检索结果-内蒙古大学图书馆

2nd International Conference on Distributed Computing and Optimization Techniques, ICDCOT 2024

作者： Uma Maheswari, S. Adnan, Myasar Mundher Lathakumari, K.R. Pradeep, V. Praveena, Hirald Dwaraka Karunya Institute of Technology and Sciences Data Science and Cyber Security Coimbatore India The Islamic University Department of Computer Science Najaf Iraq Nitte Meenakshi Institute of Technology Dept of Ece Bengaluru India Alva's Institute of Engineering and Technology Department of Information Science and Engineering Mangalore India School of Engineering Mohan Babu University Erstwhile Sree Vidya Nikethan Engineering College Department of Ece Andhra Pradesh India

ISBN: (纸本)9798350382952

The lung cancer generally presented a pulmonary nodule on images of diagnostic and correct estimation in malignant pulmonary nodules is difficult for diagnosis and protecting of lung cancer. However, the existing methods has less classification accuracy because of irrelevant features on classification of lung cancer. In this research, proposed a Monarch Butterfly Optimization based Multi-Convolution Neural Network (MBO based M-CNN) method for lung cancer classification. dataset used for lung cancer classification is LIDC-IDRI dataset and it is pre-processed by Histogram Equalization (HE). Then, the features are extracted by Gray Level Co-occurrence Matrix (GLCM) and extracted features are selected by Monarch Butterfly Optimization (MBO) algorithm. Then, the lung cancer classification is performed by Multi-Convolution Neural Network (M-CNN). The performance of proposed algorithm is tested in terms of accuracy, precision, sensitivity and specificity. Developed technique obtained accuracy 99.71%, precision 95.39%, sensitivity 99.46%, f1-score 97.45% that is better than previous method like Adaptive Boosting Self-Normalized Multiview Convolution Neural Network (AdaBoost-SNMV-CNN). © 2024 IEEE.

关键词： Adaptive boosting

来源：评论

学校读者我要写书评

暂无评论

Enhancing Thematic Holy Quran Verse Retrieval through Vector Space Model and Query Expansion for Effective Query Answering 9

Enhancing Thematic Holy Quran Verse Retrieval through Vector...

引用

9th IEEE Information Technology International Seminar, ITIS 2023

作者： Al Haromainy, Muhammad Muharrom Sari, Anggraini Puspita Prasetya, Dwi Arman Subhan, Muh Lisdiyanto, Angga Septianto, Tri UPN Veteran Jawa Timur Department of Informatics Surabaya Indonesia UPN Veteran Jawa Timur Department of Data Science Surabaya Indonesia Politeknik Negeri Fakfak Department of Management Informatic Papua Barat Indonesia Nahdlatul Ulama University Sidoarjo Department of Computer Science Sidoarjo Indonesia Politeknik Negeri Madiun Department of Engineering Madiun Indonesia

ISBN: (纸本)9798350306835

The Quran provides valuable insights into various aspects of life, including information about the natural world, such as animals and plants. Developing an information retrieval system can greatly facilitate the search for specific content within the Quran. This system utilizes a cosine similarity approach and query expansion techniques to rank and expand the retrieved documents according to the user's needs. In this research, the Term Frequency-Inverse Document Frequency (TF-IDF) weighting method is employed, following preprocessing steps for both the documents and queries. The primary objectives of this study are to assess accuracy, precision, and recall values. Notably, the inclusion of query expansion has led to an improved recall value compared to previous research, resulting in a 99.32% accuracy rate, a 5.83% increase in precision (from 83.93%), and a system's ability to capture 8.75% of the relevant verse (compared to the previous 93.51%). This signifies that all verses can be invoked, displayed, or expanded, providing a more comprehensive query experience. © 2023 IEEE.

关键词： Information retrieval

来源：评论

学校读者我要写书评

暂无评论

On the Noise Robustness of In-Context Learning for Text Generation

arXiv

引用

arXiv 2024年

作者： Gao, Hongfu Zhang, Feipeng Jiang, Wenyu Shu, Jun Zheng, Feng Wei, Hongxin Department of Statistics and Data Science Southern University of Science and Technology China School of Economics and Finance Xi’an Jiaotong University China National Key Laboratory for Novel Software Technology Nanjing University China School of Mathematics and Statistics Xi’an Jiaotong University China Department of Computer Science and Engineering Southern University of Science and Technology China

Large language models (LLMs) have shown impressive performance on downstream tasks by in-context learning (ICL), which heavily relies on the quality of demonstrations selected from a large set of annotated examples. Recent works claim that in-context learning is robust to noisy demonstrations in text classification. In this work, we show that, on text generation tasks, noisy annotations significantly hurt the performance of in-context learning. To circumvent the issue, we propose a simple and effective approach called Local Perplexity Ranking (LPR), which replaces the "noisy" candidates with their nearest neighbors that are more likely to be clean. Our method is motivated by analyzing the perplexity deviation caused by noisy labels and decomposing perplexity into inherent perplexity and matching perplexity. Our key idea behind LPR is thus to decouple the matching perplexity by performing the ranking among the neighbors in semantic space. Our approach can prevent the selected demonstrations from including mismatched input-label pairs while preserving the effectiveness of the original selection methods. Extensive experiments demonstrate the effectiveness of LPR, improving the EM score by up to 18.75 on common benchmarks with noisy annotations. Our code is available at https://***/ml-stat-Sustech/Local-Perplexity-Ranking © 2024, CC0.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

