Search

Your search keyword '"Chen, Wenqi"' showing total 303 results
Start Over Author "Chen, Wenqi" Search Limiters Available in Library Collection
303 results on '"Chen, Wenqi"'

1. Preparation and properties of polyacrylonitrile pre-oxidized fiber/silicone rubber ablation composites

Author

YANG Rui, LIU Leichun, SHI Sheng, XU Haijian, CHEN Wenqi, and CHEN Yi

Subjects
polyacrylonitrile pre-oxidized fiber, silicone rubber, interfacial adhesion, ablation, antioxidation, Materials of engineering and construction. Mechanics of materials, TA401-492
Abstract

Polyacrylonitrile pre-oxidized fiber（PANOF）is an emerging flame retardant material with low cost and high performance，showing a promising application potential as thermal protection materials in the aerospace field.Pretreated PANOF（t-PANOF）was obtained by the process of oil removing with acetone and surface modification using silane coupling agent KH570 in turn.Scanning electron microscopy（SEM），Fourier transform infrared spectroscopy（FT-IR）and X-ray photoelectron spectroscopy（XPS）were employed to figure out the changes of morphology and chemical structure of the fibers before and after pretreatment and their interfacial properties with silicone rubber（SR）.Effects of t-PANOF on the mechanical and thermal performances of SR were further investigated. With the presence of ceramic forming agent（CFA），the ablation responsive behaviors of CFA/t-PANOF/SR ternary composites were researched as well.The results show that the level of interfacial adhesion between fibers and rubber is significantly enhanced contributed from the pretreatment of PANOF，and the tensile strength of the composites is improved accordingly.However，with the increasing content of t-PANOF，the comprehensive mechanical properties of t-PANOF/SR composites show the trend that increase first and then decrease，indicating that the optimal dosage of the fibers is 5 to 10 phr.t-PANOF skeleton helps to not only enhance the dimensional stability of SR in high temperature and oxygen-rich environment，but also improve the ablation performance of CFA/t-PANOF/SR composites.The antioxidation ceramic layer with fiber skeleton as the main structure is formed on the surface of CFA/t-PANOF/SR composites under the high temperature aerobic gas with the heat flux density of 2 MW/m2，the corresponding linear and mass ablation rate are only 0.02 mm/s and 0.02 g/s respectively，which is expected to be utilized as an ablation material to satisfy the demands of the new generation of aerospace vehicles.

Published
2024
Full Text
View/download PDF

3. Representing Affect Information in Word Embeddings

Author

Zhang, Yuhan, Chen, Wenqi, Zhang, Ruihan, and Zhang, Xiajie

Subjects
Computer Science - Computation and Language
Abstract

A growing body of research in natural language processing (NLP) and natural language understanding (NLU) is investigating human-like knowledge learned or encoded in the word embeddings from large language models. This is a step towards understanding what knowledge language models capture that resembles human understanding of language and communication. Here, we investigated whether and how the affect meaning of a word (i.e., valence, arousal, dominance) is encoded in word embeddings pre-trained in large neural networks. We used the human-labeled dataset as the ground truth and performed various correlational and classification tests on four types of word embeddings. The embeddings varied in being static or contextualized, and how much affect specific information was prioritized during the pre-training and fine-tuning phase. Our analyses show that word embedding from the vanilla BERT model did not saliently encode the affect information of English words. Only when the BERT model was fine-tuned on emotion-related tasks or contained extra contextualized information from emotion-rich contexts could the corresponding embedding encode more relevant affect information.

Published
2022

6. DeepAID: Interpreting and Improving Deep Learning-based Anomaly Detection in Security Applications

Author

Han, Dongqi, Wang, Zhiliang, Chen, Wenqi, Zhong, Ying, Wang, Su, Zhang, Han, Yang, Jiahai, Shi, Xingang, and Yin, Xia

Subjects
Computer Science - Cryptography and Security, Computer Science - Machine Learning
Abstract

Unsupervised Deep Learning (DL) techniques have been widely used in various security-related anomaly detection applications, owing to the great promise of being able to detect unforeseen threats and superior performance provided by Deep Neural Networks (DNN). However, the lack of interpretability creates key barriers to the adoption of DL models in practice. Unfortunately, existing interpretation approaches are proposed for supervised learning models and/or non-security domains, which are unadaptable for unsupervised DL models and fail to satisfy special requirements in security domains. In this paper, we propose DeepAID, a general framework aiming to (1) interpret DL-based anomaly detection systems in security domains, and (2) improve the practicality of these systems based on the interpretations. We first propose a novel interpretation method for unsupervised DNNs by formulating and solving well-designed optimization problems with special constraints for security domains. Then, we provide several applications based on our Interpreter as well as a model-based extension Distiller to improve security systems by solving domain-specific problems. We apply DeepAID over three types of security-related anomaly detection systems and extensively evaluate our Interpreter with representative prior works. Experimental results show that DeepAID can provide high-quality interpretations for unsupervised DL models while meeting the special requirements of security domains. We also provide several use cases to show that DeepAID can help security operators to understand model decisions, diagnose system mistakes, give feedback to models, and reduce false positives., Comment: Accepted by 2021 ACM SIGSAC Conference on Computer and Communications Security (CCS '21)

