Cross Projects Research Articles

Early Software Defects Density Prediction: Training the International Software Benchmarking Cross Projects Data Using Supervised Learning

Early Software Defects Density Prediction: Training the International Software Benchmarking Cross Projects Data Using Supervised Learning

Using Cost-cognitive Bagging Ensemble to Improve Cross-project Defects Prediction

<p>Cross-project defect prediction (CPDP) is a field of study that allows predicting defects in software projects for which the availability of data is limited and produces generalizable prediction models. Due to the heterogeneity of cross projects, CPDP is particularly challenging and several methods have been employed to address this problem. Nevertheless, the class-imbalanced characteristic of the cross-project defect data also increases the learning difficulty of such a task but has not been investigated in depth. This paper proposed a novel, cost-cognitive ensemble method for CPDP, which includes four phases: bagging balanced resampling phase, base classifiers learning phase, cost value cognitive phase, and base classifiers ensemble phase. These phases create a composition of classifiers that are used for predicting defects. Results of an empirical evaluation on 10 datasets from the PROMISE repository indicated that our method achieves the best overall performance with respect to conventional methods. Moreover, our method could cognize the cost value automatically during the model training, it is shown to be more effective and practical.</p> <p> </p>

Heterogeneous Fault Prediction Using Feature Selection and Supervised Learning Algorithms

Software Fault Prediction (SFP) is the most persuasive research area of software engineering. Software Fault Prediction which is carried out within the same software project is known as With-In Fault Prediction. However, local data repositories are not enough to build the model of With-in software Fault prediction. The idea of cross-project fault prediction (CPFP) has been suggested in recent years, which aims to construct a prediction model on one project, and use that model to predict the other project. However, CPFP requires that both the training and testing datasets use the same set of metrics. As a consequence, traditional CPFP approaches are challenging to implement through projects with diverse metric sets. The specific case of CPFP is Heterogeneous Fault Prediction (HFP), which allows the program to predict faults among projects with diverse metrics. The proposed framework aims to achieve an HFP model by implementing Feature Selection on both the source and target datasets to build an efficient prediction model using supervised machine learning techniques. Our approach is applied on two open-source projects, Linux and MySQL, and prediction is evaluated based on Area Under Curve (AUC) performance measure. The key results of the proposed approach are as follows: It significantly gives better results of prediction performance for heterogeneous projects as compared with cross projects. Also, it demonstrates that feature selection with feature mapping has a significant effect on HFP models. Non-parametric statistical analyses, such as the Friedman and Nemenyi Post-hoc Tests, are applied, demonstrating that Logistic Regression performed significantly better than other supervised learning algorithms in HFP models.

A CUPID Li2100MoO4 scintillating bolometer tested in the CROSS underground facility

A scintillating bolometer based on a large cubic Li2100MoO4 crystal (45 mm side) and a Ge wafer (scintillation detector) has been operated in the CROSS cryogenic facility at the Canfranc underground laboratory in Spain. The dual-readout detector is a prototype of the technology that will be used in the next-generation 0ν2β experiment CUPID . The measurements were performed at 18 and 12 mK temperature in a pulse tube dilution refrigerator. This setup utilizes the same technology as the CUORE cryostat that will host CUPID and so represents an accurate estimation of the expected performance. The Li2100MoO4 bolometer shows a high energy resolution of 6 keV FWHM at the 2615 keV γ line. The detection of scintillation light for each event triggered by the Li2100MoO4 bolometer allowed for a full separation (∼8σ) between γ(β) and α events above 2 MeV . The Li2100MoO4 crystal also shows a high internal radiopurity with 228Th and 226Ra activities of less than 3 and 8 μBq/kg, respectively. Taking also into account the advantage of a more compact and massive detector array, which can be made of cubic-shaped crystals (compared to the cylindrical ones), this test demonstrates the great potential of cubic Li2100MoO4 scintillating bolometers for high-sensitivity searches for the 100Mo 0ν2β decay in CROSS and CUPID projects.

An empirical analysis of the statistical learning models for different categories of cross-project defect prediction

Currently, the research community is addressing the problem of defect prediction with the availability of project defect data. The availability of different project data leads to extend the research on cross projects. Cross-project defect prediction has now become an accepted area of software project management. In this paper, an empirical study is carried out to investigate the predictive performance of availability within project and cross-project defect prediction models. Furthermore, different categories of cross-project data are taken for training and testing to analyse various statistical models. In this paper data models are analysed and compared using various statistical performance measures. The findings during the empirical analysis of the data models state that gradient boosting predictor outperforms in the cross-project defect prediction scenario. Results also infer that cross-project defect prediction is comparable with project defect prediction and has statistical significance.

