检索结果-内蒙古大学图书馆

1st international conference on Machine Intelligence and computer science Applications, ICMICSA 2022

作者： Knouzi, Maryam Ennaji, Fatima Zohra Hafidi, Imad Laboratory of Process Engineering Computer Science and Mathematics Department of Mathematics and Informatics National School of Applied Sciences University Sultan Moulay Slimane Beni-Mellal Morocco

ISBN: (数字)9783031293139

ISBN: (纸本)9783031288456

This article evaluates techniques for obtaining facial images from convolutional neural networks to identify emotion expressions. The primary goal of the paper is to discuss the most often used approaches to analyzing and recognizing emotional expressions on the face. The works that were reviewed might be divided into two major trends: the traditional techniques and those that were especially developed using neural *** analysis revealed that using CNNs improved level of performance. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Multi-task fine-tuning for generating keyphrases in a scientific domain 9

Multi-task fine-tuning for generating keyphrases in a scient...

引用

9th IEEE International Conference on Information Technology and Nanotechnology, ITNT 2023

作者： Glazkova, Anna Morozov, Dmitry University of Tyumen Institute of Mathematics and Computer Science Tyumen Russia Novosibirsk State University Laboratory of Applied Digital Technologies Novosibirsk Russia

ISBN: (纸本)9798350397338

Automatic selection of keyphrases (keywords) is a major challenge to finding and systematizing scholarly documents. This paper investigates the efficiency of using titles of scientific papers as additional information for keyphrase generation. We propose an approach to multi-task fine-tuning the BART model using control codes1. It is shown that the suggested approach can improve the performance of BART for the task of keyphrase generation. In some cases, the presented model outperforms state-of-the-art models for keyphrase extraction. Moreover, the results have demonstrated that multitask fine-tuning also increases the performance of title generation. © 2023 IEEE.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

A Novel Multiple Dependent State Sampling Plan Based on Time Truncated Life Tests Using Mean Lifetime

引用

computers, Materials & Continua 2022年第12期73卷 4611-4626页

作者： Pramote Charongrattanasakul Wimonmas Bamrungsetthapong Poom Kumam Division of Mathematics Department of Mathematics and Computer ScienceFaculty of Science and TechnologyRajamangala University of Technology KrungthepBangkok10120Thailand Division of Applied Statistics Department of Mathematics and Computer ScienceFaculty of Science and TechnologyRajamangala University of Technology ThanyaburiPathum Thani12110Thailand Center of Excellence in Theoretical and Computational Science(TaCS-CoE)&KMUTT Fixed Point Research Laboratory Room SCL 802 Fixed Point LaboratoryScience Laboratory BuildingDepartments of MathematicsFaculty of ScienceKing Mongkut’s University of Technology Thonburi(KMUTT)Bangkok10140Thailand

The design of a new adaptive version of the multiple dependent state(AMDS)sampling plan is presented based on the time truncated life test under the Weibull *** achieved the proposed sampling plan by applying the concept of the double sampling plan and existing multiple dependent state sampling plans.A warning sign for acceptance number was proposed to increase the probability of current lot *** optimal plan parameters were determined simultaneously with nonlinear optimization problems under the producer’s risk and consumer’s risk.A simulation study was presented to support the proposed sampling plan.A comparison between the proposed and existing sampling plans,namely multiple dependent state(MDS)sampling plans and a modified multiple dependent state(MMDS)sampling plan,was considered under the average sampling number and operating characteristic curve *** addition,the use of two real datasets demonstrated the practicality and usefulness of the proposed sampling *** results indicated that the proposed plan is more flexible and efficient in terms of the average sample number compared to the existing MDS and MMDS sampling plans.

关键词： Adaptive version of multiple dependent state sampling plan time truncated life test quality level weibull distribution mean lifetime

来源：评论

学校读者我要写书评

暂无评论

The Impact of Predictive Maintenance on the Performance of Industrial Enterprises

引用

SN computer science 2025年第1期6卷 1-34页

作者： Er-Ratby, Mohamed Kobi, Abdessamad Sadraoui, Youssef Kadiri, Moulay Saddik LaSTI Laboratory of Science and Engineering Techniques National School of Applied Sciences of Khouribga Sultan Moulay Slimane University Khouribga Morocco LARIS Angevin Laboratory for Systems Engineering Research Polytech Angers University of Angers Angers France LIPIM Laboratory of Process Engineering Computer Science and Mathematics National School of Applied Sciences of Khouribga Sultan Moulay Slimane University Khouribga Morocco

