检索结果-内蒙古大学图书馆

IEEE Congress on Evolutionary Computation

作者： Li, Yamin Ma, Jinru Zhao, Qiuxia Agr Univ Hebei Artificial Intelligence Res Ctr Baoding 071001 Peoples R China Agr Univ Hebei Coll Foreign Language Baoding 071001 Peoples R China

ISBN: (纸本)9781424418220

A new classification algorithm for mufti-image classification in genetic programming (GP) is introduced, which is the centered dynamic class boundary determination with quick-decreasing power value of arithmetic progression. In the classifier learning process using GP for mufti-image classification, different sets of power values are tested to achieve a more suitable range of margin values for the improvement of the accuracy of the classifiers. In the second development, the program size is introduced into the fitness function to control the size of program growth during the evolutionary learning process. The approach is examined on a Chinese character image data set and a grass leaves data set, both of which have four or more classes. The experimental results show that while dealing with complicated problems of mufti-image classification, the new approach can he used for more accurate classification and work better than the previous algorithms of either static or dynamic class boundary determination. With the fitness function, the size of the programs in the population can be controlled effectively and shortened considerably during evolution. Thus, the readability of the programs could be seemingly improved.

关键词： Genetic programming (GP) Image classification classification algorithm Fitness function construction

来源：评论

学校读者我要写书评

暂无评论

Study on Decision classification Fusion Model

Study on Decision Classification Fusion Model

引用

9th International Conference on Hybrid Intelligent Systems (HIS 2009)

作者： Zhang, Xiao-dan Niu, Zhen-dong Beijing Inst Technol Univ Sch Comp Sci & Technol Beijing Peoples R China

ISBN: (纸本)9780769537450

In this paper, a general decision layer classification fusion model, based on information fusion for improving classification precision, is proposed, that is, different multi-classification algorithms as the feature layer doing respective classification, and the results of classification algorithms are Input into decision level, the last classification result is output. This model is applied into improving precision of text classification. And the model is used to the Computer Center of some department. Through the experiment, the text classification fusion model can improve the classification precision effectively

关键词： text classification decision layer classification fusion model classification algorithm information fusion

来源：评论

学校读者我要写书评

暂无评论

classification of heart diseases from ECG signals using wavelet transform and kNN classifier

Classification of heart diseases from ECG signals using wave...

引用

International Conference on Computing, Communication & Automation (ICCCA)

作者： Saini, Ridhi Bindal, Namita Bansal, Puneet Kurukshetra Univ Univ Inst Engn & Technol Elect & Commun Engn Dept Kurukshetra Haryana India

ISBN: (纸本)9781479988907

Heart is the most vital organ which circulates blood along with nutrients and oxygen throughout the body. There are number of reasons which may affect its normal working. In this paper ten heart diseases, as well as normal, have been classified by extracting features from original ECG (electrocardiogram) signals and sixth level wavelet transformed ECG signals. The results have been compared and improved accuracy has been obtained using wavelet transformed signals.

关键词： classification algorithm Discrete wavelet transforms Electrocardiography Principal component analysis Statistical analysis

来源：评论

学校读者我要写书评

暂无评论

Should the Same Learners Be Used Both Within Wrapper Feature Selection and for Building classification Models?

Should the Same Learners Be Used Both Within Wrapper Feature...

引用

25th IEEE International Conference on Tools with Artificial Intelligence (ICTAI)

作者： Wald, Randall Khoshgoftaar, Taghi Napolitano, Amri Florida Atlantic Univ Boca Raton FL 33431 USA

ISBN: (纸本)9781479929719

Due to the problem of high-dimensionality (datasets which contain many independent attributes or features), feature selection has become an important part of data mining research. One popular form of feature selection, wrapper selection, chooses the best features by directly addressing the question of which features build the best models. Various feature subsets are used to build classification models, and the performance of these models is the score of each feature subset. The feature subset with the best score is then used to build the final classification model. As wrappers use a classification algorithm (learner) both to select the features and to build a predictive model, it has been traditional to use the same learner for both, such that the features chosen will be those which optimize that model's performance. However, no research has considered whether having different learners operate inside and outside the wrapper (that is, for selecting the features and for building the final model) might actually result in improved classification performance. In this work, we consider five learners both inside the wrapper and for building the classification model, along with two datasets drawn from the domain of Twitter profile mining. By considering both the raw performance values and a statistical analysis, we find that contrary to intuition, usually the best performance for a given choice of external learner is not found by using the same learner within the wrapper. Instead, the Naive Bayes learner is usually the best choice for selecting features, regardless of which learner is used for the external model. We also find that Multi-Layer Perceptron is able to build consistent classification models for many different choices of internal learner. Finally, the 5-Nearest Neighbor learner gave poor results both inside and outside the wrapper.

