检索结果-内蒙古大学图书馆

22nd ACM Conference on Embedded Networked Sensor Systems, SenSys 2024

作者： Weng, Yuxuan Wu, Guoquan Zheng, Tianyue Yang, Yanbing Luo, Jun Department of Computer Science and Engineering Southern University of Science and Technology China Smart City Center Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology China College of Computer Science Sichuan University China College of Computing and Data Science Nanyang Technological University Singapore Singapore

ISBN: (纸本)9798400706974

Radio-Frequency (RF)-based Human Activity Recognition (HAR) rises as a promising solution for applications unamenable to techniques requiring computer visions. However, the scarcity of labeled RF data due to their non-interpretable nature poses a significant obstacle. Thanks to the recent breakthrough of foundation models (FMs), extracting deep semantic insights from unlabeled visual data become viable, yet these vision-based FMs fall short when applied to small RF datasets. To bridge this gap, we introduce FM-Fi, an innovative cross-modal framework engineered to translate the knowledge of vision-based FMs for enhancing RF-based HAR systems. FM-Fi involves a novel cross-modal contrastive knowledge distillation mechanism, enabling an RF encoder to inherit the interpretative power of FMs for achieving zero-shot learning. It also employs the intrinsic capabilities of FM and RF to remove extraneous features for better alignment between the two modalities. The framework is further refined through metric-based few-shot learning techniques, aiming to boost the performance for predefined HAR tasks. Comprehensive evaluations evidently indicate that FM-Fi rivals the effectiveness of vision-based methodologies, and the evaluation results provide empirical validation of FM-Fi's generalizability across various environments. © 2024 Copyright is held by the owner/author(s).

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Epigraph Based Multilevel Optimization (EMO) for Enhancing Chain-of-Thought Reasoning Capabilities

Epigraph Based Multilevel Optimization (EMO) for Enhancing C...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Lu, Songtao Ding, Yanna Horesh, Lior Gao, Jianxi Magdon-Ismail, Malik Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong IBM Research Thomas J. Watson Research Center Yorktown HeightsNY10598 United States Department of Computer Science Rensselaer Polytechnic Institute TroyNY12180 United States

ISBN: (纸本)9798350368741

Chain-of-thought (CoT) reasoning applies to complex tasks with multiple intermediate steps, a key feature of large language models. Recent studies have revealed CoT as a composition of in-context filtering and learning. This paper proposes a unified framework for CoT optimization that exploits the nested problem structure to formulate training as multilevel optimization. Each intermediate reasoning step is a distinct optimization level. We develop an epigraph-based multilevel optimization (EMO) method to iteratively find the optimal solution for this class of problems. Experiments using GPT-2 show that the proposed EMO achieves the lowest generalization errors across all intermediate steps compared to state-of-the-art, highlighting the importance of nested optimization approaches for CoT reasoning. © 2025 IEEE.

关键词： Chain-of-thought (CoT) epigraph-based multilevel optimization (EMO) large language models (LLMs) multilevel optimization (MLO)

来源：评论

学校读者我要写书评

暂无评论

Real Transportation Data Benchmark: Advancing Efficiency through ML-Based Waiting Time Prediction 27

Real Transportation Data Benchmark: Advancing Efficiency thr...

引用

27th IEEE International Conference on Intelligent Transportation Systems, ITSC 2024

作者： Klinger, Julian Mohebbi, Mohammad Reza Doller, Mario University of Applied Sciences Kufstein Tirol Kufstein Austria University of Passau Department of Computer Science Passau Germany Josef Ressel Center Vision2Move University of Applied Science S Kufstein Tirol Kufstein Austria

ISBN: (纸本)9798331505929

In transport and supply chain management (SCM), accurate prediction of waiting times is pivotal for optimizing resource allocation, minimizing costs, and operational efficiency. However, this task is filled with challenges due to numerous unknown variables, including congestion, weather conditions, and varying operational requirements. This paper addresses these challenges by predicting the waiting times of transporters in transport management and SCM. By analyzing key performance indicators in the planning process of production measures and delivery cycles, we have identified an important research direction for improving the accuracy of forecasting waiting times. Therefore, we introduce a novel dataset based on real data, focusing on delivery cycles, by analyzing GPS-Trajectories and planning data of 165 transporters on 550 production measures in a span of one year and condensing them into a total of 6144 delivery records. To set a performance baseline, we analyze this dataset and apply several Machine Learning models, such as AdaBoost, SVR and MLP, for waiting time prediction by providing different metrics such as MAE, MSE and the R2-Score. We conclude that, with our approach, it is possible to successfully predict waiting time in the form of trends already in the planning phase of transport management. The dataset is available on GitHub: ***/KlingerjU/Real-Transportation-Data-Benchmark © 2024 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

