An empirical framework for defect prediction using machine learning techniques with Android software

Structural or/and functional MRI-based machine learning techniques for attention-deficit/hyperactivity disorder diagnosis: A systematic review and meta-analysis

Software defect prediction models help testers find program modules that have a high probability of having defects. A method-calling network can express the dependencies between methods in a program. Existing approaches do not sufficiently utilize method-calling network to characterize the structural features between methods. To address this problem, in this study, it is proposed for the first time that the characteristics of methods in a program are obtained by analyzing the method-calling network, and a new approach is proposed for defect prediction at the method-level. Specifically in this study, the method-calling network of the program was first constructed and the network metrics of the method-calling network were obtained. Next, the new network embedding technique (node2vec) was used to automatically encode the method-calling network structure into a low-dimensional vector to obtain the network embedding metrics. Finally, they were combined with code metrics to construct defect prediction models. We evaluated our approaches on 13 open-source software systems. The experimental results show that the proposed method improved the values of area under the receiver operating characteristic curve by 2.5% to 6.7% and Matthews correlation coefficient by 13% to 178.4% compared to the baselines. Therefore the method-calling network contains rich structural features between methods, and the structural features extend the features used for method-level defect prediction and further improve the performance of defect prediction models.

In the age of social media, every second thousands of messages are exchanged. Analyzing those unstructured data to find out specific emotions is a challenging task. Analysis of emotions involves evaluation and classification of text into emotion classes such as Happy, Sad, Anger, Disgust, Fear, Surprise, as defined by emotion dimensional models which are described in the theory of psychology (www 1; Russell, 2005). The main goal of this paper is to cover the COVID-19 pandemic situation in India and its impact on human emotions. As people very often express their state of the mind through social media, analyzing and tracking their emotions can be very effective for government and local authorities to take required measures. We have analyzed different machine learning classification models, such as Naïve Bayes, Support Vector Machine, Random Forest Classifier, Decision Tree and Logistic Regression with 10-fold cross validation to find out top ML models for emotion classification. After tuning the Hyperparameter, we got Logistic regression as the best suited model with accuracy 77% with the given datasets. We worked on algorithm based supervised ML technique to get the expected result. Although multiple studies were conducted earlier along the same lines, none of them performed comparative study among different ML techniques or hyperparameter tuning to optimize the results. Besides, this study has been done on the dataset of the most recent COVID-19 pandemic situation, which is itself unique. We captured Twitter data for a duration of 45 days with hashtag #COVID19India OR #COVID19 and analyzed the data using Logistic Regression to find out how the emotion changed over time based on certain social factors. Keywords: classification, COVID-19, emotion, emotion analysis, Naïve Bayes, Pandemic, Random Forest, SVM.

Machine Learning (MD) has experienced significant growth in research in recent decades. Anticipated is a substantial impact of this technology on many facets of human existence. ML has revolutionized our interactions with the world, impacting industries ranging from healthcare to commerce, and from entertainment to transportation. Researchers are coming up with new algorithms and techniques to deal with complex issues, which means that ML is an essential tool in today’s technological landscape. As ML develops, its applications are projected to grow further leading to innovations and improved efficiency across various domains. This paper looks into recent and relevant scientific articles on ML approaches for weather variables analysis and numerical weather prediction. As meteorological data from several decades is recorded, we can predict wind speed, pressure, precipitation and temperature that are common studied meteorological parameters by utilizing ML techniques such as Deep Learning, Random Forest, Artificial Neural Networks, Support Vector Machine etc. This is crucial for improving the accuracy and dependability of weather forecast. As it has been found in literature review, the diversity and complexity of meteorology data means that there is no any single ML technique which serves all parameters hence different methods should be used to make accurate analysis and predictions. This demonstrates the agility and flexibility of ML as it tackles various problems thus making it a key tool in modern scientific studies and technological development.

This conference note describes a Deep Reinforcement Learning architecture specifically designed to improve wireless network performance over a heterogeneous airborne wireless network consisting of multiple waveforms, antennas, platforms, link protocols, frequencies, spatial transmission, and codes. The cooperative optimization of this high dimensional space is a difficult problem which obviously has a highly correlated characterization where human network operators cannot possibly capture these correlations. Model-free Reinforcement Learning techniques represent a potential solution to our problem. Specifically, we use Deep Q-Learning Networks (DQN) to improve networking performance. We have developed a high-fidelity network simulation tool we call Tactical Airborne Network Simulator (TANS) which we use to train our neural network before deploying to the field where the asset is deployed to some mission which is hopefully somewhat similar to the scenarios used for training. By utilizing the model developed under the TANS training scenarios for the target mission scenario our learning technique gets a head start, rather than using a truly model-free approach. Our technique is codified in the ML community as "Deep Transfer Learning" [4] where terms and metrics have been examined. This paper represents an initial investigation into both the decision support agent architecture and the ML technique. Upcoming research will be described below including our vision for an expanded agent architecture as well as ideas for improved ML techniques which ultimately will result in better wireless network performance. Here we demonstrate a minor throughput performance improvement of 4% using a proof of concept agent over the use of a standard unassisted network. We improved the throughput from 309kbps to 324 kbps.

