检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2024年第9期5卷 4428-4442页

作者： Hancer, Emrah Xue, Bing Zhang, Mengjie Victoria University of Wellington Center for Data Science and Artificial Intelligence School of Engineering and Computer Science Wellington6140 New Zealand Mehmet Akif Ersoy University Department of Software Engineering Burdur15039 Turkey

Multilabel learning is an emergent topic that addresses the challenge of associating multiple labels with a single instance simultaneously. Multilabel datasets often exhibit high dimensionality with noisy, irrelevant, and redundant features. In recent years, multilabel feature selection (MLFS) has gained prominence as a crucial and emerging machine learning task due to its ability to handle such data effectively. However, existing approaches for MLFS often prioritize top-ranked features based on intrinsic data criteria, disregarding relationships within the feature subset. Additionally, compared with conventional feature selection, multiobjective evolutionary algorithms (MOEAs) have not been widely explored in the context of MLFS. This study aims to address these gaps by proposing a multimodal multiobjective evolutionary algorithm (MMOEA) called MMDE_SICD which incorporates a preelimination scheme, an improved initialization scheme, an exploration scheme inspired by genetic operations and a statistically inspired crowding distance scheme. The results show that the proposed MMDE_SICD algorithm can outperform a variety of MOEAs and MMOEAs as well as conventional MLFS algorithms. Notably, this study is the first of its kind to consider MLFS as a multimodal multiobjective problem. © 2020 IEEE.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

MFR-CNN: A modified faster R-CNN approach based on bounding box and reliable score for cloth image retrieval

引用

Multimedia Tools and Applications 2024年 1-29页

作者： Kaur, Navneet Pandey, Shreelekha Kalra, Nidhi Department of Computer Science and Engineering Thapar Institute of Engineering & ampTechnology Punjab Patiala147004 India School of Computer Science and Engineering RV University South Karnataka Bangalore560059 India

Today, machine learning is used in a broad variety of applications. Convolution neural networks (CNN), in particular, are widely used to analyze visual data. The fashion industry is catching up to the growing usage of machine learning in a variety of fields throughout the world. The proposed model named MFR-CNN, which is a modified Faster R-CNN approach, has been presented in this manuscript. The effectiveness of an algorithm is validated on cloth attributes by established datasets such as the Deep Fashion dataset and the Fashion Landmark dataset, alongside a new dataset created for the same manually. The proposed model detects clothes using bounding boxes and further classifies them. Apart from this, retrieval functionality has also been added. The effectiveness of the proposed model is evident through the learning rate analysis done for the modules present in the same. On the metrics of accuracy, execution time, and total loss, the model proposed here is contrast with existing techniques named R-CNN, Fast R-CNN, and Faster R-CNN. It is found that MFR-CNN is 50% more accurate in contrast to R-CNN. Also, the accuracy of MFR-CNN is 30% and 7% more accurate than the Fast R-CNN and Faster R-CNN, respectively. Moreover, MFR-CNN is quite effective in contrast to other approaches as it takes less time for execution. It consumes 33 seconds less in contrast to R-CNN. Also, the time taken by the MFR-CNN is 17 seconds and 3 seconds less in comparison to Fast R-CNN and Faster R-CNN, respectively. From the experimental results it is being inferred that the MFR-CNN approach for detecting and classifying clothes is both efficient and effective. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Image retrieval

来源：评论

学校读者我要写书评

暂无评论

An Empirical Study on Cross-lingual Vocabulary Adaptation for Efficient Language Model Inference

An Empirical Study on Cross-lingual Vocabulary Adaptation fo...

引用

2024 Findings of the Association for Computational Linguistics, EMNLP 2024

作者： Yamaguchi, Atsuki Villavicencio, Aline Aletras, Nikolaos School of Computer Science University of Sheffield United Kingdom Department of Computer Science Institute of Data Science and Artificial Intelligence University of Exeter United Kingdom The Alan Turing Institute United Kingdom

