检索结果-内蒙古大学图书馆

17th Conference on Industrial Electronics and Applications (ICIEA)

作者： Yang, Xuqiang Li, Yanbin Zhang, Yan Yang, Chunling Li, SuYing Harbin Inst Technol Sch Elect Egineering Harbin Peoples R China

ISBN: (纸本)9781665409841

Infrared target detection techniques have been widely used in infrared alarming and reconnaissance, and the detection task is often done by machine learning. Feature extraction and classification algorithm are two core components of the machine learning. The selection of features and algorithm have a determinative effect on the detection result. Traditional target classification techniques only focus on only limited combinations of features and classification algorithm, which may result in a poor detection result. Based on this, this paper selects the gray features, statistical features, frequency domain features, and graphic features of the infrared target, and compares the three classification algorithms of KNN, Bayes, and SVM to give the optimal combination of features and classification algorithms by experiment.

关键词： Infrared target detection Feature extraction classification algorithm

来源：评论

学校读者我要写书评

暂无评论

Research on the Multimedia Data Mining and classification algorithm based on the Database Optimization Techniques

引用

International Journal of Technology Management 2015年第11期 58-60页

作者： Hu Xiu Computer Engineering College Jingchu University of Technology JingmenHubei448000China

In this research article, we analyze the multimedia data mining and classification algorithm based on database optimization techniques. Of high performance application requirements of various kinds are springing up constantly makes parallel computer system structure is valued by more and more common but the corresponding software system development lags far behind the development of the hardware system, it is more obvious in the field of database technology application. Multimedia mining is different from the low level of computer multimedia processing technology and the former focuses on the extracted from huge multimedia collection mode which focused on specific features of understanding or extraction from a single multimedia objects. Our research provides new paradigm for the methodology which will be meaningful and necessary.

关键词： Data Mining classification algorithm Database Optimization Multimedia Source.

来源：评论

学校读者我要写书评

暂无评论

Consumer credit risk assessment: A review from the state-of-the-art classification algorithms, data traits, and learning methods

引用

EXPERT SYSTEMS WITH APPLICATIONS 2024年第PartA期237卷

作者： Zhang, Xiaoming Yu, Lean Jiangxi Univ Finance & Econ Sch Informat Management Nanchang 330032 Peoples R China Sichuan Univ Business Sch Chengdu 610065 Peoples R China Shenzhen Inst Technol Sch Business Shenzhen 518116 Peoples R China

Credit risk assessment is a crucial element in credit risk management. With the extensive research on consumer credit risk assessment in recent decades, the abundance of literature on this topic can be overwhelming for researchers. Therefore, this article aims to provide a more systematic and comprehensive analysis from three perspectives: classification algorithms, data traits, and learning methods. Firstly, the state-of-the-art classification algorithms are categorized into traditional single classifiers, intelligent single classifiers, hybrid and ensemble multiple classifiers. Secondly, considering the diversity of data traits in the credit dataset, data traits are divided into external structure information traits, data quality traits, data quantity traits, and internal information traits. Data traits-driven modeling framework based on multiple classifiers is proposed for solving credit risk assessment. Thirdly, considering the differences in data modeling methods, learning methods are classified into data status, label status, and structure form. Furthermore, model interpretability, model bias, model multi-pattern, and model fairness are discussed. Finally, the limitations and future research directions are presented. This review article serves as a helpful guide for researchers and practitioners in the field of credit risk modeling and analysis.

