检索结果-内蒙古大学图书馆

Self-stacking random weight neural network with multi-layer features fusion

INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS 2022年第7期13卷 1945-1957页

作者： Fu, Aimin Liu, Jingna Zhang, Tian-Lun Hebei Univ Baoding 071002 Peoples R China Shijiazhuang Tiedao Univ State Key Lab Mech Behav & Syst Safety Traff Engn Shijiazhuang 050043 Hebei Peoples R China Shijiazhuang Tiedao Univ Coll Elect & Elect Engn Shijiazhuang 050043 Hebei Peoples R China Shenzhen Univ Shenzhen 518061 Peoples R China

Randomized methods are practical and efficient for training the connectionist models. In this paper, we contribute to develop a self-stacking random weight neural network. Two different methods of feature fusion are proposed in this paper. The first one inter-connects the coarse and high level features to make the classification decision more diverse by using the proposed hierarchical network architecture with dense connectivity. On the other hand, the different decisions all-throughout the network are incorporated by a novel non-linear ensemble learning in an end-to-end manner. Through experiments, we verified the effectiveness of random features fusion, and even if each hierarchical branch in the network has very unfavorable accuracy, the proposed ensemble learning presents the impressive performance to boost the classification results. Moreover, the proposed connectionist model is applied to address one practice engineering problem of gearbox fault diagnosis, and the simulation demonstrates that our method has better robust to the noise in vibration signal of working gearbox.

关键词： Random weight neural network Stacking ensemble learning Feature fusion classification algorithm Gearbox fault diagnosis

来源：评论

学校读者我要写书评

暂无评论

Construction and enumeration of self-orthogonal and self-dual codes over Galois rings of even characteristic

引用

DESIGNS CODES AND CRYPTOGRAPHY 2024年第2期92卷 303-339页

作者： Yadav, Monika Sharma, Anuradha IIIT Delhi Dept Math Delhi 110020NEW India

Let e >= 2 and r >= 1 be integers, and let R-e,R-r denote the Galois ring of characteristic 2(e) and cardinality 2(er). The Teichmuller set T-r of the Galois ring R-e,R-r can be viewed as the finite field of order 2(r) under the addition operation. and the multiplication operation of R-e,R-r, where for a, b is an element of Tr, a circle plus b is the unique element in T-r satisfying a circle plus b = (a + b) (mod 2). Now a linear code C of length n over T-r is said to be k-doubly even if it has a k-dimensional linear subcode C-0 satisfying c center dot c = 0 (mod 4) for all c is an element of C-0, where each c is an element of C-0 is viewed as an element of R-e,r(n) and center dot denotes the Euclidean bilinear form on R-e,r(n). A k-doubly even code of length n and dimension k over T-r is simply called a doubly even code. In this paper, we count all doubly even codes over T-r and their two special classes, viz. the codes containing the all-one vector and the codes that do not contain the all-one vector by studying the geometry of a certain special quadratic space over T-r. We further provide a recursive method to construct self-orthogonal and self-dual codes of the type {k(1), k(2),..., k(e)} and length n over R-e,R-r from a (k(1)+ k(2)+ ...+ k(left) (perpendiculare/2right perpendicular))-doubly even self-orthogonal code of the same length n and dimension k(1) + k(2) + ... + k(inverted right perpendiculare/2inverted left perpendicular) over T-r, where n is a positive integer and k(1), k(2),..., k(e) are non-negative integers satisfying 2k(1) + 2k(2) + ... + 2k(e-i+1) + k(e-i+2) + k(e-i+3) + ... + k(i) <= n for inverted right perpendiculare+1/2inverted left perpendicular <= i <= e, (here left perpendicuar center dot right perpendicuar denotes the floor function and inverted right perpendicular center dot inverted left perpendicular denotes the ceiling function). With the help of this recursive construction method and the enumeration formulae for doubly even codes ove

关键词： classification algorithm Galois rings Doubly even codes Witt decomposition

来源：评论

学校读者我要写书评

暂无评论

Software to quantify and map vegetative cover in fallow fields for weed management decisions