ISBN: (纸本)9798891761681

The development of state-of-the-art generative large language models (LLMs) disproportionately relies on English-centric tokenizers, vocabulary and pre-training data. Despite the fact that some LLMs have multilingual capabilities, recent studies have shown that their inference efficiency deteriorates when generating text in languages other than English. This results in increased inference time and costs. Cross-lingual vocabulary adaptation (CVA) methods have been proposed for adapting models to a target language aiming to improve downstream performance. However, the effectiveness of these methods on increasing inference efficiency of generative LLMs has yet to be explored. In this paper, we perform an empirical study of five CVA methods on four generative LLMs (including monolingual and multilingual models) across four typologically-diverse languages and four natural language understanding tasks. We find that CVA substantially contributes to LLM inference speedups of up to 271.5%. We also show that adapting LLMs that have been pre-trained on more balanced multilingual data results in downstream performance comparable to the original models. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Recognition of Critical Built-Up Areas Located on High Hill-Slope Regions using Spatial data Mining Technique 6

Recognition of Critical Built-Up Areas Located on High Hill-...

引用

6th IEEE International Conference on Emerging Smart Computing and Informatics, ESCI 2024

作者： Kodge, B.G. School of Science Department of Computer Science GITAM University Hyderabad India

ISBN: (纸本)9798350306613

In mountainous places, structures are being built for residential or commercial uses without the necessary safety precautions. Every year, landslides, torrential downpours, severe snowfall, earthquakes, volcanic eruptions, and floods cause buildings to collapse. The bulk of them are found in high-hill slope areas with loose soil types, close to river flows and other sorts of water sources. Therefore, these incidents have claimed thousands of lives. This paper deals with the process of automatic identification of critical buildings (residential/commercial) located in mountainous area which are on high-hill-slope, close to river flows, having loose soil type and high variations in land elevation contours. This study use the primary data like, built-up/residential area and water body areas which are extracted from sample LULC (Land Use and Land Cover) using image classification techniques, and another important data like slope map and land elevation contour maps which are generated from DEM (Digital Elevation Model). In addition, the supplementary data like, river maps, soil maps and other base maps, are also collected. All the data are integrated and taken into consideration for the identification and extraction of critical residential/build-up areas using spatial data mining technique. © 2024 IEEE.

关键词： Critical residential area identification DEM Image Segmentation LUCL Spatial data Mining

来源：评论

学校读者我要写书评

暂无评论

Parallelization of LSTM-GRU Architectures for Multivariate Prediction of Stock Prices 9

Parallelization of LSTM-GRU Architectures for Multivariate P...

引用

9th International Conference on Information Management and Technology, ICIMTech 2024

作者： Tarigan, Gabriel Asael Hermawan, Eric Girsang, Abba Suganda Bina Nusantara University BINUS Graduate Program - Master of Computer Science Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9798350390025

Anticipating stock market movements is challenging, as hasty prediction risks large financial losses. Stock price prediction is complex due to the high volatility and many influencing external factors. Recurrent Neural Network (RNN) models are made to learn time series data making it ideal for stock prediction based on records. Well-known RNN architectures for forecasting problems are Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU). However, past studies only explored standalone models and have not generated convincing performance results. By integrating the efficiency of GRU to effectively handle shorter data sequences with LSTM's ability to capture long-term dependencies, our model aims to improve stock price forecasting accuracy by collaborating the strength between both architectures. The proposed model was used to forecast multivariate stock prices and evaluated using Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). We found that parallelization of LSTM-GRU offers better prediction, up to 9.1% improvement compared to sequential LSTM-GRU and Bidirectional LSTM (Bi-LSTM). © 2024 IEEE.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

A Discrete-Time Collaborative Neurodynamic Approach to Distributed Global Optimization 13

A Discrete-Time Collaborative Neurodynamic Approach to Distr...

引用

13th International Conference on Intelligent Control and Information Processing, ICICIP 2025

作者： Huang, Haoen Zeng, Zhigang Wang, Jun School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China Department of Computer Science Department of Data Science City University of HongKong Kowloon Hong Kong

ISBN: (纸本)9798331516147

In this paper, a discrete-time projection neural network with an adaptive step size (DPNN) is proposed for distributed global optimization. The DPNN is proven to be convergent to a Karush-Kuhn-Tucker point. Several DPNNs are utilized in a collaborative neurodynamic framework for solving distributed global optimization problem. The efficacy of the collaborative neurodynamic approach with DPNNs is demonstrated through simulation results. © 2025 IEEE.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Automated heart disease prediction using improved explainable learning-based technique

