检索结果-内蒙古大学图书馆

An Abstractive Summarization Technique with Variable Length Keywords as per Document Diversity

computers, Materials & Continua 2021年第3期66卷 2409-2423页

作者： Muhammad Yahya Saeed Muhammad Awais Muhammad Younas Muhammad Arif Shah Atif Khan M.Irfan Uddin Marwan Mahmoud Department of Software Engineering Government College UniversityFaisalabadFaisalabad38000Pakistan Department of IT and Computer Science Pak-Austria Fachhochschule Institute of Applied Sciences and TechnologyHaripurPakistan Department of Computer Science Islamia College PeshawarPeshawarPakistan Institute of Computing Kohat University of Science and TechnologyKohatPakistan Faculty of Applied Studies King Abdulaziz UniversityJeddahSaudi Arabia

Text Summarization is an essential area in text mining,which has procedures for text *** natural language processing,text summarization maps the documents to a representative set of descriptive ***,the objective of text extraction is to attain reduced expressive contents from the text *** summarization has two main areas such as abstractive,and extractive *** text summarization has further two approaches,in which the first approach applies the sentence score algorithm,and the second approach follows the word embedding *** such text extractions have limitations in providing the basic theme of the underlying *** this paper,we have employed text summarization by TF-IDF with PageRank keywords,sentence score algorithm,and Word2Vec word *** study compared these forms of the text summarizations with the actual text,by calculating cosine ***,TF-IDF based PageRank keywords are extracted from the other two extractive *** intersection over these three types of TD-IDF keywords to generate the more representative set of keywords for each text document is *** technique generates variable-length keywords as per document diversity instead of selecting fixedlength keywords for each *** form of abstractive summarization improves metadata similarity to the original text compared to all other forms of summarized *** also solves the issue of deciding the number of representative keywords for a specific text *** evaluate the technique,the study used a sample of more than eighteen hundred text *** abstractive summarization follows the principles of deep learning to create uniform similarity of extracted words with actual text and all other forms of text *** proposed technique provides a stable measure of similarity as compared to existing forms of text summarization.

关键词： Metadata page rank sentence score word2vec cosine similarity This

来源：评论

学校读者我要写书评

暂无评论

Covid-19 Data Preprocessing Approach in Machine Learning for Prediction 1st

Covid-19 Data Preprocessing Approach in Machine Learning for...

引用

1st EAI International Conference on Safe, Secure, Ethical, Responsible Technologies and Emerging Applications, SAFER-TEA 2023

作者： Kotva Goudoungou, Samuel Dayang, Paul Tchomte, Naomi Dassi Ngossaha, Justin Moskolaï Moffo, Fabien Mouomene Mitton, Nathalie Department of Mathematics and Computer Science Faculty of Science The University of Ngaoundere Ngaoundere Cameroon Department of Mathematics and Computer Science University Institute of Technology The University of Ngaoundere Ngaoundere Cameroon Department of Computer Science Faculty of Science The University of Douala Douala Cameroon National Institute for Research in Digital Science and Technology University of Lille Villeneuve D’ascq59650 France

ISBN: (纸本)9783031563959

Artificial Intelligence (AI) is a vast field that allows the development of programs capable of simulating human intelligence. One of the most used AI techniques that is very important the preparation of raw data which is called data preprocessing in Machine Learning. Data preprocessing is a technique that makes the data clean for machine learning and improves the performance of the model. With the different steps that we found already defined we set up a simpler architecture for data preprocessing. For that purpose, we worked with 5644 data simple of Covid-19. After analyzing the contours of our data;we have as steps of this architecture first the collection of data, the cleaning of these data, the encoding, the normalization of the data, the imputation to manage the missing values, then the selection of variables and finally the selection of features. We obtain an accuracy of 78% as a result of our experimentation with a very small error rate. This Covid-19 detection prediction result shows that data preprocessing plays an important role in Machine Learning. © ICST Institute for computer sciences, Social Informatics and Telecommunications Engineering 2024.

关键词： Intelligent systems

来源：评论

学校读者我要写书评

暂无评论

Comparative study and review of grid, cloud, utility computing and software as a service for use by libraries

引用

Library Hi Tech News 2011年第3期28卷 25-32页

作者： Patel, Ahmed Seyfi, Ali Tew, Yiqi Jaradat, Ayman School of Computer Science Faculty of Information Science and Technology Universiti Kebangsaan Malaysia (The National University of Malaysia) Bangi Malaysia Computing Information Systems and Mathematics Kingston University Kingston upon Thames United Kingdom Faculty of Computer Science and Information Technology University of Malaya Kuala Lumpur Malaysia Computer and Information Technology Al-Madinah International University Shah Alam Malaysia