Taxonomy of machine learning algorithms in software fault prediction using object oriented metrics

Software testing is a time-consuming and expensive process. Software fault prediction models are used to identify fault-prone classes automatically before system testing. These models can reduce the testing duration, project risks, resource and infrastructure costs. In this study, we propose a novel fault prediction model to improve the testing process. Chidamber-Kemerer Object-Oriented metrics and method-level metrics such as Halstead and McCabe are used as independent metrics in our Artificial Immune Recognition System based model. According to this study, class-level metrics based model which applies AIRS algorithm can be used successfully for fault prediction and its performance is higher than J48 based approach. A fault prediction tool which uses this model can be easily integrated into the testing process.

The selection of software metrics for building software quality prediction models is a search‐based software engineering problem. An exhaustive search for such metrics is usually not feasible due to limited project resources, especially if the number of available metrics is large. Defect prediction models are necessary in aiding project managers for better utilizing valuable project resources for software quality improvement. The efficacy and usefulness of a fault‐proneness prediction model is only as good as the quality of the software measurement data. This study focuses on the problem of attribute selection in the context of software quality estimation. A comparative investigation is presented for evaluating our proposed hybrid attribute selection approach, in which feature ranking is first used to reduce the search space, followed by a feature subset selection. A total of seven different feature ranking techniques are evaluated, while four different feature subset selection approaches are considered. The models are trained using five commonly used classification algorithms. The case study is based on software metrics and defect data collected from multiple releases of a large real‐world software system. The results demonstrate that while some feature ranking techniques performed similarly, the automatic hybrid search algorithm performed the best among the feature subset selection methods. Moreover, performances of the defect prediction models either improved or remained unchanged when over 85were eliminated. Copyright © 2011 John Wiley & Sons, Ltd.

Software fault prediction plays a crucial role in enhancing software quality and reliability by enabling early detection of defect-prone modules. This study proposes a machine learning-based predictive framework to identify software faults using historical software metrics and defect data. The objective is to evaluate and compare the performance of various machine learning algorithms—Random Forest, XGBoost, Support Vector Machine (SVM), and Neural Networks—for their effectiveness in fault detection. Public benchmark datasets comprising object-oriented metrics such as CK, Halstead, and McCabe were used after undergoing pre-processing steps including normalization; class balancing using SMOTE, and feature selection through PCA and RFE. Models were trained and tested using stratified k-fold cross-validation, and evaluated using metrics such as Accuracy, Precision, Recall, F1-score, and ROC-AUC. Among the models, XGBoost outperformed others, achieving the highest F1-score and ROC-AUC, followed by Random Forest. Feature importance analysis and SHAP value interpretation provided insights into key predictors contributing to software faults, enhancing the model's transparency and practical usability. The study demonstrates that ensemble learning methods significantly improve the fault prediction process, making them suitable for deployment in real-world software quality assurance workflows. These findings contribute to the ongoing advancement of predictive software engineering and open new avenues for integrating intelligent fault detection systems in continuous integration and DevOps environments.

In recent years, data mining algorithm have been successfully applied in software defects and metrics prediction throughout the software development life cycle (SDLC). Features dependencies and class imbalance are the two main issues in the traditional Bayesian models for defect prediction. Traditional supervised models build high prediction rate with limited single product defect and metric data. Various models have been introduced in the literature to establish a common framework for defect and metric prediction in academia, standards and industry, but none of them are accepted as a common tool in real-time software products. Also, predicting the defects cum metrics in multiple products is one of the common problems in real time associated products. In this proposed model, a new ensemble defect prediction classification model was implemented on multiple associated products to predict metric relationship, along with defects patterns. Experimental results show that proposed model has high true positive rate compare to traditional Bayesian network models.

Prediction of faults in a proposed software is helpful in deciding the amount of effort to be given for software development. We observed that, a good number of authors hypothesized that the performance of fault prediction model depends on the source code metrics which are used as input of the model. Feature selection technique is a process of selecting suitable set of source code metrics which may improve the performance of fault prediction model. In this work, genetic algorithm (GA) has been applied as feature selection technique to select the suitable set of source code metrics. This selected set of source code metrics are used as requisite input data to develop a classifier using five different classification techniques such as logistic regression, extreme learning machine, support vector machine (SVM) with three different kernel functions (linear, polynomial, and radial basis kernel functions) in order to predict the faulty and non-faulty classes. In this study, we propose a cost evaluation framework to perform cost based analysis for evaluating the effectiveness of fault prediction model. We perform experiments on thirty number of Java Open Source projects. From the obtained results, it is observed that the model developed using selected set of source code metrics obtained better result as compared to all metrics. From costs analysis framework, it is observed that the developed fault prediction model is best suitable for software with % of faulty classes less than the threshold value depending on fault identification efficiency (low-46.44%, median-45.37%, and high-36.63%).

