We present a software tool-set which combines the theoretical, optimal control view of quantum devices with the practical operation and characterization tasks required for quantum computing. In the same framework, we ...
详细信息
ISBN:
(数字)9798331531195
ISBN:
(纸本)9798331531201
We present a software tool-set which combines the theoretical, optimal control view of quantum devices with the practical operation and characterization tasks required for quantum computing. In the same framework, we perform model-based simulations to create control schemes, calibrate these controls in a closed-loop with the device (or in this demo —by emulating the experimental process) and finally improve the system model through minimization of the mismatch between simulation and experiment, resulting in a digital twin of the device. The model based simulator is implemented using TensorFlow, for numeric efficiency, scalability and to make use of automatic differentiation, which enables gradient-based optimization for arbitrary models and control schemes. Optimizations are carried out with a collection of state-of-the-art algorithms originated in the field of machine learning. All of this comes with a user-friendly Qiskit interface, which allows end-users to easily simulate their quantum circuits on a high-fidelity differentiable physics simulator.
Open source software (OSS) vulnerabilities are crucial for ensuring the security of software systems. Existing methods typically require application developers to patch their own operational environment clients, overl...
详细信息
ISBN:
(数字)9798331520298
ISBN:
(纸本)9798331520304
Open source software (OSS) vulnerabilities are crucial for ensuring the security of software systems. Existing methods typically require application developers to patch their own operational environment clients, overlooking the security responsibilities that should be assumed by the operational environment providers. This often leads to failed patch deployments and affects the stability of the software systems. To address this issue, this paper proposes AIPD, a novel method for completing patch deployment at the manufacturing side. AIPD does not require modification of the OSS source code; instead, it extracts the core APIs involved in the vulnerability fixes and their dependencies from the source version, dynamically intercepts them in the target version's system environment, and replaces them with the patched API for secure updates. Experiments show that AIPD achieves a 91.35% patching accuracy on a dataset of 104 vulnerability patches containing 10 Python packages. In compatibility testing for the patch deployment environment, the pass rate for the relevant APIs reached 97.69%, effectively reducing the compatibility risks of the system environment.
Currently, the use of neural network based solutions in medical decision support systems remains popular. In order to overcome the hidden decision-making process called “black box”, the authors propose to use the “...
详细信息
ISBN:
(数字)9798331518752
ISBN:
(纸本)9798331518769
Currently, the use of neural network based solutions in medical decision support systems remains popular. In order to overcome the hidden decision-making process called “black box”, the authors propose to use the “white box” principle based on the process mining approach. The aim of the work is to develop and implement a process mining based method of electrocardiosignal analysis and arrhythmia detection. Pan-Tompkins algorithm for electrocardiosignal signal analysis and ProM 6.11 framework for software implementation were used. As a results a new intelligent process mining based electrocardiosignal analysis method is proposed. It consists of conversion of digital data of cardiac signal into event log, construction of Petri net and construction of diagnostic decision tree. The original plugin generates the event log in the .xes format using Pan-Tompkins algorithm. ProM 6.11 AlphaMiner plugin generates a Petri net that reflects the relationship of events in the log. The resulting event log allows building a diagnostic decision tree using ProM 6.11 discovery of the process data-flow plugin. The study demonstrates the applicability of the proposed method using normal and arrhythmic cardiac signal records from PhysioNet database. The proposed solution on the base of process mining technology and ProM 6.11 framework allows implementing software for transparent analysis of cardiac signals data for medical decision support systems. The next step is to conduct statistical assessments of the effectiveness of the proposed solution and to expand the number of informative parameters used for analysis.
In order to increase automation, strengthen decision- making, and improve efficiency while reducing costs and maintaining precision, the development of intelligent machinery is becoming more and more dependent on the ...
