检索结果-内蒙古大学图书馆

Congress on Evolutionary Computation

作者： Farnaz Kheiri Azam Asilian Bidgoli Masoud Makrehchi Shahryar Rahnamayan Dept. of Electrical Computer and Software Engineering Ontario Tech University Oshawa Canada Dept. of Computer Science Wilfrid Laurier University Waterloo Canada Dept. of Engineering Brock University St. Catharines Canada

ISBN: (数字)9798350308365

ISBN: (纸本)9798350308372

The emergence of bias in deep neural models represents a significant reliability concern, which may lead to overoptimistic results on seen data while compromising the model's ability to generalize effectively on unseen datasets. Recent studies conducted on The Cancer Genome Atlas (TCGA), which is a publicly available repository of histopathology images, reveal that the TCGA cancerous features extracted by deep neural networks surprisingly are able to discriminate slides based on their origin sites. This finding undoubtedly indicates the existence of site-specific patterns embedded in the extracted features learned by deep networks rather than focusing on histomorphologic patterns. Consequently, this biased behavior raises concerns about the reliability of these networks. This observation motivates us to conduct a series of experiments in which we present two distinct evolutionary feature selection strategies, each differentiated by its objective function evaluation. The primary goal is to select the features with a minimized foot-print of data source signatures, thereby ensuring a more accurate and site-independent cancer classification. We have conducted nine comprehensive independent experiments across nine cancer types, employing each evolutionary strategy. The comparison between results obtained through evolutionary strategies and the original feature sets highlights the substantial impact of feature selection methods on bias reduction while maintaining accuracy in cancer-type discrimination. Furthermore, the comparison of the two strategies with each other demonstrates the intricate nature of bias and its integration with cancerous features during the training process.

关键词： Training Accuracy Histopathology Training data Feature extraction Prediction algorithms Data models

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis of Customer Feedback and Reviews in E-Commerce Systems

Sentiment Analysis of Customer Feedback and Reviews in E-Com...

引用

International Conference on New Trends in Computing sciences (ICTCS)

作者： Olla Bulkrock Abdallah Qusef Ahmed BaniMustafa Software Engineering Dept. Princess Sumaya University for Technology Amman Jordan Computer Information Science Dept. Higher Colleges of Technology Abu Dhabi UAE

ISBN: (数字)9798331523411

ISBN: (纸本)9798331523428

Natural Language Processing (NLP) is an effective tool for analyzing consumer satisfaction with products and services typically sold by e-businesses. This study explores the use of sentiment analysis to understand customer feedback and ratings using a dataset acquired by Amazon. That dataset covers online purchases over a period of eight years. The study aims to examine the factors that impact sentiment polarity throughout different periods and understand changes in consumer feelings and reviews towards online products and services. The applied method comprehensively studies sentiment, extracting features using the Term Frequency-Inverse Document Frequency (TF-IDF) and other text analysis and data processing procedures. Visualization is also applied to illustrate the trends, patterns, and changes in customer satisfaction over time and per product and line of products. The findings underscore the importance of sentimental research in discerning customer sentiments on online e-commerce platforms and offer practical recommendations for enhancing online and improving customer satisfaction.

关键词： Sentiment analysis Text analysis Reviews Social networking (online) Mood Customer satisfaction Market research Feature extraction Multilingual Electronic commerce

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Decision-Making in IT Projects: A Regression Application in software Estimation

Data-Driven Decision-Making in IT Projects: A Regression App...

引用

International Conference on New Trends in Computing sciences (ICTCS)

作者： Olla Bulkrock Ahmed BaniMustafa Abdallah Qusef Software Engineering Dept. Princess Sumaya University for Technology Amman Jordan Computer Information Science Dept. Higher Colleges of Technology Abu Dhabi UAE

ISBN: (数字)9798331523411

ISBN: (纸本)9798331523428

IT-related projects are among the most challenging to manage due to their inherent complexity, uncertainty, and the intangibility of their outcomes. However, a data-driven decision-making approach can help tackle these challenges and increase the chance of IT project success. Machine learning techniques can be used to predict uncertain aspects of IT projects, such as the estimation of time, cost, and effort, when applied to historical data collected from previous or similar projects. This approach can allow the project manager to make informed decisions based on facts rather than human judgment or hunch. In this work, we review the potential of data-driven approaches in the management of IT projects, and we demonstrate a use case that applies three regression algorithms, including polynomial regression, linear regression, and partial least squared (PLS), to predict software effort estimation based on the NASA93 public dataset. The best results were achieved using polynomial regression which scored 82% in R 2 , 81% in MAPE, 152 in MAE, 276 in RMSE and 76387 in MSE metrics.