V-statistics and variance estimation

引用

Journal of Machine Learning Research 2021年第1期22卷 1-48页

作者： Zhou, Zhengze Mentch, Lucas Hooker, Giles Department of Statistics and Data Science Cornell University IthacaNY14850 United States Department of Statistics University of Pittsburgh PittsburghPA15260 United States Department of Statistics University of California Berkeley BerkeleyCA94720 United States

As machine learning procedures become an increasingly popular modeling option among applied researchers, there has been a corresponding interest in developing valid tools for understanding their statistical properties and uncertainty. Tree-based ensembles like random forests remain one such popular option for which several important theoretical advances have been made in recent years by drawing upon a connection between their natural sub-sampled structure and the classical theory of U-statistics. Unfortunately, the procedures for estimating predictive variance resulting from these studies are plagued by severe bias and extreme computational overhead. Here, we argue that the root of these problems lies in the use of subsampling without replacement and that with-replacement subsamples, resulting in V -statistics, substantially alleviates these problems. We develop a general framework for analyzing the asymptotic behavior of V -statistics, demonstrating asymptotic normality under precise regularity conditions and establishing previously unreported connections to U-statistics. Importantly, these findings allow us to produce a natural and efficient means of estimating the variance of a conditional expectation, a problem of wide interest across multiple scientific domains that also lies at the heart of uncertainty quantification for supervised learning ensembles. ©2021 Zhengze Zhou, Lucas Mentch and Giles Hooker.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

PLAME: Piecewise-Linear Approximate Measure for Additive Kernel SVM (Extended abstract)

PLAME: Piecewise-Linear Approximate Measure for Additive Ker...

引用

International Conference on data Engineering

作者： Tsz Nam Chan Zhe Li Leong Hou U Reynold Cheng Department of Computer Science and Software Engineering Shenzhen University Alibaba Cloud Department of Computer and Information Science State Key Laboratory of Internet of Things for Smart City University of Macau Department of Computer Science and Musketeers Foundation Institute of Data Science The University of Hong Kong

ISBN: (数字)9798350317152

ISBN: (纸本)9798350317169

Additive Kernel SVM has been extensively used in many applications, including human activity detection and pedestrian detection. Since training an additive kernel SVM model is very time-consuming, which is not scalable to largescale datasets, many efficient solutions have been developed in the past few years. However, most of the existing methods normally fail to achieve one of these three important conditions which are (1) low classification error, (2) low memory space, and (3) low training time. In order to simultaneously fulfill these three conditions, we develop the new piecewise-linear approximate measure (PLAME) for training additive kernel SVM models. Experimental results verify that this approach can achieve the best trade-off between accuracy, memory space, and training time compared with different types of state-of-the-art methods.

关键词： Training Support vector machines Additives Pedestrians Accuracy data engineering Kernel

来源：评论

学校读者我要写书评

暂无评论

Implicit Discriminatory Bias Reflected in Yelp Reviews - A case study

Implicit Discriminatory Bias Reflected in Yelp Reviews - A c...

引用

2022 IEEE International Conference on Big data, Big data 2022

作者： Quan, Wei Li, Zhijian Kim, Wonkyung BNU-HKBU United International College Department of Statistics and Data Science Zhuhai China BNU-HKBU United International College Department of Communication Zhuhai China

ISBN: (纸本)9781665480451

Online reviews are one of the most comprehensive and valuable sources of information actively sought by past, current, and potential customers. Especially for the service industry, such as restaurants, online reviews play a pivotal role in customers' decision-making. Although the reviews contain rich information about the objective features and qualities of the restaurants, they can also reflect the particular qualities of an individual reviewer (perceptions and behaviors). This study aims to test people's different reviewing behaviors on Chinese and French restaurants from two dimensions: users' star ratings and users' sentiments. We analyze data from Yelp - a platform that publishes crowdsourced reviews about businesses. Findings from this pilot study lay the foundation for a series of follow-up research. This work uncovers subtle information extracted from Yelp reviews and derives insights into how implicit discriminatory bias can manifest within online platforms. Furthermore, we should reduce bias when utilizing computational techniques on big data. © 2022 IEEE.

关键词： Behavioral research

来源：评论

学校读者我要写书评

暂无评论

Improving Solar Efficiency via CNN-LSTM and Cloud-Integrated IoT Prediction

Improving Solar Efficiency via CNN-LSTM and Cloud-Integrated...