引用

Neural Computing and Applications 2024年第26期36卷 16289-16318页

作者： Bizimana, Pierre Claver Zhang, Zuping Hounye, Alphonse Houssou Asim, Muhammad Hammad, Mohamed El-Latif, Ahmed A. Abd School of Computer Science and Engineering Central South University Changsha410083 China School of Mathematics and Statistics Central South University Changsha410008 China EIAS Data Science Lab College of Computer and Information Sciences and Center of Excellence in Quantum and Intelligent Computing Prince Sultan University Riyadh11586 Saudi Arabia School of Computer Science and Technology Guangdong University of Technology Guangzhou510006 China Department of Mathematics and Computer Science Faculty of Science Menoufia University Shebin El-Koom Egypt Information Technology Department Faculty of Computers and Information Menoufia University Shebin El-Koom Egypt

Heart disease (HD) stands as a major global health challenge, being a predominant cause of death and demanding intricate and costly detection methods. The widespread impact of heart failure, contributing to increased rates of morbidity and mortality, underscores the urgency for accurate and timely prediction and diagnosis. This is crucial for effective prevention, early detection, and treatment, thereby reducing the threat to individual health. However, the early and precise prediction of HD remains a significant challenge. The complexity of medical data poses a considerable challenge for healthcare professionals, who are required to interpret and utilize this information swiftly for effective intervention. Addressing this gap, our study introduces a novel Improved Explainable Learning-Based Technique (IELBT) for HD prediction. This technique harnesses a strategic combination of feature selection, Venn diagrams, data normalization methods, optimized parameters, and machine learning algorithms, specifically tailored for predicting HD. We evaluated the performance of our model using the Alizadeh Sani HD dataset, aiming to accurately detect the presence or absence of the condition. Our results demonstrate that the IELBT, employing a support vector machine with a robust scaling approach, optimal parameterization, and a data split ratio of 70:30, achieves an impressive accuracy rate of 96.00%. Beyond achieving high accuracy, the IELBT outperforms similar models in existing literature and provides significant interpretability and explanation, essential elements in the field of medical diagnosis. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

An overview on IRS-enabled sensing and communications for 6G: architectures, fundamental limits, and joint beamforming designs

引用

science China(Information sciences) 2025年第5期68卷 170-193页

作者： Xianxin SONG Yuan FANG Feng WANG Zixiang REN Xianghao YU Ye ZHANG Fan LIU Jie XU Derrick Wing Kwan NG Rui ZHANG Shuguang CUI School of Science and Engineering (SSE) Shenzhen Future Network of Intelligence Institute (FNii-Shenzhen)and Guangdong Provincial Key Laboratory of Future Networks of Intelligence The Chinese University of Hong Kong Department of Electrical Engineering City University of Hong Kong School of Information Engineering Guangdong University of Technology Key Laboratory of Wireless-Optical Communications Chinese Academy of Sciences School of Information Science and Technology University of Science and Technology of China Computer School Beijing Information Science and Technology University National Mobile Communications Research Laboratory School of Information Science and Engineering Southeast University School of Electrical Engineering and Telecommunications University of New South Wales School of Science and Engineering Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Department of Electrical and Computer Engineering National University of Singapore

This study presents an overview on intelligent reflecting surface(IRS)-enabled sensing and communication for the forthcoming sixth-generation(6G) wireless networks, in which IRSs are strategically deployed to proactively reconfigure wireless environments to improve both sensing and communication(S&C) performance. First, we exploit a single IRS to enable wireless sensing in the base station's(BS's) non-line-of-sight(NLoS) area. In particular, we present three IRS-enabled NLoS target sensing architectures with fully-passive, semi-passive, and active IRSs, respectively. We compare their pros and cons by analyzing the fundamental sensing performance limits for target detection and parameter estimation. Next, we consider a single IRS to facilitate integrated sensing and communication(ISAC), in which the transmit signals at the BS are used for achieving both S&C functionalities, aided by the IRS through reflective beamforming. We present joint transmit signal and receiver processing designs for realizing efficient ISAC, and jointly optimize the transmit beamforming at the BS and reflective beamforming at the IRS to balance the fundamental performance tradeoff between S&C. Furthermore, we discuss multi-IRS networked ISAC, by particularly focusing on multi-IRS-enabled multi-link ISAC, multi-region ISAC, and ISAC signal routing, respectively. Finally, we highlight various promising research topics in this area to motivate future work.

