检索结果-内蒙古大学图书馆

35th International Conference on Software Engineering and Knowledge Engineering, SEKE 2023

作者： Grabner, Georg Ahmed, Ali Baghaei, Nilufar School of Computing University of Liverpool United Kingdom School of Engineering and Computer Science Victoria University of Wellington New Zealand School of Information Technology and Electrical Engineering University of Queensland Australia

In the era of cloud computing, the focus of cyber criminals has shifted from traditional IT infrastructures to the cloud. Cyber forensic investigations in the cloud present several challenges due to legal obligations, technical limitations, and the dynamic nature of cloud resources. For instance, access to digital assets is crucial for practical analysis but is often hindered, as investigators may not have complete access to disk images or log files. The mainstream forensic frameworks and solutions for identifying, retrieving, and preserving evidence in the cloud are lacking. Yet, NIST has identified 65 cloud forensics challenges, with 13 related to log files alone. The preservation and integrity of forensic evidence are the backbones of a digital forensic investigation, and the chain of custody proves the authenticity of the evidence. However, cloud environments are not designed to handle digital evidence with integrity, leading to severe issues when presenting evidence in a court of law. This paper delves into the challenges of cloud forensics investigation, focusing on log files, and aims to identify potential solutions and frameworks to support the chain of custody. For that purpose, the authors evaluate the suitability of blockchain technology for maintaining a proper chain of custody of log files in the cloud. A prototype is implemented to serve as a proof of concept. © 2023 Knowledge Systems Institute Graduate school. All rights reserved.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Feature Reduction of Relational Oil Drilling Data Before Propositionalization and Harmonization by Measuring Relational Data Missingness

Feature Reduction of Relational Oil Drilling Data Before Pro...

引用

ASEAN Australian Engineering Congress, AAEC 2022

作者： Ting, Clement Pek Wen Then, Patrick Hang Hui Centre for Digital Futures School of Information and Communications Technologies Faculty of Engineering Computing and Science Swinburne University of Technology Sarawak Campus Sarawak Kuching Malaysia

ISBN: (纸本)9789819955466

Knowledge Discovery in Database (KDD) is a process undertaken by many data scientists who aim to extract knowledge from previously collected, secondary data. KDD is difficult when the scientist must analyze big data without prior due to the amount of knowledge required to understand and select relevant features for propositionalization, which is the conversion of raw relational data (stored in several tables) to propositional data (stored as one table) which is more compatible with established feature selection and data mining techniques. In this paper, we describe the Database Attribute Health Feature Reduction process (DAHFR) which is used as a filter-method feature selection metric for Feature Selection before Propositionalization (FSbP) to reduce the scope of features for observation, analysis, harmonization, and propositionalization before data mining. We employ FSbP using DAHFR during the KDD of multi-sourced oil drilling data which is contained in about 20,000 features in about 700 tables in 1 database per source. Each database follows an identical schema but has varying feature-wide missingness and semantic patterns which makes effective propositionalization difficult. DAHFR is a semi-automated process which allows the user to measure relational missingness of each feature for feature reduction and compare missingness patterns across databases. DAHFR is an improvement to existing feature selection processes because it requires minimum initial knowledge of the database’s business rules and could be used with relational data without the need for direct data observation nor prior propositionalization. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

An optimized Yolov5s approach and its application to image detection problems 6

An optimized Yolov5s approach and its application to image d...

引用

2023 IEEE 6th International Conference on information Systems and Computer Aided Education, ICISCAE 2023

作者： Song, Jiayu Zhu, Zhiquan Qiang, Yufei Jiangsu University School of Computer Science and Communication Engineering Zhenjiang China Southwest Jiaotong University School of Computing and Artificial Intelligence Chengdu China Nanjing University of Information Science and Technology School of Computer Science Nanjing China

ISBN: (纸本)9798350313444

Traditional object detection algorithms cannot cope with motion blur and have poor classification accuracy for small cluster targets, which poses challenges to the development of image classification technology. In order to address these issues, this paper adopts a multi-strategy integrated optimization framework and optimizes the training speed and classification accuracy of the model based on the YOLOv5s network model. Firstly, the CBAM attention mechanism is introduced to mitigate motion blur. Then, the Focal-EIoU loss function is used to enhance the classification of low-accuracy objects. Next, the SiLU activation function is employed to enhance the model's representation capability, and the SGD optimizer is replaced with the Adam optimizer to improve training efficiency. Finally, the improved model underwent 500 rounds of training on a subset of the BDD100K dataset. The results show that the classification accuracy of the model, after multiple strategy optimizations, increased from 46.5% to 74.7%. The YOLOv5s model, optimized by multiple strategies, can effectively solve problems such as motion blur, thereby improving the accuracy and stability of the model. © 2023 IEEE.

关键词： adam optimizer attention mechanism image classification yolov5s

来源：评论

学校读者我要写书评

暂无评论

GP-Net: Global Perceptual Network with Adaptive Gradient Clipping for Emotional Image Classification 8

GP-Net: Global Perceptual Network with Adaptive Gradient Cli...

