检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Poulis, Angelos Tsalapati, Eleni Koubarakis, Manolis Dept. of Computer Science Boston University United States Athens Technology Center Greece AI Team Dept. of Informatics and Telecommunications National and Kapodistrian University of Athens Greece Archimedes Athena RC Greece

Recent advancements in transformer-based language models have sparked research into their logical reasoning capabilities. Most of the benchmarks used to evaluate these models are simple: generated from short (fragments of) first-order logic sentences with only a few logical operators and quantifiers. We construct the natural language dataset, DELTAD, using the expressive description logic language ALCQ. DELTAD comprises 384K examples and increases in two dimensions: i) reasoning dept., and ii) linguistic complexity. In this way, we systematically investigate the logical reasoning capabilities of a supervised fine-tuned DeBERTa-based model and two large language models (GPT-3.5, GPT-4) with few-shot prompting. We show that the DeBERTa-based model fine-tuned on our dataset can master the entailment checking task. Moreover, the performance of GPTs can improve significantly even when a small number of samples is provided (9 shots). We open-source our code and datasets. © 2024, CC BY.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Synergistic Integration of Photovoltaic and the Grid: An Intelligent ANFIS-Governed Interleaved Boost Converter Approach

Synergistic Integration of Photovoltaic and the Grid: An Int...

引用

2024 International Conference on Advancement in Renewable Energy and Intelligent Systems, AREIS 2024

作者： Balaji, V. Ganga, M. Barathi, K. Gouda, Pramod Kumar Prasad, Rajesh Venkatesh, Mammula MAI-NEFHI College of Engineering and Technology Department of Electrical and Computer Engineering Asmara Eritrea Hindusthan College of Engineering and Technology Department of Biomedical Engineering Coimbatore India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Dept. of Eee Tamil Nadu India Aditya Institute of Technology and Management Dept. of Eee Andhra Pradesh Srikakulam India Stella Mary's College of Engineering Dept. of Eee Tamil Nadu India Avanthi's Research and Technological Academy Dept. of Eee A. P Vizianagaram531162 India

ISBN: (纸本)9798350387230

The transition towards Renewable Energy Sources (RES), mainly solar power is crucial in addressing the depletion of fossil fuel-based energy supplies and mitigating environmental concerns. This paper proposes an innovative approach to enhance efficiency of PhotoVoltaic (PV) systems by integrating a two-phase Interleaved Boost (IB) converter with an Adaptive Neuro Fuzzy Inference System (ANFIS) controller. The ANFIS controller optimizes the operation of the IB converter by adjusting the duty cycle based on two key input parameters: solar irradiation and temperature. This adaptive control method effectively transforms solar energy into electrical power, maximizing the PV system's power production. By using the ANFIS mechanism, the suggested system improves total output power, decreases overshoot, and produces a higher voltage output than conventional techniques, all of which lead to increased performance and dependability. The IB converter operates with an efficiency of 96.15% and minimized settling time. © 2024 IEEE.

关键词： Adaptive control systems

来源：评论

学校读者我要写书评

暂无评论

Prediction of Continuous Targets by Explainable Imbalanced Regression from Omics Data in Childhood Obesity 1

Prediction of Continuous Targets by Explainable Imbalanced R...

引用

1st Workshop on Explainable Artificial Intelligence for the Medical Domain, EXPLIMED 2024

作者： Arteaga, María Torres-Martos, Álvaro Anguita-Ruiz, Augusto Alcalá-Fdez, Jesús Alcalá, Rafael Gacto, María José Dept. of Computer Science and Artificial Intelligence Univ. of Granada Granada Spain Center of Biomedical Research Univ. of Granada Inst. de investigación Biosanitaria ibs.GRANADA Granada Spain Inst. de Salud Carlos III Madrid Spain Barcelona Spain Univ. of Granada Granada Spain Dept. of Software Engineering Univ. of Granada Granada Spain