关键词： Measurement Analytical models Reviews Computational modeling Decision making software algorithms Estimation Project management Polynomials software

来源：评论

学校读者我要写书评

暂无评论

Sinhala Sign Language Learning System for Hearing Impaired Community

Sinhala Sign Language Learning System for Hearing Impaired C...

引用

2023 4th International Informatics and software engineering Conference (IISEC)

作者： Idushan K. A. S Dilshan P. G. T Jayasundera P. S Madhushan K. L. B Jenny Krishara Dinuka Wijendra Dept. of Computer Science and Software Engineering Sri Lanka Institute of Information Technology Malabe Sri Lanka

This research aims to develop a comprehensive system for Sinhala Sign Language (SSL) that includes a learning system, dynamic sign detection, audio/video to sign conversion, and vocal training. SSL plays a crucial role in facilitating communication for individuals who are deaf or hard of hearing in Sri Lanka. The learning system provides a platform for learning SSL and includes a text-to-sign language interpreter. The dynamic sign detection system uses computer vision techniques to identify and interpret dynamic signs accurately. The audio/video to sign conversion system bridges the gap between spoken language and SSL by converting auditory information into visual representations. The vocal training system focuses on enhancing the vocal skills of cochlear implanted children. This research contributes to the development of effective communication and language skills for SSL users.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Wildfire Spread Prediction Using Attention Mechanisms in U-NET

Wildfire Spread Prediction Using Attention Mechanisms in U-N...

引用

Electrical, computer, Communications and Mechatronics engineering (ICECCME), International Conference on

作者： Kamen Shah Maria Pantoja Dept. Computer Science and Software Engineering California Polytechnic State University San Luis Obispo USA

Global warming has increased large-scale natural disasters at an alarmingly greater frequency. These natural disasters affect humans and the ecosystems that all species rely upon for food and shelter. A significant byproduct of global warming is the emergence of mega-fires, which are wildfires that burn more than 100,000 acres. Mega fires have exhibited the capacity to destroy entire ecosystems and urban areas. Researchers forecast a significant increase in wildfires during the next decade. This is a fundamentally growing problem that mandates modern solutions. Currently, simulation models are employed to predict wildfire propagation and aid firefighters in suppressing them. However, these methods are computationally expensive and often inaccurate. A recent advance in this area is the usage of neural networks (NN), which are systematically more accurate and computationally efficient. This paper focuses on predicting wildfire spread using NN with an attention mechanism to improve spatial recognition and overall performance in neural networks for wildfire spread prediction. We train different models on 12 distinct observational variables derived from the Google Earth Engine catalog. Evaluation is conducted with accuracy, Dice coefficient score, ROC-AUC, and F1-score. Results show that when augmenting segmentation models with attention mechanisms, the attention component improves feature suppression and recognition, improving overall performance. Furthermore, we use ensemble modeling to reduce bias and variation, leading to more consistent and accurate predictions. The architecture presented in this research achieved a ROC-AUC score of 86.2% and an accuracy of 82.1% when inferencing on wildfire propagation at 30-minute intervals.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improving Fake News Detection with a Mixture of Experts: Leveraging Sentiment Analysis and Sarcasm Detection as Core Expert Models

Improving Fake News Detection with a Mixture of Experts: Lev...

