检索结果-内蒙古大学图书馆

Exploring Multimodal Framework of Optimized Feature-Based Machine Learning to Revolutionize the Diagnosis of Parkinson’s Disease: AI-Driven Insights

引用

Biomedical Materials and Devices 2025年 1-20页

作者： Khanom, Fahmida Mostafiz, Rafid Uddin, Khandaker Mohammad Mohi Department of Mathematics American International University-Bangladesh 408/1 Kuratoli Khilkhet Dhaka-1229 Dhaka Bangladesh Institute of Information Technology Noakhali Science and Technology University Noakhali Bangladesh Department of Computer Science and Engineering Southeast University Dhaka China

Parkinson’s disease (PD), a progressive neurodegenerative disorder, poses significant challenges to healthcare systems worldwide due to its increasing prevalence and need for early detection to improve patient outcomes. This study introduces a robust diagnostic framework leveraging a multimodal feature selection approach, combining Minimum Redundancy Maximum Relevance (mRMR) with Recursive Feature Elimination (RFE), to optimize predictive accuracy in detecting PD. Using a dataset of 894 instances from the UCI Machine Learning Repository, the framework implemented comprehensive data preprocessing and feature selection techniques to refine input features. The Extra Trees classifier, integrated with the mRMR-RFE multimodal feature selection method, demonstrated superior performance over nine other machine learning models, achieving a perfect accuracy of 100%, with high specificity, sensitivity, precision, and F1 scores. This framework highlights the efficacy of ensemble learning methods and feature optimization in healthcare, ensuring computational efficiency and high predictive reliability. By addressing the challenges of dimensionality reduction and ensuring precise predictions, the proposed model sets a benchmark for machine learning applications in PD diagnosis. The results emphasize the potential of this framework to facilitate early detection and support clinical decision making, with significant implications for advancing AI-driven healthcare systems. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Extra trees classifier Machine learning Parkinson’s disease Recursive feature elimination (RFE)

来源：评论

学校读者我要写书评

暂无评论

Latent Dirichlet Allocation (LDA) Topic Modeling and Sentiment Analysis for Myanmar Coup Tweets 8

Latent Dirichlet Allocation (LDA) Topic Modeling and Sentime...

引用

8th International Conference on Information Technology, Information Systems and Electrical Engineering, ICITISEE 2024

作者： Manuaba, Ida Bagus Kerthyayana Karim, Mochammad Faisal School of Computing and Creative Arts Bina Nusantara University Computer Science Department Jakarta Indonesia University Islam International Indonesia Faculty of Social Science Depok Indonesia

ISBN: (纸本)9798350368970

Myanmar's coup in 2021 has produced public attention and discussions. Studying the trends and patterns of this event, a technological approach such as Natural Language Processing (NLP) could be used to analyze text data and develop insights. It can classify keywords and also group them into relevant topics, and then highlight aspects of the coup. Latent Dirichlet Allocation (LDA) is an NLP method that is widely used for information extraction and text classification. This paper focuses on the discussion about the utilization of LDA models in classifying the Topic Modeling for the N-best topic related to Myanmar's coup tweets data. This study also discussed the sentiment analysis model to visualize the trends and patterns related to public opinion based on the classified topics. The results showed the best-classified topics are 12 topics with the highest coherence score of 4.448, and these topics also can be grouped into three big themes and showed several negative sentiments for specific identified topics. To analyze further, the ANOVA test has been used to analyze further the patterns and trends of this public opinion based on the classified topics and sentiment scores © 2024 IEEE.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Linear Hashing with l8guarantees and two-sided Kakeya bounds 63

Linear Hashing with l8guarantees and two-sided Kakeya bounds

引用

63rd IEEE Annual Symposium on Foundations of computer science, FOCS 2022

作者： Dhar, Manik Dvir, Zeev Princeton University Department of Computer Science Princeton United States Princeton University Department of Computer Science Department of Mathematics Princeton United States

ISBN: (纸本)9781665455190

We show that a randomly chosen linear map over a finite field gives a good hash function in the l8 sense. More concretely, consider a set S ? Fqn and a randomly chosen linear L: Fqn Fqt with qt taken to be sufficiently smaller than |S|. Let US denote a random variable distributed uniformly on S. Our main theorem shows that, with high probability over the choice of L, the random variable L(US) is close to uniform in the l8 norm. In other words, every element in the range Fqt has about the same number of elements in S mapped to it. This complements the widely-used Leftover Hash Lemma (LHL) which proves the analog statement under the statistical, or l1, distance (for a richer class of functions) as well as prior work on the expected largest 'bucket size' in linear hash functions [1]. By known bounds from the load balancing literature [2], our results are tight and show that linear functions hash as well as truly random function up to a constant factor in the entropy loss. Our proof leverages a connection between linear hashing and the finite field Kakeya problem and extends some of the tools developed in this area, in particular the polynomial method. © 2022 IEEE.

