检索结果-内蒙古大学图书馆

NFCL: Simply interpretable neural networks for a short-term multivariate forecasting

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Jo, Wonkeun Kim, Dongil Department of Computer Science and Engineering Chungnam National University Yuseong–gu Daejeon Korea Republic of Department of Data Science Ewha Womans University Seodaemun–gu Seoul Korea Republic of

Multivariate time-series forecasting (MTSF) stands as a compelling field within the machine learning community. Diverse neural network based methodologies deployed in MTSF applications have demonstrated commendable efficacy. Despite the advancements in model performance, comprehending the rationale behind the model’s behavior remains an enigma. Our proposed model, the Neural ForeCasting Layer (NFCL), employs a straightforward amalgamation of neural networks. This uncomplicated integration ensures that each neural network contributes inputs and predictions independently, devoid of interference from other inputs. Consequently, our model facilitates a transparent explication of forecast results. This paper introduces NFCL along with its diverse extensions. Empirical findings underscore NFCL’s superior performance compared to nine benchmark models across 15 available open datasets. Notably, NFCL not only surpasses competitors but also provides elucidation for its predictions. In addition, Rigorous experimentation involving diverse model structures bolsters the justification of NFCL’s unique configuration. © 2024, CC BY.

关键词： Forecasting

Detection of IoT Botnet Based on Convolutional Neural Network and Linear Support Vector Machine 23

学校读者我要写书评

暂无评论

Detection of IoT Botnet Based on Convolutional Neural Networ...

Proceedings of the 2023 13th International Conference on Communication and Network Security

作者： Changping Lai Yinan Yao Yuzhong Chen Xintian Liang Tianqi Cai Yiqing Shi College of Computer and Data Science/College of Software Fuzhou University China College of Photonic and Electronic Engineering Fujian Normal University China

ISBN: (纸本)9798400707964

The development of machine learning has brought new methods for botnet detection. Traditional machine learning methods and deep learning methods are used in botnet detection, but the former requires prior knowledge to select features. And deep learning methods solve this problem. This paper proposes a new botnet detection model that combines Convolutional Neural Network (CNN) with Support Vector Machine (SVM). This approach directly acquires network traffic data, preprocesses the data to obtain input suitable for CNN, utilizes CNN for feature extraction after preprocessing, and feeds the extracted features through two convolutional layers. The obtained features are then input to a SVM for classification. This method leverages the powerful feature extraction capabilities of CNN and the faster computational speed of linear SVM, resulting in faster training and excellent classification performance. Experiments show that the method has good performance on botnet detection and reduces training time compared to the CNN model.

关键词： Botnet Detection Convolutional Neural Network Deep Learning Support Vector Machine

Attention-based Multimodal Fusion with Adversarial Network for In-Hospital Mortality Prediction

学校读者我要写书评

暂无评论

Attention-based Multimodal Fusion with Adversarial Network f...

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Ying An Ruping Qiu Lin Guo Xianlai Chen Big Data Institute Central South University Changsha China School of Computer Science and Engineering Central South University Changsha China

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

It is crucial to predict the in-hospital mortality for improving clinical decision-making and optimizing hospital resource allocation. Many recent studies have attempted to integrate multimodal data, such as digital time series and clinical notes from electronic health records (EHRs), to improve the performance of mortality prediction. Although current methods show good performance, it is often difficult to effectively capture fine-grained inter-modal correspondences. In addition, the different distributions and heterogeneous properties of the various modalities lead to modality gaps that severely affect the effectiveness of modal fusion. To overcome these limitations, we propose an Attention-based Multimodal Fusion with Adversarial network (AMFA) for in-hospital mortality prediction. In AMFA, feature extraction is first performed to obtain modality-specific features, and then an attention mechanism is employed to capture inter-modal interrelationships. The main difference with existing methods is that AMFA achieves this by calculating the relative importance of individual features focusing attention on the most relevant features and eliminating irrelevant features, and then reassigning attention between relevant features to obtain finer semantic relevance. Subsequently, we introduce a discriminator network that addresses the modality gap by adjusting the distribution of various modal representations through adversarial training. We evaluate AMFA on two publicly available datasets, and experimental results show that AMFA outperforms several state-of-the-art models in the task of in-hospital mortality prediction.

