检索结果-内蒙古大学图书馆

8th International Conference on Information and Communication Technology for Intelligent Systems, ICTIS 2024

作者： Takale, Dattatray G. Gawali, Piyush P. Deshmukh, Gopal B. Kashid, Shraddha S. Mahalle, Parikshit N. Sule, Bipin Patil, Sambhaji A. Chitale, Swati S. Department of Computer Engineering Vishwakarma Institute of Information Technology SPPU Pune India Department of Computer Engineering MIT ADT University Pune India Department of Artificial Intelligence and Data Science Vishwakarma Institute of Information Technology SPPU Pune India and Humanities Vishwakarma Institute of Technology Pune India Department of Electronic and Telecommunication PGM College of Engineering Wagholi SPPU Pune India Pune India

ISBN: (纸本)9789819766772

Global health challenges such as skin cancer, which arises from uncontrolled cell growth, pose a significant threat. The most prevalent form of skin cancer manifests in cells throughout the body, notably in the surface layer. Because of its widespread nature, early detection is urgently needed. An innovative approach is developed to segment and detect skin cancer by integrating Fuzzy Cognitive Maps (FCM) with Deep-Seg Nets. Adaptive bilateral filtering, preprocessing, and optimizing input skin cancer images are the first steps. In the following step, DeepSegNet, a hybrid of DeepJoint and SegNet, is used to segment the images. The segmented image is then enhanced using enhancement techniques, followed by extraction of diverse features such as texture, statistical attributes, Local Neighborhood Difference Patterns (LNDPs), and a novel contribution, a discrete wavelet transform (DWT)-based Local Directional Pattern. In skin cancer detection, FCM shows noteworthy proficiency. As a result of the evaluation of the proposed methodology, DeepSegNet scored 0.920 for dice coefficient, while the DeepSegNet+FCM model scored 97% accurate, 93% negative predictive value, 94% positive predictive value, 98% sensitivity, and 96% specificity. The combination of FCM and Deep-Seg Net offers a potential path toward more accurate and efficient skin cancer diagnosis. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Adaptive filtering

来源：评论

学校读者我要写书评

暂无评论

Large-scale Multiple Testing: Fundamental Limits of False Discovery Rate Control and Compound Oracle

arXiv

引用

arXiv 2023年

作者： Nie, Yutong Wu, Yihong The Department of Statistics and Data Science Yale University New HavenCT United States

The false discovery rate (FDR) and the false non-discovery rate (FNR), defined as the expected false discovery proportion (FDP) and the false non-discovery proportion (FNP), are the most popular benchmarks for multiple testing. Despite the theoretical and algorithmic advances in recent years, the optimal tradeoff between the FDR and the FNR has been largely unknown except for certain restricted classes of decision rules, e.g., separable rules, or for other performance metrics, e.g., the marginal FDR and the marginal FNR (mFDR and mFNR). In this paper, we determine the asymptotically optimal FDR-FNR tradeoff under the two-group random mixture model when the number of hypotheses tends to infinity. Distinct from the optimal mFDR-mFNR tradeoff, which is achieved by separable decision rules, the optimal FDR-FNR tradeoff requires compound rules even in the large-sample limit and for models as simple as the Gaussian location model. This suboptimality of separable rules also holds for other objectives, such as maximizing the expected number of true discoveries. Finally, to address the limitation of the FDR which only controls the expectation but not the fluctuation of the FDP, we also determine the optimal tradeoff when the FDP is controlled with high probability and show it coincides with that of the mFDR and the mFNR. Extensions to models with a fixed non-null proportion are also obtained. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Quantum support vector data description for anomaly detection

arXiv

引用

arXiv 2023年

作者： Oh, Hyeondo Park, Daniel K. Department of Statistics and Data Science Yonsei University Seoul Korea Republic of Department of Applied Statistics Yonsei University Seoul Korea Republic of

