检索结果-内蒙古大学图书馆

2024 international conference on Advances in data Engineering and Intelligent Computing Systems, ADICS 2024

作者： Priya Bhaminie, S. Yogesh, C. School Of Computer Science And Engineering Vellore Institue Of Technology-Chennai Chennai India

ISBN: (纸本)9798350364828

This paper presents an advanced approach in skin disease classification using a modified ResNet-50 architecture, applied to a specific subset from the ISIC 2019 dataset focusing on Benign Keratosis, Basal Cell Carcinoma, and Melanoma. The study integrates deep learning and medical imaging to enhance diagnostic accuracy in dermatology. It leverages transfer learning, adapting the well-established ResNet-50 model, and modifies it by incorporating global average pooling and dense layers, thus tailoring it for dermatological use. This method is inspired by significant strides in skin cancer detection using ResNet and innovative transfer learning techniques in skin disease classification. A notable aspect of this research is the evaluation of data augmentation's impact on model generalization and robustness. The findings of this study contribute to the growing field of deep learning in skin disease classification, providing a refined and effective tool for potential applications in dermatology. © 2024 IEEE.

关键词： Dermatology

来源：评论

学校读者我要写书评

暂无评论

Homomorphism Autoencoder - learning Group Structured Representations from Observed Transitions 40

Homomorphism Autoencoder - Learning Group Structured Represe...

引用

40th international conference on machine learning

作者： Keurti, Hamza Pan, Hsiao-Ru Besserve, Michel Grewe, Benjamin Schoelkopf, Bernhard Max Planck Inst Intelligent Syst Tubingen Germany Swiss Fed Inst Technol Inst Neuroinformat Zurich Switzerland Max Planck ETH Ctr Learning Syst Zurich Switzerland

How can agents learn internal models that veridically represent interactions with the real world is a largely open question. As machine learning is moving towards representations containing not just observational but also interventional knowledge, we study this problem using tools from representation learning and group theory. We propose methods enabling an agent acting upon the world to learn internal representations of sensory information that are consistent with actions that modify it. We use an autoencoder equipped with a group representation acting on its latent space, trained using an equivariance-derived loss in order to enforce a suitable homomorphism property on the group representation. In contrast to existing work, our approach does not require prior knowledge of the group and does not restrict the set of actions the agent can perform. We motivate our method theoretically, and show empirically(1) that it can learn a group representation of the actions, thereby capturing the structure of the set of transformations applied to the environment. We further show that this allows agents to predict the effect of sequences of future actions with improved accuracy.

关键词： Group theory

来源：评论

学校读者我要写书评

暂无评论

Mapping mangrove functional traits from Sentinel-2 imagery based on hybrid models coupled with active learning strategies

引用

international JOURNAL OF applied EARTH OBSERVATION AND GEOINFORMATION 2024年 130卷

作者： Jia, Mingming Guo, Xianxian Zhang, Lin Wang, Mao Wang, Wenqing Lu, Chunyan Zhao, Chuanpeng Zhang, Rong Wang, Ming Yan, Hengqi Wang, Zongming Verrelst, Jochem Int Res Ctr Big Data Sustainable Dev Goals Beijing 100094 Peoples R China Xiamen Univ Coll Environm & Ecol Key Lab Minist Educ Coastal & Wetland Ecosyst Xiamen 361102 Peoples R China Fujian Agr & Forestry Univ Coll Comp & Informat Sci Fuzhou 350002 Peoples R China Chinese Acad Sci Northeast Inst Geog & Agroecol State Key Lab Black Soils Conservat & Utilizat Changchun 130102 Peoples R China Jilin Earthquake Agcy Changchun 130117 Peoples R China Univ Valencia Image Proc Lab IPL Parc Cient Valencia Spain