An empirical analysis of the statistical learning models for different categories of cross-project defect prediction

Currently, the research community is addressing the problem of defect prediction with the availability of project defect data. The availability of different project data leads to extend the research on cross projects. Cross-project defect prediction has now become an accepted area of software project management. In this paper, an empirical study is carried out to investigate the predictive performance of availability within project and cross-project defect prediction models. Furthermore, different categories of cross-project data are taken for training and testing to analyse various statistical models. In this paper data models are analysed and compared using various statistical performance measures. The findings during the empirical analysis of the data models state that gradient boosting predictor outperforms in the cross-project defect prediction scenario. Results also infer that cross-project defect prediction is comparable with project defect prediction and has statistical significance.

The risk of involving foreign contractors in the Norwegian infrastructure market

Abstract As a result of globalization, high investment volume in infrastructure projects, and more use of standardization, foreign construction companies are establishing in Norway. The purpose of this paper is to explore the risks of involving foreign contractors and developing suggestions for future projects. In addition to a literature review, two Norwegian bridge projects built by Chines International Contractors studied through a document study and semi-structured interviews with key personnel. The key personnel is from a public project owner, two Chinese construction companies, and subcontractors involved in the two case projects. Some of the risks identified are language barriers, weather conditions, difficulty in managing subcontractors, and different perceptions of safety levels. In the future, project owners should consider preparing the contract documents, handbooks, and standards in both Norwegian and English. If the language risk is managed properly, it influences the other risks as well. Cross projects experience sharing will also help in identifying risks early and taking mitigation measures.

Boosted Relief Feature Subset Selection and Heterogeneous Cross Project Defect Prediction using Firefly Particle Swarm Optimization

The exponential growth in the field of information technology, need for quality-based software development is highly demanded. The important factor to be focused during the software development is software defect detection in earlier stages. Failure to detect hidden faults will affect the effectiveness and quality of the software usage and its maintenance. In traditional software defect prediction models, projects with same metrics are involved in prediction process. In recent years, active topic is dealing with Cross Project Defect Prediction (CPDP) to predict defects on software project from other software projects dataset. Still, traditional cross project defect prediction approaches also require common metrics among the dataset of two projects for constructing the defect prediction techniques. Suppose if cross project dataset with different metrics has to be used for defect prediction then these methods become infeasible. To overcome the issues in software defect prediction using Heterogeneous cross projects dataset, this paper introduced a Boosted Relief Feature Subset Selection (BRFSS) to handle the two different projects with Heterogeneous feature sets. BRFSS employs the mapping approach to embed the data from two different domains into a comparable feature space with a lower dimension. Based on the similarity measure the difference among the mapped domains of dataset are used for prediction process. This work used five different software groups with six different datasets to perform heterogeneous cross project defect prediction using firefly particle swarm optimization. To produce optimal defect prediction in the Heterogeneous environment, the knowledge of particle swarm optimization by inducing firefly algorithm. The simulation result is compared with other standard models, the outcome of the result proved the efficiency of the prediction process while using firefly enabled particle swarm optimization.

Substantiation of multinomial classification using ensemble learning approach

The software defect prediction is the most operative research domain in software engineering as it enhances its reliability. The availability of defect data related to different projects leads to cross project defect prediction an open issue. This paper instanced on multiclass/multinomial classification of defect prediction on different categories of cross projects. The ensemble learning statistical models – Random forest and Gradient Boosting are used for classification. An empirical study is carried out to determine the predictive performance of the within project and cross project prediction models. Depending on the number of defects, class level information is classified into one of three defined multiclass. The homogeneous set of object oriented metrics is used for training the model. Furthermore, k-fold cross validation is done to evaluate the training accuracy of the statistical models. Major outcome of the paper concludes that multinomial/multiclass classification is applicable on cross project data and has comparable results to within project defect data with statistical significance.

Applying Topic Modeling and Similarity for Predicting Bug Severity in Cross Projects

Recently, software has increased in complexity and been applied in various industrial fields. As a result, the presence of software bugs cannot be avoided. Various bug severity prediction methodologies have been proposed, but their performance needs to be further improved. In this study, we propose a novel technique for bug severity prediction in cross projects such as Eclipse, Mozilla, WireShark, and Xamarin by using topic modeling and similarity (i.e., KL-divergence). First, we construct topic models from bug repositories in cross projects using Latent Dirichlet Allocation (LDA). Then, we find topics in each project that contain the most numerous similar bug reports by using a new bug report. Next, we extract the bug reports belonging to the selected topics and input them to a Naive Bayes Multinomial (NBM) algorithm. Finally, we predict the bug severity in the new bug report. In order to evaluate the performance of our approach and to verify the difference between cross projects and single project, we compare it with the Naive Bayes Multinomial approach; the Lamkanfi methodology, which is a well-known bug severity prediction approach; and an emotional similarity-based bug severity prediction approach. Our approach exhibits a better performance than the compared methods.

