Automated feature learning is now possible in various fields, including healthcare, image recognition, and, more recently, feature extraction and classification of simple and complex human activity detection in mobile...
详细信息
As techniques and tools for synthetic media and Deepfakes continue to advance, it is increasingly clear that video, audio and images can no longer be relied upon as truthful recordings of reality. Every digital commun...
详细信息
Floods occur when water overflows onto normally dry land and are a destructive natural disaster. In recent times, deep learning models have demonstrated their remarkable capabilities in identifying objects and classif...
详细信息
the digital world today, with the rapidly evolving threats against information systems, has required quite sophisticated mechanisms for their detection. This work evaluates the efficiency of using machine learning alg...
详细信息
Home automation is a technology that is often viewed as a luxury and not a necessity. The means of achieving home automation is always considered expensive and inconvenient in the long run. Due to this there is no suc...
详细信息
In the last decade, there has been a significant upsurge in the demand for artificial intelligence. This remarkable growth can be attributed to the advancements in machine and deep learning techniques, coupled with th...
详细信息
The use of connected devices in the industry represents a necessity and, at the same time, a challenge. Building a network of interconnected industry assets can improve performance and scale but can lead to dangerous ...
详细信息
We present an optimal power allocation method to improve the energy efficiency (EE) of visible light communications (VLC) based on non-orthogonal multiple access (NOMA). The EE optimization model is established under ...
详细信息
ISBN:
(纸本)9798350374377
We present an optimal power allocation method to improve the energy efficiency (EE) of visible light communications (VLC) based on non-orthogonal multiple access (NOMA). The EE optimization model is established under the constraints of minimum data rate, maximum transmission power, successive interference cancellation (SIC) and direct current (DC) bias of the multiplexing light-emitting diodes (LEDs). Then an iterative power allocation algorithm is proposed to exploit the solution of the EE optimization problem. The simulation results show that the proposed power allocation algorithm can significantly improve the system EE.
Over the past few decades, classification has consistently posed a significant computational challenge. This study presents an innovative ensemble classification model designed for data classification, drawing inspira...
详细信息
Cloud application security initiates with the analysis of security requirements in DevOps. This involves gathering, managing, and tracking requirements within integrated issue-tracking systems found in repositories li...
详细信息
ISBN:
(纸本)9798350339826
Cloud application security initiates with the analysis of security requirements in DevOps. This involves gathering, managing, and tracking requirements within integrated issue-tracking systems found in repositories like GitHub. DevOps offers advantages in cloud app development, such as accelerated deployment, improved collaboration, and enhanced reliability. In DevOps, while many security verification tools are automated, security requirements analysis often relies on manual procedures. User feedback plays a pivotal role in shaping cloud application requirements, and the industry actively seeks automation solutions to expedite development. Prior research has demonstrated the limited performance of conventional NLP models trained on established datasets, such as PROMISE, when employed in the context of GitHub Issues. Recent studies have explored the integration of deep learning, particularly leveraging modern large language models and transfer learning architectures, to address requirements engineering challenges. However, a significant issue persists - the transferability of these models. While these models excel when applied to datasets similar to those they were trained on, their performance often drastically falls when dealing with external domains. In our paper, we introduce an automated method for classifying requirements within issue trackers. This method utilizes a novel dataset comprising 12,000 security and non-security issues collected from open GitHub repositories. We employed a SmallBERT-based model for training and conducted a series of experiments. Our research reaffirms the challenge related to the transferability of NLP models. Simultaneously, our model yields highly promising results when applied to GitHub Issues, even in challenging scenarios involving issues from projects that were not part of the training dataset and structured requirements texts from the PROMISE dataset. In summary, our approach significantly contributes to enhancing DevOps prac
暂无评论