Childhood obesity is a persistent challenge for society since it is highly related to insulin resistance and a wide range of other chronic diseases, which impair not only the health of the people, but also the health system itself due to extra costs of treating them. In this context, it is imperative to consider innovative approaches, like eXplainable Artificial Intelligence, to understand the factors underlying childhood obesity. Data imbalance has been thoroughly studied in classification, but scarcely studied in regression since classes do not exist. However, extreme values in continuous domains, which are usually minority are often the most clinically relevant. Facing the imbalance in regression while obtaining explainable models has never been studied. In this application work, we adopt a Machine Learning approach to obtain explainable regression models for imbalanced continuous targets in childhood obesity, as HOMA-IR and Waist Circumference. We consider 79 variables, including targeted metabolomics, targeted proteomics, exposomic data (e.g., the physical activity subdomain), hematological parameters and anthropometry of Spanish children from 3 to 18 years with significant imbalance ratios around 12% in HOMA-IR and Waist Circumference. Even though we mainly focus on extensions of linguistic fuzzy rule-based systems, particularly designed for explainability, we also consider highly accurate complementary approaches as Random Forest that could additionally provide contrast interesting information by post-hoc SHAP. The models so obtained are better considering the relevant minority information (17% improvements in F1). Moreover, they seem to properly explain biologically meaningful relations, as in the case of physical activity data or the one with the follicle stimulating hormone, among others. © 2024 Copyright for this paper by its authors.

关键词： Nutrition

来源：评论

学校读者我要写书评

暂无评论

Quality Grading of Beef using Deep Convolutional Neural Network 9

Quality Grading of Beef using Deep Convolutional Neural Netw...

引用

9th International Conference on Information Technology Trends, ITT 2023

作者： Sharma, Aarav Choudhury, Tanupriya Sarkar, Tanmay Mohanty, Madhusmita Bansal, Nikunj Mohanty, Sachi Nandan Informatics Cluster School of Computer Science Uttarakhand Dehradun2480007 India Cse Dept. Uttarakhand Dehradun India Graphic Era Hill University Cse Dept. Uttarakhand Dehradun248002 India Malda Polytechnic West Bengal State Council of Technical Education Govt. of West Bengal Department of Food Processing Technology Malda732102 India Andhra Pradesh Amaravati India VIT-AP University School of Computer Science & Engineering Andhra Pradesh Amaravati5222237 India

ISBN: (纸本)9798350327502

Beef is one of the most widely consumed meat, being an organic substance it is prone to degradation over time. In our paper we have proposed a Convolutional Neural Network model to grade a given sample of Beef and predict its freshness as good or bad. We have proposed a CNN model architecture along with a data (image) pre-processing pipeline that can be used to classify a given Beef sample. We have done comparative analysis between different pre-processing pipelines as well as other CNN based classification models and after training have achieved 99.63% test accuracy. © 2023 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

A Framework for Digital Twin Collaboration 24

A Framework for Digital Twin Collaboration

引用

Proceedings of the Winter Simulation Conference

作者： Zhengchang Hua Karim Djemame Nikos Tziritas Georgios Theodoropoulos Dept. of Computer Science and Eng. Southern University of Science and Technology (SUSTech) Shenzhen CHINA and School of Computing University of Leeds Leeds UK School of Computing University of Leeds Leeds UK Dept. of Informatics and Telecommunications University of Thessaly Lamia GREECE Research Institute of Trustworthy Autonomous Systems (RITAS) SUSTech Shenzhen CHINA and Dept. of Computer Science and Eng. Southern University of Science and Technology (SUSTech) Shenzhen CHINA

ISBN: (纸本)9798331534202

Digital Twins (DTs) have emerged as a powerful tool for modeling Large Complex Systems (LCSs). Their strength lies in the detailed virtual models that enable accurate predictions, presenting challenges in traditionally centralized approaches due to the immense scale and decentralized ownership of LCSs. This paper proposes a framework that leverages the prevalence of individual DTs within LCSs. By facilitating the exchange of decisions and predictions, this framework fosters collaboration among autonomous DTs, enhancing performance. Additionally, a trust-based mechanism is introduced to improve system robustness against poor decision-making within the collaborative network. The framework's effectiveness is demonstrated in a virtual power plant (VPP) scenario. The evaluation results confirm the system's objectives across various test cases and show scalability for large deployments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Product Recommendation Using Collaborative Filtering and K-Means Clustering

Product Recommendation Using Collaborative Filtering and K-M...