SketchGPT: Autoregressive Modeling for Sketch Generation and Recognition

arXiv

引用

arXiv 2024年

作者： Tiwari, Adarsh Biswas, Sanket Lladós, Josep Computer Vision Center Catalonia Spain Computer Science Department Universitat Autònoma de Barcelona Catalonia Spain

We present SketchGPT, a flexible framework that employs a sequence-to-sequence autoregressive model for sketch generation, and completion, and an interpretation case study for sketch recognition. By mapping complex sketches into simplified sequences of abstract primitives, our approach significantly streamlines the input for autoregressive modeling. SketchGPT leverages the next token prediction objective strategy to understand sketch patterns, facilitating the creation and completion of drawings and also categorizing them accurately. This proposed sketch representation strategy aids in overcoming existing challenges of autoregressive modeling for continuous stroke data, enabling smoother model training and competitive performance. Our findings exhibit SketchGPT’s capability to generate a diverse variety of drawings by adding both qualitative and quantitative comparisons with existing state-of-the-art, along with a comprehensive human evaluation study. The code and pretrained models will be released on this GitHub† © 2024, CC BY-SA.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Count-Data Mixed Models of Topical Tweets: A Case of Indonesia Flood Events 8

Count-Data Mixed Models of Topical Tweets: A Case of Indones...

引用

8th International Conference on computer science and Computational Intelligence, ICCSCI 2023

作者： Hidayat, Alam Ahmad Pardamean, Bens Mathematics Department School of Computer Science Bina Nusantara University Jakarta11480 Indonesia Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta11480 Indonesia Computer Science Department BINUS Graduate Program - Master of Computer Science Program Bina Nusantara University Jakarta11480 Indonesia

Examining topic-level variability in modeling Twitter data can potentially yield more comprehensive insights into public perception during critical periods, thereby enhancing natural disaster mitigation and surveillance efforts. In this study, we utilized generalized linear mixed models (GLMMs) to illustrate the variability in tweet counts related to specific topics in Indonesia during the flood events that occurred in February 2021. The glmmTMB library in R was employed for this purpose. The data were assumed to follow two distinct exponential distributions: Poisson and Negative Binomial. To incorporate random effects, random intercepts and random slopes were introduced, allowing them to vary randomly across topics in the initial two models. Additionally, the final model addressed issues related to dispersion and zero-inflation. By evaluating the Akaike Information Criteria scores, we determined that a model based on the Negative Binomial distribution with random zero-inflation intercepts best fit the data. The chosen model formulation and the estimated parameters have the potential to forecast topic-specific trends in Indonesian flood-related Twitter data. © 2023 The Authors. Published by Elsevier B.V.

关键词： count data disaster surveillance floods mixed models Twitter

来源：评论

学校读者我要写书评

暂无评论

Data mining for the global multiplex weekly average income analysis

Data mining for the global multiplex weekly average income a...

引用

2022 International Conference on ENTERprise Information Systems, centerIS 2022 - International Conference on Project MANagement, ProjMAN 2022 and International Conference on Health and Social Care Information Systems and Technologies, HCist 2022

作者： Dominic, Nicholas Elwirehardja, Gregorius Natanael Pardamean, Bens Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta11480 Indonesia Computer Science Department BINUS Graduate Program Master of Computer Science Program Bina Nusantara University Jakarta11480 Indonesia

The box office (BO) income had significantly declined up to 80% in 2020, as the COVID-19 pandemic emerged. To minimize further financial risks, multiplex (multiple cinema complexes) owners need to analyze their potential income for each movie, each week. Therefore, we developed a proper data mining strategy that allows multiplex owners to analyze and discover insights on how successfully produced movies could be. The methodology comprises (1) data loading and exploration, (2) data cleaning, (3) data selection, integration, and transformation using Pentaho, (4) data mining in which the results were stored in the MySQL database, and (5) pattern evaluation and presentation using Qlik Sense as the Business Intelligence (BI) dashboard. Based on our data mining methodology, we revealed that drama, comedy, action, and thriller are favorite genres. We also found that DreamWorks Animation and Pixar Animation Studios are both the most popular production houses, even Apatow Productions and Escape Artists still have the biggest revenue on average. © 2022 The Authors. Published by Elsevier B.V.

关键词： Animation

来源：评论

学校读者我要写书评

暂无评论

Decision theory and risk simulation analysis for optimizing profit in paylater services

Decision theory and risk simulation analysis for optimizing ...

引用

作者： Dominic, Nicholas Pardamean, Bens Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta11480 Indonesia Computer Science Department BINUS Graduate Program Master of Computer Science Program Bina Nusantara University Jakarta11480 Indonesia

The projected increase in PayLater utilization reaches up to five million people by 2025. To optimize the yearly profit from their PayLater service, fintech companies must examine all possible risks before a unanimous decision is taken. Therefore, we proposed a unified decision framework derived from decision theory and the Monte Carlo simulation technique. Two schemes were coined: (1) a decision-making scheme, and (2) a risk simulation scheme. Throughout experiments, the framework was able to estimate several alternative decisions and their impacts, analyze the causes of failure and delays in the development of the PayLater service, and execute Monte Carlo simulations in up to 10,000 trials. Outputs of this study will benefit decision-makers in the fintech initiative before launching their PayLater products. © 2023 Elsevier B.V.. All rights reserved.