关键词： integrated sensing and communication (ISAC) intelligent reflecting surface (IRS) non-line-of-sight (NLoS) sensing sensing and communication tradeoff

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Machine Learning Algorithms and data Balancing Techniques for Credit Card Fraud Detection 9th

Comparative Analysis of Machine Learning Algorithms and Data...

引用

9th International Conference on computer science and Computational Intelligence, ICCSCI 2024

作者： Wijaya, Michael Geraldin Pinaringgi, Muhammad Faza Zakiyyah, Alfi Yusrotis Meiliana Computer Science Department School of Computer Science Bina Nusantara University Jakarta11480 Indonesia Mathematics Department School of Computer Science Bina Nusantara University Jakarta11480 Indonesia

The rising prevalence of digital transactions in contemporary financial systems has heightened the risk of credit card fraud (CCF), necessitating effective detection techniques. Machine learning (ML) technologies show promise, but dataset imbalance impedes fraud detection accuracy. This study investigates the effectiveness of ensemble learning techniques and data balancing methods for detecting CCF in imbalanced datasets. We compare four ML algorithms-Logistic Regression, Decision Tree, Random Forest, and XGBoost-and evaluate data balancing techniques including SMOTE, ADASYN, Random Oversampling, CNN, NCR, and Random Undersampling, along with hybrid methods like SMOTE with NCR and ADASYN with NCR. Our experiments use a dataset of credit card transactions made by European cardholders, which includes 30 attributes and exhibits severe class imbalance, with fraud events accounting for only 0.172% of total transactions. Applying balancing techniques led to significant improvements: Random Oversampling with XGBoost achieved the highest F1-Score of 92.43%, with Precision of 97.383% and Recall of 88.418%. Random Forest with Random Oversampling followed with an F1-Score of 91.554%, and NCR with Random Forest achieved an F1-Score of 91.508%. These findings highlight the effectiveness of oversampling techniques in enhancing model performance for imbalanced datasets. Precision, Recall, and F1-Score were emphasized, while specificity and sensitivity were not directly used;macro averages ensured balanced evaluation across all classes. This study concludes that combining Random Forest or XGBoost classifiers with oversampling techniques like Random Oversampling or undersampling methods such as NCR provides accurate and unbiased results for detecting CCF in imbalanced datasets, contributing to reliable fraud detection systems for real-world applications. © 2024 The Authors.

关键词： Random forests

来源：评论

学校读者我要写书评

暂无评论

Optimizing Coverage in Wireless Sensor Networks Using the Tumbleweed Algorithm

Journal of Network Intelligence

引用

Journal of Network Intelligence 2025年第1期10卷 250-261页

作者： Zhang, Mengfei Chu, Shu-Chuan Liu, Nengxian Huang, Kuan-Chun Pan, Jeng-Shyang College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China College of Computer and Data Science Fuzhou University Fuzhou350108 China Department of Industrial and Systems Engineering Chung Yuan Christian University Taoyuan320314 Taiwan School of Artificial Intelligence Nanjing University of Information Science and Technology Nanjing210044 China Department of Information Management Chaoyang University of Technology Taichung41349 Taiwan

Wireless sensor networks (WSNs) have found extensive applications across various fields, significantly enhancing the convenience in our daily lives. Hence, an in-creasing number of researchers are directing their attention toward the performance of sensor nodes, with a particular emphasis on addressing the coverage challenges in WSNs. Node coverage plays a crucial role in determining the detection performance of WSNs, making it imperative to attain maximum coverage with a given number of sensor nodes. This paper applies the Tumbleweed Algorithm (TA) to the coverage problem in WSNs and validates its effectiveness through simulation experiments. The results demonstrate its unique advantages in global search and avoiding local optima, providing a new solution for the WSN coverage problem and offering valuable references for future research and practical applications. The TA algorithm simulates the lifecycle of tumbleweed from seedling to mature plant and includes the concept of growth cycles. This paper compares the TA algorithm with PSO, GWO, DE, BOA, SCA, and AOA, setting different numbers of wireless sensor nodes and varying sensing radii for the experiments. The experimental results show that the TA algorithm significantly improves coverage rates under various configurations and consistently outperforms other commonly used algorithms. © 2025, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Sensor nodes

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：