检索结果-内蒙古大学图书馆

Data science, Agents & Artificial Intelligence (ICDSAAI), International Conference on

作者： S Senthilkumar Kishore D Veeramalai Sankaradass A Merwin Blessy Yuvadarshan M R. Ponnusamy Dept. of Computer Science and Engineering Chennai Institute of Technology Chennai Tamil Nadu India

ISBN: (数字)9798331537555

ISBN: (纸本)9798331537562

One of the pressing issues on user-generated platforms and social media is hate speech, a threat to online communities. Conventional detection methods, such as keyword matching or basic classification, often miss hate speech's subtleties and contextual dependencies. This paper introduces a pioneering deep learning approach, leveraging transformer architectures and attention mechanisms, to advance the state-of-the-art in hate speech detection. Our model's unique feature is its ability to focus on words that carry more contextual relevance, thereby interpreting the intended meaning of the words more accurately. The multi-head self-attention mechanism allows for differential weighting of various input parts, thereby enhancing the model's ability to distinguish between hostile and hateful language, even in cases of word similarity. We explain the architecture of our model with further explanation of the calculation involved in queries, keys, and values, followed by a description of the mathematical foundation of the attention mechanism. In fine-tuning the pre-trained transformer model, say BERT, from the hate speech dataset, our optimised model uses the Adam optimiser. We run experiments to check the performance of the proposed model by using accuracy, precision, recall, and F1-score to prove that our proposed model outperforms the base models like CNNs and RNNs. Our attention-based approach successfully highlights hate speech in subtle, context-sensitive ways. As this is one of the first studies where attention has been explored in the context of hate speech detection, it provides a good platform for further work on hate speech and multimodal inputs in many other languages. The results of our experiments indicate the promise of attention-based models as a powerful and scalable solution to the real-world challenges of hate speech detection.

关键词： Deep learning Attention mechanisms Accuracy Sensitivity Social networking (online) Hate speech Transformers Mathematical models Robustness Context modeling

来源：评论

学校读者我要写书评

暂无评论

A Framework for Structural Damage detection system in automobiles for flexible Insurance claim using IOT and Machine Learning

A Framework for Structural Damage detection system in automo...

引用

2022 International Mobile and Embedded Technology Conference, MECON 2022

作者： Agarwal, Deepshikha Tripathi, Khushboo Indian Institute of Information Technology Dept. of Information Technology Lucknow India Amity University Dept. of Computer Science and Engineering Haryana Gurgaon India

ISBN: (纸本)9781665420204

Automobiles are required in everyday life to commute from destination to other. But, due to faulty road conditions, heavy traffic and lack of adequate driver skills, crash incidents happen almost every hour of the day. With every vehicle, companies offer insurance which people can claim to get the damaged parts recovered. However, during the recovery, it solely depends on the insurance companies investigating officer to give an adequate claim to the vehicle owner. Due to inadequate visual precipitance, at times, the owner is not given the full insurance claim. The proposed framework offers automated damage detection system by which better and correct claim can be given to such an owner. The proposed framework is based on Machine Learning and IOT approach for transport network scenario. This paper presents the idea of implementing proposed framework to the researchers in future to detect damages and calculate the correct claim to the owner for Insurance claim. © 2022 IEEE.

关键词： Automobiles

来源：评论

学校读者我要写书评

暂无评论

Communication-Avoiding Recursive Aggregation

Communication-Avoiding Recursive Aggregation

引用

IEEE International Conference on Cluster Computing

作者： Yihao Sun Sidharth Kumar Thomas Gilray Kristopher Micinski Dept. Electrical Engineering and Computer Science Syracuse University Syracuse USA Dept. Computer Science University of Illinois at Chicago Chicago USA Dept. Computer Science University of Alabama at Birmingham Birmingham USA Dept. of Electrical Engineering and Computer Science Syracuse University Syracuse USA

