Software Defect Prediction Method Based on Stable Learning

The purpose of software defect prediction is to identify defect-prone code modules to assist software quality assurance teams with the appropriate allocation of resources and labor.In previous software defect prediction studies,transfer learning was effective in solving the problem of inconsistent project data distribution.However,target projects often lack sufficient data,which affects the performance of the transfer learning model.In addition,the presence of uncorrelated features between projects can decrease the prediction accuracy of the transfer learning model.To address these problems,this article propose a software defect prediction method based on stable learning(SDP-SL)that combines code visualization techniques and residual networks.This method first transforms code files into code images using code visualization techniques and then constructs a defect prediction model based on these code images.During the model training process,target project data are not required as prior knowledge.Following the principles of stable learning,this paper dynamically adjusted the weights of source project samples to eliminate dependencies between features,thereby capturing the“invariance mechanism”within the data.This approach explores the genuine relationship between code defect features and labels,thereby enhancing defect prediction performance.To evaluate the performance of SDP-SL,this article conducted comparative experiments on 10 open-source projects in the PROMISE dataset.The experimental results demonstrated that in terms of the F-measure,the proposed SDP-SL method outperformed other within-project defect prediction methods by 2.11%-44.03%.In cross-project defect prediction,the SDP-SL method provided an improvement of 5.89%-25.46% in prediction performance compared to other cross-project defect prediction methods.Therefore,SDP-SL can effectively enhance within-and cross-project defect predictions.

Application of Hyperspectral Imaging Technology in Rapid Detection of Preservative in Milk

To ensure the quality and safety of pure milk,detection method of typical preservative-potassium sorbate in milk was researched in this paper.Hyperspectral imaging technology was applied to realize rapid detection.Influence factors for hyperspectral data collection for milk samples were firstly researched,including height of sample,bottom color and sample filled up container or not.Pretreatment methods and variable selection algorithms were applied into original spectral data.Rapid detection models were built based on support vector machine method(SVM).Finally,standard normalized variable(SNV)-competitive adaptive reweighted sampling(CARS)and SVM model was chosen in this paper.The accuracies of calibration set and testing set were 0.97 and 0.97,respectively.Kappa coefficient of the model was 0.93.It could be seen that hyperspectral imaging technology could be used to detect for potassium sorbate in milk.Meanwhile,it also provided methodological supports for the rapid detection of other preservatives in milk.

A new method of searching for concealed Au deposits by using the spectrum of arid desert plant species

With the increase of exploration depth,it is more and more difficult to find Au deposits.Due to the limitation of time and cost,traditional geological exploration methods are becoming increasingly difficult to be effectively applied.Thus,new methods and ideas are urgently needed.This study assessed the feasibility and effectiveness of using hyperspectral technology to prospect for hidden Au deposits.For this purpose,48 plant(Seriphidium terrae-albae)and soil(aeolian gravel desert soil)samples were first collected along a sampling line that traverses an Au mineralization alteration zone(Aketasi mining region in an arid region of China)and were used to obtain soil Au contents by a chemical analysis method and the reflectance spectra of plants obtained with an Analytical Spectral Device(ASD)FieldSpec3 spectrometer.Then,the corresponding relationship between the soil Au content anomaly and concealed Au deposits was investigated.Additionally,the characteristic bands were selected from plant spectra using four different methods,namely,genetic algorithm(GA),stepwise regression analysis(STE),competitive adaptive reweighted sampling(CARS),and correlation coefficient method(CC),and were then input into the partial least squares(PLS)method to construct a model for estimating the soil Au content.Finally,the quantitative relationship between the soil Au content and the 15 different plant transformation spectra was established using the PLS method.The results were compared with those of a model based on the full spectrum.The results obtained in this study indicate that the location of concealed Au deposits can be predicted based on soil geochemical anomaly information,and it is feasible and effective to use the full plant spectrum and PLS method to estimate the Au content in the soil.The cross-validated coefficient of determination(R2)and the ratio of the performance to deviation(RPD)between the predicted value and the measured value reached the maximum of 0.8218 and 2.37,respectively,with a minimum value of 6.56μg/kg for the root-mean-squared error(RMSE)in the full spectrum model.However,in the process of modeling,it is crucial to select the appropriate transformation spectrum as the input parameter for the PLS method.Compared with the GA,STE,and CC methods,CARS was the superior characteristic band screening method based on the accuracy and complexity of the model.When modeling with characteristic bands,the highest accuracy,R2 of 0.8016,RMSE of 7.07μg/kg,and RPD of 2.20 were obtained when 56 characteristic bands were selected from the transformed spectra(1/lnR)'(where it represents the first derivative of the reciprocal of the logarithmic spectrum)of sampled plants using the CARS method and were input into the PLS method to construct an inversion model of the Au content in the soil.Thus,characteristic bands can replace the full spectrum when constructing a model for estimating the soil Au content.Finally,this study proposes a method of using plant spectra to find concealed Au deposits,which may have promising application prospects because of its simplicity and rapidity.