引用

Computing, Power and Communication Technologies (IC2PCT), IEEE International Conference on

作者： M Ranjith Kumar S Vishnu G Roshen D Niranjan Kumar P Revathi David Ricardo Ledo Baster Department of Mathematics Amrita School of Engineering Amrita Vishwa Vidyapeetham Chennai India Dept. of Computer Science & Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Chennai India Dept. of Computer Science & Engineering C. Abdul Hakeem College of Engineering and Technology Melvisharam Vellore India Dept. of Informatics and Mathematics University of Holguin Cuba

The growth of e-commerce has altered how consumers shop, providing a digital space where convenience, vast product offerings, and competitive pricing converge. In today’s world, e-commerce websites are transitioning from traditional search-driven methods to customized and intuitive approaches via product suggestions. Product recommendation systems are vital in e-commerce, from bringing new business to retaining existing ones. Our three-part recommendation system is designed so that new users have a great and engaging experience as the Product Popularity -Based System shows them carefully chosen products that are in demand. Collaborative Filtering Recommendations are highly personalized recommendations given to people who have already made their first purchases based on their prior actions and preferences. The K-Means Clustering-Based Recommendation System uses textual clustering analysis to deliver contextually relevant recommendations. We use a variety of evaluation metrics, such as click-through rates, user engagement, and the Silhouette Score, to assess the effectiveness and accuracy of our recommendation systems. Our findings show significant increases in user engagement, conversion rates, and relevant recommendations. Our findings demonstrate the transformative power of well-designed recommendation systems, which improve user experiences and retention and provide invaluable solutions for businesses entering the e-commerce space. This paper provides an in-dept. examination of the multifaceted landscape of e-commerce recommendations, shedding light on their far-reaching implications for customer acquisition and retention in this dynamic digital era.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Maintenance of Model Resilience in Distributed Edge Learning Environments

Maintenance of Model Resilience in Distributed Edge Learning...

引用

International Conference on Intelligent Environments (IE)

作者： Qiyuan Wang Christos Anagnostopoulos Jordi Mateo Fornes Kostas Kolomvatsos Andreas Vrachimis School of Computing Science University of Glasgow Department of Computer Science University of Lleida Dept. of Informatics & Telecommunications University of Thessaly School of Computing Science Glasgow University

Distributed Machine Learning (DML) at the edge of the network involves model learning and inference across networking nodes over distributed data. One type of model learning could be the delivery of predictive analytics services to formulate intelligent environments, however, those environments heavily rely on real-time inference and are significantly influenced by changes in the underlying data (concept drifts). Moreover, the quality of service and availability in DML environments are directly tied to each node’s reliability, since such environments are highly susceptible to the impact of node failures. Even if such challenges can be tackled with distributed resilience mechanisms, their effectiveness and efficiency, due to concept drifts, should be maintained to ensure continuous and sustained quality of service. DML systems operate in dynamic environments, thus, they require their models to be updated according to the novel trends embedded in the new data they encounter. We, therefore, introduce several model maintenance mechanisms to ensure resilient DML systems in the long term when concept drifts emerge. We provide a comprehensive experimental evaluation of our resilience maintenance mechanisms over synthetic and real data showcasing their importance and applicability in edge learning environments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Carotid plaque stroke risk assessment using multiscale AM-FM analysis based on DoG filterbanks

Carotid plaque stroke risk assessment using multiscale AM-FM...