Empirical validation of object-oriented metrics on cross-projects with different severity levels

An object-oriented (OO) metrics has become crucial desideratum for software effort and fault predictions. To strengthen the adequacy of object-oriented metrics, it becomes important to know relationship between OO metrics and fault proneness at different levels of severity. It is inconceivable to build model of accurate estimate due to the inherent uncertainty in development projects. Empirical validation of software metrics is essential issue to determine applicability of prediction model. In this study, empirical validation is done on OO metrics given by Chidamber and Kemerer (CK suite) for predicting faults at different severity levels. This paper also instanced on defect prediction using cross-projects (CP) because of the unpredictability in selection of software attributes by analogy-based approach that deliver imprecise and ambiguous solution. This paper depicts detection of fault-proneness by extracting the relevant OO metrics and such models helps to focus on fault-prone modules of new projects by allocating more resources to them with use of regression and other machine learning methods. Combination of CP data with regression techniques improves effectiveness of prediction by extracting similar features impacted by all datasets.

Empirical validation of object-oriented metrics on cross-projects with different severity levels

An object-oriented (OO) metrics has become crucial desideratum for software effort and fault predictions. To strengthen the adequacy of object-oriented metrics, it becomes important to know relationship between OO metrics and fault proneness at different levels of severity. It is inconceivable to build model of accurate estimate due to the inherent uncertainty in development projects. Empirical validation of software metrics is essential issue to determine applicability of prediction model. In this study, empirical validation is done on OO metrics given by Chidamber and Kemerer (CK suite) for predicting faults at different severity levels. This paper also instanced on defect prediction using cross-projects (CP) because of the unpredictability in selection of software attributes by analogy-based approach that deliver imprecise and ambiguous solution. This paper depicts detection of fault-proneness by extracting the relevant OO metrics and such models helps to focus on fault-prone modules of new projects by allocating more resources to them with use of regression and other machine learning methods. Combination of CP data with regression techniques improves effectiveness of prediction by extracting similar features impacted by all datasets.

Prediction of Cross Project Defects using Ensemble based Multinomial Classifier

BACKGROUND: The availability of defect related data of different projects leads to cross project defect prediction an open issue. Many studies have focused on analyzing and improving the performance of Cross project defect prediction. OBJECTIVE: The multinomial classification has not been much explored. This paper instanced on multiclass/multinomial classification of defect prediction of cross projects. METHOD: The ensemble based statistical models – Gradient Boosting and Random Forest are used for classification. An empirical study is carried out to determine the performance of multinomial classification for cross project defect prediction. Depending on the number of defects, class level information is classified into one of three defined multiclass class 0, class 1, and class 2. RESULTS & CONCLUSION: Major outcome of the paper concludes that multinomial/multiclass classification is applicable on cross project data and has comparable results to within project defect data.

Neural Network based Refactoring Area Identification in Software System with Object Oriented Metrics

Objectives of the Study (a) To study previously designed models for identification of refactoring area in Object Oriented Software Systems. (b) To design a general framework or model that helps to easily identify the software code smells for a good quality of coding. (c) To identify the bad smells in the code with a design of neural network based model with the help of object-oriented metrics and further to predict the performance of the proposed model using various evaluation parameters of confusion matrix. Analysis/Methods: In this study, two different versions of Rhino (1.7r1 and 1.7r2) were taken as dataset. Object –Oriented metrics were taken as input data and the probability factor (occurrence or non-occurrence of a bad smell as output. Presence of a bad smell was considered as 1 and 0 means absence of bad smell. If there was at least one bad smell present in the code in a class, it was marked as smelly class. The tool used to extract the databases for collected object –oriented metrics and bad smells of these Rhino versions is PTIDEJ. Further, the data was tested on neural networks for different epochs to predict their performance. Findings: a) Bad Smell Analysis: Twelve design smells were considered to detect the presence of bad smell in code. If there was at least one bad smell present in the code in a class, it was marked as smelly class. b) Neural Network Model Table: Weight and bias factor for various predictors were calculated for different epochs (500, 1000, and 2000). It shows the weights assigned from input layer to hidden layer and from hidden layer to output neurons layer. After the training, the weights were tested on various datasets. C) Performance Tables and Graphs: In this, the Neural network proposed model was trained using different number of epochs to examine if the number of epochs used in training has any impact on the results or not. Further, the results for the accuracy of these models were shown. Novelty/Improvement: When the data was highly trained then the results were better. When the data was trained with 500 epochs, it was suitable for only with-in company projects but when the data was more trained than the model was also appropriate for cross projects. It was seen that when the data was trained with 1000 and 2000 epochs, the results of the proposed model were improved.

