Cross-Project Software Defect Prediction Based on SMOTE and Deep Canonical Correlation Analysis

Cross-Project Defect Prediction(CPDP)is a method that utilizes historical data from other source projects to train predictive models for defect prediction in the target project.However,existing CPDP methods only consider linear correlations between features(indicators)of the source and target projects.These models are not capable of evaluating non-linear correlations between features when they exist,for example,when there are differences in data distributions between the source and target projects.As a result,the performance of such CPDP models is compromised.In this paper,this paper proposes a novel CPDP method based on Synthetic Minority Oversampling Technique(SMOTE)and Deep Canonical Correlation Analysis(DCCA),referred to as S-DCCA.Canonical Correlation Analysis(CCA)is employed to address the issue of non-linear correlations between features of the source and target projects.S-DCCA extends CCA by incorporating the MlpNet model for feature extraction from the dataset.The redundant features are then eliminated by maximizing the correlated feature subset using the CCA loss function.Finally,cross-project defect prediction is achieved through the application of the SMOTE data sampling technique.Area Under Curve(AUC)and F1 scores(F1)are used as evaluation metrics.This paper conducted experiments on 27 projects from four public datasets to validate the proposed method.The results demonstrate that,on average,our method outperforms all baseline approaches by at least 1.2%in AUC and 5.5%in F1 score.This indicates that the proposed method exhibits favorable performance characteristics.

Numerical Analysis of Hydroforming Deep Drawing of Conical Cup

Aided by the FE-code. analysis is carried to find the proper hydroforming deep-drawing condition for the perfect forming of a conical cup that can not be drawn successfully by conventional deep drawing method. Hydraulic counter pressure must be reasonably controlled, otherwise defects such as fracture and wrinkling can not be avoided. Therefore, the forming procedure is divided into three stages, and the counter pressure is adjusted intentionally to make the blank clamped onto the punch at a suitable time, then deformation at dangerous area is resisted by the effect of the counter pressure and the conical cup can be formed without defects.

Geometric Scaling Analysis of Deep Inelastic Scattering Data Including Heavy Quarks

An analytic massive total cross section of photon proton scattering is derived, which has geometric scaling. A geometric scaling is used to perform a global analysis of the deep inelastic scattering data on inclusive structure function F2 measured in lepton-hadron scattering experiments at small values of Bjorken x. It is shown that the descriptions of the inclusive structure function F2 and longitudinal structure function FL are improved with the massive analytic structure function, which may imply the gluon saturation effect dominating the parton evolution process at HERA. The inclusion of the heavy quarks prevent the divergence of the lepton-hadron cross section, which plays a significant role in the description of the photoproduction region.

Customized Convolutional Neural Network for Accurate Detection of Deep Fake Images in Video Collections

The motivation for this study is that the quality of deep fakes is constantly improving,which leads to the need to develop new methods for their detection.The proposed Customized Convolutional Neural Network method involves extracting structured data from video frames using facial landmark detection,which is then used as input to the CNN.The customized Convolutional Neural Network method is the date augmented-based CNN model to generate‘fake data’or‘fake images’.This study was carried out using Python and its libraries.We used 242 films from the dataset gathered by the Deep Fake Detection Challenge,of which 199 were made up and the remaining 53 were real.Ten seconds were allotted for each video.There were 318 videos used in all,199 of which were fake and 119 of which were real.Our proposedmethod achieved a testing accuracy of 91.47%,loss of 0.342,and AUC score of 0.92,outperforming two alternative approaches,CNN and MLP-CNN.Furthermore,our method succeeded in greater accuracy than contemporary models such as XceptionNet,Meso-4,EfficientNet-BO,MesoInception-4,VGG-16,and DST-Net.The novelty of this investigation is the development of a new Convolutional Neural Network(CNN)learning model that can accurately detect deep fake face photos.

Analysis of the Diffractive Deep Inelastic Scattering Data with Running Coupling and Gluon Number Fluctuations

We study the effects of running coupling and gluon number fluctuations in the latest diffractive deep inelastic scattering data. It is found that the description of the data is improved once the running coupling and gluon number fluctuations are included with x2/d.o.f. = 0.867, x2/d.o.f. = 0.923 and x2/d.o.f. = 0.878 for three different groups of experimental data. The values of diffusive coefficient subtracted from the fit are smaller than the ones obtained by considering only the gluon number fluctuations in our previous studies. The smaller values of the diffusive coefficient are in agreement with the theoretical predictions, where the gluon number fluctuations are suppressed by the running coupling which leads to smaller values of the diffusive coefficient.

Numerical Simulation of Installation Process and Uplift Resistance for An Integrated Suction Foundation in Deep Ocean

A concept design, named integrated suction foundation, is proposed for a tension leg platform（TLP） in deep ocean. The most important improvement in comparing with the traditional one is that a pressure-resistant storage module is designed. It utilizes the high hydrostatic pressure in deep ocean to drive water into the module to generate negative pressure for bucket suction. This work aims to further approve the feasibility of the concept design in the aspect of penetration installation and the uplift force in-place. Seepage is generated during suction penetration, and can have both positive and negative effects on penetration process. To study the effect of seepage on the penetration process of the integrated suction foundation, finite element analysis（FEA） is carried out in this work. In particular, an improved methodology to calculate the penetration resistance is proposed for the integrated suction foundation with respect to the reduction factor of penetration resistance. The maximum allowable negative pressure during suction penetration is calculated with the critical hydraulic gradient method through FEA. The simulation results of the penetration process show that the integrated suction foundation can be installed safely. Moreover, the uplift resistance of the integrated suction foundation is calculated and the feasibility of the integrated suction foundation working on-site is verified. In all, the analysis in this work further approves the feasibility of the integrated suction foundation for TLPs in deep ocean applications.

