Filters








72 Hits in 6.2 sec

An Automated Approach for the Prediction of the Severity Level of Bug Reports Using GPT-2

Mohsin kamal, Sikandar Ali, Anam Nasir, Ali Samad, Samad Basser, Azeem Irshad, Muhammad Arif
2022 Security and Communication Networks  
For bug severity report prediction, numerous automated strategies have been proposed in addition to manual ones.  ...  We use GPT-2's features (limiting overfitting and supplying sequential predictors rather than weight computation) to develop a new approach for predicting the severity level of bug reports in this study  ...  [19] presented a positioning-based future determination technique for execution enhancement of bug seriousness forecast as it saved investment. e exploration work tested the technique on web crawlers  ... 
doi:10.1155/2022/2892401 fatcat:azqruegu7nekrbclxsi4octeti

DRAST – A Deep Learning and AST Based Approach for Bug Localization [article]

Shubham Sangle, Sandeep Muvva, Sridhar Chimalakonda, Karthikeyan Ponnalagu, Vijendran Gopalan Venkoparao
2020 arXiv   pre-print
Conclusions: This paper presents a novel bug localization approach that works on C and Java projects and a bug localization C dataset along with a novel source code representation.  ...  We also tested DRAST on Tomcat and AspectJ, projects from benchmark dataset with better results at accuracy@1, MAP and MRR when compared with state-of-the-art.  ...  In the C project, each C file is divided into code blocks/vectors based on the function definition, whereas in a Java project, each java file is divided into code blocks/vectors based on function and class  ... 
arXiv:2011.03449v1 fatcat:lmhs4qfvird6xjj4obvojun3am

Maintenance & Extension of Scikit-learn: Machine Learning in Python [article]

Thomas Fan
2021 figshare.com  
In addition, with DataFrame propagation, it will be more accessible to integrate scikit-learn with workflows based around DataFrames.  ...  This backlog includes essential bug reports, bug fixes, performance regression reports, and feature contributions.  ... 
doi:10.6084/m9.figshare.16528449.v1 fatcat:emnynyh2cvcn3hrrtxemwxnz4a

An Assessment of Eclipse Bugs' Priority and Severity Prediction Using Machine Learning

Mohammed Q. Shatnawi, Batool Alazzam
2022 International Journal of Communication Networks and Information Security  
The aim is to build a tool that automatically classifies software bugs according to the severity and priority of the bugs and makes predictions based on the most representative features and bug report  ...  Therefore, in this paper, we present a novel automatic assessment tool using Machine Learning algorithms, for assessing bugs' reports based on several features such as hardware, product, assignee, OS,  ...  Bug severity is divided into seven types, which are major, blocker, critical, normal, enhancement, minor, and trivial [5] . • Major: is a major loss of function. • Blocker: blocks the work of testing  ... 
doi:10.17762/ijcnis.v14i1.5266 fatcat:xmf4khnclzafrnpd4xa5p6yt6i

Extracting Arguments based on User Decisions in App Reviews

Anang Kunaefi, Masayoshi Aritsugi
2021 IEEE Access  
Users abandoned apps based on common arguments, such as the existence of bugs, errors, and crashes.  ...  However, XGBoost outperformed other classifiers based on the F1-score and runtime assessment.  ... 
doi:10.1109/access.2021.3067000 fatcat:xsz6azwbvjf4falrl36ymbdz7u

Software Defect Prediction using Ensemble Learning: A Systematic Literature Review

Faseeha Matloob, Taher M. Ghazal, Nasser Taleb, Shabib Aftab, Munir Ahmad, Muhammad Adnan Khan, Sagheer Abbas, Tariq Rahim Soomro
2021 IEEE Access  
In [60] , ELBlocker employed an ensemble of multiple classifiers to predict the likelihood of a blocking bug.  ...  Using a random forest, these classifiers were combined to compute the likelihood score for bugs to be blocking bugs.  ... 
doi:10.1109/access.2021.3095559 fatcat:72divlxlbjdirpmpotdeyxndi4

Test case prioritization using test case diversification and fault-proneness estimations [article]

Mostafa Mahdieh, Seyed-Hassan Mirian-Hosseinabadi, Mohsen Mahdieh
2021 arXiv   pre-print
The diversification of test cases is preserved by incorporating fault-proneness on a clustering-based approach scheme.  ...  To evaluate this approach we study its performance on real-world open-source projects. Method: The bug history is used to estimate the fault-proneness of source code areas.  ...  However, the clustering algorithms are completely language-neutral and the defect prediction procedure is mostly based on language-independent features.  ... 
arXiv:2106.10524v2 fatcat:yrih7ozqzfc7lhkjtomrsxmoou

Automated Identification of Security Discussions in Microservices Systems: Industrial Surveys and Experiments [article]

Ali Rezaei Nasab, Mojtaba Shahin, Peng Liang, Mohammad Ehsan Basiri, Seyed Ali Hoseyni Raviz, Hourieh Khalajzadeh, Muhammad Waseem, Amine Naseri
2021 arXiv   pre-print
We applied these models on a manually constructed dataset consisting of 4,813 security discussions and 12,464 non-security discussions.  ...  This work is partially sponsored by the National Key R&D Program of China with Grant No. 2018YFB1402800.  ...  ML models include Random Forest, Decision Tree, XGBoost, and SVM-LR algorithms based on three text feature selection techniques: BoW, TF-IDF, and GloVe.  ... 
arXiv:2107.10059v1 fatcat:qmxvint675ax3crc7ouglmeoai