引用

COMPUTERS AND ELECTRONICS IN AGRICULTURE 2011年第1期78卷 106-115页

作者： Wiles, L. J. USDA ARS Water Management Res Unit Ft Collins CO 80526 USA

Mapping weed cover during the fallow period of dryland crop rotations would be valuable for weed management in subsequent crops and could be done with low cost color digital cameras, however most managers lack the specialized software and expertise needed to create a map from the images. A system of software was developed to quantify weed cover in fallow fields in digital images and to simplify and automate the most challenging tasks that non-GIS professionals confront in creating and using maps derived from a large number of images. A GIS file of image locations is created with inexpensive consumer software. Images are classified, a GIS file is generated and the map is displayed in a simple GIS viewer with free software we developed. A map can be generated from 1000 images and 5000 GPS coordinates in 30 min, including image classification. The classified and original images for all locations can be viewed together easily from the map application. The accuracy of estimating weed cover was evaluated using images collected in 15 fields under natural light with a consumer grade camera mounted on an ATV driving 8-11 km h(-1). Weed cover was estimated with 96% accuracy for images, regardless of the amount of crop residue, unless part of the image was shaded by the camera. In those images, accuracy was 90% or better. This system will work with many professional and consumer digital cameras and GPS units and the classification algorithm can be easily modified for other applications. Published by Elsevier B.V.

关键词： Digital images Image analysis Weed map classification algorithm Site-specific weed management

来源：评论

学校读者我要写书评

暂无评论

Optimal selection of potential customer range through the union sequential pattern by using a response model

引用

EXPERT SYSTEMS WITH APPLICATIONS 2011年第6期38卷 7451-7461页

作者： Chen, Wen-Chin Hsu, Chiun-Chieh Hsu, Jing-Ning Natl Taiwan Univ Sci & Technol Dept Informat Management Taipei Taiwan Chunghwa Telecom Co Ltd Telecommun Labs Business & Mkt Strategy Res Dept Taipei Taiwan Chunghwa Telecom Co Ltd Data Commun Business Grp Dept Mkt Taipei Taiwan

Direct marketing is a common data mining application. Previous studies largely adopt the approach that, as the subjects for response model predictions, the entire customer population or filtered through certain attribute values is based on recommendations made from sales professionals. However, such methods may reduce response rates due to an oversized potential customer population, thus diminishing the accuracy of the prediction model. In resolve this problem, this work presents proposes a novel forecasting method that integrates the union sequential pattern with classification algorithms to facilitate the construction of customer response models. Based on use of a union sequential pattern, the potential customer size is established by identifying attributes with a high level of association. The prediction model is then constructed using classification algorithms such as support vector machines and logistic regression. Consequently, the problem involving the setting of range for potential customers can be solved, as well as the time spent on processing extended lists of customers during prediction. Finally, predicted potential Internet-phone customers and churning mobile-phone customers of a telecommunication company in Taiwan as are taken as an illustrative example, based on the proposed prediction model. The proposed method more accurately predicts potential customers than those of previous studies. Crown Copyright (c) 2010 Published by Elsevier Ltd. All rights reserved.

关键词： Response model Union sequential pattern classification algorithm Support vector machine Logistic regression

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Machine Learning Framework for Predicting Students' Performance in Virtual Learning Environment

INTERNATIONAL JOURNAL OF EMERGING TECHNOLOGIES IN LEARNING

引用

INTERNATIONAL JOURNAL OF EMERGING TECHNOLOGIES IN LEARNING 2021年第24期16卷 255-272页

作者： Evangelista, Edmund Zayed Univ Abu Dhabi Campus Abu Dhabi U Arab Emirates

Virtual Learning Environments (VLE), such as Moodie and Blackboard, store vast data to help identify students' performance and engagement. As a result, researchers have been focusing their efforts on assisting educational institutions in providing machine learning models to predict at-risk students and to improve their performance. However, it requires an efficient approach to construct a model that can ultimately provide accurate predictions. Consequently, this study proposed a hybrid machine learning framework to predict students' performance using eight classification algorithms and three ensemble methods (Bagging, Boosting, Voting) to determine the best-performing predictive model. In addition, this study used filter-based and wrapper-based feature selection techniques to select the best features of the dataset related to students' performance. The obtained results reveal that the ensemble methods recorded higher predictive accuracy when compared to single classifiers. Furthermore, the accuracy of the models improved due to the feature selection techniques utilized in this study.