详细信息
ISBN:
(数字)9798331522667
ISBN:
(纸本)9798331522674
In order to increase automation, strengthen decision- making, and improve efficiency while reducing costs and maintaining precision, the development of intelligent machinery is becoming more and more dependent on the integration of data, embedded software, and modern technologies. Reliability of embedded software becomes essential for guaranteeing high system performance and an improved user experience as technology and hardware merge to build smart devices. This work explores methods to quantify software reliability, which measures the likelihood of residual failures in a system. The objective is to build a statistical relationship between quality parameters and product or process metrics through the analysis of software metrics produced from development data. These indicators aid in evaluating the software's dependability and performance, enabling developers to spot and fix possible problems early on. Using pattern recognition algorithms to pinpoint software's fault-prone areas is a crucial strategy for enhancing software reliability prediction. These algorithms enable developers to focus attention on software areas that have a higher likelihood of failure by identifying patterns in the product's code or structure that suggest increased risk. In summary, these methods focus on enhancing the reliability of intelligent machines by equipping developers with advanced tools to reduce defects, improve dependability, and ensure that smart systems perform optimally. This approach aligns with modern trends in using machine learning and data-driven models to tackle complex technological challenges.
Early defect detection and prediction are essential in software engineering to reduce costs and improve quality. This study presents an AI-driven approach for fault detection and prediction employing ML techniques on ...
详细信息
ISBN:
(数字)9798331529246
ISBN:
(纸本)9798331529253
Early defect detection and prediction are essential in software engineering to reduce costs and improve quality. This study presents an AI-driven approach for fault detection and prediction employing ML techniques on the CM1 dataset from NASA’s software metrics program. The data preprocessing steps include normalisation, feature selection, and the use of the ADASYN algorithm to enhance minority class detection. Faults are predicted using ML classifiers: Extra Trees (ET), Support Vector Machine (SVM), KNN, RF, MLP, and NB. F1 score, recall, accuracy, and precision are a metrics utilised to assess a models. When it comes to defect prediction, the Extra Trees Classifier (ETC) model outperforms all of the others with an accuracy91.51%, precision91.76%, and recall99%. Although the Naive Bayes model shows lower accuracy (77.61%) and efficiency, it offers a simpler approach for fault detection. Future research in this area should be guided by these findings, which highlight the promise of AI-driven approaches to enhance frameworks for software defect prediction and quality assurance.
Defect prediction plays a crucial role for any software system across various domains, as its failure may cause unavoidable and undeniable scenarios. For reliable software, defect-free is considered as one of the most...
详细信息
ISBN:
(数字)9798331506940
ISBN:
(纸本)9798331506957
Defect prediction plays a crucial role for any software system across various domains, as its failure may cause unavoidable and undeniable scenarios. For reliable software, defect-free is considered as one of the most important criteria. This research aims to design a hybrid swarm-optimized machine learning software defect prediction (HSoMLSDP) framework to predict software defects. We strive to do this by designing a swarm-optimized machine learning defect prediction (SoMLDP) model within the HSoMLSDP framework. In pursuit of enhancing the defect prediction accuracy of the SoMLDP model, this paper introduces a hybrid swarm optimization algorithm (SOA) referred to as the gravitational force Lévy flight grasshopper optimization algorithm-artificial bee colony (GFLFGOA-ABC) algorithm. By combining the enhanced exploration feature of the gravitational force Lévy flight grasshopper optimization algorithm (GFLFGOA) with the robust exploitation capacity of the artificial bee colony (ABC), the GFLFGOA-ABC algorithm is proposed. Prior to validating the HSoMLSDP framework, the LFGFGOA-ABC algorithm’s performance is first confirmed by experiments on 6 benchmark functions (BFs) to assess its mean and convergence rate. Following BF verification, the second experiment tunes the hyperparameters of ML models (ANN, GB, XGB) to improve the defect accuracy of the SoMLDP model. As an enhancement of accuracy justifies the correctness of the SoMLDP model, thus validating the HSoMLSDP framework.
The field of Artificial Intelligence (AI) has been witnessing a huge demand in the field of research, tools development, and applications of deployment. There are multiple software companies which are shifting their f...
详细信息
ISBN:
(数字)9798331527495
ISBN:
(纸本)9798331527501
The field of Artificial Intelligence (AI) has been witnessing a huge demand in the field of research, tools development, and applications of deployment. There are multiple software companies which are shifting their focus to developing AI systems, many are also deploying the AI paradigms to their current system. Correspondingly, the academic research community has also used AI paradigms to solve the traditional software engineering problems. AI has become very useful to the software engineering community. The paper also uses a survey of the most commonplace methods of using AI into the software Engineering, that covers the methods for SE phases such as Requirements, Design, Development, Testing, Release, and Maintenance. The paper also aims in answering to the sufficient intelligence in the SE lifecycle, this evaluates the overlapping of the SE and AI domains, and challenge the current conventional wisdom of the state-of-the-art.
software Bug Prediction (SBP) is essential to control overall software quality. As systems become complex, faults increase, and early prediction is needed to enhance quality and reduce expenses. This paper provides a ...
