计算机科学 ›› 2016, Vol. 43 ›› Issue (Z11): 486-489.doi: 10.11896/j.issn.1002-137X.2016.11A.109
张晓风,张德平
ZHANG Xiao-feng and ZHANG De-ping
摘要： 软件缺陷预测是软件可靠性研究的一个重要方向。由于影响软件失效的因素有很多,相互之间关联关系复杂,在分析建模中常用联合分布函数来描述,而实际应用中难以确定,直接影响软件失效预测。基于拟似然估计提出一种软件失效预测方法,通过主成分分析筛选影响软件失效的主要影响因素,建立多因素软件失效预测模型,利用这些影响因素的数字特征(均值函数和方差函数)以及采用拟似然估计方法估计出模型参数,进而对软件失效进行预测分析。基于两个真实数据集Eclipse JDT和Eclipse PDE,与经典Logistic回归和Probit回归预测模型进行实验对比分析,结果表明采用拟似然估计对软件缺陷预测具有可行性,且预测精度均优于这两种经典回归预测模型。
| [1] Catal C.Performance Evaluation Metrics for Software FaultPrediction Studies[J].Acta Polytechnica Hungarica,2012,9(4):193-206 [2] 杨晓杏.基于度量元的软件缺陷预测技术[D].合肥:中国科学技术大学,2014:1-11 [3] Zhang F,Mockus A,Keivanloo I,et al.Towards building a universal defect prediction model[C]∥Working Conference on Mining Software Repositories.2014:182-191 [4] Suresh Y,Kumar L,Rath S K.Statistical and Machine Learning Methods for Software Fault Prediction Using CK Metric Suite:A Comparative Analysis[J].International Scholarly Research Notices,2014:1-15 [5] Basili B V R,Briand L,Melo W L.A Validation of Object-O-riented Design Metrics As Quality Indicators[J].IEEE Transactions on Software Engineering,1996,2(10):751-756 [6] Briand L C,Wüst J,Daly J W,et al.Exploring the relationships between design measures and software quality in object-oriented systems[J].Journal of Systems & Software,2000,51(3):245-273 [7] Olague H M,Etzkorn L H,Gholston S,et al.Empirical Validation of Three Software Metrics Suites to Predict Fault-Proneness of Object-Oriented Classes Developed Using Highly Iterative or Agile Software Development Processes[J].IEEE Tran-sactions on Software Engineering,2007,33(6):402-419 [8] Kanmani S,Uthariaraj V R,Sankaranarayanan V,et al.Object oriented software quality prediction using general regression neural networks[J].Acm Sigsoft Software Engineering Notes,2004,29(5):1-6 [9] Nagappan N,Williams L,Vouk M,et al.Early estimation ofsoftware quality using in-process testing metrics:A controlled case study[J].Acm Sigsoft Software Engineering Notes,2005,30(4):1-7 [10] D’Ambros M,Lanza M,Robbes R.An extensive comparison of bug prediction approaches[C]∥IEEE Working Conference on Mining Software Repositories.2010:31-41 [11] Wedderburn R W M.Quasi-likelihood function,generalized linear models and the Gauss-Newton method[J].Biometrika.,1974,1(3):439-447 [12] Xia T,Jiang X,Wang X.Strong consistency of the maximumquasi-likelihood estimator in quasi-likelihood nonlinear models with stochastic regression[J].Statistics & Probability Letters,2015,103(4):391-400 [13] Su F,Chan K S.Quasi-likelihood estimation of a threshold diffusion process[J].Journal of Econometrics,2015,189(2):473-484 [14] Hu H C,Song L.Quasi-Maximum Likelihood Estimators inGeneralized Linear Models with Autoregressive Processes[J].数学学报(英文版),2014,30(12):2085-2102 [15] Liang K Y,Zeger S L.Logitudinal data analysis using generali-zed linear models[J].Biometrika,1986,73(1):13-22 [16] 陈夏,陈希孺.广义线性模型极大拟似然估计的强相合性和渐进正态性[J].应用概率统计,2005,1(3):251-263 [17] 张三国,廖源.关于广义线性模型拟似然估计弱相合性的几个问题[J].中国科学A辑:数学,2007,37(11):1369-1377 [18] 朱仲义.异方差非线性模型中的拟似然估计的渐进性质[J].河海大学学报,1999,24(2):110-115 [19] Ahmed S E,Fallahpour S.Shrinkage estimation strategy in quasi-likelihood models[J].Statistics & Probability Letters,2012,82(12):2170-2179 [20] 高启兵,吴耀华.广义线性回归拟似然估计的渐进正态性[J].系统科学与数学,2005,5(6):738-745 |
| No related articles found! |
|
||
首页