Optimizing maintenance and enhancing the performance of businesses are significant concerns in the modern industrial world. Predictive maintenance is emerging as an innovative approach to address these challenges, allowing companies to shift from corrective maintenance to preventive maintenance. Predictive maintenance relies on the use of advanced technologies such as the Internet of Things (IoT), Artificial Intelligence (AI), and Machine Learning (ML) to collect and analyse real-time data from equipment. Through this predictive analysis, it becomes possible to identify early warning signals of failures, enabling the anticipation of potential issues and the proactive planning of maintenance interventions. In this study, we will thoroughly examine the impact of predictive maintenance on the performance of businesses. We will explore the benefits and opportunities it offers in terms of reducing downtime, optimizing maintenance costs, and enhancing productivity. We will also investigate the various technologies and methods used in the implementation of predictive maintenance, along with potential challenges and best practices for successful adoption. This research focuses on studying the application of predictive maintenance within a company using data science and machine learning methods. Predictive maintenance represents an innovative approach aimed at anticipating equipment failures by leveraging real-time collected data. Through the analysis of this data using sophisticated algorithms, it becomes possible to identify early signals of potential problems and implement preventive maintenance actions before breakdowns occur. This approach not only reduces unexpected downtime but also optimizes maintenance operations by avoiding unnecessary interventions and maximizing resource utilization. The ultimate goal is to improve equipment availability, optimize operational performance, and maximize the overall yield of the company. © The Author(s), under exclusive licence to S

关键词： Artificial intelligence (AI) and machine learning (ML) Data science Optimizing maintenance Performance indicators Predictive maintenance Production

来源：评论

学校读者我要写书评

暂无评论

A Self-Expressive Feature Selection Algorithm Based on Self-Representation of Data

A Self-Expressive Feature Selection Algorithm Based on Self-...

引用

2024 International Symposium on Digital Home, ISDH 2024

作者： Qiu, Jie Chen, Zuoyun Chen, Jianmei Lan, Wei Ding, Xiaojun Computer Science and Engineering College Yulin Normal University Center for Applied Mathematics of Guangxi Guangxi Yulin537000 China Guangxi University Guangxi Key Laboratory of Multimedia Communications and Network Technology Center for Applied Mathematics of Guangxi Guangxi Nanning537000 China

ISBN: (纸本)9798331509873

Many biological data exhibit characteristics such as high dimensionality, large scale, high complexity, and high noise. In many cases, it is necessary to eliminate noise and identify truly useful features for better data analysis and modeling. In labeled data, algorithms such as correlation analysis or embedded selection methods can be used to select features most relevant to the labels. However, these algorithms are not effective for unlabeled data. Currently, common methods involve selecting features with higher variance or, based on feature correlation, identifying central features from a group of related features. This paper proposes a new method for feature selection using self-representation of data. The idea is that features in high-dimensional biological big data exhibit some redundancy, and redundant features can be represented by other features. Therefore, it is only necessary to find a subset of features, from which the entire dataset can be generated. This subset may represent more essential and important features. Using this subset for subsequent work can resist certain noise and reduce complexity, thereby improving algorithm performance. The paper presents a framework that can use linear models and non-linear neural network models to select feature subsets. Experiments compared the proposed method with popular unsupervised feature selection methods, using KMeans, KNN, and SVM as classification methods to classify the original data and the data after feature selection, and analyzed the results. The experiments ultimately demonstrated that the proposed method in this paper outperforms other popular unsupervised feature selection methods on multiple biological datasets. © 2024 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Feature Selection for Text Classification Using Genetic Algorithm 1

引用

1st international conference on Machine Intelligence and computer science Applications, ICMICSA 2022

作者： Belkarkor, Salma Hafidi, Imad Nachaoui, Mourad Laboratory of Process Engineering Computer Science and Mathematics at the National School of Applied Sciences Sultan Moulay Slimane University Khouribga Morocco Laboratory of Mathematics and Applications Team of Mathematics and Interactions at the Faculty of Science and Technology Sultan Moulay Slimane University Beni-Mellal Morocco

ISBN: (数字)9783031293139

ISBN: (纸本)9783031288456

Today’s large amount of text data makes feature classification a difficult text processing challenge. High dimensionality is the primary challenge in text processing, and feature selection is a common method for reducing dimensions. The most crucial factors in text classification are a strong text representation and a very accurate classifier. As a result, choosing the right features is essential for using machine learning algorithms effectively. Different optimization techniques, including the Genetic Algorithm (GA), have been effectively used for dimensionality reduction in the field of text classification. To evaluate the performance of GA for Feature Selection (FS), we compared the GA for FS with other filtering methods to prove the efficiency of the GA for FS, for that, we used the NB classifier and three benchmark document collections: SMS, BBC, and 20Newsgroups. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Throughput Prediction Using LS-SVM For 4g-LTE Network and Beyond

Throughput Prediction Using LS-SVM For 4g-LTE Network and Be...

引用

Telecommunications and Intelligent Systems (ICTIS), International Conference on

作者： Mostefa Hamdani Ahmed Sahnoune Mokhtar Merzougui Abdelmalek Douara Fouad Khatemi Department of Mathematics and Computer Science Laboratory of Electronics Computer Science and Applied Mathematics Faculty of Science and Technology University El Wancharissi Department of Science and Technology Laboratory of Electronics Computer Science and Applied Mathematics Faculty of Science and Technology University El Wancharissi

ISBN: (数字)9798331527396

ISBN: (纸本)9798331527402

The integration of Artificial Intelligence(AI) with Long-Term Evolution (LTE) networks offers substantial potential for improving communication infrastructure. By harnessing AI algorithms, it is possible to dynamically optimize network parameters, proactively predict and mitigate equipment failures, and enhance Quality of Service (QoS) by prioritizing critical services through real-time analysis of key performance indicators. This study focuses on predicting network throughput using Least Square Support Vector Machine (LS-SVM). The predictive models developed can trigger adaptive network reconfigurations, ensuring that the network responds promptly to changing conditions. Simulation results reveal that the integration of LS-SVM significantly improves prediction accuracy, enabling more effective implementation of self-organizing network strategies in real-time.

