检索结果-内蒙古大学图书馆

Enhancing Offensive Language Detection with Data Augmentation and Knowledge Distillation

Research 2024年第2期6卷 29-40页

作者： Jiawen Deng Zhuang Chen Hao Sun Zhexin Zhang Jincenzi Wu Satoshi Nakagawa Fuji Ren Minlie Huang The CoAl group DCSTInstitute for Artificial InteligenceState Key Lab of Intelligent Technology and SystemsBeijing National Research Center for Information Science and TechnologyTsinghua UniversityBeijing 100084China Graduate School of Information Science&Technology The University of Tokyo.Tokyo 1138654Japan School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengduSichuan611731China

Offensive language detection has received important attention and plays a crucial role in promoting healthy communication on social platforms,as well as promoting the safe deployment of large language *** data is the basis for developing detectors;however,the available offense-related dataset in Chinese is severely limited in terms of data scale and coverage when compared to English *** significantly affects the accuracy of Chinese offensive language detectors in practical applications,especially when dealing with hard cases or out-of-domain *** alleviate the limitations posed by available datasets,we introduce AugCOLD(Augmented Chinese Offensive Language Dataset),a large-scale unsupervised dataset containing 1 million samples gathered by data crawling and model ***,we employ a multiteacher distillation framework to enhance detection performance with unsupervised *** is,we build multiple teachers with publicly accessible datasets and use them to assign soft labels to *** soft labels serve as a bridge for knowledge to be distilled from both AugCOLD and multiteacher to the student network,i.e.,the final offensive *** conduct experiments on multiple public test sets and our well-designed hard tests,demonstrating that our proposal can effectively improve the generalization and robustness of the offensive language detector.

关键词： COLD dealing limitations

来源：评论

学校读者我要写书评

暂无评论

引用

15th International Conference on Information and Communication technology Convergence, ICTC 2024

作者： Fuji, Takumi Natsume, Kiyohisa Graduate School of Life Science and Systems Engineering Kyushu Institute of Technology Kikyushu City Fukuoka Japan

ISBN: (纸本)9798350364637

In future, humans live with autonomous robot, and cooperate with it. In this study an autonomous robot and a human were asked to cooperate in an exploration task to reach a goal in a large-scale maze. While they were doing a task, electroencephalogram (EEG) was measured from the human. The human was asked to choose the direction of travel at intersections, and 30% of the choose, the robot traveled in a different direction from that specified by the human. Error-related potentials (ErrPs) occurred in the human brain at the time, and that was consistent with that reported in the previous studies. Time-frequency analysis of EEG revealed an increase in θ-frequency power and a decrease in β-frequency power. Some participants felt 'bored' during the task. As if to reflect this, in the second half of the task, the amplitude of ErrPs decreased and θ-frequency power also decreased. When the ErrPs of the first and second halves were discriminated using a convolutional neural network, the accuracy rate of the correct responses decreased in the second half compared to the first half. These results suggest that ErrP changes with boredom, and there is a possibility to detect the boredom using EEG. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

AFuNet: an attention-based fusion network to classify texts in a resource-constrained language

引用

Neural Computing and Applications 2025年第9期37卷 6725-6748页

作者： Hossain, Md. Rajib Hoque, Mohammed Moshiul Dewan, M. Ali Akber Hoque, Enamul Siddique, Nazmul Department of Computer Science & Engineering Chittagong University of Engineering & Technology Chittagong4349 Bangladesh School of Computing and Information Systems Faculty of Science and Technology Athabasca University AthabascaABT9S 3A3 Canada School of Information Technology York University TorontoONM3J 1P3 Canada School of Computing Engineering and Intelligent Systems Ulster University Coleraine United Kingdom

