检索结果-内蒙古大学图书馆

22nd International symposium INFOTEH-JAHORINA, INFOTEH 2023

作者： Lojić, Ana Jukić, Samed International Burch University Department of Information Technology Sarajevo Bosnia and Herzegovina

ISBN: (纸本)9781665475464

In research to determine the degree of interest in enrolling students in certain high schools, predictive analysis and comparison models are rarely used when classifying and processing different data. All this leads to large fluctuations in enrolment in secondary schools, where certain schools are unable to enrol numerous students who show an interest in a particular field. On the other hand, students lose interest in certain schools, which leads to the discontinuation of certain courses necessary for the needs of today's labour market. Institutions responsible for organizing the educational process do not sufficiently compare and connect teaching and non-teaching activities when analysing the talents and interests of elementary school students from different fields. The goal of this work is to predict the enrolment of students in secondary schools, using the classifiers of programming languages, based on the results that students express during regular classes in elementary *** results show that the accuracy of the data during the training of the Random Forest predictor is 52%, while in Wolfram Alpha it is 62% © 2023 ieee.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

NQ-PROTONET: NOISY QUERY PROTOTYPICAL NETWORK FOR FEW-SHOT REMOTE SENSING SCENE CLASSIFICATION

NQ-PROTONET: NOISY QUERY PROTOTYPICAL NETWORK FOR FEW-SHOT R...

引用

ieee International Geoscience and Remote Sensing symposium (IGARSS)

作者： Lin, Weiquan Tang, Xu Han, Xiao Ma, Jingjing Zhang, Xiangrong Jiao, Licheng Xidian Univ Sch Artificial Intelligence Xian 710071 Shaanxi Peoples R China Geovis Spatial Technol Co Ltd Beijing Peoples R China

ISBN: (数字)9781665427920

ISBN: (纸本)9781665427920

Few-shot remote sensing scene classification, which aims to recognize unseen classes given only a few labeled samples, is a challenge task due to the complex content contained in remote sensing scenes. In this end, we propose a noisy query prototypical network (NQ-ProtoNet), which uses query-mix module (QM) to produce extra query samples with interference information for classification and thus implicitly enhance the feature learning ability of model. Our method alleviates the problem of large intraclass variances and interclass similarity of remote sensing scenes to some extent, and the positive experimental results on UC Merced and NWPU data sets show that it outperforms several few-shot learning methods.

关键词： Few-shot remote sensing scene classification query-mix module large intraclass variances and interclass similarity

来源：评论

学校读者我要写书评

暂无评论

Uncertainty Unveiled: Revealing the Uncertainty of Distribution visualization Through Cross Reality

Uncertainty Unveiled: Revealing the Uncertainty of Distribut...

引用

ieee International symposium on Mixed and Augmented Reality Workshops (ISMARW)

作者： Alexander Gall Anja Heim Bernhard Fröhler Christoph Heinzl Sciences Upper University of Applied TU Wien Austria University Passau Sciences Upper University of Applied Austria

In many scientific fields, the analysis of distributions is essential to unveil patterns in order to set up, confirm or reject hypothesis. However, the application of specific visualization techniques may introduce uncertainties along the analytical pipeline. These so-called visualization mirages lead to cognitive biases and misinterpretations during the analysis, which users are often not aware of and which may lead to incorrect decisions. Complementary interfaces are a reasonable solution to uncover and address these uncertainties without disrupting the analysis workflow. Cross-reality allows to integrate various technologies along the reality-virtuality continuum (RVC) to augment visualization systems. This facilitates users to interact naturally and intuitively with a 2D system, such as augmented reality (AR) devices, and be aware of the relationship between data and its representation, avoiding misinterpretations through uncertainties.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Biglog: Unsupervised large-scale Pre-training for a Unified Log Representation 31

Biglog: Unsupervised Large-scale Pre-training for a Unified ...