Purpose: Grid computing, cloud computing (CC), utility computing and software as a service are emerging technologies predicted to result in massive consolidation as meta-level computing services of everything beneath one umbrella in the future. The purpose of this study is to foster the understanding and differentiation, by using the three aforementioned types of computing technologies and software, as a service by both public and private libraries to meet their expectations and strategic objectives. Design/methodology/approach: The approach in this study is a review based on comparing the four computing technologies with a brief analysis for researching and designing the mind map of a new meta-level computing service approach, taking into consideration the need for new economic tariff and pricing models as well as service-level agreements. Findings: Since it is anticipated that there will be likely potential consolidation and integration of computing services, a study of these four most advanced computing technologies and their methodologies is presented through their definition, characteristics, functionalities, advantages and disadvantages. This is a well-timed technological advancement for libraries. Practical implications: It appears that the future of library services will become even more integrated, running over CC platforms based on usage rather than just storage of data. Social implications: Libraries will become an open useful resource to all and sundry in a global context, and that will have huge societal benefits never imagined before. Originality/value: Concisely addresses the strategies, functional characteristics, advantages and disadvantages by comparing these technologies from several service aspects with a view to assisting in creating the next generation outer space computing. © Emerald Group Publishing Limited.

关键词： Communication technologies Libraries Metadata Pricing Service levels

来源：评论

学校读者我要写书评

暂无评论

Octahedron-shaped Convolution for Refining Aorta Semantic Segmentation 5

Octahedron-shaped Convolution for Refining Aorta Semantic Se...

引用

5th Asian Conference on Artificial Intelligence technology, ACAIT 2021

作者： Xiang, Xi Luo, Gongning Zhao, Pengfei Wang, Wei Wang, Kuanquan Harbin Institute of Technology Faculty of Computing Harbin China Harbin Institute of Technology School of Computer Science and Technology Shenzhen China

ISBN: (纸本)9781665426305

Refining 3D aorta segmentation is essential for clinical aorta analysis. The small tubular diameter of the aorta branches and the discontinuity of neighboring information make it difficult to get a continuous semantic segmentation map. In this paper, we propose a novel octahedron-shaped convolution (OSC) based on VNet and signed distance map (SDM). OSC aimes to aggregate more contextual information for each sample point in the aortic branches with small tubular diameters. SDM is introduced as auxiliary information to measure the similarity of neighboring points, and it provides weights for feature fusion. Furthermore, we embed the learned 3D offset field into OSC to avoid inaccurate segmentation on the region around the narrow tubular structures. The OSC module prolongs the predicted length of small aorta branches and then improves the tubular continuity of the aorta segmentation map. We evaluate the OSC module on our-collected dataset and MICCAI ASOCA2020 coronary artery dataset. Our method has achieved the state-of-the-art results in terms of Dice and Jaccard metrics. © 2021 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

A hybrid swarm optimization for neural network training with application in stock price forecasting

A hybrid swarm optimization for neural network training with...

引用

2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016

作者： Pan, Jianjia Tang, Yuan Yan Yulong, Wang Xianwei, Zheng Luo, Huiwu Yuan, Haoliang Wang, Patrick Shen Pei Department of Computer and Information Science Faculty of Science and Technology University of Macau China College of Computer and Information Science Northeastern University BostonMA02115 United States

ISBN: (纸本)9781509018970

A improved swarm optimization method based on particle swarm optimization (PSO) and simplified swarm optimization (SSO) is proposed to adjust the weight in artificial neural network. This method is a modification of traditional PSO and SSO, and combines them to a new optimization method (PSOSSO for short). The proposed method overcomes some of the drawbacks of SSO and improves its ability to train the weight of ANN. In the experiments, the PSOSSO is employed to train fuzzy wavelet neural network (FWNN) forecasting model to predict the prices of Hong Kong Hang Seng Index. The experimental results present that the PSOSSO is more efficient than traditional PSO and SSO methods. © 2016 IEEE.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Finding exemplars in dense data with affinity propagation on clusters of GPUs

Finding exemplars in dense data with affinity propagation on...

引用

作者： Kurdziel, Marcin Boryczko, Krzysztof Department of Computer Science Faculty of Computer Science Electronics and Telecommunications AGH University of Science and Technology al. A. Mickiewicza 30 30-059 Krakow Poland

This work presents an efficient implementation of affinity propagation (AP) on clusters of graphical processing units (GPUs). AP is a state-of-the-art method for finding exemplars in data sets described by similarity matrices. It is typically employed in crisp clustering applications. However, when finding exemplars in an n-pattern data set with dense, non-metric similarities, AP performs iterative processing of three n × n floating point matrices. One of them stores the similarities, and the other two store the values that will ultimately pinpoint the exemplars. For large similarity matrices, AP is therefore computationally expensive. Although matrix operations of AP are well suited for GPUs, its memory footprint limits the size of tasks that can be solved on one unit. We present, however, a decomposition scheme for AP that distributes the calculations over multiple GPUs, with low communication-to- computation ratio. Because of this favorable communication pattern, our implementation finds exemplars in large, dense similarity data efficiently, even when GPUs are connected by a slow network. Furthermore, by combining global device memory of multiple GPUs, it can solve problems that would not fit in a single unit. Copyright © 2012 John Wiley & Sons, Ltd.