Resilient modernisation: Morocco’s agency in Red Cross projects from Hassan I to the Rif Republic, 1886-1926

From the end of the 19 th to the beginning of the 20 th century, Morocco’s modernizing projects included plans for signing up to the Geneva Convention and creating a local Red Cross/Crescent society. These plans initially stemmed from the convergence of Moroccan administrative/military reforms and Spanish “regenerationist” interventions. They ran parallel to developments in leading Islamic countries such as Egypt, Persia and the Ottoman Empire though they would have to wait till the country’s independence from Franco-Spanish domination in 1956 to become a reality. Beyond their lack of actual results, those early initiatives would serve as legal ground for Morocco’s “humanitarian sovereignty”, tacitly confirmed by the provisions of the Algeciras Act of 1906. In the following two decades, the resilience of this sovereignty would reveal itself in the sustained competition between the Spanish and the French Red Cross for “humanitarian hegemony” in the country, and also in the repeated and nearly successful demands to establish a local Red Cross/Crescent that were made to the International Committee of the Red Cross by the insurgent leader Abdelkrim during the so-called Rif War.

Tuesday Morning September 11, 2001

Abstract This article sites several post-attack, national surveys indicating negative mental health consequences as a context for considering the recent mobilization of the Green Cross Projects (GCP) in New York's lower Manhattan area. The GCP is a humanitarian, disaster mental health organization that provides crisis intervention to organizations (public and private) affected by traumatic events. After noting the history, purpose, and structure of the GCP, the article describes its recently completed mobilization beginning with the invitation by a large, local service employee union. Based on the invitation an agreed upon mission was established with six measurable objectives. This is followed by a description of what happened, who and how many were served and trained. The final section of the paper discusses the lessons applied from traumatology and the lessons learned that should be applied to future community-based, organizational assistance following a major disaster.

Lessons learned from Green Cross projects deployment to New York City.

Lessons learned from Green Cross projects deployment to New York City.

CO in Alberta-A Vision of the Future

Abstract An exciting new future of large-scale carbon dioxide enhanced oil recovery awaits Alberta's depleted oil fields. This paper presents the potential and how the necessary infrastructure could be developed. Such an infrastructure currently exists in the Permian Basin of West Texas and Southeast New Mexico, where over 28.17 106m3/day of carbon dioxide is injected for enhanced oil recovery. A number of factors have come, or are coming, together to form the environment where a comparable infrastructure may be possible in Alberta. A scoping evaluation of carbon dioxide sources and locations, reservoir locations, potential oil recovery, and capital costs to develop the infrastructure is carried out. The factors leading to a large-scale carbon dioxide industry are compared to the factors in place during the 1970s and 1980s, when most of the hydrocarbon miscible floods were initiated in Alberta. Finally, conclusions are drawn regarding the viability of the infrastructure proposed in this paper. Introduction Carbon dioxide (CO2), a simple molecule composed of one carbon and two oxygen atoms, has been the subject of interest by the oil industry for over 25 years. Original interest was in its potential for enhanced oil recovery (EOR). More recently, that interest has been eclipsed by its perceived role in global climate change, and the potentially negative business and economic implications of emitting CO2 into the atmosphere. A win-win situation could develop by constructing a province-wide infrastructure for CO2 collection and transmission that will address both interests: increased light oil recovery through EOR, and reduced CO2 emissions. Factors leading in this direction include:Declining reserves of light-medium oil. Alberta has enjoyed a long history of light oil production. However, new discoveries are not replacing production and reserves are declining, as illustrated in Figure 1. Current reserves are now approximately one third of proved reserves in 1977.FIGURE 1: Historical light-medium oil reserves for the province of Alberta. (Available in full paper)Environmental issues related to CO2 emissions into the atmosphere. There is an increasing concern by the public and government about controlling CO2 emissions to the atmosphere. The government of Canada showed its concern on the subject by signing the Kyoto Accord, which, if ratified, will commit Canada to reducing greenhouse gas (GHG) emissions to 94% of 1990 levels by 2012.Buoyant natural gas markets. Use of natural gas in hydrocarbon miscible floods (HCMF) was feasible in the past due to lack of markets and low prices. Current markets and prices most likely preclude its use in miscible flooding.Application of a Proven Technology. CO2 flooding has been used commercially since 1972 in the Permian Basin of Texas and New Mexico, when injection began into the SACROC and North Cross projects(1). By early 1998 there were over 40 CO2 projects operating in the Permian Basin, with total CO2 injection of 28.17 106m3/d and incremental production of 23,850 m3/day. Factors that led to the success of CO2 in the Permian Basin include:Well developed CO2 infrastructureHigh quality, high productivity CO2 sources