In the era of widespread Internet use and extensive social media interaction, the digital realm is accumulating vast amounts of unstructured text data. This unstructured data often contain undesirable information, necessitating time-consuming manual classification efforts. An intelligent text classification system capable of automatically categorizing digitized texts based on semantic meaning is crucial. However, this task is particularly challenging for low-resource languages like Bengali due to a shortage of annotated corpora, issues with out-of-vocabulary words, lack of domain-specific hyperparameter tuning, limited ability to extract generalized text features, and class imbalances within the corpus. AFuNet: an attention-based fusion network to classify texts in a resource-constrained language. AFuNet undergoes a comprehensive four-phase experimental process, including baseline model evaluation and hyperparameter tuning, late fusion and model selection, attention-based early fusion and model identification, and an ablation study with impact analysis. Fine-tuned based on five Bengali text classification corpora, AFuNet achieves impressive accuracies: 96.60 ± 0.2 (BTCC11), 85.37 ± 0.2 (OSBC), 97.35 ± 0.2 (BARD), 93.74 ± 0.2 (IndicNLP), and 96.51 ± 0.2 (ProthomAlo). In comparison with previous state-of-the-art models on these corpora, AFuNet demonstrates significant accuracy improvements ranging from 0.54% to 4.49%, showcasing its effectiveness in advancing text classification capabilities for the Bengali language. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Robust Self-repairing PID Control Against Measurement Noise

IEEJ Transactions on Electronics, Information and Systems

引用

IEEJ Transactions on Electronics, Information and systems 2025年第3期145卷 287-292页

作者： Kuroda, Kakeru Hara, Masayoshi Takahashi, Masanori Graduate School of Engineering Oita University 700 Dannoharu Oita870-1192 Japan Faculty of Science and Technology Oita University 700 Dannoharu Oita870-1192 Japan

Our previous work has developed a self-repairing PID control (SRPIDC) for sensor failures. This detection method uses the I-controller (integrator) of the PID control and an auxiliary signal to detect failures of a stuck type. Since the auxiliary signal varies in the same direction (sign) as the feedback signal from the sensor, the magnitude of the integrator output increases when the feedback signal gets stuck. Such an unstable behavior is found by a threshold. However, when noise is mixed in the feedback signal, this detection function does not work well because the sign varies. In this paper, we quantitatively analyze the degradation of the detection performance of the SRPIDC due to noise and propose two methods using an on-off type dead zone and a piecewise constant switching signal, that can make the sign of the shaped signal fixed on periods to detect sensor failures in the presence of noise. © 2025 The Institute of Electrical Engineers of Japan.

关键词： Proportional control systems

来源：评论

学校读者我要写书评

暂无评论

INFLUENCE OF SITE ENVIRONMENT ON DETERIORATION OF REINFORCED CONCRETE BUILDINGS WITH INHERENT SALINITY UNDER SALTY ENVIRONMENT

引用

AIJ Journal of technology and Design 2024年第75期30卷 604-607页

作者： Nerome, Yasufumi Imamoto, Kei-Ichi Taga, Noriaki Ridengoaoqier, E. Kiyohara, Chizuru Dept. of Architecture Graduate School of Engineering Tokyo Univ. of Science Japan Dept. of Architecture Faculty of Engineering Tokyo Univ. of Science Japan Research Institute for Science and Technology Tokyo Univ. of Science Japan

This report presents the results of a survey on salt-damaged buildings in Okinawa. The findings showed that concrete used that period was often made of sea sand that had not been de-salted, and that the wind direction caused significant deterioration of the structures due to the effects of airborne salt and moisture content. However, the influence of wind direction also indicated the importance of considering changes in the surrounding site environment, especially for buildings. © 2024 Architectural Institute of Japan. All rights reserved.

关键词： Deterioration

来源：评论

学校读者我要写书评

暂无评论

Exploring Respiratory Parameters Related to Psychophysiological Indexes of Mental Health

Exploring Respiratory Parameters Related to Psychophysiologi...

引用

2023 IEEE International Symposium on Medical Measurements and Applications, MeMeA 2023

作者： Suzuki, Miyu Sato, Hiroki Shibaura Institute of Technology Saitama Systems Engineering and Science Graduate School of Engineering and Science Japan Shibaura Institute of Technology Saitama College of Systems Engineering and Science Department of Bioscience and Engineering Japan

ISBN: (纸本)9781665493840

Psychophysiological indexes based on questionnaire scores and heart rate variability analysis are widely used as mental health indexes. In this study, we focused on respiratory information and aimed to explore respiratory parameters related to psychophysiological indexes of mental health. Seven respiratory parameters were used: respiratory rate, inhalation time, exhalation time, exhalation-inhalation ratio, inhalation time variability, exhalation time variability, and exhalation-inhalation ratio variability. Furthermore, the seven respiratory parameters were analyzed using representative psychophysiological indexes: relaxation scale, anxiety scale (STAI), sleep scale (PSQI-J), and pulse wave, EEG. Twenty-eight university students participated in the experiment. Simultaneous measurements with a band-type respiration sensor, an optical pulse wave sensor, and an electrode-type EEG sensor were performed for 5 min in a resting state after obtaining answers to each questionnaire, and finally, answers to the questionnaires were obtained again. A correlation analysis of the respiratory parameters and psychophysiological indexes revealed a positive correlation (r=0.41 and r=0.44, respectively) between state anxiety and the variability of inhalation and exhalation times (p © 2023 IEEE.