引用

International Conference on Digital Signal Processing (DSP)

作者： Kyriacos P. Constantinou Ioannis P. Constantinou Marios S. Pattichis Efthyvoulos Kyriacou Constantinos S. Pattichis Dept. of Computer Science University of Cyprus Nicosia Cyprus Istognosis Ltd Nicosia Cyprus Dept. of Electrical and Computer Engineering University of New Mexico Albuquerque NM USA Dept. of Electrical Engineering Computer Engineering and Informatics Cyprus University of Technology Limassol Cyprus

The objective of this work was the investigation of multiscale Amplitude Modulation - Frequency Modulation (AM-FM) analysis based on Difference of Gaussians (DoG) filterbanks representations in order to predict the risk of stroke by analysing carotid plaques ultrasound images of individuals with asymptomatic carotid stenosis. We computed the instantaneous amplitude, instantaneous phase and the magnitude of instantaneous frequency to extract histogram features on each plaque region. The Support Vectors Machine classifier was implemented to classify asymptomatic versus symptomatic plaques. A dataset of 100 carotid plaque images (50 asymptomatic and 50 symptomatic) were tested, and showed that the AM-FM features based on DoG filterbanks and simple histograms performed better than the traditional AM-FM features. Best results were obtained when an eight scale filterbank with a combination of scales was used reaching the accuracy of 75%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

REVISITING SILHOUETTE AGGREGATION

arXiv

引用

arXiv 2024年

作者： Pavlopoulos, John Vardakas, Georgios Likas, Aristidis Department of Informatics Athens University of Economics and Business Archimedes Athena RC Greece Dept. of Computer Science & Engineering University of Ioannina IoanninaGR 45110 Greece

Silhouette coefficient is an established internal clustering evaluation measure that produces a score per data point, assessing the quality of its clustering assignment. To assess the quality of the clustering of the whole dataset, the scores of all the points in the dataset are typically (micro) averaged into a single value. An alternative path, however, that is rarely employed, is to average first at the cluster level and then (macro) average across clusters. As we illustrate in this work with a synthetic example, the typical micro-averaging strategy is sensitive to cluster imbalance while the overlooked macro-averaging strategy is far more robust. By investigating macro-Silhouette further, we find that uniform sub-sampling, the only available strategy in existing libraries, harms the measure’s robustness against imbalance. We address this issue by proposing a per-cluster sampling method. An experimental study on eight real-world datasets is then used to analyse both coefficients in two clustering tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： Cluster analysis

来源：评论

学校读者我要写书评

暂无评论

TRAP: Tailored Robustness Assessment for Index Advisors via Adversarial Perturbation

TRAP: Tailored Robustness Assessment for Index Advisors via ...

引用

International Conference on Data Engineering

作者： Wei Zhou Chen Lin Xuanhe Zhou Guoliang Li Tianqing Wang Shanghai Artificial Intelligence Laboratory School of Informatics Xiamen University China Dept. of Computer Science Tsinghua University Beijing China Huawei Company Beijing China

ISBN: (数字)9798350317152

ISBN: (纸本)9798350317169

Many index advisors have recently been proposed to build indexes automatically to improve query performance. However, they mainly consider performance improvement in static scenarios. Their robustness, i.e., stable performance in dynamic scenarios (e.g., with minor workload changes), has not been well investigated. This paper addresses the challenges of assessing the index advisor's robustness from the following aspects. First, we introduce perturbation-based workloads for robustness assessment and identify three typical perturbation constraints that occur in real scenarios. Second, with the perturbation constraints, we formulate the generation of perturbed queries as a sequence-to-sequence problem and propose TRAP (Tailored Robustness assessment via Adversarial Perturbation) to pinpoint the performance loopholes of index advisors. Third, to generalize to various index advisors, we place TRAP in an opaque-box setting (i.e., with little knowledge of the index advisors' internal design), and we propose a two-phase training paradigm to efficiently train TRAP without elaborately annotated data. Fourth, we conduct comprehensive robustness assessments on standard benchmarks and real workloads for ten existing index advisors. Our findings reveal that these index advisors are vulnerable to the workloads generated by TRAP. Finally, based on the assessment results, we shed light on insights to enhance the robustness of different index advisors. For example, learning-based index advisors can benefit from adopting a fine-grained state representation and a candidate pruning strategy.

关键词： Training Perturbation methods Benchmark testing Syntactics Data engineering Robustness Indexes

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：