引用

31st ieee/ACM International symposium on Quality of Service, IWQoS 2023

作者： Tao, Shimin Liu, Yilun Meng, Weibin Ren, Zuomin Yang, Hao Chen, Xun Zhang, Liang Xie, Yuming Su, Chang Oiao, Xiaosong Tian, Weinan Zhu, Yichen Han, Tao Qin, Ying Li, Yun University of Science and Technology of China China Huawei China Media Group China

ISBN: (纸本)9798350399738

Automated log analysis has been widely applied in modern data-center network, performing critical tasks such as log parsing, log anomaly detection and log-based failure prediction. However, existing approaches rely on hand-crafted features or domain-specific vectors to represent logs, which are either laborious in manual efforts or ineffective facing multiple domains in a system. Furthermore, general-purpose word embeddings are not optimized for log data, thus are data-inefficient in handling complex log analysis tasks. In this paper, we present a pre-training phase for language models to understand both in-sentence and cross-sentence features of logs, resulting in a unified representation of logs that is well-suited for various downstream analysis tasks. The pre-training phase is unsupervised, utilizing 0.45 billion logs from 16 diverse domains. Experiments on 12 publicly available evaluation datasets across 3 tasks indicate superiority of our approach against existing approaches, especially in online scenarios with limited historical logs. Our approach also exhibits remarkable few-shot learning ability and domain-adaptiveness, which not only outperforms existing approaches using only 0.0025% of their required training data, but also adapts into new domains via only a few in-domain logs. We release our code and pre-trained model. © 2023 ieee.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Multi-Agent Actor-Critic Generative AI for Query Resolution and analysis

IEEE Transactions on Artificial Intelligence

引用

ieee Transactions on Artificial Intelligence 2025年

作者： Ur Rahman, Mohammad Wali Nevarez, Ric Mim, Lamia Tasnim Hariri, Salim University of Arizona Department of Electrical and Computer Engineering TucsonAZ85721 United States Trustweb New YorkNY10001 United States Avirtek Inc. United States

In this paper, we introduce MASQRAD (Multi-Agent Strategic Query Resolution and Diagnostic tool), a transformative framework for query resolution based on the actor-critic model, which utilizes multiple generative AI agents. MASQRAD is excellent at translating imprecise or ambiguous user inquiries into precise and actionable requests. This framework generates pertinent visualizations and responses to these focused queries, as well as thorough analyses and insightful interpretations for users. MASQRAD addresses the common shortcomings of existing solutions in domains that demand fast and precise data interpretation, such as their incapacity to successfully apply AI for generating actionable insights and their challenges with the inherent ambiguity of user queries. MASQRAD functions as a sophisticated multi-agent system but "masquerades"to users as a single AI entity, which lowers errors and enhances data interaction. This approach makes use of three primary AI agents: Actor Generative AI, Critic Generative AI, and Expert analysis Generative AI. Each is crucial for creating, enhancing, and evaluating data interactions. The Actor AI generates Python scripts to generate data visualizations from large datasets within operational constraints, and the Critic AI rigorously refines these scripts through multi-agent debate. Finally, the Expert analysis AI contextualizes the outcomes to aid in decision-making. With an accuracy rate of 87% when handling tasks related to natural language visualization, MASQRAD establishes new benchmarks for automated data interpretation and showcases a noteworthy advancement that has the potential to revolutionize AI-driven applications. © 2025 ieee.

关键词： Actor-Critic Model data visualization Deep Learning Generative AI Multi-Agent System Natural Language Processing Query Resolution

来源：评论

学校读者我要写书评

暂无评论

Design of Intelligent Processing Platform for Electric Big data in Hadoop 3

Design of Intelligent Processing Platform for Electric Big D...