关键词： Hash functions

来源：评论

学校读者我要写书评

暂无评论

Salp Swarm Incorporated Adaptive Dwarf Mongoose Optimizer with Lévy Flight and Gbest-Guided Strategy

引用

Journal of Bionic Engineering 2024年第4期21卷 2110-2144页

作者： Gang Hu Yuxuan Guo Guanglei Sheng Department of Applied Mathematics Xi’an University of TechnologyXi’an710054People’s Republic of China School of Computer Science and Engineering Xi’an University of TechnologyXi’an710048People’s Republic of China Department of Electronics and Information Engineering Bozhou UniversityBozhou236800People’s Republic of China

In response to the shortcomings of Dwarf Mongoose Optimization(DMO)algorithm,such as insufficient exploitation capability and slow convergence speed,this paper proposes a multi-strategy enhanced DMO,referred to as ***,we propose an improved solution search equation that utilizes the Gbest-guided strategy with different parameters to achieve a trade-off between exploration and exploitation(EE).Secondly,the Lévy flight is introduced to increase the diversity of population distribution and avoid the algorithm getting stuck in a local *** addition,in order to address the problem of low convergence efficiency of DMO,this study uses the strong nonlinear convergence factor Sigmaid function as the moving step size parameter of the mongoose during collective activities,and combines the strategy of the salp swarm leader with the mongoose for cooperative optimization,which enhances the search efficiency of agents and accelerating the convergence of the algorithm to the global optimal solution(Gbest).Subsequently,the superiority of GLSDMO is verified on CEC2017 and CEC2019,and the optimization effect of GLSDMO is analyzed in *** results show that GLSDMO is significantly superior to the compared algorithms in solution quality,robustness and global convergence rate on most test ***,the optimization performance of GLSDMO is verified on three classic engineering examples and one truss topology optimization *** simulation results show that GLSDMO achieves optimal costs on these real-world engineering problems.

关键词： Dwarf mongoose optimization algorithm Gbest-guided Lévy flight Adaptive parameter Salp swarm algorithm Engineering optimization Truss topological optimization

来源：评论

学校读者我要写书评

暂无评论

On a Numerical Investigation of MHD Flow of a Hybrid Nanofluid with Rotation and Thermal Buoyancy Force 2nd

On a Numerical Investigation of MHD Flow of a Hybrid Nan...

引用

2nd International Conference on Nonlinear Dynamics and Applications, ICNDA 2024

作者： Ahmedai, Salma Sibanda, Precious Goqo, Sicelo P. Noreldin, Osman A. I. School of Mathematics Statistics and Computer Science University of KwaZulu-Natal Pietermaritzburg South Africa Faculty of Mathematical Sciences and Statistics Al Neelain University Khartoum Sudan

ISBN: (纸本)9783031691331

A magnetohydrodynamic flow of aluminum oxide, and copper nanoparticles in water based hybrid nanofluid with thermal radiation and rotation is investigated. The numerical solution of the model describing the flow is achieved through the implementation of the improved an overlapping spectral simple iteration method (O-SSIM). The accuracy is enhanced by the overlapping grid, which reduces the density of the coefficient matrix. The objective is to assess the efficiency of the O-SSIM on a system of nonlinear ordinary differential equations that describe the hybrid nanofluid flow. The O-SSIM algorithm is implemented in MATLAB and tested for its convergence and accuracy. The accuracy of the O-SSIM are proved using the residual error analysis. The results are compared with previously published findings that were obtained using the overlapping spectral local linearization method. The nonlinear differential equations are linearised using the simple iteration method. The study aims to examine how various fluid parameters influence the characteristics of the hybrid nanofluid. The results presented the influence the stretching ratio, rotation, magnetic, and heat generation and absorption parameters on the temperature profile. The skin friction coefficients and Nusselt number were calculated and analyzed. It is found among other results, that the heat transfer rate of the hybrid nanofluid increased with both the volume fraction of copper nanoparticles and heat generation parameters. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Biomedical Image Segmentation Using Integrated FCM Clustering Modified with Regularized Level Set Method

Biomedical Image Segmentation Using Integrated FCM Clusterin...