关键词： Credit risk assessment classification algorithm Data trait Learning method

来源：评论

学校读者我要写书评

暂无评论

On the Generalizability of Machine Learning classification algorithms and Their Application to the Framingham Heart Study

引用

INFORMATION 2024年第5期15卷 252-252页

作者： Kahouadji, Nabil NE Illinois Univ Coll Arts & Sci Dept Math 5500 N St Louis Ave Chicago IL 60625 USA

The use of machine learning algorithms in healthcare can amplify social injustices and health inequities. While the exacerbation of biases can occur and be compounded during problem selection, data collection, and outcome definition, this research pertains to the generalizability impediments that occur during the development and post-deployment of machine learning classification algorithms. Using the Framingham coronary heart disease data as a case study, we show how to effectively select a probability cutoff to convert a regression model for a dichotomous variable into a classifier. We then compare the sampling distribution of the predictive performance of eight machine learning classification algorithms under four stratified training/testing scenarios to test their generalizability and their potential to perpetuate biases. We show that both extreme gradient boosting and support vector machine are flawed when trained on an unbalanced dataset. We then show that the double discriminant scoring of type 1 and 2 is the most generalizable with respect to the true positive and negative rates, respectively, as it consistently outperforms the other classification algorithms, regardless of the training/testing scenario. Finally, we introduce a methodology to extract an optimal variable hierarchy for a classification algorithm and illustrate it on the overall, male and female Framingham coronary heart disease data.

关键词： machine learning classification algorithm health disparities variable selection methodology optimal variable hierarchy

来源：评论

学校读者我要写书评

暂无评论

Music Emotion classification Based on Heterogeneous Graph Neural Networks

引用

IEEE ACCESS 2025年 13卷 76473-76480页

作者： Guo, Jing-Ying Wang, Peng Coventry Univ Sch Media & Performing Arts Coventry CV1 5RW England Beijing City Univ Performance Dept Beijing 101309 Peoples R China

The classification of musical emotions is crucial for the indexing, structuring, searching, and recommending of tracks and albums across various music platforms. Consequently, the automated categorization of musical emotions has become a vital element in nearly all music applications. Recent studies have mainly concentrated on utilizing textual, audio, or multimodal data for genre classification, frequently neglecting the impact of singers, composers, and listener preferences. In practice, composers possess unique compositional styles, listeners have varied musical preferences, and singers focus on particular music genres. These different viewpoints offer significant insights into the classification of musical emotions, greatly enhancing the effectiveness of classification performance. In this paper, we introduce a novel heterogeneous graph neural network (HGN) that models the relationships of music emotion preferences among singers, composers, and listeners, in order to generate accurate node feature representations for downstream tasks. The experimental results show that our model significantly outperforms current state-of-the-art (SOTA) methods on two datasets for music emotion classification.

关键词： Feature extraction Music Vectors Semantics Accuracy Multiple signal classification Data models Spectrogram Attention mechanisms Reviews Music emotion classification algorithm heterogeneous graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Sample Size Requirements for Popular classification algorithms in Tabular Clinical Data: Empirical Study

引用

JOURNAL OF MEDICAL INTERNET RESEARCH 2024年 26卷 e60231页

作者： Silvey, Scott Liu, Jinze Virginia Commonwealth Univ Sch Publ Hlth Dept Biostat 830 East Main St Richmond VA 23219 USA

Background: The performance of a classification algorithm eventually reaches a point of diminishing returns, where the additional sample added does not improve the results. Thus, there is a need to determine an optimal sample size that maximizes performance while accounting for computational burden or budgetary concerns. Objective: This study aimed to determine optimal sample sizes and the relationships between sample size and dataset-level characteristics over a variety of binary classification algorithms. Methods: A total of 16 large open-source datasets were collected, each containing a binary clinical outcome. Furthermore, 4 machine learning algorithms were assessed: XGBoost (XGB), random forest (RF), logistic regression (LR), and neural networks (NNs). For each dataset, the cross-validated area under the curve (AUC) was calculated at increasing sample sizes, and learning curves were fit. Sample sizes needed to reach the observed full-dataset AUC minus 2 points (0.02) were calculated from the fitted learning curves and compared across the datasets and algorithms. Dataset-level characteristics, minority class proportion, full-dataset AUC, number of features, type of features, and degree of nonlinearity were examined. Negative binomial regression models were used to quantify relationships between these characteristics and expected sample sizes within each algorithm. A total of 4 multivariable models were constructed, which selected the best-fitting combination of dataset-level characteristics. Results: Among the 16 datasets (full-dataset sample sizes ranging from 70,000-1,000,000), median sample sizes were 9960 (XGB), 3404 (RF), 696 (LR), and 12,298 (NN) to reach AUC stability. For all 4 algorithms, more balanced classes (multiplier: 0.93-0.96 for a 1% increase in minority class proportion) were associated with decreased sample size. Other characteristics varied in importance across algorithms-in general, more features, weaker features, and more complex relationsh