Published
2021
Full Text
View/download PDF

11. Evaluating and Improving Adversarial Robustness of Machine Learning-Based Network Intrusion Detectors

Author

Han, Dongqi, Wang, Zhiliang, Zhong, Ying, Chen, Wenqi, Yang, Jiahai, Lu, Shuqiang, Shi, Xingang, and Yin, Xia

Subjects
Computer Science - Cryptography and Security, Computer Science - Machine Learning, Computer Science - Networking and Internet Architecture
Abstract

Machine learning (ML), especially deep learning (DL) techniques have been increasingly used in anomaly-based network intrusion detection systems (NIDS). However, ML/DL has shown to be extremely vulnerable to adversarial attacks, especially in such security-sensitive systems. Many adversarial attacks have been proposed to evaluate the robustness of ML-based NIDSs. Unfortunately, existing attacks mostly focused on feature-space and/or white-box attacks, which make impractical assumptions in real-world scenarios, leaving the study on practical gray/black-box attacks largely unexplored. To bridge this gap, we conduct the first systematic study of the gray/black-box traffic-space adversarial attacks to evaluate the robustness of ML-based NIDSs. Our work outperforms previous ones in the following aspects: (i) practical-the proposed attack can automatically mutate original traffic with extremely limited knowledge and affordable overhead while preserving its functionality; (ii) generic-the proposed attack is effective for evaluating the robustness of various NIDSs using diverse ML/DL models and non-payload-based features; (iii) explainable-we propose an explanation method for the fragile robustness of ML-based NIDSs. Based on this, we also propose a defense scheme against adversarial attacks to improve system robustness. We extensively evaluate the robustness of various NIDSs using diverse feature sets and ML/DL models. Experimental results show our attack is effective (e.g., >97% evasion rate in half cases for Kitsune, a state-of-the-art NIDS) with affordable execution cost and the proposed defense method can effectively mitigate such attacks (evasion rate is reduced by >50% in most cases)., Comment: This article has been accepted for publication by IEEE JSAC

Published
2020
Full Text
View/download PDF

26. Combined Exposure to High-Cholesterol Diet and PM 2.5 : Brain Injury and Regulatory Mechanism of HIF-1α in ApoE −/− Female Mice.

Author

Chen, Wenqi, Chen, Shanshan, Bai, Lirong, and Li, Ruijin

Abstract

High-cholesterol diet (HCD) and fine particulate matter (PM2.5) are related to stroke. However, little is known about the combined effects of stroke, especially for females. This study investigated the brain injuries in Apolipoprotein E−/− (ApoE−/−) female mice exposed to HCD plus PM2.5 for 6 months. The protein levels of the genes related to stroke and the blood–brain barrier (BBB) in different groups of mice were measured. The molecular regulation mechanisms were explored. The results showed that HCD and PM2.5 co-exposure altered brain–body weight ratio, behavior, brain pathology, and inflammatory markers in mice relative to exposure to HCD or PM2.5 alone. Co-exposure significantly changed the expressions of HIF-1α and the key genes in its signaling pathway in the brains of mice compared to the single exposure. It suggests that the HIF-1α pathway exerts an important regulatory role in brain injury and behavioral abnormality in female mice after 6-month exposure to HCD plus PM2.5, which are potential mechanisms for HCD and PM2.5-triggering stroke in female individuals. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

31. CSES-01 Electron Density Background Characterisation and Preliminary Investigation of Possible Ne Increase before Global Seismicity

Author

Chen, Wenqi, primary, Marchetti, Dedalo, additional, Zhu, Kaiguang, additional, Sabbagh, Dario, additional, Yan, Rui, additional, Zhima, Zeren, additional, Shen, Xuhui, additional, Cheng, Yuqi, additional, Fan, Mengxuan, additional, Wang, Siyu, additional, Wang, Ting, additional, Zhang, Donghua, additional, Zhang, Hanshuo, additional, and Zhang, Yiqun, additional

Published
2023
Full Text
View/download PDF

35. Pollution Characteristics of Different Components of PM 2.5 in Taiyuan during 2017–2020 Wintertime and Their Toxicity Effects on HepG2 Cells.

