The growing threat of malware,particularly in the Portable Executable(PE)format,demands more effective methods for detection and classification.Machine learning-based approaches exhibit their potential but often negle...The growing threat of malware,particularly in the Portable Executable(PE)format,demands more effective methods for detection and classification.Machine learning-based approaches exhibit their potential but often neglect semantic segmentation of malware files that can improve classification performance.This research applies deep learning to malware detection,using Convolutional Neural Network(CNN)architectures adapted to work with semantically extracted data to classify malware into malware families.Starting from the Malconv model,this study introduces modifications to adapt it to multi-classification tasks and improve its performance.It proposes a new innovative method that focuses on byte extraction from Portable Executable(PE)malware files based on their semantic location,resulting in higher accuracy in malware classification than traditional methods using full-byte sequences.This novel approach evaluates the importance of each semantic segment to improve classification accuracy.The results revealed that the header segment of PE files provides the most valuable information for malware identification,outperforming the other sections,and achieving an average classification accuracy of 99.54%.The above reaffirms the effectiveness of the semantic segmentation approach and highlights the critical role header data plays in improving malware detection and classification accuracy.展开更多
Using a simple damped slab model, it was possible to show that a local wind induced 88% （15 of 17） of the near-inertial oscillations （NIO） observed in the mixed layer near the east coast of Korea from 1999 to 2004...Using a simple damped slab model, it was possible to show that a local wind induced 88% （15 of 17） of the near-inertial oscillations （NIO） observed in the mixed layer near the east coast of Korea from 1999 to 2004. The model, however, overestimated the energy level in about two-thirds of the simulated cases, because the slab model was forced with winds whose characteristic period was shorter than the damping time scale of the model at 1.5 d. At the observation site, due to typhoons and orographic effects, high-frequency wind forcing is quite common, as is the overestimation of the energy level in the slab model results. In short, a simple slab model with a damping time-scale of about 1.5 d would be enough to show that the local wind was the main energy source of the near-inertial energy in this area, but the model could not be used to accurately estimate the amount of the work done by the wind to the mixed layer.展开更多
文摘The growing threat of malware,particularly in the Portable Executable(PE)format,demands more effective methods for detection and classification.Machine learning-based approaches exhibit their potential but often neglect semantic segmentation of malware files that can improve classification performance.This research applies deep learning to malware detection,using Convolutional Neural Network(CNN)architectures adapted to work with semantically extracted data to classify malware into malware families.Starting from the Malconv model,this study introduces modifications to adapt it to multi-classification tasks and improve its performance.It proposes a new innovative method that focuses on byte extraction from Portable Executable(PE)malware files based on their semantic location,resulting in higher accuracy in malware classification than traditional methods using full-byte sequences.This novel approach evaluates the importance of each semantic segment to improve classification accuracy.The results revealed that the header segment of PE files provides the most valuable information for malware identification,outperforming the other sections,and achieving an average classification accuracy of 99.54%.The above reaffirms the effectiveness of the semantic segmentation approach and highlights the critical role header data plays in improving malware detection and classification accuracy.
基金The Agency for Defense Development under contract Nos 609-83-01532,UD000008BD and UD970022ADKorea Institute of Science and Technology Evaluation and Planning under contract No.2000-N-NL-01-C-012+3 种基金the Korean Ministry of Environments under contract No.121-041-033Korean Ministry of Education under the BK21 ProgramKorea Research Foundation under the Free-doctoral scholars programKorean Ministry of Oceans and Fisheries under the projects"Development of Korea Operational Oceanographic System(KOOS)"and"Development of Technology for CO2Marine Geological Storage"
文摘Using a simple damped slab model, it was possible to show that a local wind induced 88% （15 of 17） of the near-inertial oscillations （NIO） observed in the mixed layer near the east coast of Korea from 1999 to 2004. The model, however, overestimated the energy level in about two-thirds of the simulated cases, because the slab model was forced with winds whose characteristic period was shorter than the damping time scale of the model at 1.5 d. At the observation site, due to typhoons and orographic effects, high-frequency wind forcing is quite common, as is the overestimation of the energy level in the slab model results. In short, a simple slab model with a damping time-scale of about 1.5 d would be enough to show that the local wind was the main energy source of the near-inertial energy in this area, but the model could not be used to accurately estimate the amount of the work done by the wind to the mixed layer.