关键词： Cluster analysis

来源：评论

学校读者我要写书评

暂无评论

A novel hybrid algorithm for lucky imaging

引用

Research in Astronomy and Astrophysics 2021年第5期21卷 155-167页

作者： Jin-Liang Wang Bin-Hua Li Xi-Liang Zhang Faculty of information engineering and automation Kunming University of Science and TechnologyKunming 650500China Key Laboratory of Applications of Computer Technologies of the Yunnan Province Kunming University of Science and TechnologyKunming 650500China Yunnan Observatories Chinese Academy of SciencesKunming 650011China

Lucky imaging is a high-resolution astronomical image recovery technique with two classic implementation algorithms,*** selecting,shifting and adding in image space,and data selecting and image synthesizing in Fourier *** paper proposes a novel lucky imaging algorithm where with space-domain and frequency-domain selection rates as a link,the two classic algorithms are combined successfully,making each algorithm a proper subset of the novel hybrid *** results show that with the same experiment dataset and platform,the high-resolution image obtained by the proposed algorithm is superior to that obtained by the two classic *** paper also proposes a new lucky image selection and storage scheme,which can greatly save computer memory and enable lucky imaging algorithm to be implemented in a common desktop or laptop with small memory and to process astronomical images with more frames and larger *** addition,through simulation analysis,this paper discusses the binary star detection limits of the novel lucky imaging algorithm and traditional ones under different atmospheric conditions.

关键词： methods:data analysis techniques:high angular resolution techniques:image processing

来源：评论

学校读者我要写书评

暂无评论

Random projection initialization for deep neural networks 25

Random projection initialization for deep neural networks

引用

25th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2017

作者： Wójcik, Piotr Iwo Kurdziel, Marcin AGH University of Science and Technology Faculty of Computer Science Electronics and Telecommunications Department of Computer Science al. A. Mickiewicza 30 Krakow30-059 Poland

ISBN: (纸本)9782875870391

In this work we propose to initialize rectifier neural networks with random projection matrices. We focus on Convolutional Neural Networks and fully-connected networks with pretraining. Our results show, that in convolutional networks a well designed random projection initialization can perform better than the current state-of-the-art He's initialization. Specifically, in our evaluation, initialization based on the Subsampled Randomized Hadamard Transform consistently outperformed He's initialization on several evaluated image classification datasets. © ESANN 2017 - Proceedings, 25th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Protein structure prediction with improved quantum immune algorithm

Protein structure prediction with improved quantum immune al...

引用

International Conference on Intelligent Network and Intelligent Systems

作者： Zhu, Hongbing Wu, Jun Wu, Jianguo College of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Dept. of Computer Science Faculty of Informatics Hiroshima Kokusai Gakuin University Hiroshima Japan College of Information Science and Engineering Wuhan University of Science and Technology Wuhan China

ISBN: (纸本)9780769542492

A novel hybrid algorithm Quantum Immune(QI), which combines Quantum Algorithm (QA) and Immune Clonal Selection(ICS) Algorithm, has been presented for dealing with multi-extremum and multi-parameter problem based on AB off-lattice model in the predicting 2D protein folding structure. Clonal Selection Algorithm was introduced into the hypermutation operators in the Quantum Algorithm to improve the local search ability, and double chains quantum coded was designed to enlarge the probability of the global optimization solution. It showed that the solution mostly trap into the local optimum, to escape the local best solution the aging operator is introduced to improve the performance of the algorithm. Experimental results showed that the lowest energies and computing-time of the improved Quantum Clonal Selection(QCS) algorithm were better than that of the previous methods, and the QCS was further improved by adding aging operator to combat the premature convergence. Compared with previous approaches, the improved QCS algorithm remarkably enhanced the convergence performance and the search efficiency of the immune optimization algorithm. © 2010 IEEE.

关键词： Global optimization

来源：评论

学校读者我要写书评

暂无评论

A named entity recognition approach for tweet streams using active learning

A named entity recognition approach for tweet streams using ...

引用

作者： Tran, Van Cuong Hoang, Dinh Tuyen Nguyen, Ngoc Thanh Hwang, Dosam Department of Computer Engineering Yeungnam University Gyeongsan Korea Republic of Faculty of Engineering and Technology Quang Binh University Dong Hoi Viet Nam Faculty of Computer Science and Management Wroclaw University of Science and Technology Wroclaw Poland

In recent years, information extraction from tweets has been challenging for researchers in the fields of knowledge discovery and data mining. Unlike formal text, such as news articles and pieces of longer content, tweets are of a specific nature: Short, noisy, and with dynamic content. Thus, it is difficult to apply the traditional natural language processing algorithms to analyze them. Active learning is well-suited to many problems in natural language processing, especially when unlabeled data may be abundant, but labeled data is limited. The method proposed here aims to minimize annotation costs while maximizing the desired performance from the model. The method recognizes named entities from tweet streams on Twitter by using an active learning method with different query strategies. The tweets are queried for labeling by a human annotator based on query-by-committee, uncertainty-based sampling, and diversity-based sampling. The experimental evaluations of the proposed method on tweet data achieved better results than random sampling. © 2017 - IOS Press and the authors. All rights reserved.

关键词： Data mining

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：