引用

3rd International symposium on Advances in Informatics, Electronics and Education, ISAIEE 2022

作者： Gao, Guanhao Tianjin University of Technology Tianjin China

ISBN: (纸本)9781665463577

In order to improve the capabilities of parallel processing and planning of large volumes of electricity data, an intelligent platform for processing large volumes of electricity data developed on the basis of the Hadoop framework is proposed. The research includes a data processing module, a data loading module, a bus control module, and a human-machine interaction module to build an integrated database model for the parallel processing of big data in the field of electricity. The results of the simulation tests show that the intelligent big electricity data processing platform developed in this study can efficiently combine data mining and resource planning capabilities, further improve the integrated big electric data processing capability, and provide a new idea for an efficient solution for the efficient management and analysis of big electricity data, which is crucial for the development of electric systems. © 2022 ieee.

关键词： data mining

来源：评论

学校读者我要写书评

暂无评论

CSFlow: Learning Optical Flow via Cross Strip Correlation for Autonomous Driving

CSFlow: Learning Optical Flow via Cross Strip Correlation fo...

引用

33rd ieee Intelligent Vehicles symposium (ieee IV)

作者： Shi, Hao Zhou, Yifan Yang, Kailun Yin, Xiaoting Wang, Kaiwei Zhejiang Univ State Key Lab Modern Opt Instrumentat Hangzhou Zhejiang Peoples R China Shanghai Artificial Intelligence Lab Shanghai Peoples R China Karlsruhe Inst Technol Inst Anthropomat & Robot Karlsruhe Germany

ISBN: (纸本)9781665488211

Optical flow estimation is an essential task in self-driving systems, which helps autonomous vehicles perceive temporal continuity information of surrounding scenes. The calculation of all-pair correlation plays an important role in many existing state-of-the-art optical flow estimation methods. However, the reliance on local knowledge often limits the model's accuracy under complex street scenes. In this paper, we propose a new deep network architecture for optical flow estimation in autonomous driving-CSFlow, which consists of two novel modules: Cross Strip Correlation module (CSC) and Correlation Regression Initialization module (CRI). CSC utilizes a striping operation across the target image and the attended image to encode global context into correlation volumes, while maintaining high efficiency. CRI is used to maximally exploit the global context for optical flow initialization. Our method has achieved state-of-the-art accuracy on the public autonomous driving dataset KITTI-2015. Code is publicly available at https://***/MasterHow/CSFlow.

关键词： Optical losses Strips visualization Image motion analysis Correlation Estimation Network architecture

来源：评论

学校读者我要写书评

暂无评论

Exploiting Reduced Precision for GPU-based Time Series Mining 36

Exploiting Reduced Precision for GPU-based Time Series Minin...

引用

36th ieee International Parallel and Distributed Processing symposium (ieee IPDPS)

作者： Ju, Yi Raoofy, Amir Yang, Dai Laure, Erwin Schulz, Martin Max Plank Comp & Data Facil Munich Germany Tech Univ Munich Munich Germany NVIDIA GmbH Wurselen Germany

ISBN: (纸本)9781665481069

The mining of multi-dimensional time series is a crucial step in gaining insights into data obtained from physical systems and from monitoring infrastructures. A widely accepted approach for this challenge is the matrix profile, which, however, is computationally very expensive. It relies on calculating large correlation matrices coupled with sort operations across all dimensions of the data, as well as on performing inclusive scans. All of these steps are inherently data parallel and can, therefore, benefit from execution on GPUs, and even more so from horizontal scaling on multiple GPUs. In addition, the nature of the matrix profile calculation allows the exploitation of reduced precision on GPUs. This offers further improvements to enable the analysis of ever growing data sets in real-world scenarios. Based on these motivations, we introduce the first parallel algorithm for multi-dimensional matrix profile on multiple GPUs exploiting reduced precision modes and provide a highly optimized implementation using novel optimization techniques. On one NVIDIA A100 GPU, our implementation achieves a 54x performance improvement in comparison to an optimized singlenode execution on a state-of-the-art CPU-based implementation relying on double-precision computation and an additional factor of 1.4x when switching to reduced precision while maintaining sufficient accuracy. We study the accuracy and performance trade-offs for our proposed algorithm in detail and present synthetic and real-world case studies to demonstrate how the reduced precision improves the performance, while accomplishing sufficiently accurate results.