关键词： Wrapper feature selection classification algorithm learner Twitter social bots

来源：评论

学校读者我要写书评

暂无评论

An efficient approach for clustering and classification for fraud detection using bankruptcy data in IoT environment

引用

International Journal of Information Technology (Singapore) 2021年第6期13卷 2497-2503页

作者： Srikanth, Panigrahi Department of CSE GMR Institute of Technology GMR Nagar Rajam 532127 Andhra Pradesh India

The fraud identification area is distinguished by a classify the data (also known as fraud/legitimate) which are not representative general of the population due to sampling bias. To deal with unsupervised fraud detection (FD) methods, which are uncommon in policies fraud detection studies, we use with these problems with the domain. Our approach consists of applying FD approaches to companies in the same sector and facilitates an effective auditing technique that can be implemented by policy authorities. Therefore, new technology must be researched and evaluated in order to leverage the insight gained from the study for the prediction of fraud and for the creation of systems to avoid fraud with additional checks for minimizing suspicious occurrences. Increased reliance on emerging technology such as Internet of Things (IoT) with FD has compounded the issue in recent years. In this context, the key challenges are to determine the proposed novel similarity measure to satisfy properties, to generate clusters with the proposed measure of similarity and to minimize dimensionality, and to recognize fraud to increase machine efficiency. Our administration involves the regulation issuers to effectively identify illegal activities to protect consumer rights and reduce fraud risks and compliance costs. Our proposed methods are highly efficient and successful have demonstrated real time polish companies data to improve high accuracy and system performance. © 2021, Bharati Vidyapeeth's Institute of Computer Applications and Management.

关键词： Bankruptcy data classification algorithm Clustering algorithm Dimensionality reduction Fraud detection with IoT Similarity measure

来源：评论

学校读者我要写书评

暂无评论

Quantum Entanglement Inspired Correlation Learning for classification 26th

Quantum Entanglement Inspired Correlation Learning for Class...

引用

26th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD)

作者： Zhang, Junwei Li, Zhao Wang, Juan Wang, Yinghui Hu, Shichang Xiao, Jie Li, Zhaolin Tianjin Univ Coll Intelligence & Comp1 Tianjin Peoples R China Zhejiang Univ Hangzhou Zhejiang Peoples R China Alibaba Grp Hangzhou Zhejiang Peoples R China Hangzhou Yugu Technol Co Ltd Hangzhou Peoples R China Tianjin Xiniu Huaan Technol Co Ltd Tianjin Peoples R China

ISBN: (纸本)9783031059360;9783031059353

Correlation is an important information resource, which is often used as a fundamental quantity for modeling tasks in machine learning. Since correlation between quantum entangled systems often surpasses that between classical systems, quantum information processing methods show superiority that classical methods do not possess. In this paper, we study the virtue of entangled systems and propose a novel classification algorithm called Quantum Entanglement inspired the classification algorithm (QECA). Particularly, we use the joint probability derived from entangled systems to model correlation between features and categories, that is, Quantum Correlation (QC), and leverage it to develop a novel QC-induced Multi-layer Perceptron framework for classification tasks. Experimental results on four datasets from diverse domains show that QECA is significantly better than the baseline methods, which demonstrates that QC revealed by entangled systems can improve the classification performance of traditional algorithms.

关键词： Quantum correlation Quantum-inspired algorithms classification algorithm

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Two Matrix Factorization Methods Applied to the classification of Gene Expression Data

A Comparative Study of Two Matrix Factorization Methods Appl...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Nikulin, Vladimir Huang, Tian-Hsiang McLachlan, Geoffrey J. Univ Queensland Dept Math Brisbane Qld Australia Natl Cheng Kung Univ Inst Informat Management Tainan Taiwan Univ Queensland Inst Mol Biosci Brisbane Qld Australia

ISBN: (纸本)9781424483075

In microarray data analysis, dimension reduction is an important consideration in the construction of a successful classification algorithm. As an alternative to feature selection, we use a well-known matrix factorisation method. For example, we can employ the popular singular-value decomposition (SVD) or nonnegative matrix factorization. In this paper, we consider a novel algorithm for gradient-based matrix factorisation (GMF). We compare GMF and SVD in their application to five gene expression datasets. The experimental results show that our method is faster, more stable, and sensitive.