引用

Advancements in Computing (ICAC), International Conference on

作者： Tharindu Gunasekara Disira Thihan Sansika Kodithuwakku Dinuka Dissanayake Harinda Fernando Samitha Vidhanaarachchi Archchana Sindhujan dept. of Software Engineering Sri Lanka Institute of Information Technology Malabe Sri Lanka dept. of Computer Computer Systems Engineering Sri Lanka Institute of Information Technology Malabe Sri Lanka dept. of Computer Science and Electronic Engineering University of Surrey United Kingdom

ISBN: (数字)9798331517878

ISBN: (纸本)9798331517885

Fake news continues to proliferate, posing an increasing threat to public discourse. The paper proposes a framework of a Mixture of Experts, Sentiment Analysis, and Sarcasm Detection experts for improved fake news detection. This approach captures the emotional cues in the text through a Sentiment Analysis expert, which is based on bidirectional encoder representations from Transformers (BERT) models with sentiment vectors generated using SentiWordNet and Integrated Gradients. It combines a sarcasm detection expert based on BERT, recognizing sarcasm and its type to help classify fake news. By fusing these experts through a Mixture of Experts gateway, subtle linguistic cues often found in fake news are more effectively analyzed, leading to improved accuracy in detecting misinformation. Experimental results are presented as 96% for the Sarcasm expert with the BERT base model and 83% for the Sentiment Analysis expert with the distilled version of the BERT (DistilBERT) base model, proving the effectiveness of the proposed approach in beating traditional methods.

关键词： Sentiment analysis Analytical models Accuracy Bidirectional control Logic gates Linguistics Transformers Encoding Vectors Fake news

来源：评论

学校读者我要写书评

暂无评论

Optimizing Endotracheal Suctioning Classification: Leveraging Prompt engineering in Machine Learning for Feature Selection

Optimizing Endotracheal Suctioning Classification: Leveragin...

引用

Activity and Behavior Computing (ABC), International Conference on

作者： Mahera Roksana Islam Anik Mahmud Ferdous Shahera Hossain Md Atiqur Rahman Ahad Fady Alnajjar Dept. of Electrical and Electronic Engineering University of Dhaka Dhaka Bangladesh Dept. of Computer Science and Digital Technologies University of East London London UK Dept. of Computer Science and Software Engineering College of Information Technology UAE University Al Ain UAE

ISBN: (数字)9798350375503

ISBN: (纸本)9798350375510

In a world with an overgrowing elderly population, there exists a critical need for a greater number of skilled individuals in the nursing industry. AI-based systems can be useful, compared to traditional ones with require in-person assistance, to accurately identify nursing activities and assess the nursing trainees to help them become proficient. This paper addresses classifying activities in one such procedure, endotracheal suctioning, using skeletal keypoint data of the subject performing the procedure. A multi-step structured prompt engineering method was established and utilized on several LLMs to select or calculate key features from the data. Then the features are passed onto several tuned machine learning models to obtain results. A tuned XGBoost prevailed across all models, achieving 90% accuracy on the validation set.

关键词： Training Accuracy Computational modeling Machine learning Medical services Predictive models Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Malware Detection and Classification for Internet of Things Using Self-Attention Based Long Short-Term Memory 3

Malware Detection and Classification for Internet of Things ...

引用

3rd IEEE International Conference on Distributed Computing and Electrical Circuits and Electronics, ICDCECE 2024

作者： Sudhakar, K. Husain, Saif O. Babu, D. Raja Kumar, M. Ashok Dineshkumar, R. Nitte Meenakshi Institute Of Technology Dept. Of Ai & Ds Bengaluru India College Of Technical Engineering The Islamic University Department Of Computers Techniques Engineering Najaf Iraq Sr University Department Of Electrical And Electronics Engineering Warangal India Skyline University Nigeria Department Of Computer Science Software Engineering And Cyber Security Kano State Nigeria Saveetha School Of Engineering Saveetha Institute Of Medical And Technical Sciences Saveetha University Department Of Electronics And Communication Engineering Chennai India