BP神经网络结合变量选择方法在牛奶蛋白质含量检测中的应用

牛奶中的蛋白质含量会影响牛奶的品质,利用高光谱图像的光谱特征信息研究对牛奶蛋白质含量预测的可行性。本文提出一种基于竞争性自适应重加权算法(competitive adaptive reweighted sampling, CARS)和连续投影算法(successive projections algorithm, SPA)结合多层前馈神经网络(back propagation, BP)的预测建模方法,实验以含有不同浓度蛋白质的牛奶为对象,利用可见光/近红外高光谱成像系统共采集到5种牛奶共计250组高光谱数据,通过实验对比选择采用标准化方法对获取到的吸收光谱预处理,然后采用CARS结合SPA筛选特征波长,得到18个特征波长,建立CARS-SPA-BP模型,经过试验,CARS-SPA-BP模型的训练集决定系数和测试集决定系数R;和R;分别达到0.971和0.968,训练集均方根误差(root mean square error of calibration,RMSEC)和测试集均方根误差(root mean square error of prediction,RMSEP)达到了0.033和0.034。研究发现,采用CARS结合SPA筛选的牛奶特征波长建立的多层前馈神经网络模型,其模型预测结果与全波长建模相比并没有明显降低,因此将CARS结合SPA用于波长筛选并且结合BP神经网络基本可以完成对牛奶蛋白质含量的预测。为验证CARS-SPA-BP模型的预测能力,在相同数据环境下,使用较为传统的偏最小二乘回归(partial least squares regression, PLSR)进行建模,实验结果表明,CARS-SPA-BP相较于PLSR,R;和RMSEP均有明显提升。研究表明,CARS-SPA-BP可充分利用牛奶光谱特征信息实现较高精度的牛奶蛋白质含量检测。

Comparison and rapid prediction of lignocellulose and organic elements of a wide variety of rice straw based on near infrared spectroscopy

Rice straw is a major kind of biomass that can be utilized as lignocellulosic materials and renewable energy.Rapid prediction of the lignocellulose(cellulose,hemicellulose,and lignin)and organic elements(carbon,hydrogen,nitrogen,and sulfur)of rice straw would help to decipher its growth mechanisms and thereby improve its sustainable usages.In this study,364 rice straw samples featuring different rice subspecies(japonica and indica),growing seasons(early-,middle-,and late-season),and growing environments(irrigated and rainfed)were collected,the differences among which were examined by multivariate analysis of variance.Statistic results showed that the cellulose content exhibited significant differences among different growing seasons at a significant level(p<0.01),and the contents of cellulose and nitrogen had significant differences between different growing environments(p<0.01).Near infrared reflectance spectroscopy(NIRS)models for predicting the lignocellulosic and organic elements were developed based on two algorithms including partial least squares(PLS)and competitive adaptive reweighted sampling-partial least squares(CARS-PLS).Modeling results showed that most CARS-PLS models are of higher accuracy than the PLS models,possibly because the CARS-PLS models selected optimal combinations of wavenumbers,which might have enhanced the signal of chemical bonds and thereby improved the predictive efficiency.As a major contributor to the applications of rice straw,the nitrogen content was predicted precisely by the CARS-PLS model.Generally,the CARS-PLS models efficiently quantified the lignocellulose and organic elements of a wide variety of rice straw.The acceptable accuracy of the models allowed their practical applications.

Rapid fatty acids detection of vegetable oils by Raman spectroscopy based on competitive adaptive reweighted sampling coupled with support vector regression

Objectives:The composition and content of fatty acids are critical indicators of vegetable oil quality.To overcome the drawbacks of traditional detection methods,Raman spectroscopy was investigated for the fast determination of the fatty acids composition of oil.Materials and Methods:Rapeseed and soybean oil at different depths of the oil tank at different storage times were collected and an eighth-degree polynomial function was used to fit the Raman spectrum.Then,the multivariate scattering correction,standard normal variable transformation(SNV),and Savitzky–Golay convolution smoothing methods were compared.Results:Polynomial fitting combined with SNV was found to be the optimal pretreatment method.Characteristic wavelengths were selected by competitive adaptive reweighted sampling.For monounsaturated fatty acids(MUFAs),polyunsaturated fatty acids(PUFAs),and saturated fatty acids(SFAs),44,75,and 92 characteristic wavelengths of rapeseed oil,and 60,114,and 60 characteristic wavelengths of soybean oil were extracted.Support vector regression was used to establish the prediction model.The R^(2)values of the prediction results of MUFAs,PUFAs,and SFAs for rapeseed oil were 0.9670,0.9568,and 0.9553,and the root mean square error(RMSE)values were 0.0273,0.0326,and 0.0340,respectively.The R^(2)values of the prediction results of fatty acids for soybean oil were respectively 0.9414,0.9562,and 0.9422,and RMSE values were 0.0460,0.0378,and 0.0548,respectively.A good correlation coefficient and small RMSE value were obtained,indicating the results to be highly accurate and reliable.Conclusions:Raman spectroscopy,based on competitive adaptive reweighted sampling coupled with support vector regression,can rapidly and accurately analyze the fatty acid composition of vegetable oil.