关键词： Multi-GPU algorithms data mining reduced precision multi-dimensional time series matrix profile

来源：评论

学校读者我要写书评

暂无评论

Applying visual analytics to firearms trafficking trails for the extraction of strategic intelligence

Applying visual analytics to firearms trafficking trails for...

引用

ieee International Conference on Big data

作者： E. Adamakis E. Sykianaki G. Margetis S. Ntoa C. Stephanidis Institute of Computer Science Foundation for Research and Technology Hellas Heraklion Greece Department of Computer Science University of Crete Heraklion Greece

ISBN: (数字)9798350362480

ISBN: (纸本)9798350362497

Big data analysis and insight extraction are critical in contemporary large-scale criminal investigations. Analyzing large sets of information related to criminal activities can assist in the identification of correlations between records which can be utilized as strategic intelligence information for the coordination of law enforcement efforts. A Law Enforcement Agency (LEA) officer tasked with this analysis requires tools that can enable Big data exploration while being intuitive and without requiring the need for prior knowledge. A common practice of data analysis software is the ability to create dashboards that can provide all the necessary visualizations and filters to support such an analysis. However, when it comes to dashboard-wide data exploration through drill-down or roll-up analysis; review and evaluation of correlations between data records; and creation of collections with information important to an investigation, current solutions fall short. In this paper, we present a system that addresses these limitations by facilitating the visualization and exploration of diverse data sources, allowing for individual or concurrent analysis via user-configured dashboards. Individual data records of interest to the analysis can be reviewed and their correlations evaluated by the user, allowing for the creation of collections of records and their relationships based on formalized types of correlations.

关键词： Correlation Filters data analysis Law enforcement Reviews Visual analytics Soft sensors data visualization Big data data mining

来源：评论

学校读者我要写书评

暂无评论

Taking a Deeper Look at the Brain: Predicting Visual Perceptual and Working Memory Load From High-Density fNIRS data

引用

ieee JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 2022年第5期26卷 2308-2319页

作者： Wang, Jiyang Grant, Trevor Velipasalar, Senem Geng, Baocheng Hirshfield, Leanne Syracuse Univ Dept Elect Engn & Comp Sci Syracuse NY 13244 USA Univ Colorado Inst Cognit Sci Boulder CO 80309 USA Univ Alabama Birmingham Dept Comp Sci Birmingham AL 35294 USA

Predicting workload using physiological sensors has taken on a diffuse set of methods in recent years. However, the majority of these methods train models on small datasets, with small numbers of channel locations on the brain, limiting a model's ability to transfer across participants, tasks, or experimental sessions. In this paper, we introduce a new method of modeling a large, cross-participant and cross-session set of high density functional near infrared spectroscopy (fNIRS) data by using an approach grounded in cognitive load theory and employing a Bi-Directional Gated Recurrent Unit (BiGRU) incorporating attention mechanism and self-supervised label augmentation (SLA). We show that our proposed CNN-BiGRU-SLA model can learn and classify different levels of working memory load (WML) and visual processing load (VPL) across participants. Importantly, we leverage a multi-label classification scheme, where our models are trained to predict simultaneously occurring levels of WML and VPL. We evaluate our model using leave-one-participant-out (LOOCV) as well as 10-fold cross validation. Using LOOCV, for binary classification (off/on), we reached an F1-score of 0.9179 for WML and 0.8907 for VPL across 22 participants (each participant did 2 sessions). For multi-level (off, low, high) classification, we reached an F1-score of 0.7972 for WML and 0.7968 for VPL. Using 10-fold cross validation, for multi-level classification, we reached an F1-score of 0.7742 for WML and 0.7741 for VPL.

关键词： Functional near-infrared spectroscopy visualization Task analysis Brain modeling Adaptive systems Spatial resolution data models Cognitive load working memory load workload classification deep learning fNIRS self-supervision

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：