this proposal outlines the development of a comprehensive educational platform aimed at bridging the Information Technology knowledge gap among Tamil-speaking students in Sri Lanka. the platform is designed to enhance...
详细信息
ISBN:
(数字)9798331517878
ISBN:
(纸本)9798331517885
this proposal outlines the development of a comprehensive educational platform aimed at bridging the Information Technology knowledge gap among Tamil-speaking students in Sri Lanka. the platform is designed to enhance programming skills through personalized quizzes, hands-on projects, and adaptive learning models. the application focuses on two primary components: the first component emphasizes project-based learning and assessments, leveraging advanced Artificial Intelligence models for generating educational content. this model is combined with Retrieval-Augmented Generation, which retrieves relevant external information to refine and tailor the educational content to each student's needs. this component keeps students engaged and challenged at appropriate levels, promoting deeper understanding and practical Information Technology skills. the second component involves a Level-based IT Fundamental Knowledge Evaluation system, which incorporates machine learning algorithms to predict student performance, recommend customized learning paths, and assess overall technical knowledge. the platform also integrates the OpenAI Application programming Interface, which provides real-time Tamil translation, making the content more accessible to the target demographic. the platform empowers educators to better address individual student needs with real-time feedback and tailored experiences. Its innovative approach enhances Information Technology education, equipping Sri Lankan students with essential programming skills to succeed globally.
We present a no-reference image-quality - assessment algorithm based on active reasoning module. this algorithm has three modules: the feature extraction module, the active reasoning module, and the quality assessment...
We present a no-reference image-quality - assessment algorithm based on active reasoning module. this algorithm has three modules: the feature extraction module, the active reasoning module, and the quality assessment module. the active reasoning module incorporates the generator component of the generative adversarial network and enhances its structure withthe Res2Net architecture. By integrating this module into the backbone feature extraction network, we improve the receptive field of each convolutional layer, enabling the network to capture information at different scales of the image. To preserve the texture information of the image, we input the gradient map of the distorted image, the distorted image itself, and the image features generated by the generator into the quality assessment module, which has a multi-feature regression networks. this module establishes a mapping model from image-feature to image-quality scores. We conducted experimental analysis of this algorithm on three widely used public datasets, confirming the excellent performance and superiorities of the presented algorithm. the results validate the effectiveness of the designed algorithm and its ability to assess image quality accurately.
Renewable energy is the fundamental approach to reducing the carbon emission of power systems. To fully utilize renewable energy sources in the planning stage, a co-optimization of generation and transmission expansio...
详细信息
this research addresses the optimization of routes and the selection of locations for multiple electric logistics vehicle charging stations. Electric logistics vehicles, as a key component of new energy vehicles, offe...
详细信息
ISBN:
(数字)9798350374315
ISBN:
(纸本)9798350374322
this research addresses the optimization of routes and the selection of locations for multiple electric logistics vehicle charging stations. Electric logistics vehicles, as a key component of new energy vehicles, offer significant environmental benefits such as zero emissions and high energy efficiency, making them ideal for urban logistics. However, efficient operation depends on the strategic placement of charging stations and effective route planning. this research introduces a mixed-integer programming model incorporating multi-vehicle configuration and genetic algorithms to solve the charging station location and path optimization problems. the model aims to minimize transportation costs, charging costs, time window penalties, environmental costs, and charging station construction costs. By factoring in battery capacity and time window constraints, the model ensures vehicles complete delivery tasks efficiently. the proposed Pareto ranking-based GA shows robust performance in generating high-quality solutions, as demonstrated by small, medium and large-scale examples. Sensitivity analysis confirms the model's practicality and reliability, showing the total delivery cost stabilizes as battery capacity increases. this study contributes to enhancing the operational efficiency and adoption of electric logistics vehicles through optimized charging infrastructure and route planning.
the article presents a practical solution in the field of integration of neural network models in spatial data analysis systems. the article shows that the deep neural network model can to be differentiated by topolog...
详细信息
In this journal-first paper, we present an overview of our novel formalism of Attack-Fault-Maintenance Trees (AFMTs). Detailed version of work is available in [3]. AFMTs enable practitioners to quantify the disruption...