Author

Bai, Lirong, Zhang, Mei, Chen, Shanshan, Chen, Wenqi, Li, Zhiping, Yue, Jianwei, Dong, Chuan, and Li, Ruijin

Subjects
POLLUTION, LIPID metabolism disorders, AIR pollution, LIPID synthesis, LIPID metabolism
Abstract

Fine particulate matter (PM2.5) is a common pollutant, and its health risk has attracted much attention. Studies have shown that PM2.5 exposure is associated with liver disease. The composition of PM2.5 is complex, and its hepatotoxic effects and lipid metabolism process are not well understood. In this study, we detected the concentrations of PM2.5 and its components (metals and polycyclic aromatic hydrocarbon (PAHs)) in the winter in Taiyuan, Shanxi Province, China, from 2017 to 2020 and then assessed the health risks. We also investigated the effects of different components (whole particles (WP), water-soluble particles (WSP), organic particles (OP)) of PM2.5 on the cytotoxicity and lipid metabolism in human liver cell line (HepG2) after 24 h of treatment. The changes in cytotoxicity indexes (LDH, IL-6, reactive oxygen species (ROS)) and lipids (triglyceride (TG), free fatty acid (FFA)) were measured after 24 h. The mRNA expression of lipid metabolism-related factors (SREBP1, CD36, MTTP) was determined by real-time quantitative RT-qPCR. Finally, the correlation between metals and PAHs with higher PM2.5 content in 4 years and biomarkers was analyzed. The results showed that: (1) The PM2.5 pollution was severe in Taiyuan during winter in 2017 and the subsequent four years. The calculation results of the metal enrichment factor (EF) value and PAHs characteristic ratio of PM2.5 showed that PM2.5 pollution sources differed in different years. (2) Exposure to metals and PAHs in PM2.5 did not cause a non-carcinogenic risk. Metals had no cancer risk, while PAHs inhaled in PM2.5 in 2017 and 2018 had a potential cancer risk. The atmospheric PM2.5 pollution in Taiyuan has had a downward trend, but the PAHs in the PM2.5 of 2017–18, when the pollution is relatively serious, have a potential carcinogenic risk. (3) WP, OP and WSP inhibited cell survival rate from 2017 to 2020, and OP had higher cytotoxicity at the same concentration. (4) WP, OP and WSP increased the levels of LDH, IL-6, TNF-α, ROS, MDA, TG and FFA, and inhibited SOD activity in a dose-effect relationship. The organic components in PM2.5 are more toxic to HepG2 cells and affect the expression of lipid metabolism-related factors at the transcriptional level. (5) The mRNA expressions of factors related to lipid synthesis, uptake, oxidation and output were up-regulated after treatment with PM2.5 and its components, suggesting a lipid metabolism disorder. (6) The biomarkers were associated with certain metals (Zn, Pb, Cu and Cr) and PAHs in PM2.5. These suggested that PM2.5- and PM2.5-bound organic matter caused HepG2 cytotoxicity and affected lipid metabolism. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

45. Construction of High Spatiotemporal Continuity Surface Water Bodies Dataset in the Haihe River Basin

Author

Chen, Wenqi Li, Bo Gao, Huili Gong, and Beibei

Subjects
water body mapping, spatiotemporal correlation model, high spatiotemporal continuity, Haihe River Basin
Abstract

Surface water bodies exhibit high dynamic variability on seasonal and interannual scales, and high spatiotemporal resolution water bodies extent data are crucial for studying surface water bodies’ evolution. Existing surface water bodies datasets are mainly based on optical data acquisition, which has the advantages of long temporal coverage and convenience but is susceptible to cloud contamination, leading to low spatiotemporal continuity. Although microwave remote sensing data are not affected by clouds, early SAR acquisition and short temporal coverage limit its use. Therefore, existing surface water bodies datasets face the problem of insufficient spatiotemporal resolution or low continuity. This research integrates Sentinel-2 optical data and Sentinel-1 Synthetic Aperture Radar (SAR) observations to reconstruct the surface water bodies dataset with a 6-day and 10-meter spatiotemporal resolution. Then, the proposed method introduces a spatiotemporal correlation model and predicts the land cover (water or land) of Sentinel-2 cloudy pixels, which improves the spatiotemporal continuity of the reconstructed surface water bodies dataset further. Based on the proposed method, we construct the Haihe River Water Dataset (HRWD) from 2016 to 2020 with a 6-day and 10-meter spatiotemporal resolution. Compared with the European Commission’s Joint Research Centre’s (JRC’s) Global Surface Water Explorer and Global Surface Water Extent Dataset (GSWED), the HRWD shows a rational accuracy (e.g., the overall accuracy of the HRWD is more than 93%) and a better spatiotemporal continuity, which provide an improved performance in identifying and monitoring surface water bodies in the Haihe River Basin. This indicates that the proposed method can improve the spatiotemporal continuity of surface water body mapping and meet the needs of accurate and long-term quantitative observation of the distribution of large-scale and high spatiotemporal continuity surface water bodies.

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results