Accurately quantifying functional traits across large scales is considered fundamental for the management and conservation of existing mangrove ecosystems. In recent years, hybrid models, which combine radiative transfer model simulations with machine learning regression algorithms (MLRA), have been effectively employed in satellite-based estimations of plant functional traits across diverse ecosystems. Nevertheless, the inevitable data redundancy stemming from heavy-parameterization radiative transfer models restricts the application of the hybrid model. Previous studies have indicated that active learning (AL) strategies can mitigate this redundancy through smart sampling selection criteria. While many studies have attempted to investigate mangrove functional traits using various models, there is limited understanding of the performance of hybrid models coupled with active learning strategies in retrieving the traits. In recent years, Sentinel-2 has become mainstream for retrieving detailed and reliable information across diverse ecosystems. The aim of this study is to utilize a retrieval scheme to extract four mangrove functional traits from Sentinel-2 imagery: leaf area index (LAI), leaf chlorophyll content (Cab), leaf dry matter content (Cm), and leaf equivalent water thickness (Cw). In order to achieve this goal, we systematically evaluated 36 different MLRA-AL models, which were combinations of six MLRAs and six AL strategies. Retrieval results showed that GPR (Gaussian processes regression)-ABD (anglebased diversity) and GPR-PAL (variance-based pool of regressors) yielded the highest accuracies for LAI (R 2 = 0.68, NRMSE = 10.488 %) and Cw (R 2 = 0.47, NRMSE =13.868 %), respectively. GPR-EBD (Euclidean distancebased diversity) had the highest accuracies of Cm (R 2 = 0.54, NRMSE = 11.695 %) and Cab (R 2 = 0.71, NRMSE = 13.764 %). The retrieval models were subsequently applied to produce distribution pattern maps of four mangrove functional traits within a Ram

关键词： Mangrove functional traits Active learning machine learning regression algorithm Radiative transfer model Sentinel-2

来源：评论

学校读者我要写书评

暂无评论

Beyond the Password: A data-Driven Analysis of Biometric Security in the Age of AI

Beyond the Password: A Data-Driven Analysis of Biometric Sec...

引用

2024 IEEE international conference on Big data and machine learning, ICBDML 2024

作者： Gupta, Pratik Jain, Vipin Gandhi, Dhruv Alkeshkumar Patheja, Pushpinder Singh SCSE VIT Bhopal University Madhya Pradesh Kothri Kalan466 114 India

ISBN: (纸本)9798350374100

This research presents a comprehensive bibliometric analysis of Biometric research in the cybersecurity domain from 2017 to 2024, utilizing the Scopus database. The study explores 40 publications to discern trends, influential authors, affiliated institutions, countries of origin, and subject areas. Employing bibliometric methods, the analysis includes publication count, co-authorship, keyword co-occurrence, and citation analyses. Results illustrate a rise in publications until 2019, followed by a subsequent decline. Notable authors like Gavrilova, M. and collaborative networks were identified. The research predominantly originates from institutions in the United States, Canada, India, Saudi Arabia, Egypt, and Pakistan. Computer science emerges as the primary subject area, indicating a predominant interest among researchers. A word cloud depicts prominent terms associated with cybersecurity and biometrics, highlighting key focus areas in the field. © 2024 IEEE.

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

SocialGAIL: Faithful Crowd Simulation for Social Robot Navigation

SocialGAIL: Faithful Crowd Simulation for Social Robot Navig...

引用

IEEE international conference on Robotics and Automation (ICRA)

作者： Ling, Bo Lyu, Yan Li, Dongxiao Gao, Guanyu Shi, Yi Xu, Xueyong Wu, Weiwei Southeast Univ Nanjing Peoples R China Nanjing Univ Sci & Technol Nanjing Peoples R China North Informat Control Res Acad Grp Co Ltd Nanjing Peoples R China

ISBN: (纸本)9798350384581;9798350384574

Navigation through crowded human environments is challenging for social robots. While reinforcement learning has been adopted for its capacity to capture complex interactions, the training process often relies on simulators to replicate realistic crowd behaviors, ensuring cost-efficiency. Existing crowd simulation methods typically rely on either handcrafted rules, which may lead to overly aggressive navigation, or learning from human trajectory demonstrations, which can be challenging to generalize effectively. In this paper, we introduce a data-driven crowd simulation method called SocialGAIL, which leverages Generative Adversarial Imitation learning (GAIL) to emulate real pedestrian navigation in crowded environments. SocialGAIL utilizes an attention-based graph neural network to encode observations and employs a generator-discriminator architecture to closely mimic pedestrian behavior. We propose a set of metrics to evaluate the faithfulness of crowd simulation. Experimental results demonstrate that SocialGAIL outperforms baseline methods in terms of goal-reaching, intermediate state faithfulness, trajectory faithfulness, and adherence to global trajectory patterns. The code of our approach is available at https://***/William-island/SocialGAIL.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Optimizing IT Governance and Project Management in Software Development through AI Integration and COBIT 2019 Framework 2

Optimizing IT Governance and Project Management in Software ...