详细信息
ISBN:
(数字)9783031086793
ISBN:
(纸本)9783031086793;9783031086786
In this journal-first paper, we present an overview of our novel formalism of Attack-Fault-Maintenance Trees (AFMTs). Detailed version of work is available in [3]. AFMTs enable practitioners to quantify the disruption scenarios by answering several safety-security metrics. Alongside, it provides an informed decision on optimal maintenance policies by suggesting preventive component repairs and inspection frequencies. We answer the aforementioned metrics through "what-if" and "scenario analysis". the models are supported by a graphical friendly tool of PASST. the tool's front-end is a drawing canvas that provides the different syntactic elements used to design a well-formed AFMT model. the back-end of the tool is based on the statistical-model checking techniques. From the practitioner perspective, once the AFMT is designed and input parameters on component failure, detection rates, inspection rates are provided, the entire analysis can be then done as push-button technology using model-checking techniques
Emerging applications such as the metaverse, telesurgery or cloud computing require increasingly complex operational demands on networks (e.g., ultra-reliable low latency). Likewise, the ever-faster traffic dynamics w...
详细信息
ISBN:
(纸本)9783903176515
Emerging applications such as the metaverse, telesurgery or cloud computing require increasingly complex operational demands on networks (e.g., ultra-reliable low latency). Likewise, the ever-faster traffic dynamics will demand network control mechanisms that can operate at short timescales (e.g., sub-minute). In this context, Traffic engineering (TE) is a key component to efficiently control network traffic according to some performance goals (e.g., minimize network congestion). this paper presents Routing By Backprop (RBB), a novel TE method based on Graph Neural Networks (GNN) and differentiable programming. thanks to its internal GNN model, RBB builds an end-to-end differentiable function of the target TE problem (MinMaxLoad). this enables fast TE optimization via gradient descent. In our evaluation, we show the potential of RBB to optimize OSPF-based routing (similar to 25% of improvement with respect to default OSPF configurations). Moreover, we test the potential of RBB as an initializer of computationally-intensive TE solvers. the experimental results show promising prospects for accelerating this type of solvers and achieving efficient online TE optimization.
As a key component of the sixth generation (6G) communication network, satellite network has attracted extensive attention due to its advantages of wide coverage and high capacity. However, the current limited resourc...
详细信息
the proceedings contain 37 papers. the topics discussed include: machine learning prediction of TBI from mobility, gait and balance patterns;improve image-based skin cancer diagnosis withgenerative self-supervised le...
ISBN:
(纸本)9781665439657
the proceedings contain 37 papers. the topics discussed include: machine learning prediction of TBI from mobility, gait and balance patterns;improve image-based skin cancer diagnosis withgenerative self-supervised learning;RT-ACL: identification of high-risk youth patients and their most significant risk factors to reduce anterior cruciate ligament reinjury risk;detection and analysis of interrupted behaviors by public policy interventions during COVID-19;information extraction from patient care reports for intelligent emergency medical services;high-confidence data programming for evaluating suppression of physiological alarms;EDA-based data stream pattern analysis and peak detection algorithm for substance users;and sensor-based human activity recognition for elderly in-patients with a Luong self-attention network.
In the context of rapid development, remote sensing imaging technology is widely used in environmental monitoring, disaster warning and other fields. As an important component of remote sensing image processing, seman...
详细信息
ISBN:
(数字)9798350353174
ISBN:
(纸本)9798350353181
In the context of rapid development, remote sensing imaging technology is widely used in environmental monitoring, disaster warning and other fields. As an important component of remote sensing image processing, semantic segmentation of remote sensing images is also highly valued. Traditional remote sensing image segmentation methods are mostly based on manual features and simple classifiers, but they are limited by the complexity, diversity, and scale changes of remote sensing images, making it difficult for these methods to achieve ideal results. this article proposes a method for semantic segmentation of remote sensing images based on an improved Deeplabv3+ network. Based on the Keras deep learning framework
[1]
, a convolutional neural network based on Deeplabv3+
[2] [3]
is designed, and many small modifications are made to the network to improve the segmentation results. Comparative experiments were conducted between the proposed method and common semantic segmentation methods
[4]–[5]
, and the experimental results showed that the proposed method has relatively excellent mIOU and Kappa coefficients, and has excellent semantic segmentation performance, making it suitable for training on small-scale datasets. Meanwhile, due to the high bias of target categories in small-scale datasets, the classification accuracy for certain categories is relatively low. Further research is needed, such as using GAN methods to enhance segmentation accuracy.
暂无评论