关键词： machine learning Weka predictive model ensemble student performance prediction classification algorithm virtual learning

来源：评论

学校读者我要写书评

暂无评论

Classified JPEG coding of mixed document images for printing

引用

IEEE TRANSACTIONS ON IMAGE PROCESSING 2000年第4期9卷 716-720页

作者： Ramos, MG de Queiroz, RL Cornell Univ Ithaca NY 14853 USA Xerox Corp Webster NY 14580 USA

This paper presents a modified JPEG coder that is applied to the compression of mixed documents (containing test, natural images, and graphics) for printing purposes. The modified JPEG;coder proposed in this paper takes advantage of the distinct perceptually significant regions in these documents to achieve higher perceptual quality than the standard JPEG coder. The region-adaptivity. is performed via classified thresholding being totally compliant with the baseline standard. ii computationally efficient classification algorithm is presented, and the improved performance of the classified JPEG coder is verified.

关键词： .Mixed Document JPEG coder classification algorithm JPEG baseline

来源：评论

学校读者我要写书评

暂无评论

Detection of waterlogging stress based on hyperspectral images of oilseed rape leaves (Brassica napus L.)

引用

COMPUTERS AND ELECTRONICS IN AGRICULTURE 2019年 159卷 59-68页

作者： Xia Ji'An Cao HongXin Yang YuWang Zhang WeiXin Wan Qian Xu Lei Ge DaoKuo Zhang WenYu Ke YaQi Huang Bo Nanjing Univ Sci & Technol Sch Comp Sci & Engn Nanjing 210094 Jiangsu Peoples R China Jiangsu Acad Agr Sci Inst Agr Informat Nanjing 210014 Jiangsu Peoples R China

The objective of this study was to investigate the utility of hyperspectral images (HSIs) for the detection of oilseed rape waterlogging stress. We assessed two oilseed rape varieties, the non-hybrid NingYou 22 (NY 22) and hybrid NingZa 19 (NZ 19), and HSIs of oilseed rape leaves under different durations of waterlogging stress (0, 3, and 6 days) were collected to build three datasets (NY 22, NZ 19, and both combined). We extracted red-green blue (RGB) images and visible and near-infrared (VNIR 400-1000 nm) spectra from a region of interest (ROI) in each HSI. Quadratic discriminant analysis (QDA), k-nearest neighbor (KNN), and support vector machine (SVM) classifiers were used to build classification models for comparing images and spectra of samples under different waterlogging levels among the three datasets, and to conduct training and prediction. From each dataset, 70% of the images were used for training, and the remaining 30% were used for testing. In the classification of full-wavelength HSIs (400-1000 nm), QDA and SVM exhibited high multivariate classification accuracy, reaching 77.37% and 95.90% accuracy, respectively. In contrast, KNN displayed low accuracy, but good identification and prediction ability for variety NZ 19. Six optimal wavebands of 529, 641, 698, 749, 856, and 979 nm were used as input for successive projections algorithm (SPA) classification and analysis. The QDA mode had better classification performance, with identification accuracies of 100% and 94.44%, respectively. Overall, the VNIR classification results exceeded those of image classification. These results show that hyperspectral imaging technology is feasible and useful for the detection of oilseed rape waterlogging stress.

