Pengwei Li

Pengwei Li

Auto-generated profile, based on publicly available information

Pengwei Li

School of Information EngineeringNanjing Audit UniversityNanjing211815China

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
23 May 2023 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
XXXiuxin Xia
Xiuxin Xia
YSYan Shi
Yan Shi
PLPengwei Li
Pengwei Li
XLXiaosong Liu
Xiaosong Liu
Jingjing LiuJingjing Liu
Jingjing Liu
HMHong Men
Hong Men
At present, the sensory evaluation of food mostly depends on artificial sensory evaluation and machine perception, but artificial sensory evaluation is greatly interfered with by subjective factors, and machine perception is difficult to reflect human feelings. In this article, a frequency band attention network (FBANet) for olfactory electroencephalogram (EEG) was proposed to distinguish the difference in food odor. First, the olfactory EEG evoked experiment was designed to collect the olfactory EEG, and the preprocessing of olfactory EEG, such as frequency division, was completed. Second, the FBANet consisted of frequency band feature mining and frequency band feature self-attention, in which frequency band feature mining can effectively mine multiband features of olfactory EEG with different scales, and frequency band feature self-attention can integrate the extracted multiband features and realize classification. Finally, compared with other advanced models, the performance of the FBANet was evaluated. The results show that FBANet was better than the state-of-the-art techniques. In conclusion, FBANet effectively mined the olfactory EEG data information and distinguished the differences between the eight food odors, which proposed a new idea for food sensory evaluation based on multiband olfactory EEG analysis.

8 September 2020 in World Wide Web
Verified
Journal Article
Closed access
PDF file
JHJiajia Huang
Jiajia Huang
Min PengMin Peng
Min Peng
PLPengwei Li
Pengwei Li
HZHu Zhiwei
Hu Zhiwei
Chao XuChao Xu
Chao Xu

24 March 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
XXXiuxin Xia
Xiuxin Xia
QWQun Wang
Qun Wang
HWHe Wang
He Wang
CLChenrui Liu
Chenrui Liu
PLPengwei Li
Pengwei Li
HMHong Men
Hong Men
The taste electroencephalogram (EEG) evoked by the taste stimulation can reflect different brain patterns and be used in applications such as sensory evaluation of food. However, considering the computational cost and efficiency, EEG data with many channels has to face the critical issue of channel selection. This paper proposed a channel selection method called class activation mapping with attention (CAM-Attention). The CAM-Attention method combined a convolutional neural network with channel and spatial attention (CNN-CSA) model with a gradient-weighted class activation mapping (Grad-CAM) model. The CNN-CSA model exploited key features in EEG data by attention mechanism, and the Grad-CAM model effectively realized the visualization of feature regions. Then, channel selection was effectively implemented based on feature regions. Experimental results showed that the proposed CAM-Attention method achieved an accuracy of 97.85% and an F1-score of 97.74% when the selected channel number was 12, which were only 0.25% and 0.33% lower, respectively, compared to using all channels. This demonstrates that the CAM-Attention method can significantly reduce computational burden while maintaining excellent classification performance. In short, it has excellent recognition performance and provides effective technical support for taste sensory evaluation.

1 December 2018 in IEICE Transactions on Information and Systems
Verified
Journal Article
Open access
PDF file
BCBinlin Cheng
Binlin Cheng
PLPengwei Li
Pengwei Li
Japan's largest platform for academic e-journals: J-STAGE is a full text database for reviewed academic papers published by Japanese societies

1 September 2019 in Chinese Journal of Electronics
Verified
Journal Article
Open access
PDF file
PLPengwei Li
Pengwei Li
JFJianming Fu
Jianming Fu
CXChao Xu
Chao Xu
BCBinlin Cheng
Binlin Cheng
HZHuanguo Zhang
Huanguo Zhang
Security-sensitive operations in Android applications (apps for short) can either be benign or malicious. In this work, we introduce an approach of static program analysis that extracts fisecond-step behavior featuresfl, i.e., what was triggered by the security-sensitive operation, to assist app analysis in differentiating between malicious and benign operations. Firstly, we summarized the characteristics of malicious operations, such as spontaneity, independence, stealthiness and continuity, which can be used to classify the malicious operations and benign ones. Secondly, according to these characteristics, Second step behavior features (SSBFs for short) have been presented, including structural features and semantic features. Thirdly, an analysis prototype named SSdroid has been implemented to automatically extract SSBFs of security-sensitive operations. Finally, experiments on 9285 operations from both benign and malicious apps show that SSBFs are effective and usefulness. Our evaluation results suggest that the second-step behavior can greatly assist in Android malware detection.