發(fā)布時(shí)間：2018-08-04 18:30

【摘要】：軟件缺陷預(yù)測(cè)是根據(jù)歷史數(shù)據(jù)以及已經(jīng)發(fā)現(xiàn)的缺陷等軟件度量數(shù)據(jù)預(yù)測(cè)未來(lái)開(kāi)發(fā)軟件缺陷性的技術(shù)。本文針對(duì)軟件缺陷預(yù)測(cè)做了以下工作:首先,針對(duì)缺陷檢出率和誤報(bào)率這兩個(gè)目標(biāo),已有的研究主要集中于多目標(biāo)微粒群算法,但該算法生成規(guī)則的分類器包含多條規(guī)則,每條規(guī)則分布不均勻,規(guī)則之間有重復(fù)區(qū)域覆蓋,且生成的規(guī)則需要進(jìn)行組合才能對(duì)軟件缺陷進(jìn)行預(yù)測(cè)。這些問(wèn)題在很大程度上影響算法性能。為此,論文引入多目標(biāo)定向布谷鳥(niǎo)算法用以優(yōu)化預(yù)測(cè)模型的參數(shù),布谷鳥(niǎo)算法的個(gè)體當(dāng)前位置即為其個(gè)體歷史最優(yōu)位置,這一特性使得該多目標(biāo)算法生成的規(guī)則分布較為均勻,從而改善了算法性能。為了驗(yàn)證該算法,論文選擇了NASA數(shù)據(jù)庫(kù)的八個(gè)數(shù)據(jù)集,并與八個(gè)比較算法進(jìn)行比較,實(shí)驗(yàn)結(jié)果表明,基于MOOCS的多目標(biāo)軟件缺陷預(yù)測(cè)技術(shù)的平均性能較優(yōu)。其次,我們對(duì)NASA數(shù)據(jù)庫(kù)的六個(gè)數(shù)據(jù)集中的模塊代碼行數(shù)分布不均衡性分布進(jìn)行了分析,結(jié)果發(fā)現(xiàn)大部分軟件模塊的代碼行數(shù)較少(稱為小模塊),僅有少量模塊的代碼行數(shù)較長(zhǎng)(稱為大模塊)。根據(jù)Arisholm的研究結(jié)論:軟件模塊的測(cè)試資源大小和軟件模塊的代碼行數(shù)成正比,大模塊在缺陷預(yù)測(cè)過(guò)程中產(chǎn)生的誤報(bào)率會(huì)嚴(yán)重浪費(fèi)測(cè)試資源。為了降低這種由于大模塊誤報(bào)而導(dǎo)致的資源浪費(fèi),我們將一個(gè)數(shù)據(jù)集按照代碼長(zhǎng)度分為大模塊和小模塊,并各自分配一個(gè)支持向量分類器,以達(dá)到降低總體測(cè)試資源浪費(fèi)的目標(biāo)。為了進(jìn)一步驗(yàn)證技術(shù)性能,論文將該算法與九個(gè)算法進(jìn)行比較,結(jié)果表明了基于多目標(biāo)雙支持向量機(jī)軟件缺陷預(yù)測(cè)技術(shù)的有效性。最后,基于雙支持向量機(jī)的多目標(biāo)缺陷預(yù)測(cè)需要將一個(gè)數(shù)據(jù)集依據(jù)代碼行數(shù)分割為大模塊和小模塊,這個(gè)比例的選擇對(duì)缺陷預(yù)測(cè)的效果影響很大,因此我們采用黃金分割法對(duì)五個(gè)數(shù)據(jù)集的分割比例做了選擇,實(shí)驗(yàn)結(jié)果表明,該比例在40%-80%之間,效果較優(yōu)。
[Abstract]:Software defect prediction is a technology to predict future development software defects based on historical data and the detected defects, such as defects. In this paper, the following work is done for software defect prediction: firstly, the research focuses on the two targets of defect detection rate and false alarm rate, which are mainly focused on the multi-objective particle swarm optimization algorithm, but this calculation is calculated. The classifier of the rule generating rules contains many rules, each rule is not evenly distributed, and the rules have repeated area coverage, and the rules that are generated need to be combined to predict the software defects. These problems affect the performance of the algorithm to a great extent. The individual current position of the cuckoo algorithm is the optimal location of individual history, which makes the rule distribution of the multi-objective algorithm more uniform and thus improves the performance of the algorithm. In order to verify the algorithm, eight data sets of the NASA database are selected and compared with the eight comparison algorithms. The experimental results are compared. It shows that the average performance of the MOOCS based multi target software defect prediction technology is better. Secondly, we analyze the distribution of the unbalance distribution of the number of code lines in the module of the six data sets of the NASA database. The results show that most of the software modules have less code lines (called small modules), and only a small number of modules have a long number of code lines. According to the research conclusion of Arisholm: the test resource size of the software module is proportional to the number of code lines in the software module, the false alarm rate generated by the large module in the defect prediction process is a serious waste of test resources. In order to reduce the resource waste caused by the large module misinformation, we put a data set in accordance with the code. The length is divided into large modules and small modules, and a support vector classifier is allocated each to reduce the waste of the overall test resources. In order to further verify the technical performance, the paper compares the algorithm with the nine algorithms. The results show the effectiveness of the software defect prediction technology based on the multi target dual support vector machine software. Finally, The multi target defect prediction based on dual support vector machine (SVM) needs to divide a data set into large modules and small modules based on the number of code lines. The ratio selection has a great influence on the effect of defect prediction. Therefore, we choose the proportion of five data sets by the golden segmentation method. The experimental results show that the ratio is in 40%-80%. The effect is better.
【學(xué)位授予單位】：太原科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP311.5