Sentiment Analysis of Code-Mixed Bambara-French Social Media Text Using Deep Learning Techniques

The global growth of the Internet and the rapid expansion of social networks such as Facebook make multilingual sentiment analysis of social media content very necessary. This paper performs the first sentiment analysis on code-mixed Bambara-French Facebook comments. We develop four Long Short-term Memory（LSTM）-based models and two Convolutional Neural Network（CNN）-based models, and use these six models, Na？ve Bayes, and Support Vector Machines（SVM） to conduct experiments on a constituted dataset. Social media text written in Bambara is scarce. To mitigate this weakness, this paper uses dictionaries of character and word indexes to produce character and word embedding in place of pre-trained word vectors. We investigate the effect of comment length on the models and perform a comparison among them. The best performing model is a one-layer CNN deep learning model with an accuracy of 83.23 %.

Online Content Impact on Young people’s Values Transformation

The investigation that underpins the present article interprets the gaps of the social data continuum.It is designed to select a set of images from the“media noise”of the information society,and then describe those that characterize the visual conceptualization of the ideas.The authors present the results of their 14-year research based on the original research methodology,and carried out in several stages(2006,2012,2017).The study is called“Fictional creatures of the mass media era.Russia,21 century”.In 2017,it is assumed that the overall youth international value agenda,an essential feature of which is the further reduction of the impact of advertising and brand communications,has been formed.Specific data are given in the article.

Statistical key variable analysis and model-based control for improvement performance in a deep reactive ion etching process

This paper proposes to develop a data-driven via＇s depth estimator of the deep reactive ion etching process based on statistical identification of key variables.Several feature extraction algorithms are presented to reduce the high-dimensional data and effectively undertake the subsequent virtual metrology（VM） model building process.With the available on-line VM model,the model-based controller is hence readily applicable to improve the quality of a via＇s depth.Real operational data taken from a industrial manufacturing process are used to verify the effectiveness of the proposed method.The results demonstrate that the proposed method can decrease the MSE from 2.2×10^（-2） to 9×10^（-4） and has great potential in improving the existing DRIE process.

Atlas-based deep gray matter and white matter analysis in Alzheimer's disease: diffusion abnormality and correlation with cognitive function

Objective To identify the diffusion alterations of deep gray matter(GM)and white matter(WM)among Alzheimer’s disease(AD),mild cognitive impairment(MCI)and healthy people by atlas-based analysis(ABA),and to investigate the respective relationship with cognitive function.Methods Twenty-one AD patients(AD group),8 MCI patients(MCI group)and

Improving the Hydromechanical Deep-Drawing Process Using Aluminum Tailored Heat Treated Blanks

The present work demonstrates the effectiveness of combining the hydromechanical deep-drawing process with the Tailored Heat Treated Blank（THTB） technique. In the hydromechanical deep-drawing process, the fluid pressure is used for postponing the fracture occurrence in the blank, while the THTB technique allows to create a material property gradient through a suitable artificial aging treatment carried out prior to the forming process. Since the number of process variables is large, in the present work the authors propose an optimization loop for the determination of the parameters controlling the extension of the blank regions to be subjected to the aging treatment and the temperature levels to be set during the heat treatment. The proposed methodology couples a simple finite element model（Abaqus） with a multiobjective optimization platform（mode FRONTIER）. A preliminary experimental campaign was carried out for determining the effect of the aging treatment on the mechanical（through tensile tests） and deformative（through formability tests）behavior of the AC170 PX aluminum alloy. Optimization results prove the effectiveness of the adopted methodology and put in evidence that the adoption of properly aged blanks in the hydromechanical deep drawing allows to increase the limit drawing ratio and to simplify the process since it is conducted at room temperature.

Soil liquefaction assessment by using hierarchical Gaussian Process model with integrated feature and instance based domain adaption for multiple data sources

For soil liquefaction prediction from multiple data sources,this study designs a hierarchical machine learning model based on deep feature extraction and Gaussian Process with integrated domain adaption techniques.The proposed model first combines deep fisher discriminant analysis(DDA)and Gaussian Process(GP)in a unified framework,so as to extract deep discriminant features and enhance the model performance for classification.To deliver fair evalu-ation,the classifier is validated in the approach of repeated stratified K-fold cross validation.Then,five different data resources are presented to further verify the model’s robustness and generality.To reuse the gained knowledge from the existing data sources and enhance the generality of the predictive model,a domain adaption approach is formu-lated by combing a deep Autoencoder with TrAdaboost,to achieve good performance over different data records from both the in-situ and laboratory observations.After comparing the proposed model with classical machine learn-ing models,such as supported vector machine,as well as with the state-of-art ensemble learning models,it is found that,regarding seismic-induced liquefaction prediction,the predicted results of this model show high accuracy on all datasets both in the repeated cross validation and Wilcoxon signed rank test.Finally,a sensitivity analysis is made on the DDA-GP model to reveal the features that may significantly affect the liquefaction.