检索结果-内蒙古大学图书馆

15th International Conference on Information and Communication Technology Convergence, ICTC 2024

作者： Bae, Sungho Seo, Giup Hwang, Euiseok School of Elect. Eng. & Comp. Sci. Gwangju Inst. Sci. & Tech. Gwangju Korea Republic of

ISBN: (纸本)9798350364637

In this study, we propose a novel method for predicting the number of household members using smart meter electricity consumption data. Utilizing various feature selection techniques, including SHAP values, mutual information, feature importance, and a proposed method involving pairwise grouping and selection based on importance testing, we aimed to enhance the model's predictive accuracy. The proposed method involves selecting time points based on specific time ranges and overlapping important features from other techniques. The analysis revealed that time periods such as morning, evening, and before bedtime are most informative for estimating household member numbers, while periods with lower activity, like lunchtime to early evening, are less useful. © 2024 IEEE.

关键词： feature selection mutual information predictive modeling SHAP values smart meter data socio-demographic information

来源：评论

学校读者我要写书评

暂无评论

Characterizing Human and Zero-Shot GPT-3.5 Object-Similarity Judgments

Characterizing Human and Zero-Shot GPT-3.5 Object-Similarity...

引用

2024 Findings of the Association for comp.tational Linguistics: NAACL 2024

作者： McKnight, D. Estes Fyshe, Alona Dept. Computing Science University of Alberta Alberta Machine Intelligence Inst. Canada Depts. Comp. Sci. and Psychology University of Alberta Alberta Machine Intelligence Inst. Canada

ISBN: (纸本)9798891761193

Recent advancements in large language models' (LLMs) capabilities have yielded few-shot, human-comp.rable performance on a range of tasks. At the same time, researchers expend significant effort and resources gathering human annotations. At some point, LLMs may be able to perform some simple annotation tasks, but studies of LLM annotation accuracy and behavior are sparse. In this paper, we characterize OpenAI's GPT-3.5's judgment on a behavioral task for implicit object categorization. We characterize the embedding spaces of models trained on human vs. GPT responses and give similarities and differences between them, finding many similar dimensions. We also find that despite these similar dimensions, augmenting humans' responses with GPT ones drives model divergence across the sizes of datasets tested. © 2024 Association for comp.tational Linguistics.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Kubernetes Edge-Powered Vision-Based Navigation Assistance System for Robotic Vehicles 18

Kubernetes Edge-Powered Vision-Based Navigation Assistance S...

引用

18th IEEE International Wireless Communications and Mobile comp.ting, IWCMC 2022

作者： Tian, Jung-Syuan Huang, Szu-Chieh Yang, Shun-Ren Lin, Phone National Tsing Hua University Dept. of Comp. Sci. Inst. of Info. Sys. and App. Inst. of Commun. Eng. Hsinchu Taiwan National Taiwan University Dept. of Comp. Sci. and Info. Eng. Taipei Taiwan

ISBN: (纸本)9781665467490

In recent years, more and more developers have been investigating robotic vehicles. Generally, the operations of robotic vehicles rely on navigation assistance systems, which make recommendations to guide robotic vehicles step-by-step using low-cost cameras until reaching the destinations. Some developers have gradually transferred relevant comp.ting tasks of robotic vehicles without powerful comp.ting power and training models to the edge comp.ting platforms. However, none of such existing edge comp.ting based navigation assistance systems can immediately scale a sufficient number of inst.nces to adapt to the changing requested loads of robotic vehicles. In this paper, we propose a Kubernetes edge-powered vision-based navigation assistance system with a novel auto-scaling algorithm, allowing robotic vehicles to request navigation-related services. Once the requested load does not match the current load, the number of inst.nces can be auto-scaled on demand. In order to evaluate the performance of our auto-scaling algorithm, we comp.re it with two selected auto-scaling algorithms. The experiment results demonstrate that our algorithm can immediately scale up to the most appropriate number of inst.nces to reduce the latency of requests. © 2022 IEEE.

关键词： Navigation

来源：评论

学校读者我要写书评

暂无评论

Socio-Demographic Information Estimation via Smart Meter Data Analysis and Feature Selection

Socio-Demographic Information Estimation via Smart Meter Dat...