详细信息
ISBN:
(数字)9798331530952
ISBN:
(纸本)9798331530969
software Bug Prediction (SBP) is essential to control overall software quality. As systems become complex, faults increase, and early prediction is needed to enhance quality and reduce expenses. This paper provides a technical study of thirteen Machine Learning (ML) techniques for predicting software bugs by utilizing NASA's Promise Repository datasets. The study examines the effectiveness of these classification techniques focusing on evaluation metrics such as Precision, Recall, F-measure, Accuracy, Error Rate, FPR, FNR, ROC, and MCC. The results indicate that Random Forest (RF) achieved the highest accuracy on PC1 (94.96%) and CM1 (96.60%), while Multilayer Perceptron (MLP) scored 85.6% on KC1, and Logistic Regression (LR) got 86.25% on KC2. Regarding error rates, RF had the lowest on PC1 (0.050), LR on KC2 (0.137). For FPR, SVM, MLP, LR, RBF, ANN, and AdaBoost excelled on PC1 (0.00). For FNR, Decision Tree (DT) performed best on PC1 (0.611), KC2 (0.518), and LDA on CM1 (0.500).
Artificial Intelligence (AI) has revolutionized and transformed the landscape of software development and testing by introducing new efficiencies and capabilities through advancements like Generative AI (GenAI) and La...
详细信息
ISBN:
(数字)9798331507695
ISBN:
(纸本)9798331507701
Artificial Intelligence (AI) has revolutionized and transformed the landscape of software development and testing by introducing new efficiencies and capabilities through advancements like Generative AI (GenAI) and Large Language Models (LLMs). While these technologies bring major benefits in terms of productivity, personalization, and innovation, they also raise critical ethical challenges, such as biases, lack of transparency, data privacy concerns, and potential negative societal impacts. This paper examines the ethical considerations involved in developing such advanced AI systems as well using AI systems within software development and testing. It explores existing ethical frameworks and principles provided by leading organizations, emphasizing core concepts like human-centered design, accountability, transparency, fairness, and privacy. Practical strategies for integrating ethical practices throughout the AI development lifecycle are discussed, with a strong emphasis on the need for continuous ethical evaluation. The paper explores the ethical landscape of AI in software development, addressing challenges like algorithmic bias, data security, and broader societal impacts. Real-world case studies presented in the paper demonstrate the consequences of neglecting ethical considerations. Looking forward, the paper suggests future directions, including the development of unified ethical standards, collaborative ethical auditing, regulatory advancements, and higher societal engagement.
Focus of the paper is on predicting software defects (SD) based on Machine learning (ML) techniques which is a challenging research area because of unbalanced nature of data sets. In general, a set of design metrics i...
详细信息
ISBN:
(数字)9798331522100
ISBN:
(纸本)9798331522117
Focus of the paper is on predicting software defects (SD) based on Machine learning (ML) techniques which is a challenging research area because of unbalanced nature of data sets. In general, a set of design metrics is used for fault prediction and for identifying the fault-proneness in software and recent studies shows that ML techniques is being applied for defect prediction. However, some ML techniques cannot produce the needed results when dealing with unbalanced dataset and the results produced are not certainly inferred by the developers by observing these factors, we propose a unique approach for fault prediction which is based on feature selection technique which improves the overall performance by using attribute selection when predicting defects. The ML concentrates on the algorithms entirely centred on statistical methods and data mining techniques for classifying and predicting the defects and these statistical methods followed are quite similar to regression based methods which we used earlier to the ML. When more data is available, ML algorithms behave as dynamic algorithms to improve their performance significantly. The DT ML technique is providing good accuracy compared to other NB and SVM technique. In this model is simulated python language and calculated simulation parameter i.e. precision, recall and accuracy.
暂无评论