Recursive aggregation has been of considerable interest due to its unifying a wide range of deductive-analytic workloads, including social-media mining and graph analytics. For example, Single-Source Shortest Paths (SSSP), Connected Components (CC), and PageRank may all be expressed via recursive aggregates. Implementing recursive aggregation has posed a serious algorithmic challenge, with state-of-the-art work identifying sufficient conditions (e.g., pre-mappability) under which implementations may push aggregation within recursion, avoiding the serious materialization overhead inherent to traditional reachability-based methods (e.g., Datalog).State-of-the-art implementations of engines supporting recursive aggregates focus on large unified machines, due to the challenges posed by mixing semi-naïve evaluation with distribution. In this work, we present an approach to implementing recursive aggregates on high-performance clusters which avoids the communication overhead inhibiting current-generation distributed systems to scale recursive aggregates to extremely high process counts. Our approach leverages the observation that aggregators form functional dependencies, allowing us to implement recursive aggregates via a high-parallel local aggregation to ensure maximal throughput. Additionally, we present a dynamic join planning mechanism, which customizes join order per-iteration based on dynamic relation sizes. We implemented our approach in PARALAGG, a library which allows the declarative implementation of queries which utilize recursive aggregates and executes them using our MPI-based runtime. We evaluate PARALAGG on a large unified node and leadership-class supercomputers, demonstrating scalability up to 16,384 processes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Epileptic Seizure Prediction: Exploring Diverse Pre-Processing Techniques and Machine Learning Models 10

Comparative Analysis of Epileptic Seizure Prediction: Explor...

引用

10th IEEE Uttar Pradesh Section International Conference on Electrical, Electronics and computer engineering, UPCON 2023

作者： Hasan Talukder, Md. Simul Sulaiman, Rejwan Bin Chhabra, Gunjan Yadav, Suman Avdhesh Haque, Rezaul Islam, Md Babul Kaushik, Keshav Nuclear Safety Security and Safegurad Division Bangladesh Atomic Energy Regulatory Authority Dhaka Bangladesh School of Computer Science and Technology Northumbria University Newcastle United Kingdom Graphic Era Hill University Dept. of Cse Dehradun India Amity School of Engineering & Technology Department of Computer Science and Engineering Noida India East West University Dept of Computer Science and Engineering Dhaka1212 Bangladesh Calabria Italy School of Computer Science University of Petroleum and Energy Studies Dehradun India

ISBN: (纸本)9798350382471

Precise prediction and detection of epilepsy, a common neurological condition marked by sporadic and repeated seizures, is necessary for efficient management and patient care. To do accurate and reliable seizure detection, machine learning (ML) on electroencephalogram (EEG) recordings is a good choice because of its capacity to provide insightful information about brain activity during seizures. In this work, we do a thorough comparative analysis of five machine learning methods for the prediction of epileptic seizures using EEG data: Random Forest (RF), Decision Tree (DT), Extra Trees (ET), Logistic Regression (LR), and Gradient Boosting (GB). Firstly, to assure the quality of the data and enable precise model training, the dataset underwent extensive preprocessing, which included cleaning, normalization, treatment of outliers, and oversampling. These preprocessing techniques significantly enhanced the models' performance. In our study, we evaluate the accuracy of each model;the LR classifier obtains 56.95% accuracy, while both GB and DT achieve 97. 17% accuracy. Outperforming the findings of previous investigations, the ET model showed the best performance with an accuracy of 99.29%. Achieving 98.99% accuracy was another accomplishment for the RF model. Our results imply that the ET model not only performed better than the other models in the comparison analysis but also more effectively than the most advanced outcomes from earlier research. Regarding the accurate and dependable prediction of epileptic seizures using EEG data, the ET model is a solid option due to its remarkable performance. © 2023 IEEE.

关键词： EEG epileptic seizure Extra Tree hyperparameter tuning machine learning

来源：评论

学校读者我要写书评

暂无评论

Stacking Ensemble Framework for Hate Speech Detection in Bangla Videos

Stacking Ensemble Framework for Hate Speech Detection in Ban...

引用

Computing, applications and Systems (COMPAS), IEEE International Conference on

作者： Shamrin Jahan Mim Tanjim Mahmud Md Hasan Ali Mohammad Tarek Aziz Department of Computer Science and Engineering Rangamati Science and Technology University Rangamati Bangladesh Dept. of Computer Science and Engineering Rangamati Science and Technology University Rangamati Bangladesh Department of Computer Science and Engineering Chittagong University of Engineering & Technology Chittagong Bangladesh