关键词： Training Hospitals Time series analysis MIMICs Semantics Mortality Predictive models Feature extraction Resource management Biomedical imaging

A Proximal DC Algorithm for Sample Average Approximation of Chance Constrained Programming

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Peng Jiang, Rujun Kong, Qingyuan Balzano, Laura Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor United States School of Data Science Fudan University Shanghai China

Chance constrained programming (CCP) refers to a type of optimization problem with uncertain constraints that are satisfied with at least a prescribed probability level. In this work, we study the sample average approximation (SAA) method for chance constraints, which is an important approach to CCP in the data-driven setting where only a sample of multiple realizations of the random vector in the constraints is available. The SAA method approximates the underlying distribution with an empirical distribution over the available sample. Assuming that the functions in the chance constraints are all convex, we reformulate the SAA of chance constraints into a difference-of-convex (DC) form. Additionally, by assuming the objective function is also a DC function, we obtain a DC constrained DC program. To solve this reformulation, we propose a proximal DC algorithm and show that the subproblems of the algorithm are suitable for off-the-shelf solvers in some scenarios. Moreover, we not only prove the subsequential and sequential convergence of the proposed algorithm but also derive the iteration complexity for finding an approximate Karush-Kuhn-Tucker point. To support and complement our theoretical development, we show via numerical experiments that our proposed approach is competitive with a host of existing approaches. Our publicly available code repository can be found at https://***/peng8wang/2024.0648. Copyright © 2023, The Authors. All rights reserved.

关键词： Constrained optimization

Teranga Go!: Carpooling Collaborative Consumption Community with multi-criteria hesitant fuzzy linguistic term set opinions to build confidence and trust

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Montes, Rosana Morales, Jeovani Zermeño, Noe Duran, Jeronimo Herrera, Francisco Department of Software Engineering University of Granada Spain Department of Computer Science and Artificial Intelligence University of Granada Spain Andalusian Research Institute in Data Science and Computational Intelligence DaSCI Spain

Classic Delphi and Fuzzy Delphi methods are used to test content validity of a data collection tools such as questionnaires. Fuzzy Delphi takes the opinion issued by judges from a linguistic perspective reducing ambiguity in opinions by using fuzzy numbers. We propose an extension named 2-Tuple Fuzzy Linguistic Delphi method to deal with scenarios in which judges show different expertise degrees by using fuzzy multigranular semantics of the linguistic terms and to obtain intermediate and final results expressed by 2-tuple linguistic values. The key idea of our proposal is to validate the full questionnaire by means of the evaluation of its parts, defining the validity of each item as a Decision Making problem. Taking the opinion of experts, we measure the degree of consensus, the degree of consistency, and the linguistic score of each item, in order to detect those items that affect, positively or negatively, the quality of the instrument. Considering the real need to evaluate a b-learning educational experience with a consensual questionnaire, we present a Decision Making model for questionnaire validation that solve it. Additionally, we contribute to this consensus reaching problem by developing an online tool under GPL v3 license. The software visualizes the collective valuations for each iteration and assists to determine which parts of the questionnaire should be modified to reach a consensual solution. © 2024, CC BY-NC-ND.

关键词： Semantics

Cross Space and Time: A Spatio-Temporal Unitized Model for Traffic Flow Forecasting

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ruan, Weilin Wang, Wenzhuo Zhong, Siru Chen, Wei Liu, Li Liang, Yuxuan The Hong Kong University of Science and Technology Guangzhou China The Computer Science Department Jinan University Guangzhou China The School of Big Data and Software Engineering Chongqing University Chongqing China

Predicting spatio-temporal traffic flow presents significant challenges due to complex interactions between spatial and temporal factors. Existing approaches often address these dimensions in isolation, neglecting their critical interdependencies. In this paper, we introduce the Spatio-Temporal Unitized Model (STUM), a unified framework designed to capture both spatial and temporal dependencies while addressing spatio-temporal heterogeneity through techniques such as distribution alignment and feature fusion. It also ensures both predictive accuracy and computational efficiency. Central to STUM is the Adaptive Spatio-temporal Unitized Cell (ASTUC), which utilizes low-rank matrices to seamlessly store, update, and interact with space, time, as well as their correlations. Our framework is also modular, allowing it to integrate with various spatio-temporal graph neural networks through components such as backbone models, feature extractors, residual fusion blocks, and predictive modules to collectively enhance forecasting outcomes. Experimental results across multiple real-world datasets demonstrate that STUM consistently improves prediction performance with minimal computational cost. These findings are further supported by hyperparameter optimization, pre-training analysis, and result visualization. We provide our source code for reproducibility at https://***/r/STUM-E4F0. Copyright © 2024, The Authors. All rights reserved.

关键词： Graph neural networks

Actuation Sequence Tampering Attack Detection on Field-Level Micro-Electrode Dot Arrays Based on 2-Layer Authentication 23

学校读者我要写书评

暂无评论

Actuation Sequence Tampering Attack Detection on Field-Level...