关键词： Sleep research

来源：评论

学校读者我要写书评

暂无评论

Interface engineering and Anion engineering of Mo-Based Heterogeneous Electrocatalysts for Hydrogen Evolution Reaction

引用

Energy & Environmental Materials 2023年第1期6卷 87-104页

作者： Zehao Yin Xuan Liu Siru Chen Tingli Ma Yanqiang Li State Key Laboratory of Fine Chemicals School of Chemical EngineeringDalian University of TechnologyPanjin CampusPanjin 124221China School of Material and Chemical Engineering Centre for Advanced Materials ResearchZhongyuan University of TechnologyZhengzhou 450007China Department of Materials Science and Engineering China Jiliang UniversityHangzhou 310018China Graduate School of Life Science and Systems Engineering Kyushu Institute of TechnologyKitakyushu Fukuoka 808-0196Japan

Development and utilization of hydrogen energy is an effective way to achieve carbon neutrality,only hydrogen production through electrolytic water splitting meets the goal of zero carbon *** facilitate the large-scale commercialization of water splitting devices,the development of highly efficient and low-cost catalysts to reduce the energy consumption is ***_(2)has been regarded as a promising electrocatalyst to replace platinum in hydrogen evolution reaction due to its low price and unique 2D layered ***,the poor conductivity and inert basal planes of MoS_(2)limited its wide-spread ***,researches have demonstrated that the conductivity and active sites of MoS_(2)can be improved by heteroatoms doping or constructing of heterogeneous *** this review,the recent progress of Mo-based catalysts are summarized centered on MoS_(2)based on interface engineering and anion engineering,including MoS_(2)–MoO_(2),MoS_(2)–Mo_(2)C,MoS_(2)–MoN_(x),MoS_(2)–MoP,and MoS_(2)–MoSe_(2).The preparation method,structure,and performance of the catalysts are introduced and the possible mechanism behind the improved catalytic activity are revealed to give readers an overall comprehension on the progress of the Mo-based electrocatalysts for hydrogen evolution *** addition,an outlook on future opportunities and development directions of Mo-based catalysts are proposed to facilitate the development of Mo-based catalysts for hydrogen production.

关键词： electrocatalysis heterostructure hydrogen evolution reaction Mo-based catalysts

来源：评论

学校读者我要写书评

暂无评论

Fatigue Crack Propagation in High-Entropy Alloy CrMnFeCoNi Fabricated by Mixed Sintering of Nitrided Powders

引用

Zairyo/Journal of the Society of Materials science, Japan 2024年第12期73卷 912-918页

作者： Kubo, Ryosuke Matsuura, Hidenori Fujita, Keisuke Kawaguchi, Takahiko Kikuchi, Shoichi Department of Engineering Graduate School of Integrated Science and Technology Shizuoka University Johoku Chuo-ku Hamamatsu432-8561 Japan Department of Mechanical Engineering Faculty of Engineering Shizuoka University Johoku Chuo-ku Hamamatsu432-8561 Japan Educational Division Graduate School of Science and Technology Shizuoka University Johoku Chuo-ku Hamamatsu432-8561 Japan Department of Electronics and Materials Faculty of Engineering Shizuoka University Johoku Chuo-ku Hamamatsu432-8561 Japan