Anomaly detection is a critical problem in data analysis and pattern recognition, finding applications in various domains. We introduce quantum support vector data description (QSVDD), an unsupervised learning algorithm designed for anomaly detection. QSVDD utilizes a shallow-depth quantum circuit to learn a minimum-volume hypersphere that tightly encloses normal data, tailored for the constraints of noisy intermediate-scale quantum (NISQ) computing. Simulation results on the MNIST and Fashion MNIST image datasets demonstrate that QSVDD outperforms both quantum autoencoder and deep learning-based approaches under similar training conditions. Notably, QSVDD offers the advantage of training an extremely small number of model parameters, which grows logarithmically with the number of input qubits. This enables efficient learning with a simple training landscape, presenting a compact quantum machine learning model with strong performance for anomaly detection. Copyright © 2023, The Authors. All rights reserved.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Maximum likelihood based estimation with quasi oppositional chemical reaction optimization algorithm for speech signal enhancement

引用

International Journal of Information Technology (Singapore) 2022年第6期14卷 3265-3275页

作者： Kaur, Inderjeet Nassa, Vinay Kumar Kavitha, T. Mohan, Prakash Velmurugan, S. Department of Computer Science and Engineering Ajay Kumar Garg Engineering College Uttar Pradesh Ghaziabad India Department of Artificial Intelligence & Data Science Sharad Institute of Technology College of Engineering Maharashtra Yadrav-Kohiapur India Department Computer Applications Kongu Engineering College Perundurai Tamil Nadu Erode India Data Science and Analytics Center Karpagam College of Engineering Tamil Nadu Coimbatore India Department of Computer Science and Engineering Vel Tech Multi Tech Dr. RR & Dr.SR Engineering College Tamil Nadu Chennai India

In recent times, speech enhancement (SE) becomes a significant process in the field of speech signal processing. Since the speech signal in real time gets affected by background noise, the efficacy of the speech-based applications gets severely affected. The SE techniques have gained considerable attention in the usage of speaker recognition, video conferences, speech broadcast, speech-enabled biometric systems, smartphones, hearing aid, and so on. Numerous conventional SE and machine learning (ML) models have been employed for processing and removing the additive noise from the speech signal. This paper presents a novel Maximum Likelihood Based Estimation with Quasi Oppositional Chemical Reaction Optimization (MLE-QOCRO) algorithm for speech signal enhancement. The presented MLE-QOCRO algorithm follows three-stage processes namely preprocessing, mask generation, and spectral filtering. Besides, the mask generation process involves QOCRO algorithm to classify the speech signal into noise or speech frames. In addition, MLE based spectral filtering technique is employed at the final stage to get enhanced speech signals. A set of experiments were performed to highlight the improved performance of the proposed MLE-QOCRO algorithm, and the results are investigated in terms of several performance measures. Speech signal augmentation using the unique MLE-QOCRO algorithm, which combines maximum likelihood estimation with quasi-opposing chemical reaction optimization, is the subject of this paper. The QOCRO algorithm is used in the mask generation process in the model to distinguish between noise and speech frames in the input speech signal. A spectral filtering method based on MLE is also utilized to improve the voice signal at the very end. The resultant experimental values pointed out the enhanced outcome of the proposed MLE-QOCRO algorithm over the other compared techniques. © 2022, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of computer Appl

关键词： Chemical reaction optimization Maximum likelihood Spectral filtering Speech signal enhancement

来源：评论

学校读者我要写书评

暂无评论

Pandemic Bonds Issued by the Chinese Government Supported Post-Disaster Recovery from COVID-19 Pandemic

引用

Journal of Social Computing 2022年第2期3卷 158-170页

作者： Feng Liu Deli Kong Jinhua Kong Shiying Lu Zilong Xiao Qing Xu Aimin Zhou Jiayin Qi the Institute of AI for Education and also with the School of Computer Science and TechnologyEast China Normal UniversityShanghai 200062China the School of Business and Management Shanghai International Studies UniversityShanghai 201620China the Institute of Artificial Intelligence and Change Management Shanghai University of International Business and EconomicsShanghai 200336China the School of Statistics and Information Shanghai University of International Business and EconomicsShanghai 201620China the Department of Economics and Management Changsha University of Science and Technology410000 ChangshaChina the Department of Data Science and Computer Science Sun Yat-sen UniversityGuangzhou 510080China the Department of Software Beijing University of Posts and TelecommunicationsBeijing 100876China