ISBN: (数字)9798331529765

ISBN: (纸本)9798331529772

The proliferation of internet usage and social media platforms has significantly enhanced the ability of individuals to express their opinions on various topics. However, this freedom of expression sometimes morphs into a vehicle for disseminating hate speech, leading to increased incidents of cyberbullying, violations, and conflicts. Particularly on video-sharing websites, which have become a prominent stage for such activities due to their widespread use and the dynamic nature of video content. This study aims to address the issue of hate speech in video content by developing a robust method for detecting hate speech in Bangla language videos. The focus is on the spoken content within these videos, which is a primary vector for the transmission of harmful messages. We constructed a comprehensive dataset by extracting and converting audio from a collection of videos into text. Utilizing this dataset, we applied machine learning techniques and deep learning models to analyze and classify the content. Specifically, our approach involves a stacking ensemble model that combines the strengths of Convolutional Neural Networks (CNN) and Bidirectional Gated Recurrent Units (BiGRU) with the analytical capabilities of customized feature extraction applied to Multinomial Naive Bayes and Random Forest classifiers serving as a meta-model. The proposed stacking ensemble model demonstrates a high level of efficacy, achieving an accuracy rate of 96% in detecting hate speech within the tested video content. This performance indicates a significant advancement over existing methods, underlining the effectiveness of our hybrid, multi-model approach.

关键词： Analytical models Computational modeling Hate speech Stacking Logic gates Vectors Bayes methods Convolutional neural networks Vehicle dynamics Random forests

来源：评论

学校读者我要写书评

暂无评论

Design and Investigation of 8×10Gbps WDM System using Different Modulators

Design and Investigation of 8×10Gbps WDM System using Diffe...

引用

2024 Asia Pacific Conference on Innovation in Technology, APCIT 2024

作者： Kumari, Meet Sharma, Neha Chauhan, Rahul Joshi, Kireet Kumar, Amit UIE Chandigarh University Dept. of ECE Mohali140413 India Chitkara University Chitkara University Institute of Engineering and Technology Punjab Rajpura India Graphic Era Deemed to Be University Computer Science & Engineering Uttarakhand Dehradun248002 India

ISBN: (纸本)9798350361537

A high speed 8×10Gbps wavelength division multiplexing (WDM) system over long-haul fiber is designed and investigated. The system performance is measured for district optical modulators or varied fiber range and throughput. Simulation results depict that faithful fiber range of 55km can be obtained by incorporating single drive mach-zehnder modulator than electroabsorption and MZM modulators. Besides, the fiber range can be extended upto 120km at 5Gbps throughput. The obtained eye patterns are also depict the superior performance of the system for varied fiber range using single drive MZM. Besides, this work illustrates best performance as compared to existing works. © 2024 IEEE.

关键词： Electroabsorption modulators

来源：评论

学校读者我要写书评

暂无评论

Comparative Study of Novel Convolutional Neural Network and Artificial Neural Network to Enhance the Accuracy for Flaw Detection in Animal Hide

Comparative Study of Novel Convolutional Neural Network and ...

引用

Data science, Agents & Artificial Intelligence (ICDSAAI), International Conference on

作者： R. Rajarajeswari Veeramalai Sankaradass Dept. of Computer Science and Engineering Chennai Institute of Technology Chennai Tamil Nadu India

ISBN: (数字)9798331537555

ISBN: (纸本)9798331537562

This research explores how well Artificial Neural Networks (ANN) and Convolutional Neural Networks (NCNN) detect animal hide imperfections to improve quality of weather. Materials and Methods: The study extracted 3600 leather images from Kaggle and split them into two separate groups for this investigation. Group 1 used NCNN, and Group 2 used ANN. Statistical testing generated a sample size of 20 specimens at 95% confidence with 80% power in this experiment. Python became the chosen programming platform for data analysis and preprocessing. NCNN shows better results than ANN because it delivers 96.80% accuracy at 0.001 significance level (p<0.05). Conclusion: The analysis proves the Novel Convolutional Neural Network successfully finds defects in animal hides, but ANN cannot match its results.

关键词： Productivity Accuracy Data analysis Animals Statistical analysis Programming Convolutional neural networks Artificial intelligence Python Meteorology

来源：评论

学校读者我要写书评

暂无评论

Growth Rate Prediction on Fetal Brain Development using Deep Learning Techniques

Growth Rate Prediction on Fetal Brain Development using Deep...