引用

Sentiment Analysis and Deep Learning (ICSADL), International Conference on

作者： P Varun T. Sunitha Mythili Nagalingam P. Nithiya S Selvakumaran T A Mohanaprakash Department of Data Science and Business Systems School of Computing SRM Institute of Science and Technology Kattankulathur Department of Computer Science and Engineering St. Joseph's Institute of Technology OMR Chennai Department of Artificial Intelligence and Data Science Rajalakshmi Institue of Technology Chennai India Department of Computer Science and Engineering P.B college of engineering Chennai India Department of Artificial Intelligence and Data Science Velammal Engineering College Surapet Chennai Department of Computer Science and Engineering Panimalar Engineering College Chennai India

ISBN: (数字)9798350396157

ISBN: (纸本)9798350396164

Photovoltaic panel used in solar power generation is an environmentally beneficial and sustainable energy source that has been used to transform sunlight into electrical power. Arranged in large solar facilities, these panels are connected to a central inverter, which converts Direct Current (DC) to alternating current (AC) electricity with a small amount of energy loss. Clean surfaces, unhindered light exposure, and high solar irradiance are all necessary for optimal panel performance. It's critical to assess the inverter's efficiency by comparing its AC to DC power. Large-scale installations with sensor-equipped panels and inverters track performance to help with maintenance and forecasting of power generation. The Internet of Things (IoT) facilitates data accessibility and remote monitoring, which helps choose the best location for solar power generation. Smart system continuous monitoring expedites site inspections, which supports urban smart grid integration. In this research study, a hybrid machine learning model is presented by combining the attention processes, long short-term memory (LSTM) networks, and clustering approaches. This model is separated into different phases for forecasting, training, and cloud data clustering, finds pertinent historical data, builds a hybrid machine learning model, and chooses the best training model. In comparison to conventional approaches, this method significantly improves prediction accuracy, which is important for integrating photovoltaic systems into smart grids, particularly in smart cities.

关键词： Training Photovoltaic systems Cloud computing Predictive models Inverters data models Smart grids

来源：评论

学校读者我要写书评

暂无评论

Online multiple testing with e-values

arXiv

引用

arXiv 2023年

作者： Xu, Ziyu Ramdas, Aaditya Department of Statistics and Data Science Carnegie Mellon University United States Departments of Statistics and Data Science and Machine Learning Carnegie Mellon University United States

A scientist tests a continuous stream of hypotheses over time in the course of her investigation — she does not test a predetermined, fixed number of hypotheses. The scientist wishes to make as many discoveries as possible while ensuring the number of false discoveries is controlled — a well recognized way for accomplishing this is to control the false discovery rate (FDR). Prior methods for FDR control in the online setting have focused on formulating algorithms when specific dependency structures are assumed to exist between the test statistics of each hypothesis. However, in practice, these dependencies often cannot be known beforehand or tested after the fact. Our algorithm, e-LOND, provides FDR control under arbitrary, possibly unknown, dependence. We show that our method is more powerful than existing approaches to this problem through simulations. We also formulate extensions of this algorithm to utilize randomization for increased power, and for constructing confidence intervals in online selective inference. Copyright © 2023, The Authors. All rights reserved.

关键词： Inference engines

来源：评论

学校读者我要写书评

暂无评论

Edge Detection with multi-scale representation and refined Network 3

Edge Detection with multi-scale representation and refined N...

引用

3rd International Conference on Distributed Sensing and Intelligent Systems, ICDSIS 2022

作者： Elharrouss, Omar Akbari, Younes Al-Maadeed, Somaya Bouridane, Ahmed Department of Computer Science and Engineering Qatar University Qatar Center for Data Analytics and Cybersecurity University of Sharjah Ukraine

ISBN: (纸本)9781839538186

Edge detection is a representation of boundaries between objects and regions in an image. Due to the variations of types, scales, intensities as well as background, the detection of these boundaries represents a challenge for different computer vision algorithms. The exploitation of multi-scale representation using various deep learning backbones make edge detection more realistic. However, edge detection is still not effective in terms of erroneous detection in some scenarios and the quality of detected edges. In this paper, we attempted to overcome these challenges using a high-resolution and refined network by fusing the output of different scales of the network and the output of the first layers. The proposed network used horizontal and vertical blocks, while the horizontal block (HB) contains a set of convolution layers and the vertical block (VB) represent the different scale of the image using convolution and pooling layers. The output of each scale in VR is combined with the corresponding layer in HB. Then, we used an affined batch normalization layer as an erosion operation for the homogeneous region in the image before fusing these outputs for the final edge detection results. The proposed method is evaluated using the most challenging datasets including BSDS500, NYUD, and Multicue. The obtained results outperform the designed edge detection networks in terms of performance metrics and quality of output images. © IET Conference Proceedings 2022.

关键词： Edge detection

来源：评论

学校读者我要写书评

暂无评论

More powerful multiple testing under dependence via randomization

arXiv

引用

arXiv 2023年

We show that two procedures for false discovery rate (FDR) control — the Benjamini-Yekutieli procedure for dependent p-values, and the e-Benjamini-Hochberg procedure for dependent e-values — can both be made more powerful by a simple randomization involving one independent uniform random variable. As a corollary, the Hommel test under arbitrary dependence is also improved. Importantly, our randomized improvements are never worse than the originals and are typically strictly more powerful, with marked improvements in simulations. The same technique also improves essentially every other multiple testing procedure based on e-values. Copyright © 2023, The Authors. All rights reserved.

关键词： Random processes

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：