Software Defect Prediction (SDP) plays an active area in many research domain of Software Quality of Assurance (SQA). Many existing research studies are based on software traditional metric sets and defect prediction models are built in machine language to detect the bug for limited source code line. Inspired by the above existing system. In this paper, defect prediction is focused on predicting defects in source code. The objective of this thesis is to improve the software quality for accurate defect prediction is source code for file level. So, that it helps the developer to find the bug and fix the issue, to make better use of a resource which reduces the test effort, minimize the cost and improve the quality of software. A new approach is introduced to improve the prediction performance of Bidirectional RNNLM in Deep Neural Network. To build the defect prediction model a defect learner framework is proposed and first it need to build a Neural Language Model. Using this Language Model it helps to learn to deep semantic features in source code and it train & test the model. Based on language model it combined with software traditional metric sets to measure the code and find the defect. The probability of language model and metric set Cross-Entropy with Abstract Syntax Tree (CE-AST) metric is used to evaluate the defect proneness and set as a metric label. For classification the metric label K-NN classifier is used. BPTT algorithm for learning RNN will provide additional improvement, it improves the predictions performance to find the dynamic error.

The bearings of double toothed roller crushers in open pit mines are prone to failure during prolonged operation, resulting in reduced productivity and shortened equipment life. Therefore, a bearing fault prediction model based on asymmetric loss function gradient boosting algorithm and digital twin has been proposed for effective fault prediction of double toothed roller crusher bearings. First, the fault signal is extracted by Wigner-Ville time-frequency distribution and empirical mode decomposition. Then, the asymmetric loss function gradient boosting algorithm is used to model and predict the signal characteristics. Finally, a digital twin model is constructed to visualize bearing fault information by combining sensor data and historical fault data. The experimental results showed that the fault prediction model achieved a precision rate of 98%, a recall rate of 99%, and an F1 value of 98.5% in predicting bearing rolling element failures. Among them, the running time of fault prediction was 1 minute and 49s. The experimental results demonstrate that the proposed fault prediction model can learn complex feature representations from large amounts of data, with good fault prediction performance and efficiency, and has certain practical application value. The research results contribute to improving the accuracy and intelligence level of fault prediction methods, thereby ensuring the stable operation of open-pit mine double toothed roller crushers.

Background Adverse remodeling is significantly associated with heart failure, and strains derived from Cardiovascular magnetic resonance (CMR) serves as a sensitive detector in predicting this pathophysiological process after acute myocardial infarction (MI). However, strategies to identify individuals at risk of adverse remodeling focusing on segmental strains, especially strains of remote myocardium are lacking, we therefore aimed to investigate the predictive value of functional early change in infarcted, viable and remote myocardium after acute MI. Method We derived data from a cohort of 271 STEMI patients with serial CMR collected at one week and six months after. Adverse remodeling is defined as 20% increase in end-diastolic volume after 6 months of MI. According to whether the adverse remodeling occurred, stratified random sampling was used to divide 271 patients into training and testing set in 8:2 ratio. A total of 53 features (38 clinical parameters, 3 CMR traditional parameters and 12 CMR feature-tracking derived strain parameters) have been used to predict the risk of adverse remodeling. To evaluate the predictive accuracy of various machine learning techniques and feature combinations, we calculate the areas under the receiver operating characteristic curve (AUC)s, in which 1) ML with clinical features (ML-clinic), 2) ML with traditional CMR parameters (ML-LGE), and 3) ML with strain data (ML-strain), 4) ML combining clinical and traditional CMR parameters (ML-combined LGE), 5) ML combining clinical and strain data (ML-combined strain) was used as an input of five ML techniques. The SHAP (SHapley Additive exPlanations) value was used to quantity the feature importance. Results Multilayer perceptron (MLP) with residual framework perform best in ML techniques. Using MLP, addition of segmental and global strain measures to clinical data increased the AUC from 0.55 (ML-selected clinic) and 0.42 (ML-strain) to 0.83 (ML-combined strain), which is comparable to 0.73 (LGE combined model). We also found that LS in adjacent regions may serve as a more sensitive detector for adverse remodeling after a SHAP value-based comparison of global versus regional stains. Conclusion ML combined both clinical and strains were found to have high predictive accuracy for adverse remodeling, which could serve as alternative to traditional CMR parameters. Besides, we illustrated a framework for comparing the prognostic value of multiple CMR-feature tracking parameters in the context of clinical parameters.The Study Workflow SHAP feature importance from MLP model

Defect prediction approaches use software metrics and fault data to learn which software properties associate with faults in classes. Existing techniques predict fault-prone classes in the same release (intra) or in a subsequent releases (inter) of a subject software system. We propose an intra-release fault prediction technique, which learns from clusters of related classes, rather than from the entire system. Classes are clustered using structural information and fault prediction models are built using the properties of the classes in each cluster. We present an empirical investigation on data from 29 releases of eight open source software systems from the PROMISE repository, with predictors built using multivariate linear regression. The results indicate that the prediction models built on clusters outperform those built on all the classes of the system.