引用

8th International Conference on Intelligent computing and Signal Processing, ICSP 2023

作者： Wang, Xiaohua Yang, Jie Hu, Min Wu, Xia Hefei University of Technology Anhui Province Key Laboratory of Affective Computing and Advanced Intelligent Machine School of Computer Science and Information Engineering Hefei China

ISBN: (纸本)9798350302455

Traditional image emotion recognition focuses only on the emotion information embedded in the subject or part of the image, while ignoring the global emotion information. In this paper, we propose a new network based on Vision Transformer: a global perception network, containing a main module and a background module. The main module enhances the weighting of objects by showing the interdependencies between modelled channels. The background module further searches for emotional information in the background. In addition, we employ multi-task learning and Adaptive Gradient Clipping to improve the network. Extensive experiments show that our proposed method outperforms the state-of-the-art methods for image sentiment analysis. © 2023 IEEE.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Sparsing and Smoothing for the seq2seq Models

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2023年第3期4卷 464-472页

作者： Zhao, Shuai Liang, Zhuoqian Wen, Jinming Chen, Jie College of Information Science and Technology Jinan University Guangzhou511400 China Pazhou Lab Guangzhou510330 China Shanghai Key Laboratory of Trustworthy Computing School of Software Engineering East China Normal University Shanghai200062 China

Current neural language models are trained to minimize cross-entropy and use softmax to compute the locally normalized probabilities over the target. While this setup provides solid results in several natural language processing (NLP) tasks, one unsatisfying aspect is its dense output. This density is wasteful, making models hard to interpret and assigning probability mass to many implausible outputs. To overcome this problem, we propose T-softmax, a simple but effective method to draw considerably sparse probability out of neural language models than softmax. Our method avoids dense output by truncating the unreliable tail of the probability distribution to improve the model's performance. In addition, we generalize logits with temperature, a critical regularization technique, from the softmax to T-softmax. To show our approach as a drop-in replacement for softmax, we evaluate them on three NLP tasks: summary generation, question answer, and math word problem. Experimental results show that our proposed model significantly improves performance without sacrificing speed;notably, in all experiments, our method outperforms the softmax. © 2020 IEEE.

关键词： Probability distributions

来源：评论

学校读者我要写书评

暂无评论

Optimizing Fabric Welding Using Image Processed Laser Inspection 5th

Optimizing Fabric Welding Using Image Processed Laser Inspec...

引用

5th International Conference on Data science, Machine Learning and Applications, ICDSMLA 2023

作者： Ansari, Tabish Sekhar, Malepati Chandra Alam, Intekhab Bargavi, Manju Department of Civil Engineering School of Engineering and Computing Dev Bhoomi Uttarakhand University Navgaon Manduwala Dehradun India School of Computer Science and Engineering Presidency University Karnataka Bengaluru India Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh Lucknow India Karnataka Bengaluru India

ISBN: (纸本)9789819780426

Material welding is used in expanding industrial and industrial applications and depends on accuracy and precision to ensure satisfaction and protection. Laser inspection technology has been developed to help enhance the procedure's accuracy and performance by presenting rapid, reliable imaging that may be processed to decide where changes are wished. Through photograph processing, particular modifications to cloth welding can be recognized quickly and effectively. Additionally, photograph-processed laser inspection can hit upon any irregularities or defects in the welding method that can cause premature failure or decreased functionality. With this generation, designers and manufacturers can also develop higher control structures for his or her tasks, making more accurate and price-effective use of their assets. Sooner or later, the laser inspection era can offer actual-time remarks for automatic tactics, allowing changes to be made quickly with minimal disruption. Using this image processing technology, fabric welding groups can improve exceptional accuracy and value across merchandise. Optimizing material welding through image-processed laser inspection is a valuable method for controlling the pleasantness of welds. This technique uses a laser scanner, mild source, and a digital camera to seize detailed photos of the welds before and after welding. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Inspection

来源：评论

学校读者我要写书评

暂无评论

A Video Summarization Method for Movie Trailer-Genre Classification Based on Emotion Analysis 9th

A Video Summarization Method for Movie Trailer-Genre Classif...

引用

9th International Conference on computing and Informatics, ICOCI 2023

作者： Ng, Wan En Pozi, Muhammad Syafiq Mohd Omar, Mohd Hasbullah Katuk, Norliza Raziff, Abdul Rafiez Abdul Faculty of Computer Science and Information Technology Universiti Malaya Kuala Lumpur Kuala Lumpur50603 Malaysia School of Computing Universiti Utara Malaysia Kedah Sintok06010 Malaysia Kulliyyah of Information and Communication Technology International Islamic University Malaysia Gombak Kuala Lumpur50728 Malaysia

ISBN: (纸本)9789819995882

We live in an information world where visual data undergo exponential growth within a very short time window. With diverging content diversity, we simply have no capacity to keep track of those data. While short video platforms (such as TikTok™ or YouTube Shorts™) can helped users viewing relevant videos within the shortest time possible, those videos might have misleading information, primarily if it is derived from long videos. Here, we analyzed several short videos (in terms of movie trailers) from YouTube and established a correlation between one movie trailer and the classified movie genre based on the emotion found in the trailer. This paper contributes to (1) an efficient framework to process the movie trailer and (2) a correlation analysis between the movie trailer and movie genre. We found that every movie genre can be represented by two unique emotions. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Performance of ResNet50 in Classifying Artificial Intelligence Generated Arts 21

Performance of ResNet50 in Classifying Artificial Intelligen...