Proceedings of the 2023 13th International Conference on Communication and Network Security

作者： Xiaodong Guo Chenxi Lyu Zhenyi Chen Xiaoding Wang Ruishen Bao Chen Dong College of Computer and Big Data Fuzhou University China Department of Computer Science and Engineering University of South Florida USA College of Computer and Cyber Security Fujian Normal University China

ISBN: (纸本)9798400707964

As a next-generation digital microfluidic biochip (DMFB) platform, micro-electrode-dot -array (MEDA) play an important role in the biochemical industry with more advanced fluidic manipulation and higher reliability. As the MEDA biochip market has grown by leaps and bounds, there are still some issues with MEDA that have not kept pace in terms of security. One of them is that an attacker of a MEDA biochip can change the protocol parameters by tampering with the actuation sequence (AS) in the field before performing the bioassay, which can lead to catastrophic consequences. The attackers could be untrustworthy third parties introduced by the distributed supply chain and malicious remote parties at the time of cyber-physical MEDA biochip deployment. In this paper, a two-layer authentication solution for AS tampering attack detection is proposed, based on SHA-256 hash function to ensure the security of the authentication process and at the same time, it is able to detect and locate the tampered AS with simplicity, security and efficiency. Experiments demonstrate the obvious advantages in security, performance and applicability within a controllable time complexity.

关键词： actuation sequence micro-electrode-dot-array security tempering attack

A Deep Reinforcement Learning-Based Task Offloading Framework for Edge-Cloud Computing

学校读者我要写书评

暂无评论

A Deep Reinforcement Learning-Based Task Offloading Framewor...

Integrated Circuits and Communication Systems (ICICACS), IEEE International Conference on

作者： Ponugoti Kalpana Muntather Almusawi Yerrolla Chanti V Sunil Kumar M Varaprasad Rao Department of Computer Science and Engineering AVN Institute of Engineering & Technology Telangana India The Islamic University Najaf Iraq School of Computer Science & Artificial Intelligence SR University Warangal India Department of AI and ML Nitte Meenakshi Institute of Technology Bengaluru India Department of Data Science CVR College of Engineering Hyderabad India

By introducing the processing and resources storage to the network edge, edge computing is a new, promising computing paradigm that drastically decreases network traffic and service latency. Many edge computing applications consist of interdependent tasks, wherein the results of one task are the inputs of another. The important and difficult problem of where to position each running task to optimise Quality-of-Service (QoS) is how to offload these tasks to edge of network. In this work, implemented a novel Deep Reinforcement Learning based Task Offloading (DRLTO) method utilized as the intelligent task offloading that uses a Directed Acyclic Graph (DAG) to represent the dependent tasks and off-policy reinforcement learning powered by a Sequence-to-Sequence (S2S) neural network. This research outcomes show that the DRLTO achieved less cloud processing time, and number of single terminal tasks, and percentage of failed task when compared to Deep reinforcement learning-based cloud-edge collaborative mobile computation offloading (DRL-CCMCO) and DeepEdge.

关键词：

Identification of Breast Cancer using Jelly fish Based Deep Wavelet Autoencoder

学校读者我要写书评

暂无评论

Identification of Breast Cancer using Jelly fish Based Deep ...

International Conference on Advanced Computing and Communication Systems (ICACCS)

作者： K. Anbumani R. Balamanigandan T. Rajesh Kumar A. Sasi Kumar Mahaveerakannan. R Department of EIE Sri Sai Ram Engineering College Chennai Computer Science and Engineering Saveetha School of Engineering Computer Science and Engineering Saveetha School of Engineering SIMATS Chennai Tamil Nadu India Department of Cloud Technology & Data Science Institute of Engineering & Technology Srinivas University Mangalore

Mammography, an imaging technology used to detect breast cancer, has recently received more interest from scientists. Screening mammography is the best way to detect breast cancer and other abnormalities. Breast tumours often have two distinct characteristics: mass lesions and microcalcification. A gathering of appropriate picture pre-processing, proceduresare essential for accurate cancer detection and classification in mammograms. Image processing is widely employed in the medical field to improve the diagnostic and treatment processes. When it comes to classification and segmentation, deep neural networks (DNN) have excelled so far. With this in mind, the authors of this paper suggest a deep wavelet autoencoder (DWA) method for image compression by fusing the autoencoder's basic feature reduction function with the picture decomposition capabilities of the wavelet transform. The best attributes for identifying breast cancer are chosen in this study by using the Jelly Fish Search Algorithm (JFSA). When used together, they have a dramatic impact on reducing the size of the feature set required to persist through a subsequent classification job using DNN. The suggested DWA-DNN image classifier was evaluated on two public datasets of breast cancer pictures. The DWA-DNN classifier's performance criteria was compared to that of other existing classifiers, and it was found that the suggested technique excels in comparison across a range of measures.

关键词： Wavelet transforms Neural networks Merging Medical treatment Fish Turning Breast cancer