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 劉望舒;陳翔;顧慶;劉樹(shù)龍;陳道蓄;;軟件缺陷預(yù)測(cè)中基于聚類分析的特征選擇方法[J];中國(guó)科學(xué):信息科學(xué);2016年09期

2 陳翔;顧慶;劉望舒;劉樹(shù)龍;倪超;;靜態(tài)軟件缺陷預(yù)測(cè)方法研究[J];軟件學(xué)報(bào);2016年01期

3 王男帥;薛靜鋒;胡昌振;單純;李志強(qiáng);;基于遺傳優(yōu)化支持向量機(jī)的軟件缺陷預(yù)測(cè)模型[J];中國(guó)科技論文;2015年02期

4 林開(kāi)標(biāo);馬櫻;胡建強(qiáng);朱順痣;;靜態(tài)代碼屬性與軟件模塊缺陷數(shù)的偏相關(guān)分析[J];廈門(mén)理工學(xué)院學(xué)報(bào);2014年05期

5 王洪坡;周紅建;王洪雷;;基于威布爾分布的嵌入式軟件安全性量化分析[J];計(jì)算機(jī)應(yīng)用與軟件;2014年04期

6 王洪坡;周紅建;王洪雷;;基于S曲線模型的航空裝備嵌入式軟件量化安全評(píng)估[J];計(jì)算機(jī)工程與科學(xué);2014年03期

7 尹然;丁曉明;李小亮;梅瑩;;基于SA-BP神經(jīng)網(wǎng)絡(luò)的軟件缺陷預(yù)測(cè)模型的研究[J];西南師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年08期

8 于安雷;皮德常;;基于PSO-BP的軟件缺陷預(yù)測(cè)模型[J];計(jì)算機(jī)工程與應(yīng)用;2013年07期

9 仇晨曄;王春露;左興權(quán);方濱興;;基于K-Means全局引導(dǎo)策略的多目標(biāo)微粒群算法[J];北京郵電大學(xué)學(xué)報(bào);2012年05期

10 何亮;宋擒豹;沈鈞毅;;基于Boosting的集成k-NN軟件缺陷預(yù)測(cè)方法[J];模式識(shí)別與人工智能;2012年05期

相關(guān)博士學(xué)位論文 前2條

1 楊曉杏;基于度量元的軟件缺陷預(yù)測(cè)技術(shù)[D];中國(guó)科學(xué)技術(shù)大學(xué);2014年

2 馬櫻;基于機(jī)器學(xué)習(xí)的軟件缺陷預(yù)測(cè)技術(shù)研究[D];電子科技大學(xué);2012年

，

本文編號(hào)：2164739