引用

21st IEEE International Colloquium on Signal Processing and Its Applications, CSPA 2025

作者： Dioses, Isaac Angelo M. Castillo, Wendell Santiago, Patrick Neil Geronimo, Marites P. MacAso, Joel A. Domingo, Suzette Dc. School of Information Technology Mapua University Makati Philippines College of Computing Studies Information and Communication Technology Isabela State University Echague Campus Isabela Echague Philippines College of Industrial Technology Nueva Ecija University of Science and Technology Nueva Ecija Philippines Institute of Linguistics and Literature Nueva Ecija University of Science and Technology Nueva Ecija Philippines

ISBN: (纸本)9798331522193

The increasing prevalence of AI-generated art has sparked debates regarding its authenticity and impact on traditional art. This study evaluates the performance of the ResNet50 deep learning model in classifying images as either AI-generated or non-AI-generated (traditional art). A dataset collected from Kaggle was divided into two categories, and training was conducted using a learning rate of 0.001, a minibatch size of 32, and 20 epochs, with validation occurring every 50 iterations. The model achieved an overall accuracy of 95.45%, with class wise accuracies of 95.82% for Algenerated images and 95.03% for non-AI-generated images. Precision, recall, and F1 scores further confirmed the model's robust performance, with balanced metrics across both classes. The results highlight the effectiveness of ResNet50 in distinguishing between these two forms of art, demonstrating its potential as a tool for addressing controversies related to AI -generated content in art competitions and creative domains. Future research could focus on refining the model's performance by exploring more diverse datasets and advanced architectures. This study contributes to the growing body of research at the intersection of deep learning and creative industries, offering a reliable solution for classifying AI-generated and traditional art. © 2025 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

An Efficient Ensemble approach for Fake Reviews Detection 3

An Efficient Ensemble approach for Fake Reviews Detection

引用

3rd IEEE International Conference on Artificial Intelligence, ICAI 2023

作者： Iqbal, Asif Rauf, Muhammad Arslan Zubair, Muhammad Younis, Tanveer Riphah College of Computing Riphah International University Faisalabad Pakistan School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798350322125

People prefer to buy items and services online to save time in the current age of growing e-commerce. these internet purchases are heavily impacted by the reviews or opinions of people who have already purchased them. customers provide comments to businesses on how to improve product quality, develop, and monitor business strategies in order to boost sales and profits. customers may also use these comments to choose the proper items with less effort and time spent. giving fake review is the practice of fraudulent people who wants to promote or degrade products or services for illegitimate monetary gain. in this research paper, we present an ensemble machine-learning model to identify whether a review is fraudulent or authentic. to achieve this objective, amazon reviews dataset is used. the proposed ensemble model outperformed as compared to other individual classifiers. random forest provides 99% accuracy which is better than other algorithms. © 2023 IEEE.

关键词： Adaptive boosting

来源：评论

学校读者我要写书评

暂无评论

U-HRNet+: a real-time precipitation estimation method based on the fusion of temporal and spatial domain features

引用

International Journal of Remote Sensing 2024年第13期45卷 4300-4323页

作者： Runzhe Tao Yonghong Zhang Jiangeng Wang Peibei Zheng Liyang Liu a School of Computing and Software Nanjing Vocational University of Industry Technology Nanjing China b School of Automation Nanjing University of Information Science and Technology Nanjing China

ABSTRACTThe precipitation estimation with high spatial and temporal resolution over Tibet is of great significance for monitoring the stability of the plateau ecosystem and early warning of natural disasters. In response to the problem of the unsatisfactory accuracy in traditional real-time precipitation inversion methods and quantitative accuracy estimation products in this region. This paper adopts deep-learning methods to explore the corresponding relationship between precipitation and multi-band observation data from geostationary satellite imagery. The U-High Resolution Network (U-HRNet) as backbone is fully exploited to generate multi-scale features and fused using attention gates to extract high-resolution spatial representations representing precipitation clouds, thereby achieving precipitation estimation of target clouds at different scales. In addition, in response to the characteristics of high precipitation intensity and drastic cloud cluster changes in summer convective weather within a short period of time, a multi-head attention mechanism is introduced to extract temporal information from time series inputs and fuse it with spatial features to improve the accuracy of real-time precipitation estimation. Compared with the operational precipitation products and baseline deep-learning models, the classification metrics such as Probability of detection (POD), False Alarm Ratio (FAR) and Critical success index (CSI) show that the proposed model (U-HRNet+) has better performance in test set and has high generalization ability for estimating short-term heavy rainfall over Tibet and its surrounding areas in summer. Therefore, the proposed model has the potential to serve as a more accurate and reliable satellite-based precipitation estimation product.

关键词： Fengyun-4A satellite real-time precipitation estimation neural network feature fusion

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：