关键词： SVD bioinformatics biological techniques classification algorithm data reduction dimension reduction gene expression data classification gene expression datasets gradient based matrix factorisation gradient methods matrix algebra matrix factorization meth

来源：评论

学校读者我要写书评

暂无评论

classification for Trademark Image Based on Normal Cloud Model

Classification for Trademark Image Based on Normal Cloud Mod...

引用

2nd International Conference on Information Management, Innovation Management and Industrial Engineering

作者： Qi, Ya-Li Beijing Inst Graph Commun Dept Comp Beijing 102600 Peoples R China

ISBN: (纸本)9780769538761

classification is an important research topic in the field of image data mining. There have been many data classification methods studied, including decision-tree method, statistical methods, neural networks, rough sets, etc. This paper proposed a method to classify the image with normal cloud model which is an uncertainty transformation model between quantities and qualities conception. We develop the algorithm for classification based on normal cloud model. Finally we perform experiments on an artificial trademark image database. The results show the advantages of the cloud model in the process of classification.

关键词： data mining thecloud model classification algorithm trademark image retrival

来源：评论

学校读者我要写书评

暂无评论

Research on Pulsar classification Based on Machine Learning 3

Research on Pulsar Classification Based on Machine Learning

引用

3rd International Conference on Intelligent Autonomous Systems (ICoIAS)

作者： Chen, Zhiyu Xu, Aiting Zhou, Yingying Gai, Yingjie Zhejiang Gongshang Univ Coll Stat & Math 18 Xuezheng Str Hangzhou 310018 Peoples R China

ISBN: (纸本)9781728160788

This paper uses pulsar signal data for data mining, on the basis of exploratory analysis, constructs a variety of classification models, such as Random Forest, SVM, Logical Regression, K-Nearest Neighbor, Naive Bayes, Decision tree, AdaBoost classifier, GBDT and XGBoost, to classify pulsar candidate samples. It is hoped that valuable suspected pulsar samples can be effectively screened from massive data for further observation and confirmation.

关键词： pulsar machine learning classification algorithm random forest

来源：评论

学校读者我要写书评

暂无评论

Emotion classification of Songs Using Deep Learning

Emotion Classification of Songs Using Deep Learning

引用

International Conference on Green Energy, Computing and Sustainable Technology (GECOST)

作者： Mate, Nikita Akre, Durva Patil, Gaurav Sakarkar, Gopal Basuki, Thomas Anung GH Raisoni Coll Engn Dept Artificial Intelligence Nagpur Maharashtra India DY Patil Inst Master Comp Applicat & Management Pune Maharashtra India Univ Southampton Dept Comp Sci Iskandar Puteri Johor Malaysia

ISBN: (纸本)9781665486637

The sentiment prediction of songs will have a vast quantity of applications within the current era, like the Music recommendation system for request, choosing the music for public gatherings like procession or restaurants to enhance the emotional standing of a personal or cluster, seemingly folks and customers. A song is one way to say something indirectly and one of the most adventurous parts is that the one song could express more than one emotion. We can use songs to apologize, congratulate, express happiness or unhappiness, etc. Emotion is a vital part of daily life: it affects choice-making, conception, human interdependence, and human understanding. There are units of positive emotions and negative emotions;positive emotions are units a lot related to human health also as work potency, whereas negative emotions could cause health issues. In this research paper, we are going to use Deep Learning Frameworks, which are mostly used for classification techniques like Multilayer Perceptron (MLP), Gated Recurrent Units(GRU), Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM). In this paper, the dataset operated is the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS)[1] for the enactment of audio categorization. As a result, we obtained graphs for each Deep Learning framework that is for training and validation loss, training and validation accuracy. After implementation, we got the highest accuracy for CNN that is 92.71% and the least accuracy for MLP that is 52.60% and for LSTM and GRU it lies between the range of 63% - 66%.

关键词： Emotion classification Deep Learning algorithms classification algorithm

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：