ISBN: (纸本)9798350318609

The Internet of Things (IoT) connects numerous intelligent devices providing security features that interact with default settings accessed through applications. Additionally, Deep Learning (DL)-based mechanisms was involved for malware detection to identify and mitigate attacks. The Adaptive Synthetic Sampling (ADASYN) technique addresses class imbalances, especially when splitter class instances are distributed data in complex patterns. The proposed Self-Attention based Long Short-Term Memory (LSTM) classification method facilitates the handling of long-range dependencies in transferred data sequences offering high flexibility and adaptability to varying input lengths. For feature selection Particle Swarm Optimization (PSO) is utilized iteratively optimizes a solution in a high-dimensional space although it may have a low convergence rate due to its iterative process. The obtained results demonstrate the proposed Self-Attention based LSTM model achieves better accuracy reaching 98.05% on the NSL-KDD dataset and 99.01 % on the UNSW-NB15 dataset. These results ensure accurate classification compared to other existing methods, such as Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). © 2024 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Graph Learning under Distribution Shifts: A Comprehensive Survey on Domain Adaptation, Out-of-distribution, and Continual Learning

arXiv

引用

arXiv 2024年

作者： Wu, Man Zheng, Xin Zhang, Qin Shen, Xiao Luo, Xiong Zhu, Xingquan Pan, Shirui The School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China Monash University Melbourne Australia The College of Computer Science and Software Engineering Shenzhen University Guangdong China The School of Computer Science and Technology Hainan University Haikou China The Dept. of Electrical Engineering & Computer Science Florida Atlantic University United States Griffith University QLD Australia

Graph learning plays a pivotal role and has gained significant attention in various application scenarios, from social network analysis to recommendation systems, for its effectiveness in modeling complex data relations represented by graph structural data. In reality, the real-world graph data typically show dynamics over time, with changing node attributes and edge structure, leading to the severe graph data distribution shift issue. This issue is compounded by the diverse and complex nature of distribution shifts, which can significantly impact the performance of graph learning methods in degraded generalization and adaptation capabilities, posing a substantial challenge to their effectiveness. In this survey, we provide a comprehensive review and summary of the latest approaches, strategies, and insights that address distribution shifts within the context of graph learning. Concretely, according to the observability of distributions in the inference stage and the availability of sufficient supervision information in the training stage, we categorize existing graph learning methods into several essential scenarios, including graph domain adaptation learning, graph out-of-distribution learning, and graph continual learning. For each scenario, a detailed taxonomy is proposed, with specific descriptions and discussions of existing progress made in distribution-shifted graph learning. Additionally, we discuss the potential applications and future directions for graph learning under distribution shifts with a systematic analysis of the current state in this field. The survey is positioned to provide general guidance for the development of effective graph learning algorithms in handling graph distribution shifts, and to stimulate future research and advancements in this area. © 2024, CC BY.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

SCHC over DtS-IoT: Performance of SCHC Confirmation Modes in Satellite IoT

SCHC over DtS-IoT: Performance of SCHC Confirmation Modes in...

引用

IEEE World Forum on Internet of Things (WF-IoT)

作者： Rodrigo Munoz-Lara Sandra Céspedes Marcos Diaz Dept.of Electrical Engineering University of Chile Santiago Chile Dept. of Computer Science & Software Eng. Concordia University Montreal QC Canada

ISBN: (数字)9798350373011

ISBN: (纸本)9798350373028

Static Context Header Compression and Fragmentation (SCHC) is a standard defined as an adaptation layer for supporting IPv6, UDP, and CoAP protocols in low power wide area network (LPWAN) technologies. SCHC has fragmentation modes that use confirmation mechanisms to deliver reliable communications. Direct-to-satellite Internet of Things (DtS-IoT) directly communicates between end devices and the satellite in an IoT environment. If SCHC is implemented in a DtSIoT environment, applications using IPv6 could have global coverage through the Internet of Things (IoT). Unfortunately, it is not known whether SCHC can operate with the current acknowledgement mechanisms in a DtS-IoT environment, and if so, it is not known which one performs better. This work evaluates SCHC acknowledgement modes for ACK-on-Error fragmentation mode, including the new SCHC Compound ACK mode. It determines which mode has the lowest transfer delay for different LoRa data rates and frame error rates in a DtS-IoT environment with LEO satellites.

关键词： Satellites Protocols Low earth orbit satellites Delays Internet of Things Compounds Reliability IP networks Standards Low-power wide area networks

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：