhi-wu[],JING Xiao-yuan[] [],WU Fei[]. Defect Prediction of Co-training Software with Non-negative Sparse Graph[J].,2017,27(12):38.

更新日期/Last Update: 2021-12-10