引用

International Conference on Information and Communication Technology Convergence (ICTC)

作者： Sungho Bae Giup Seo Euiseok Hwang School of Elect. Eng. & Comp. Sci. Gwangju Inst. Sci. & Tech. Gwangju South Korea

ISBN: (数字)9798350364637

ISBN: (纸本)9798350364644

关键词： Meters Accuracy Electricity Predictive models Feature extraction Smart meters Robustness Information and communication technology Mutual information Testing

来源：评论

学校读者我要写书评

暂无评论

MouseScholar: Evaluating an Image+Text Search System for Biocuration

MouseScholar: Evaluating an Image+Text Search System for Bio...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Trabucco, Juan Trelles Floricel, Carla Arighi, Cecilia Shatkay, Hagit Raciti, Daniela Ringwald, Martin Marai, G. Elisabeta University of Illinois Chicago Dept. Comp. Sci. ChicagoIL60607 United States University of Delaware Dept. Comp. Info. Sci. NewarkDE19716 United States California Inst. Technology Div. Biology Biological Eng. PasadenaCA91125 United States The Jackson Laboratory Bar HarborME04609 United States

ISBN: (纸本)9798350337488

Biocuration is the process of analyzing biological or biomedical articles to organize biological data into data repositories using taxonomies and ontologies. Due to the expanding number of articles and the relatively small number of biocurators, automation is desired to improve the workflow of assessing articles worth curating. As figures convey essential information, automatically integrating images may improve curation. In this work, we inst.ntiate and evaluate a first-in-kind, hybrid image+text document search system for biocuration. The system, MouseScholar, leverages an image modality taxonomy derived in collaboration with biocurators, in addition to figure segmentation, and classifiers comp.nents as a back-end and a streamlined front-end interface to search and present document results. We formally evaluated the system with ten biocurators on a mouse genome informatics biocuration dataset and collected feedback. The results demonstrate the benefits of blending text and image information when presenting sci.ntific articles for biocuration. © 2023 IEEE.

关键词： biocuration document search

来源：评论

学校读者我要写书评

暂无评论

Blockchain and Artificial Intelligence: Synergies and Conflicts

arXiv

引用

arXiv 2024年

作者： Witt, Leon Fortes, Armando Teles Toyoda, Kentaroh Samek, Wojciech Li, Dan Dept. of Comp. Sci. & Tech. Tsinghua University Fraunhofer HHI China Dept. of Comp. Sci. & Tech. Tsinghua University China Inst. of High Performance Computing A*STAR Singapore Dept. of Elec. Eng. & Comp. Sci. TU Berlin Fraunhofer HHI Germany

Blockchain technology and Artificial Intelligence (AI) have emerged as transformative forces in their respective domains. This paper explores synergies and challenges between these two technologies. Our research analyses the biggest projects combining blockchain and AI, based on market capitalization, and derives a novel framework to categorize contemporary and future use cases. Despite the theoretical comp.tibility, current real-world applications combining blockchain and AI remain in their infancy. © 2024, CC BY.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

An efficient discrimination discovery method for fairness testing 34

An efficient discrimination discovery method for fairness te...

引用

34th International Conference on Software Engineering and Knowledge Engineering, SEKE 2022

作者： Sano, Shinya Kitamura, Takashi Takada, Shingo Dept. of Info. and Comp. Sci. Keio University Yokohama Japan Nat. Inst. of Advanced Industrial Science and Technology Tokyo Japan

ISBN: (纸本)1891706543

With the increasing use of machine learning software in our daily life, software fairness has become a growing concern. In this paper, we propose an individual fairness testing technique called KOSEI. Individual fairness is one of the central concepts in software fairness. Testing individual fairness aims to detect individual discriminations included in the software. KOSEI is based on AEQUITAS by Udeshi et al., a landmark fairness testing technique featuring a two-step search strategy of global and local search. KOSEI improves the local search part of AEQUITAS, based on our insight to overcome the limitations of the local search of AEQUITAS. Our experiments show that KOSEI outperforms AEQUITAS by orders of magnitude. KOSEI, on average, detects 5,084.8% more discriminations than AEQUITAS, in just 7.5% of the execution time. © 2022 Knowledge Systems inst.tute Graduate School. All rights reserved.

关键词： Software testing

来源：评论

学校读者我要写书评

暂无评论

EADD: An Intelligent Edge-Based Anomaly Detection Platform for Car Driving

EADD: An Intelligent Edge-Based Anomaly Detection Platform f...