关键词： medical informatics machine learning sample size research design decision trees classification algorithm clinical research learning-curve analysis analysis analyses guidelines ML decision making algorithm curve analysis dataset

来源：评论

学校读者我要写书评

暂无评论

Class imbalance in gradient boosting classification algorithms: Application to experimental stroke data

引用

STATISTICAL METHODS IN MEDICAL RESEARCH 2021年第3期30卷 916-925页

作者： Lyashevska, Olga Malone, Fiona MacCarthy, Eugene Fiehler, Jens Buhk, Jan-Hendrik Morris, Liam GMIT Enterprise Ireland Med & Engn Technol Gateway Dublin Rd Galway Ireland GMIT Marine & Freshwater Res Ctr Galway Ireland Univ Med Ctr Hamburg Eppendorf Dept Diagnost & Intervent Neuroradiol Hamburg Germany

Imbalance between positive and negative outcomes, a so-called class imbalance, is a problem generally found in medical data. Imbalanced data hinder the performance of conventional classification methods which aim to improve the overall accuracy of the model without accounting for uneven distribution of the classes. To rectify this, the data can be resampled by oversampling the positive (minority) class until the classes are approximately equally represented. After that, a prediction model such as gradient boosting algorithm can be fitted with greater confidence. This classification method allows for non-linear relationships and deep interactive effects while focusing on difficult areas by iterative shifting towards problematic observations. In this study, we demonstrate application of these methods to medical data and develop a practical framework for evaluation of features contributing into the probability of stroke.

关键词： Imbalanced data gradient boosting classification algorithm trees stroke oversampling

来源：评论

学校读者我要写书评

暂无评论

Discrimination of citrus infested with Bactrocera dorsalis (Hendel) using an olfactory detection system based on MIPs-QCM sensor array and Bayesian optimized classification algorithms

引用

SENSORS AND ACTUATORS B-CHEMICAL 2023年 385卷

作者： Wen, Tao He, Weitao Nie, Qiyi Gong, Zhongliang Li, Dapeng Wen, Liang Peng, Hailong Cent South Univ Forestry & Technol Sch Mech & Elect Engn Changsha 410004 Hunan Peoples R China Nanchang Univ Sch Chem & Chem Engn Nanchang 330031 Jiangxi Peoples R China

The characteristic substances in citrus volatile organic compounds (VOCs) are associated with the infestation of the Bactrocera dorsalis (Hendel), which provides a noninvasive evaluation method to discriminate infested citrus. This paper developed an olfactory detection system based on a quartz crystal microbalance (QCM) sensor array and classification algorithms to identify the citrus infested with B. dorsalis. Six characteristic substances, including D-limonene, myrcene, alpha-pinene, decanal, linalool, and beta-ocimene, which vary significantly after B. dorsalis infestation were selected as template molecules to prepare molecularly imprinted polymers (MIPs) and modify QCM respectively. The experimental results show that the prepared MIPs-QCM sensors had sensitivity in the range of 0.043-0.070 Hz/(mg/m3), and their stability and reproducibility were above 93.1%. Four sensors that contributed to the classification were screened by a stepwise discriminant analysis. Afterward, Bayesian optimization was employed to optimize the hyperparameters. The accuracy of the optimized support vector machine (SVM) reached 94.17%. The olfactory detection system developed in this study enables the discrimi-nation of citrus infested with B. dorsalis, which may have potential applications in the field of post-harvest treatment of citrus.