High entropy alloys (HEAs), which have five or more principal elements with an equiatomic composition, exhibit multiple excellent mechanical properties. Superior fatigue properties would be required to use HEAs as structural components in engineering fields. The purpose of this study is to develop HEAs having high crack propagation resistance via melamine reduction-nitridation process and powder metallurgy. The HEA compact was fabricated by mixed sintering of nitrided and un-nitrided CrMnFeCoNi powders to control the amount of nitrogen addition. Stress intensity factor, K, decreasing tests were conducted for CrMnFeCoNi alloys with nitrogen at force ratios, R, from 0.1 to 0.8 in air at room temperature. Threshold stress intensity range, ΔKth, of HEA compact with nitrogen was almost the same as the un-nitrided one at R = 0.1, whereas ΔKth of HEA compact with nitrogen was high in comparison to the un-nitrided one at R = 0.7. This was because the nitrogen addition increased the effective threshold stress intensity range, ΔKeff,th, of the HEA compact, but the magnitude of crack closure was reduced. The HEA compact with nitrogen shows excellent fatigue crack propagation properties, especially at high R conditions where crack closure does not occur. © 2024 Society of Materials science Japan. All rights reserved.

关键词： Sintering

来源：评论

学校读者我要写书评

暂无评论

Proposed numerical and machine learning models for fiber-reinforced polymer concrete-steel hollow and solid elliptical columns

引用

Frontiers of Structural and Civil engineering 2024年第8期18卷 1169-1194页

作者： Tang QIONG Ishan JHA Alireza BAHRAMI Haytham F.ISLEEM Rakesh KUMAR Pijush SAMUI School of Applied Technologies Qujing Normal UniversityQujing 655011China Department of Civil Engineering Indian Institute of Technology-BHUVaranasi 221005India Department of Building Engineering Energy Systems and Sustainability ScienceFaculty of Engineering and Sustainable DevelopmentUniversity of GävleGävle 80176Sweden Department of Civil Engineering National Institute of Technology-PatnaPatna 800005India

This study employs a hybrid approach,integrating finite element method(FEM)simulations with machine learning(ML)techniques to investigate the structural performance of double-skin tubular columns(DSTCs)reinforced with glass fiber-reinforced polymer(GFRP).The investigation involves a comprehensive examination of critical parameters,including aspect ratio,concrete strength,number of GFRP confinement layers,and dimensions of steel tubes used in DSTCs,through comparative analyses and parametric *** ensure the credibility of the findings,the results are rigorously validated against experimental data,establishing the precision and trustworthiness of the *** present research work examines the use of the columns with elliptical cross-sections and contributes valuable insights into the application of FEM and ML in the design and evaluation of structural systems within the field of structural engineering.

关键词： elliptical column fiber-reinforced polymer machine learning finite element method ABAQUS

来源：评论

学校读者我要写书评

暂无评论

Applying Nonnegative Matrix Factorization for Underground Mining Method Selection Based on Mining Projects' Historical Data

引用

International Journal of the Society of Materials engineering for Resources 2024年第1期26卷 1-10页

作者： Manjate, Elsa Pansilvania Andre Ohtomo, Yoko Arima, Takahiko Adachi, Tsuyoshi Bene, Bernardo Miguel Kawamura, Youhei Division of Sustainable Resources Engineering Graduate School of Engineering Hokkaido University Sapporo Japan Graduate School of International Resource Sciences Akita University Akita Japan Division of Sustainable Resources Engineering Faculty of Engineering Hokkaido University Sapporo Japan North China Institute of Science and Technology China Instituto Superior Politécnico de Tete Mozambique

Mining methods selection (MMS) is one of the most critical and complex decision-making tasks in mine planning. The selection of underground mining methods is considered to be the most problematic due to the complexity associated with the orebody geometry, geology, and geotechnical properties. This study integrated artificial intelligence and machine learning in the MMS process by introducing the recommendation systems (RS) approach in MMS through the nonnegative matrix factorization (NMF) algorithm. As such, the weighted nonnegative matrix factorization (WNMF) algorithm is applied to build a model for underground MMS. The study's input dataset is based on thirty mining projects' historical data. In the experiments, we evaluate the capability of the WNMF to predict underground mining methods using five input variables: ore strength, host-rock strength, orebody thickness, shape, and dip. The results show that the WNMF model achieved an average prediction accuracy of 67.5%, considered reasonable and realistic. Further findings reveal that the WNMF model is sensitive to the imbalanced class dataset used in the experiments, thus, suggesting the need to improve the dataset's quality. These results reveal the model's effectiveness in predicting underground mining methods;therefore, with continuous improvement, the WNMF model can be effectively applied in underground MMS. ©2024 Soc. Mater. Eng. Resour. Japan.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：