引用

Data Intelligence and Cognitive Informatics (ICDICI), International Conference on

作者： Vellampalli Krishna Sai Mahesh Veldurthi Pavan Kumar Pandiselvam Pandiyarajan Vavila Raveendra Y. Mukesh Kumar Reddy Dept. of Computer Science and Engineering Kalasalingam Academy of Research and Education Tamil Nadu India Dept. of Computer Science and Engineering Kalasalingam Academy of Research and Education Kalasalingam University Krishnankovil Tamilnadu India

ISBN: (数字)9798350389609

ISBN: (纸本)9798350389616

The development of the fetal brain is crucial for a child's cognitive and emotional maturation, highlighting the importance of early identification of neurological issues for effective prenatal care. This study introduces an innovative deep learning framework aimed at predicting fetal brain growth rates from 2D ultrasound images. We utilize a Convolutional Neural Network (CNN) that combines longitudinal data with various imaging modalities to enhance prediction accuracy. By incorporating transfer learning alongside sophisticated techniques like attention mechanisms, our model adept.y captures both spatial and temporal dynamics within the data. Results from our experiments demonstrate that this approach outperforms conventional methods, achieving an accuracy. The findings of this research hold significant promise for prenatal healthcare by enabling timely interventions for high-risk pregnancies. Ultimately, this study contributes to a deeper understanding of fetal brain development and illustrates the potential of integrating AI technologies into standard prenatal diagnostic practices.

关键词： Deep learning Pregnancy Accuracy Ultrasonic imaging Three-dimensional displays Transfer learning Imaging Brain modeling Data models Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Optimization Effects of Rotating Transmitarray for Direction-of-Arrival Estimation with a Single Antenna

引用

IEICE Communications Express 2025年第6期14卷 210-213页

作者： Ryusuke Sunagawa Hibiki Shiiba Makoto Sano Ryuji Kuse Takeshi Fukusako Dept. of Computer Science and Electrical Engineering Kumamoto University Kumamoto Japan Faculty of Advanced Science and Technology Kumamoto University Kumamoto Japan

This paper discusses the optimization effects of a transmitarray on a direction-of-arrival (DOA) estimation using the transmitarray and a single receiving antenna. The transmitarray consists of 1-bit elements with 0° and 180° transmission phases. The transmission phase distribution was optimized using a genetic algorithm and compared to a random distribution. The optimized transmitarray achieves a DOA estimation accuracy comparable to that of a conventional system using a 3-element linear array antenna. The root mean square error (RMSE) of the optimized system exhibits a decreasing trend as the signal-to-noise ratio (SNR) increases. Notably, at an SNR of 20 dB, the RMSE is 20° lower than that of the 3-element linear array antenna.

关键词： Estimation Direction-of-arrival estimation Linear antenna arrays Accuracy Antenna radiation patterns Transmitting antennas Optimization Signal to noise ratio Receiving antennas Vectors

来源：评论

学校读者我要写书评

暂无评论

Advantages of MIMO Radar for Contactless Pulse Waveform Monitoring using End-to-End Learning

Advantages of MIMO Radar for Contactless Pulse Waveform Moni...

引用

IEEE Radio and Wireless Symposium (RWS formerly RAWCON)

作者： Chandler J. Bauder Aly E. Fathy Dept. of Electrical Engineering & Computer Science University of Tennessee Knoxville Knoxville TN USA

ISBN: (数字)9798350377743

ISBN: (纸本)9798350377750

Radar-based methods for contactless pulse monitoring typically use heart rate estimation accuracy as the primary performance evaluation metric. Recently, more in-dept. metrics such as heart rate variability (HRV) have been of interest, requiring precise pulse waveform reconstructions for reliable estimations. Traditional phase variation-based vital signs signal processing techniques have proven to be susceptible to sources of noise, impacting the shape of the extracted pulse waveform. In this work, the advantages of using raw multi-channel radar data and deep learning in an end-to-end manner for precise pulse waveform reconstruction are explored. Preliminary results show electrocardiogram (ECG) waveforms can be reconstructed from raw in-phase and quadrature 60 GHz multiple-input-multiple-output (MIMO) frequency modulated continuous wave (FMCW) radar data using an end-to-end neural network. Additionally, an ablation study is performed, revealing that waveform reconstruction similarity and estimation accuracy of several HRV metrics generally improve as the number of utilized MIMO channels increases.

关键词： Estimation error Accuracy Shape Channel estimation Radar Electrocardiography MIMO Telemetry Heart rate variability Monitoring

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：