关键词： Fault tree analysis

来源：评论

学校读者我要写书评

暂无评论

Forecasting Air Quality Index in Thailand Using Ensemble Method 26

Forecasting Air Quality Index in Thailand Using Ensemble Met...

引用

26th International computer science and Engineering Conference, ICSEC 2022

作者： Lertnilkarn, Saksiri Phimoltares, Suphakant Research Center Department of Mathematics and Computer Science Faculty of Science Pathumwan Bangkok Thailand

ISBN: (纸本)9781665491983

Air pollution is one of the most serious problems in many regions of the world. Thailand also has had to face this trouble unavoidably, especially in the northern region of Thailand, the area that has been highly contaminated by air pollution for so many years. In this paper, an ensemble method was introduced to forecast the level of air quality index (AQI) in the northern part of Thailand. The ensemble method, in this study, is a technique gaining the results from majority vote of outputs of three classification models - k-nearest neighbors, random forest, and support vector machine. The proposed model compared the voted accuracy with the accuracies of existing classification models. It made use of the 2018 - 2021 data from seven stations in four provinces of Northern Thailand. In the end, the proposed model yielded 99.68% - 99.84% accuracy rate on average higher than most of the performance of the other comparative models. © 2022 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Factors Affecting the Willingness to Share Data in Mobile Health (mHealth) Apps in Indonesia 25

Factors Affecting the Willingness to Share Data in Mobile He...

引用

25th International Seminar on Intelligent Technology and Its Applications, ISITIA 2024

作者： Maulana, Fairuz Iqbal Lestari, Dian Rahayu, Anita Adi, Puput Dani Prasetyo Puspitasari, Chasandra Persada, Satria Fadil Bina Nusantara University School of Computer Science Computer Science Department Jakarta11480 Indonesia Faculty of Medicine Surabaya60132 Indonesia Research Center for Telecommunication Bandung Indonesia Bina Nusantara University Binus Business School Undergraduate Program Entrepreneurship Department Jakarta Indonesia

ISBN: (纸本)9798350378573

This study investigates the factors influencing user willingness to share data in mobile health (m-health) applications in Indonesia, integrating the Technology Acceptance Model, Theory of Planned Behavior, and Privacy Calculus Theory through Structural Equation Modeling (SEM). Data were collected via an online questionnaire from 103 Indonesian active users of m-health platforms, aged 15-55, focusing on applications that provide online consultations and health-related information. The SEM analysis, validated by indices including the Goodness of Fit Index and Root Mean Square Error of Approximation, reveals a strong model fit. The findings highlight that perceived benefits, such as improved health outcomes and personalized advice, are the primary drivers for users' willingness to share data, significantly outweighing concerns over privacy risks. Perceived usefulness and trust in the application also significantly affect data-sharing behavior, while ease of use and social influences have a moderate to minor influence, respectively. These insights are critical for application developers and policymakers in enhancing app utility and user engagement while addressing privacy concerns. The study's limitations include its demographic scope and reliance on self-administered questionnaires, pointing to directions for future research to broaden demographic representation and deepen qualitative insights. © 2024 IEEE.

关键词： data sharing mHealth privacy concerns SEM analysis structural equation modeling

来源：评论

学校读者我要写书评

暂无评论

GeoContrastNet: Contrastive Key-Value Edge Learning for Language-Agnostic Document Understanding

arXiv

引用

arXiv 2024年

作者： Biescas, Nil Boned, Carlos Lladós, Josep Biswas, Sanket Computer Vision Center Catalonia Spain Computer Science Department Universitat Autònoma de Barcelona Catalonia Spain

This paper presents GeoContrastNet, a language-agnostic framwork to structured document understanding (DU) by integrating a contrastive learning objective with graph attention networks (GATs), emphasizing the significant role of geometric features. We propose a novel methodology that combines geometric edge features with visual features within an overall two-staged GAT-based framework, demonstrating promising results in both link prediction and semantic entity recognition performance. Our findings reveal that combining both geometric and visual features could match the capabilities of large DU models that rely heavily on Optical Character Recognition (OCR) features in terms of performance accuracy and efficiency. This approach underscores the critical importance of relational layout information between the named text entities in a semi-structured layout of a page. Specifically, our results highlight the model’s proficiency in identifying key-value relationships within the FUNSD dataset for forms and also discovering the spatial relationships in table-structured layouts for RVLCDIP business invoices. Our code is accessible on this GitHub† © 2024, CC BY-SA.

关键词： Optical character recognition

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：