During the SARS-CoV-2(COIVD-19)outbreak,China repeatedly stressed that the response to the pandemic required action at all levels of government,including the issuance of Pandemic Bonds to help the country return to work and ***,studies on the effectiveness of Pandemic Bonds during that period are *** with China’s national financial bond market data after COVID-19 in 2020,this paper focuses on the correlation between the Credit Spreads of the relevant bonds and the corresponding bond market rate of return,based on the Copula *** empirical analysis is also carried out for multiple dimensional groupings such as enterprises,industries,provinces,and bond *** results show that there is a significant positive correlation between the Credit Spreads of Pandemic Bonds and market *** addition,the market correlation is higher for Pandemic Bonds issued in Hubei Province,which is at the center of the 2020 pandemic,and the shorter the maturity of the Pandemic Bond issued,the stronger the relationship with market ***,this paper provides recommendations for financial regulators and policy makers to consider in their decisions on how to build a more resilient financial system under heavy economic,fiscal,and social pressures.

关键词： COVID-19 Pandemic Bond Credit Spread Copula model

来源：评论

学校读者我要写书评

暂无评论

On a new robust method of inference for general time series models

arXiv

引用

arXiv 2025年

作者： Wang, Zihan Qiao, Xinghao Li, Dong Tong, Howell Department of Statistics and Data Science Tsinghua University Beijing China Faculty of Business and Economics The University of Hong Kong Hong Kong Department of Statistics London School of Economics London United Kingdom

In this article, we propose a novel logistic quasi-maximum likelihood estimation (LQMLE) for general parametric time series models. Compared to the classical Gaussian QMLE and existing robust estimations, it enjoys many distinctive advantages, such as robustness in respect of distributional misspecification and heavy-tailedness of the innovation, more resiliency to outliers, smoothness and strict concavity of the log logistic quasi-likelihood function, and boundedness of the influence function among others. Under some mild conditions, we establish the strong consistency and asymptotic normality of the LQMLE. Moreover, we propose a new and vital parameter identifiability condition to ensure desirable asymptotics of the LQMLE. Further, based on the LQMLE, we consider the Wald test and the Lagrange multiplier test for the unknown parameters, and derive the limiting distributions of the corresponding test statistics. The applicability of our methodology is demonstrated by several time series models, including DAR, GARCH, ARMA-GARCH, DTARMACH, and EXPAR. Numerical simulation studies are carried out to assess the finite-sample performance of our methodology, and an empirical example is analyzed to illustrate its usefulness. Copyright © 2025, The Authors. All rights reserved.

关键词： Lagrange multipliers

来源：评论

学校读者我要写书评

暂无评论

Investigating the Effectiveness of Detecting Misinformation on Social Media using Tshivenda Language

Investigating the Effectiveness of Detecting Misinformation ...

引用

IST-Africa

作者： Maduvha Malange Seani Rananga Mahlatse S Mbooi Bassey Isong Vukosi Marivate Department of Computer Science Data Science for Social Impact University of Pretoria South Africa North West University South Africa

ISBN: (数字)9781905824731

ISBN: (纸本)9798350356595

The spread of misinformation on social media poses a major challenge to information integrity and public discourse. This study examines the effectiveness of detecting misinformation in Tshivenda language, which is one of the under-represented languages in South Africa. The same applies also on social media platforms. We analyse misinformation patterns, adapt existing detection techniques, and examine the influence of Tshivenda language. Through an extensive literature review, we investigated the state of the art in misinformation detection and its applicability to languages with limited digital footprints. To address this gap, we used Long Short-Term Memory (LSTM) models, a type of recurrent neural network known for capturing long-range dependencies, for misinformation detection. Our research involved training and evaluating the LSTM model on the Tshivenda and English datasets. This comparative analysis provided valuable insights into the challenges and opportunities that linguistic diversity presents in detecting misinformation. Our results shed light on the effectiveness of using LSTM models to detect misinformation in underrepresented languages. By analysing the results from the Tshivenda and English datasets, we were able to gain valuable insight into the differences in performance and the impact of linguistic variation on the accuracy of misinformation detection.

关键词： Training Accuracy Recurrent neural networks Social networking (online) Navigation Pressing Linguistics

来源：评论

学校读者我要写书评

暂无评论

Devil in the Noise: Detecting Advanced Persistent Threats with Backbone Extraction

Devil in the Noise: Detecting Advanced Persistent Threats wi...