关键词： Support vector machines Correlation coefficient Accuracy Quality of service Predictive models Throughput Real-time systems Telecommunications Analysis of variance Testing

来源：评论

学校读者我要写书评

暂无评论

Boosting Adversarial Training with Learnable Distribution

引用

computers, Materials & Continua 2024年第3期78卷 3247-3265页

作者： Kai Chen Jinwei Wang James Msughter Adeke Guangjie Liu Yuewei Dai School of Electronics and Information Engineering Nanjing University of Information Science and TechnologyNanjing210044China Key Laboratory of Intelligent Support Technology for Complex Environments Ministry of EducationNanjing210044China School of Computer and Software Nanjing University of Information Science and TechnologyNanjing210044China Nanjing Center for Applied Mathematics Nanjing211135China

In recent years,various adversarial defense methods have been proposed to improve the robustness of deep neural *** training is one of the most potent methods to defend against adversarial ***,the difference in the feature space between natural and adversarial examples hinders the accuracy and robustness of the model in adversarial *** paper proposes a learnable distribution adversarial training method,aiming to construct the same distribution for training data utilizing the Gaussian mixture *** distribution centroid is built to classify samples and constrain the distribution of the sample *** natural and adversarial examples are pushed to the same distribution centroid to improve the accuracy and robustness of the *** proposed method generates adversarial examples to close the distribution gap between the natural and adversarial examples through an attack algorithm explicitly designed for adversarial *** algorithm gradually increases the accuracy and robustness of the model by scaling ***,the proposed method outputs the predicted labels and the distance between the sample and the distribution *** distribution characteristics of the samples can be utilized to detect adversarial cases that can potentially evade the model *** effectiveness of the proposed method is demonstrated through comprehensive experiments.

关键词： Adversarial training feature space learnable distribution distribution centroid

来源：评论

学校读者我要写书评

暂无评论

Analyzing Instagram Images to Predict Personality Traits 1

引用

1st international conference on Machine Intelligence and computer science Applications, ICMICSA 2022

作者： El Bahy, Siham Aboutabit, Noureddine Hafidi, Imad Laboratory of Process Engineering Computer Science and Mathematics National School of Applied Sciences Khouribga University Sultan Moulay Slimane Beni Mellal Morocco

ISBN: (数字)9783031293139

ISBN: (纸本)9783031288456

Our posts on social media are a way of expressing ourselves and our inclinations. In this research, we focused on Instagram (a social media platform for sharing images) to study the relationship between the personality of users and the content of photos they post. We also predicted their personalities based on the relationship between them and their images. This study is the first work that focuses on the Moroccan population. We collected a database of 316 Instagram Moroccan users, larger than the other databases used in previous works. We asked the users to fill out a NEO PI form to extract a Big Five that we chose to express their personality. And then we downloaded the images of users and extracted from them three categories of features;visual features (9 colors and lighting), image content (objects, animals, faces, people, light sources, dark, buildings), and emotional features (anger, disgust, fear, happy, neutral, sad, surprise). A root mean square error (RMSE) was used to indicate prediction accuracy, which relates to the [1,5] score scale. We succeeded in outperforming the best results obtained in the previous works, especially in predicting extraversion and neuroticism traits. We got an RMSE value of 0.83 versus 0.90 for extraversion and 0.87 versus 0.89 for neuroticism. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Light sources

来源：评论

学校读者我要写书评

暂无评论

An Unsupervised Voice Activity Detection Using Time-Frequency Features 1

引用

1st international conference on Machine Intelligence and computer science Applications, ICMICSA 2022

作者： Ait Mait, Hind Aboutabit, Noureddine Laboratory of Process Engineering Computer Science and Mathematics National School of Applied Sciences Khouribga University Sultan Moulay Slimane Beni Mellal Morocco

ISBN: (数字)9783031293139

ISBN: (纸本)9783031288456

Voice Activity Detection (VAD) is a binary classification problem for separating speech segments from background silence or noise. Over time, many features for the VAD have been proposed. In our study, we applied two features: Short Time Energy (STE) and Spectral Centroid, which belong to temporal and frequency domains, respectively. The goal of applying this VAD method is to use the speech segment extracted as an input of a phonetic segmentation method of Arabic and Moroccan dialect speech signals. We evaluated the method on 400 sentences from the Arabphone corpus recorded in noisy and noiseless environments. The results showed promising accuracy of 85%, which is comparable to some other previously proposed methods. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Speech recognition

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：