DeepCVA: Automated Commit-level Vulnerability Assessment with Deep Multi-task Learning [article]

Triet H. M. Le, David Hin, Roland Croft, M. Ali Babar
2021 arXiv   pre-print
We propose a novel Deep multi-task learning model, DeepCVA, to automate seven Commit-level Vulnerability Assessment tasks simultaneously based on Common Vulnerability Scoring System (CVSS) metrics.  ...  We show that DeepCVA is the best-performing model with 38% to 59.8% higher Matthews Correlation Coefficient than many supervised and unsupervised baseline models.  ...  This work was supported with supercomputing resources provided by the Phoenix HPC service at the University of Adelaide.  ... 
arXiv:2108.08041v1 fatcat:klz7v2uuxzbkjmsfuyy3py5hea

Anomaly Detection in Blockchain Networks: A Comprehensive Survey [article]

Muneeb Ul Hassan, Mubashir Husain Rehmani, Jinjun Chen
2022 arXiv   pre-print
For this, we first discuss how anomaly detection can aid in ensuring security of blockchain based applications.  ...  Over the past decade, blockchain technology has attracted a huge attention from both industry and academia because it can be integrated with a large number of everyday applications of modern information  ...  Another critical work to enhance bug prediction accuracy for smart contracts has been carried out by Kim et al. in [33] .  ... 
arXiv:2112.06089v3 fatcat:hiwz3s5hvzf3bbqszfdzaqpifm

Performance analysis of machine learning models for intrusion detection system using Gini Impurity-based Weighted Random Forest (GIWRF) feature selection technique

Raisa Abedin Disha, Sajjad Waheed
2022 Cybersecurity  
As the performance of IDS deteriorates with a high dimensional feature vector, an optimum set of features was selected through a Gini Impurity-based Weighted Random Forest (GIWRF) model as the embedded  ...  Based upon the importance score, 20 features were selected from UNSW-NB 15 and 10 features from the Network TON_IoT dataset.  ...  Based on the analysis performed in A block, IDS can be classified into Signature-based IDS (misuse-based) and Anomaly-based IDS.  ... 
doi:10.1186/s42400-021-00103-8 fatcat:kyr5omri2rb7dd4ydsgsrmb6pm

A Survey of DeFi Security: Challenges and Opportunities [article]

Wenkai Li, Jiuyang Bu, Xiaoqi Li, Hongli Peng, Yuanzheng Niu, Yuqing Zhang
2022 arXiv   pre-print
Then we classify and analyze real-world DeFi attacks based on the principles that correlate to the vulnerabilities.  ...  DeFi, or Decentralized Finance, is based on a distributed ledger called blockchain technology. Using blockchain, DeFi may customize the execution of predetermined operations between parties.  ...  Then, for vulnerability identification, the XGBoost model (Chen and Guestrin, 2016) , which is a gradient model based on the decision tree, was built using the features.  ... 
arXiv:2206.11821v2 fatcat:5z4ew5bbpngarabdocenndp4jm

Towards an Improved Understanding of Software Vulnerability Assessment Using Data-Driven Approaches [article]

Triet H. M. Le
2022 arXiv   pre-print
The key contributions include a systematisation of knowledge, along with a suite of novel data-driven techniques and practical recommendations for researchers and practitioners in the area.  ...  [132] compared the performance of ML-based regressors (e.g., XGBoost [78] and Linear regression) and DL-based ones (e.g., CNN [120] and LSTM [138] ) for predicting the severity score of CVSS version  ...  For the prediction layers, we raised 8.8% and 24.4% MCC of DeepCVA with Taskspecific blocks and Multi-task learning, respectively.  ... 
arXiv:2207.11708v1 fatcat:q5nhlhboyfc2tipexhnqzmz77a

Automating Intention Mining

Qiao Huang, Xin Xia, David Lo, Gail C. Murphy
2018 IEEE Transactions on Software Engineering  
Based on this manual effort, we refined the previous categories. We assess Di Sorbo et al.'s patterns on this dataset, finding that the accuracy rate achieved is low (0.31).  ...  • Qiao Huang is with the College  ...  A feature request is also known as a request for enhancement, often shortened as RFE. [29] , where they applied text mining to classify issue reports as bug reports or feature requests.  ... 
doi:10.1109/tse.2018.2876340 fatcat:jhhuou74mndk3n5b63rcotvae4

Fusing Feature Engineering and Deep Learning: A Case Study for Malware Classification [article]

Daniel Gibert, Carles Mateu, Jordi Planes, Quan Le
2022 arXiv   pre-print
While traditional feature-based approaches rely on the manual design of hand-crafted features based on experts knowledge of the domain, deep learning approaches replace the manual feature engineering process  ...  In particular, our approach relies on deep learning to extract N-gram like features from the assembly language instructions and the bytes of malware, and texture patterns and shapelet-based features from  ...  Traditionally, anti-malware engines relied on signature-based and heuristic-based methods to detect and block malware before they performed any damage.  ... 
arXiv:2206.05735v1 fatcat:6sq6t2c6rrcojdobwpdu6whjzm
« Previous Showing results 1 — 15 out of 72 results