关键词： Bactrocera dorsalis-infested citrus Molecularly imprinted polymers Quartz crystal microbalance Volatile organic compounds classification algorithm Bayesian optimization

来源：评论

学校读者我要写书评

暂无评论

Evaluation of chicken freshness using a low-cost colorimetric sensor array with AdaBoost-OLDA classification algorithm

引用

LWT-FOOD SCIENCE AND TECHNOLOGY 2014年第2期57卷 502-507页

作者： Chen, Quansheng Hui, Zhe Zhao, Jiewen Ouyang, Qin Jiangsu Univ Sch Food & Biol Engn Zhenjiang 212013 Peoples R China

This paper attempted to evaluate chicken freshness using a low-cost colorimetric sensor array with the help of a classification algorithm. We fabricated a novel and low-cost colorimetric sensors array, with a specific colorific fingerprint to volatile compounds, using printing chemically responsive dyes on a C2 reverse silica-gel flat plate. In addition, we proposed a novel classification algorithm for sensors data classification - orthogonal linear discriminant analysis (OLDA) and adaptive boosting (AdaBoost) algorithm, namely AdaBoost-OLDA. And we compared it with two classical classification algorithms - linear discriminant analysis (LDA) and back propagation artificial neural network (BP-ANN). Experimental results showed classification results by AdaBoost-OLDA algorithm is superior to BP-ANN and LDA algorithms, the classification results by which are both 100% in the calibration and prediction sets. This study sufficiently demonstrated that the colorimetric sensors array with a classification algorithm has a high potential in evaluating chicken freshness, and AdaBoost-OLDA algorithm has a strong performance in solution to a complex data classification. (C) 2014 Elsevier Ltd. All rights reserved.

关键词： Colorimetric sensor array classification algorithm Chicken Freshness

来源：评论

学校读者我要写书评

暂无评论

Burst-classifier: Automated classification of solar radio burst type II III and IV for CALLISTO spectra using physical properties maximum of solar 24

引用

ADVANCES IN SPACE RESEARCH 2024年第11期74卷 6104-6123页

作者： Afandi, N. Z. Mohd Umar, R. Sabri, N. H. Safei, S. Monstein, C. Lau, C. C. Zafar, S. N. A. Syed Univ Sultan Zainal Abidin East Coast Environm Res Inst ESERI Kuala Nerus 21300 Terengganu Malaysia Univ Malaysia Terengganu Fac Sci & Marine Environm Adv Nano Mat ANoMa Res Grp Kuala Nerus 21030 Terengganu Malaysia Univ Sultan Zainal Abidin Fac Informat & Comp Besut 22200 Terengganu Malaysia Univ Svizzera italiana USI Ist Rech Solari Aldo & Cele Dacco IRSOL Fac Informat CH-6605 Locarno Switzerland Malaysian Space Agcy MYSA Explorat & Space Sci Div 13 Jalan Tun Ismail Kuala Lumpur 50480 Malaysia

Continuous observation of solar radio bursts (SRBs) throughout the year using the CALLISTO spectrometer generates a huge volume of spectral data. This study introduces a burst-classifier algorithm, which is an automated algorithm, to classify the SRB spectrum into three solar radio bursts, namely Type II (SRBT II), Type III (SRBT III) and Type IV (SRBT IV). The proposed algorithm was designed using four characteristic parameters derived from a collection of training dataset files. The characteristic parameters were derived from the intensity bursts observed on frequency channels and timesteps of the spectrum. This dataset consisted of 50 spectra of SRBT II and SRBT III, along with 40 spectra for SRBT IV, collected during the solar maximum of 2014 (Solar Cycle 24). After observations and analysis of the training dataset, each burst type was set up with a threshold. A training dataset of 80 data spectra from 2013 to 2016 was used to test the algorithm. Accuracy of the proposed algorithm was calculated using the percentage of true positives (TP) and false positives (FP). Findings demonstrate an accuracy of X74 % with 57 out of 80 spectra classified as TP and 23 spectra as FP. (c) 2024 COSPAR. Published by Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

关键词： Solar radio burst Burst II Burst III Burst IV CALLISTO 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：