关键词： Rapeseed leaves Waterlogging stress Hyperspectral imaging Image processing Spectral analysis Machine learning classification algorithm

来源：评论

学校读者我要写书评

暂无评论

A stochastic algorithm for feature selection in pattern recognition

引用

JOURNAL OF MACHINE LEARNING RESEARCH 2007年第3期8卷 509-547页

作者： Gadat, Sebastien Younes, Laurent ENS Cachan CMLA F-94235 Cachan France Johns Hopkins Univ Ctr Imaging Sci Baltimore MD 21218 USA

We introduce a new model addressing feature selection from a large dictionary of variables that can be computed from a signal or an image. Features are extracted according to an efficiency criterion, on the basis of specified classification or recognition tasks. This is done by estimating a probability distribution P on the complete dictionary, which distributes its mass over the more efficient, or informative, components. We implement a stochastic gradient descent algorithm, using the probability as a state variable and optimizing a multi-task goodness of fit criterion for classifiers based on variable randomly chosen according to P. We then generate classifiers from the optimal distribution of weights learned on the training set. The method is first tested on several pattern recognition problems including face detection, handwritten digit recognition, spam classification and micro-array analysis. We then compare our approach with other step-wise algorithms like random forests or recursive feature elimination.

关键词： stochastic learning algorithms Robbins-Monro application pattern recognition classification algorithm feature selection

来源：评论

学校读者我要写书评

暂无评论

Developing and applying novel spectral feature parameters for classifying soil salt types in arid land

引用

ECOLOGICAL INDICATORS 2015年 54卷 116-123页

作者： Jin, Pingbin Li, Pingheng Wang, Quan Pu, Zhi Zhejiang Univ Dept Geosci Hangzhou 310027 Zhejiang Peoples R China Chinese Acad Sci Xinjiang Inst Ecol & Geog Urumqi 830011 Peoples R China Shizuoka Univ Grad Sch Agr Shizuoka 4228529 Japan

Soil salinization is a major desertification process that threatens especially the stability of arid ecosystems. There is an urgent need for intensive monitoring and quick assessment of salinization through remote sensing as a tool for combating desertification in such ecosystems. Recent researches have revealed that in order to retrieve soil salt contents accurately from hyperspectral reflectance, a pre-knowledge of salt types is required, which greatly outlines the spectral features of saline soil reflectance. In this study, a set of feature parameters have been developed after a thorough investigation of spectral responses to different soil salt types and salt contents for quick and accurate classification of soil salt types. The application has been validated using three independent datasets composed from: laboratory experiments (dataset I), insitu field measurements (dataset II), and satellite-borne Hyperion image (dataset III). For comparison, four other common classification algorithms have been validated using the same datasets. The results showed that the new approach proposed in this study performed well with not only single-type but also multipletype salts for which the four common algorithms performed rather fairly. Furthermore, validating using datasets II and III showed that the newly proposed approach had a stable performance while the other four failed, indicating the advantage of the new approach. The feature parameters developed in this study hence provide a novel and efficient approach for salt type classification from reflectance spectra, and we foresee its potential applications on large-scale soil salt type mapping towards better understanding soil salinity characterization from remote sensing data. (C) 2015 Elsevier Ltd. All rights reserved.

关键词： Arid land Soil salinity Salt type Hyperspectral Feature parameter classification algorithm

来源：评论

学校读者我要写书评

暂无评论

Superiority Demonstration of Variance-Considered Machines by Comparing Error Rate with Support Vector Machines

引用

INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS 2011年第3期9卷 595-600页

作者： Yeom, Hong-Gi Park, Seung-Min Park, Junheong Sim, Kwee-Bo Chung Ang Univ Sch Elect & Elect Engn Seoul South Korea

To improve the performance of classification algorithms, we proposed a new variance-considered machine (VCM) classification algorithm in a previous study. The study showed theoretically that VCMs have lower error probabilities than SVMs. The purpose of this paper is to experimentally demonstrate the superiority of VCMs. Therefore, we verified our proposal with several case experiments using data following a Gaussian distribution with different variances and prior probabilities. To estimate performance, the experiment for each case was executed 1000 times and the error rates were averaged for accuracy. The data of each experiment have different distances between means of data, and different ratios between training data and testing data. Thus, we proved that the error rate of VCMs is lower than the error rate of SVMs, although their performances were not similar in each case. Consequently, we expect that VCMs will be applied to a variety fields.

关键词： classification algorithm optimal hyperplane support vector machine variance-considered machine

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：