引用

IEEE Symposium on computers and Communications (ISCC)

作者： Caio M. C. Viana Carlos H. G. Ferreira Fabricio Murai Aldri Luiz Dos Santos Lourenço Alves Pereira Júnior Computer Science Division Aeronautics Institute of Technology Brazil Department of Computing and Systems Federal University of Ouro Preto Brazil Data Science & Computer Science Worcester Polytechnic Institute Worcester MA USA Department of Computer Science Federal University of Minas Gerais Brazil

ISBN: (数字)9798350354232

ISBN: (纸本)9798350354249

The use of host intrusion detection systems shows promising results in detecting APT campaigns due to the use of systems logs as source data to get more information about system environment. However, dealing with the increase of logs in time while tracking the execution context is a challenge for security analysts. Therefore, this work presents backbone extraction as a crucial preprocessing step, filtering out irrelevant logs. As the logs are modeled as provenance graphs, we discard spurious edges to detect residuals with distinctive node and edge distributions that indicate security threats. By applying our methodology to state-of-the-art benchmark datasets, we observed an increase in the performance of one-class classifiers by up to 62% on F1-score and 48% on recall in the Streamspot dataset and by up to 40% on F1-score and 33% on recall in the DARPA3 THEIA dataset. Moreover, our results indicate mitigation of the dependency explosion problem and underscore the ability of our methodology to improve the detection landscape by shrinking graph sizes without losing essential aspects to characterize attacks.

关键词： computers Filtering Image edge detection Prevention and mitigation Noise Intrusion detection Benchmark testing Explosions Security

来源：评论

学校读者我要写书评

暂无评论

Doubly Non-Central Beta Matrix Factorization for Stable Dimensionality Reduction of Bounded Support Matrix data

arXiv

引用

arXiv 2024年

作者： Albert, Anjali N. Flaherty, Patrick Schein, Aaron Department of Mathematics & Statistics University of Massachusetts AmherstMA01003 United States Department of Statistics & Data Science Institute University of Chicago ChicagoIL60637 United States

We consider the problem of developing interpretable and computationally efficient matrix decomposition methods for matrices whose entries have bounded support. Such matrices are found in large-scale DNA methylation studies and many other settings. Our approach decomposes the data matrix into a Tucker representation wherein the number of columns in the constituent factor matrices is not constrained. We derive a computationally efficient sampling algorithm to solve for the Tucker decomposition. We evaluate the performance of our method using three criteria: predictability, computability, and stability. Empirical results show that our method has similar performance as other state-of-the-art approaches in terms of held-out prediction and computational complexity, but has significantly better performance in terms of stability to changes in hyper-parameters. The improved stability results in higher confidence in the results in applications where the constituent factors are used to generate and test scientific hypotheses such as DNA methylation analysis of cancer samples. © 2024, CC BY.

关键词： Methylation

来源：评论

学校读者我要写书评

暂无评论

Halo: Learning to prune neural networks with shrinkage

Halo: Learning to prune neural networks with shrinkage

引用

2021 SIAM International Conference on data Mining, SDM 2021

作者： Seto, Skyler Wells, Martin T. Zhang, Wenyu Department of Statistics and Data Science Cornell University IthacaNY United States

ISBN: (纸本)9781611976700

Deep neural networks achieve state-of-the-art performance in a variety of tasks by extracting a rich set of features from unstructured data, however this performance is closely tied to model size. Modern techniques for inducing sparsity and reducing model size are (1) network pruning, (2) training with a sparsity inducing penalty, and (3) training a binary mask jointly with the weights of the network. We study these approaches from the perspective of Bayesian hierarchical models and present a novel penalty called Hierarchical Adaptive Lasso (HALO) which learns to sparsify weights of a given network via trainable parameters. When used to train over-parametrized networks, our penalty yields small subnetworks with high accuracy without fine-tuning. Empirically, on image recognition tasks, we find that HALO is able to learn highly sparse network (only 5% of the parameters) with significant gains in performance over state-of-the-art magnitude pruning methods at the same level of sparsity. Code1 and appendix2 are available at the provided links. © 2021 by SIAM.

关键词： Deep learning Feature selection Network pruning Penalization

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：