引用

IEEE International Conference on Communications (ICC)

作者： En-Hau Yeh Yu-Ming Chen Phone Lin Shun-Ren Yang Rongxing Lu Dept. of Comp. Sci. & Info. Eng. National Taiwan University Taipei Taiwan Dept. of Comp. Sci. & Info. Eng. GIBEBI and GINM National Taiwan University Taipei Taiwan Dept. of Comp. Sci. and the Inst. of Comm. Eng. National Tsing Hua University Hsinchu Taiwan Faculty of Comp. Sci. University of New Brunswick Fredericton Canada

ISBN: (数字)9781728190549

ISBN: (纸本)9781728190556

Detecting abnormal driving behavior is crucial for preventing traffic accidents, as they are responsible for a sig-nificant majority of incidents. However, existing methods for detection often come with high costs or execution restrictions. In this paper, we introduce EADD, an Edge-based Anomaly Detection platform for Driving behavior. EADD overcomes these limitations by detecting abnormal driving behavior without the need for additional sensors or restrictions. Additionally, EADD boasts low comp.tational requirements and enables real-time detection on mobile devices like the Raspberry Pi 3 Model B.

关键词： Costs Accuracy Image edge detection comp.tational modeling Real-time systems Mobile handsets Automobiles

来源：评论

学校读者我要写书评

暂无评论

Location Information Measurements Using Multiple Devices Toward Digital Transformation of Open Sky Museum 7

Location Information Measurements Using Multiple Devices Tow...

引用

7th IEEE International Conference on Artificial Intelligence and eXtended and Virtual Reality, AIxVR 2025

作者： Saito, Yasuyuki Uemura, Shigeki Okuzumi, Jun Suzuki, Norihiko Umezu, Nobuyuki Nat. Inst. of Tech. Kisarazu College Dept. of Inf. Eng. Kisarazu Japan Nat. Inst. of Tech. Kisarazu College Dept. of Civil Eng. Kisarazu Japan Shibayama Ancient-tomb and Clay-image Museum Shibayama-machi Board of Education Shibayama-machi Japan N&s Support Office Kurashiki Japan Ibaraki University Dept. of Comp. and Inf. Sci. Hitachi Japan

ISBN: (纸本)9798331521578

The purpose of this study is to comp.re the results of measuring location information using multiple devices. In the Open Sky Museum for digital transformation, accurate location information is necessary for presenting information on a terminal. We comp.re the location information obtained by receiving the quasi-zenith satellite system (QZSS) L6 correction signal with the Google API on Android tablets that we have used in our previous study. We demonstrate that the location information can be obtained stably using the QZSS L6 signal, and examine the measurement error using multiple Android devices of the same model. © 2025 IEEE.

关键词： Velocity measurement

来源：评论

学校读者我要写书评

暂无评论

On Poisoning Attacks and Defenses for LSTM Time Series Prediction Models: Speed Prediction as an Example

On Poisoning Attacks and Defenses for LSTM Time Series Predi...

引用

International Wireless Communications and Mobile comp.ting Conference, IWCMC

作者： Yi-Yu Chen Hui-Nien Hung Shun-Ren Yang Chia-Cheng Yen Phone Lin Dept. of Comp. Sci Inst. of Info. Sec. and Inst. of Commun. Eng National Tsing Hua University Taiwan Inst. of Stats National Yang Ming Chiao Tung University Taiwan Dept. of Comp. Sci. & Info. Eng Grad. Inst. of Biomed. Elec. & Bioinform. Grad. Inst. of Net. & MM Inst. of Med. Device & Imaging and IoT Research Center National Taiwan University Taiwan

ISBN: (数字)9798350361261

ISBN: (纸本)9798350361278

The Long Short-Term Memory (LSTM) model has significantly improved time series prediction accuracy, but also brought forth concerns regarding reliability and security with its widespread adoption, particularly in the context of poisoning attacks. While there is substantial research on attacks and defenses for LSTM models, there’s limited focus on LSTM time series prediction models. In this paper, we propose an arithmetic-based poisoning attack methodology for a demonstrative LSTM time series speed prediction model. Furthermore, we employ the “red team/blue team exercises” commonly used in network security to develop defense strategies using support vector machine and linear regression analysis methods. Through the system-level simulation experiments, we verify the effectiveness of our proposed methodology. Our experiment results indicate that, regarding attacks, our methodology can identify the optimal attacks for the representative road segments. As for defenses, we demonstrate that the defended model’s performance is close to the real model’s performance.

关键词： Support vector machines Wireless communication Roads comp.tational modeling Time series analysis Linear regression Predictive models

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：