引用

2nd international conference on Technology Innovation and Its Applications, ICTIIA 2024

作者： Teguh, Andy Slamet, Joko Saputro, Hartono Sungkono, Kelly Sarno, Riyanarto Institut Teknologi Sepuluh Nopember Department of Informatics Surabaya Indonesia

ISBN: (纸本)9798350351613

In the modern business environment, efficient time management and robust IT governance are crucial due to rapid technological advancements. Effective governance and resource management are imperative for organizations aiming to optimize performance. This research uses AI techniques and the COBIT 2019 framework to optimize IT governance in software development projects. Quantitative methods are employed to analyze factors causing scheduling delays in projects using machine learning algorithms, focusing on identifying factors that impact project timelines. data from a previous study on software cost estimation involves 120 projects from 42 organizations. data preprocessing techniques are applied to handle missing data and scale features. Feature engineering is conducted to identify critical factors affecting project completion times. We evaluate various machine learning models for robust analysis of project delay factors, and we find that CatBoost provides reliable project duration predictions. Model performance was evaluated using MAE, RMSE, R2, and MAPE. Integrating AI with COBIT 2019 enhances project duration estimation accuracy and strengthens IT governance. This combined approach improves project efficiency and aligns IT initiatives with organizational goals. © 2024 IEEE.

关键词： Project management

来源：评论

学校读者我要写书评

暂无评论

SparseTSF: Modeling Long-term Time Series Forecasting with 1k Parameters 41

SparseTSF: Modeling Long-term Time Series Forecasting with 1...

引用

41st international conference on machine learning, ICML 2024

作者： Lin, Shengsheng Lin, Weiwei Wu, Wentai Chen, Haojun Yang, Junjie School of Computer Science and Engineering South China University of Technology Guangzhou510006 China Peng Cheng Laboratory Shenzhen518066 China College of Information Science and Technology Jinan University Guangzhou510632 China

This paper introduces SparseTSF, a novel, extremely lightweight model for Long-term Time Series Forecasting (LTSF), designed to address the challenges of modeling complex temporal dependencies over extended horizons with minimal computational resources. At the heart of SparseTSF lies the Cross-Period Sparse Forecasting technique, which simplifies the forecasting task by decoupling the periodicity and trend in time series data. This technique involves downsampling the original sequences to focus on cross-period trend prediction, effectively extracting periodic features while minimizing the model's complexity and parameter count. Based on this technique, the SparseTSF model uses fewer than 1k parameters to achieve competitive or superior performance compared to state-of-the-art models. Furthermore, SparseTSF showcases remarkable generalization capabilities, making it well-suited for scenarios with limited computational resources, small samples, or low-quality data. The code is publicly available at this repository: https://***/lss-1138/SparseTSF. Copyright 2024 by the author(s)

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Fast real-time monitor of rice grains infested with Sitophilus oryzae based on terahertz imaging combined with machine learning

引用

FOOD CONTROL 2025年 176卷

作者： Yu, Jingxiao Pu, Hongbin Sun, Da-Wen South China Univ Technol Sch Food Sci & Engn Guangzhou 510641 Peoples R China South China Univ Technol Acad Contemporary Food Engn Guangzhou Higher Educ Mega Ctr Guangzhou 510006 Peoples R China Guangzhou Higher Educ Mega Ctr Engn & Technol Res Ctr Guangdong Prov Intelligent Guangzhou 510006 Peoples R China Guangzhou Higher Educ Mega Ctr Guangdong Prov Engn Lab Intelligent Cold Chain Log Guangzhou 510006 Peoples R China Natl Univ Ireland Univ Coll Dublin Agr & Food Sci Ctr Food Refrigerat & Computerized Food Technol FRCFT Dublin Ireland

