Applying Machine Learning Techniques to Predict the Maintainability of Open-Source Software

Gopal, Madhwaraj Kango; Amirthavalli, M.

doi:10.35940/ijeat.e1045.0785s319

Cited by 22 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…with the agile methodologies and its implementation in software engineering projects, which they reduced with the help of AI based machine learning simulator of MATLAB. In 2019, Gopal and Amirthavalli [37] performed an empirical analysis on case study of open-source software, with an objective to predict relationship between maintainability and metrics, and observed that Instability metric is most influencing amongst all. In 2019, Kumar and Singh proposed [38] a threshold based segmentation using MATLAB tool to automate performance evaluation of students and evaluated that ML based procedure performs more accurately.…”

Section: Empirical Discussionmentioning

confidence: 99%

Machine Learning: A Software Process Reengineering in Software Development Organization

Bhavsar¹

2019

IJEAT

View full text Add to dashboard Cite

BPR (Business Process Re-engineering) is an organizational mechanism that improves the organizational ability in responding to the challenges of qualitative result by change management and improvement in software engineering processes, productivity, product quality and competitive advantage. BPR inherits, explores and implements the building of process change, to incorporate enhancements to the essential considerations and protocols of (SEM) Software Engineering Management. Machine Learning (ML) can be the key aspect for BPR in software development organization. The goal of this research study is raising the conceptual vision about integration of automation technology like ML and its life cycle development within Software Development Life Cycle (SDLC) of the software product and highlights benefits and drawbacks ML techniques in SPM (Software Project Management), and how to implement ML in standard SEM practices. We have attempted the introduction of machine learning in SEM to determine specific performance and tasks reuse using empirical analysis and discussion on implementation of ML algorithms. The empirical study of software technologies includes control structure of an autonomous software application. In current era, ML imparts consistently promising accuracy in some SEM fields. The goal of this paper is an empirical and analytical study and literature review to propose desired level of quality software, through the comparative evaluation of existing processes and their respective support for Software Quality Engineering (SQE).

show abstract

Section: Empirical Discussionmentioning

confidence: 99%

Machine Learning: A Software Process Reengineering in Software Development Organization

Bhavsar¹

2019

IJEAT

View full text Add to dashboard Cite

show abstract

“…Several machine learning algorithms are being used today due to the benefits they offer like prediction, analysis and model building etc… Madhwaraj & Amirthavalli [3] used machine learning techniques to predict the maintainability of open source software. Viswanath et al [4] performed a case study to predict the number of deaths due to dengue disease.…”

Section: Literature Reviewmentioning

confidence: 99%

Modeling Traffic Congestion using Graph Convolutional Networks

Gopal,

Asha,

Prasad

et al. 2023

Advances in Computer Science Research

View full text Add to dashboard Cite

Congestion in a network can cause data packet losses, delays, and reduced network performance. To prevent congestion, network engineers must be able to accurately predict and manage network traffic. In this research paper, we explore the use of Graph Convolutional Networks (GCN) for predicting congestion in a network. GCN is a type of deep learning algorithm that can analyze complex network structures to predict the behavior of nodes in a network. With GCN, predicting congestion in a network, identification of potential congested areas becomes a reality. Proactive measures to prevent congestion from occurring is also been attempted in this research work. The results of our experiments demonstrate that GCN outperforms other conventional machine learning techniques in predicting network congestion with high accuracy and precision using ReLU6, which was the most suitable activation function for implementing the model. This research also demonstrates the potential of using deep learning algorithms such as GCN to improve network management and optimize network performance.

show abstract

“…LITERATURE REVIEW Today, a few machine learning algorithms are employed because of the advantages they provide, including prediction, analysis, model development, etc. Machine learning approaches were utilized by Madhwaraj & Amirthavalli [3] to forecast the maintainability of free software. To estimate the number of deaths brought on by dengue fever, Viswanath et al [1] conducted a case study.…”

Section: IImentioning

confidence: 99%

Tool to Detect Fake Accounts in Twitter

Gopal,

Asha,

Saju

et al. 2023

Advances in Computer Science Research

View full text Add to dashboard Cite

On the internet social media platforms are growing in popularity, and that popularity has raised increasing questions about security and privacy. Users of social networks face serious security risks due to fake and copied profiles. One major issue is the cloning of user profiles, when duplicate profiles are created using stolen user data and then utilized to harm the original profile owner. Threats like phishing, stalking, spamming, and others are also used to accomplish a variety of goals. A fake profile is one that is made on a social networking site using the name of an organization or person that does not exist and participates in destructive activities. In this study, a new tool is created that uses machine learning methods to verify user identification. People who utilize fake accounts may be recognized as having fake profiles in one of three ways: the number of abuse reports, daily comments, or rejected friend requests. Data from Twitter was used in a case study. The Random Forest algorithm and the Support vector machine approach offered a greater projected accuracy when detecting whether a user was a fraudulent or genuine user compared to other machine learning methods.

show abstract

Applying Machine Learning Techniques to Predict the Maintainability of Open-Source Software

Cited by 22 publications

References 11 publications

Machine Learning: A Software Process Reengineering in Software Development Organization

Machine Learning: A Software Process Reengineering in Software Development Organization

Modeling Traffic Congestion using Graph Convolutional Networks

Tool to Detect Fake Accounts in Twitter

Contact Info

Product

Resources

About