Rice grains are frequently infested by Sitophilus oryzae due to improper storage, reducing the nutritional value and posing challenges for long-term storage and food security. The current research employed terahertz (THz) imaging combined with machine learning (ML) to monitor rice grains infested with Sitophilus oryzae in real-time. For one-dimensional (1D) spectral data, the support vector machine classification (SVC) model is improved by 9.68 % after first-order derivatives (1-st der) preprocessing. The SVC model based on 1-st der spectra reached 100.00 % after the competitive adaptive reweighted sampling (CARS) extracting feature frequencies, and the predictive time was reduced by 0.79 ms. Thereby, the 1-st der-CARS-SVC model was employed to visualize THz images and successfully demonstrated the process of rice grain infested with Sitophilus oryzae. For twodimensional (2D) image data, the VGG7 model added the attention improved to 100.00 %. Then, gradient-weighted class activation mapping (Grad-CAM) indicates attention-assisted VGG7 model can mine the important information in the image more accurately and quickly. In addition, fusion models based on weighted voting that utilise both spectral and image data also perform very well. In conclusion, the current research provides important insights into the process of pest infestation of rice that can be monitored quickly, accurately, and in real-time for pest management and grain preservation.

关键词： Terahertz imaging Rice grains infested with Sitophilus oryzae Fast real-time monitor machine learning Deep learning

来源：评论

学校读者我要写书评

暂无评论

Power Sequencial data - Forecasting Trend 5th

Power Sequencial Data - Forecasting Trend

引用

5th EAI international conference on Multimedia Technology and Enhanced learning, ICMTEL 2023

作者： Li, Lejie Zhang, Lu Sun, Bin Dong, Benjie Xu, Kaining Jinan Lingsheng Info Tech. Co. Ltd. Huaiyin District Jinan250000 China School of EE University of Jinan Jinan250022 China

ISBN: (纸本)9783031505799

In reduce the use of non-renewable energy, the use of renewable energy is increasing day by day. In recent years, with the strong support of the state, renewable energy has been applied in various industries. Renewable energy generates a considerable amount of electricity, which brings us huge economic benefits but also brings certain problems. For example, the instability of the power generation system, the scheduling, and distribution of power, etc. Therefore, the analysis of the massive power data generated by the power system has become particularly important. Effective processing and forecasting of these power data can not only improve the efficiency and performance of the power system but also enable effective power dispatching and deployment. At the same time, it can ensure the safety of industrial and family users and ensure social stability. machine learning has been widely used in various fields and achieved good performance in recent years. Therefore, many researchers have begun to use machine learning to predict power data. Therefore, we provide a preliminary overview of the history and evolution of machine learning-based power data analysis and forecasting from the perspective of bibliometrics. © ICST Institute for Computer sciences, Social Informatics and Telecommunications Engineering 2024.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Farm Households Food Security Status Automation through Supervised learning Approach: A Look at Agroecological Farms 10

Farm Households Food Security Status Automation through Supe...

引用

international conference on Smart Computing (SMARTCOMP)

作者： Nikiema, Theodore Ezin, Eugene C. Chogou, Sylvain Kpenavoun Katic, Pamela G. Univ Abomey Calavi Inst Math & Phys Sci Cotonou Benin Univ Abomey Calavi Inst Training & Res Comp Sci Cotonou Benin Univ Abomey Calavi Fac Agron Sci Cotonou Benin Univ Greenwich Nat Ressources Inst London England

ISBN: (纸本)9798350349955;9798350349948

Food insecurity is a pervasive phenomenon in Africa, and the paradox is that it affects farming households more than others. Although early and accurate detection of famine in farm households is a complex challenge, it remains very important to help reduce their vulnerability and inform policy and practice. Existing food security monitoring tools focus on isolated dimensions of the problem and some remain static. In this study, we focus on a composite metric of food security which combines three key indicators: Food Consumption Score (FCS), Livelihood Coping Strategies (LCS), and Food Expenditure Share (FES). We trained and tested various automatic classifications models, applied them to Burkina Faso's 2020 Permanent Agricultural Survey data to predict food security and validated our results with data from 2019. This prediction was compared to actual food security. The approach used correctly identifies the food security status of 84% of the households. Farms which practice agroecology are slightly less affected by food insecurity than those practicing conventional agriculture. We also find that households that implement integrated livestock-agriculture systems are less affected by food insecurity than households that do not. The proposed models can speedily predict food security status using multidimensional datasets and are able to identify the risk factors associated with predicted food insecurity trends, which is key to target priority areas for intervention by policymakers. They could also be integrated into data collection tools via server communication protocols to enable real-time monitoring.

关键词： Food security machine learning Agroecology Multi-classification Farm households

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：