引用

2023 International Conference on Disruptive Technologies, ICDT 2023

作者： Mishra, Annu Gupta, Pankaj Tewari, Peeyush Bit Mesra Department of Computer Science Ranchi Noida India Bit Mesra Department of Mathematics Ranchi Jaipur India

ISBN: (纸本)9798350323887

Biomedical image segmentation is used widely for various diagnosis of various diseases and other medicinal purposes and help the radiologist and doctor fraternity to reduce their work and help them concentrate more on their research for new diseases. Researchers and medical practitioners use applications based on image segmentation for detecting abnormalities as well as analyzing the effect of certain deformations or deviations quantitatively. However, there are various issues faced while carrying out this task. The primary reason is the presence of inherent noise, the non-uniform intensity of the pixels, and other artifacts. The presence of artifacts not only limits the process of image segmentation but also increases the computational time for the segmentation process. In biomedical images, the problem is more complicated and recurrent. This is due to the different anatomical structures and multi-modal systems available. In this paper, a new algorithm is proposed where a modified fuzzy C-means (MFCM) clustering algorithm is integrated with Regularized Level set method to enhance the efficiency of the image segmentation process which improves the analysis exercise of the image processing system. The approach encompasses two crucial steps. Initially, the image is segmented using the Modified FCM. The MFCM approach has two basic updates with respect to the conventional FCM [1]. Firstly, we introduce a factor to the conventional FCM and secondly, Euclidean distance is replaced with the kernel-dependent distance measure. The factor increases the speed of computation of the FCM algorithm. Replacing the Euclidean distance with a kernel-dependent distance measure makes the algorithm more robust. After the initial segmentation, the Regularized Level Set method was used to refine the result and track the variation boundaries. The regularized level set method solves the reinitialization problem faced in the conventional level set method and enhances the capability and effici

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Hybridization of Self Supervised Learning Models for Enhancing Automatic Arabic Speech Recognition

Hybridization of Self Supervised Learning Models for Enhanci...

引用

science and Information Technology in Smart Administration (ICSINTESA), International Conference of

作者： Hiba Adreese Younis Yusra Faisal Mohammad Department of Computer Science College of Computer Science and Mathematics University of Mosul Mosul IRAQ

ISBN: (数字)9798350376111

ISBN: (纸本)9798350376128

End-to-end systems for automatic recognition of Arabic speech have gained significant importance in the last few years. Arabic is considered as one of the low resource languages that lacks the labeled data required for training large models using supervised learning. To overcome this problem self-supervised learning models appears which can improve the robustness of the system by avoiding several issues with labels resulting from corrupted audio files. The objective of this paper is to assess the effectiveness of various models for Arabic speech recognition and also to compare the base SSL models with hybrid models in terms of word error rate(WER). The methodology includes hybridization of two state of the art models (wav2vec2xlsr, HUBERT) with adapter layer of MMS model. Another hybridization was also implemented by hybridization these SSL models with four layers of transformer encoder. The MMS model was also used for the first time for fine-tuning Arabic language using different hyper parameter values and gave competing results. As optimization algorithm, the SWATS mechanism was being used with these four hybrid models to improve their performance. Results showed that hybrid hubert_transformer model gave the best results on test dataset with 23.83 WER, followed by MMS model, hybrid wave transformer model, hybrid wave_MMS and hybrid hubert_MMS model respectively.

关键词： Training Adaptation models Art System performance Supervised learning Speech recognition Self-supervised learning Switches Benchmark testing Transformers

来源：评论

学校读者我要写书评

暂无评论

Machine learning and deep learning algorithms in detecting COVID-19 utilizing medical images: a comprehensive review

引用

Iran Journal of computer science 2024年第3期7卷 699-721页

作者： Nurjahan Mahbub-Or-Rashid, Md. Satu, Md. Shahriare Tammim, Sanjana Ruhani Sunny, Farhana Akter Moni, Mohammad Ali Department of IoT and Robotics Engineering Bangabandhu Sheikh Mujibur Rahman Digital University Kaliakair Bangladesh Department of Computer Science and Engineering Bangladesh University of Business and Technology Mirpur Bangladesh Department of Management Information Systems Noakhali Science and Technology University Noakhali Bangladesh Department of Computer Science and Software Engineering Auburn University Auburn AL United States Department of Computer Science and Engineering Green University of Bangladesh Dhaka Bangladesh Faculty of Health and Behavioural Sciences School of Health and Rehabilitation Sciences The University of Queensland St Lucia Australia

The public’s health is seriously at risk from the coronavirus pandemic. Millions of people have already died as a result of this devastating illness, which affects countless people daily worldwide. Unfortunately, no specific therapeutic drugs or vaccines are available to cure patients completely. Therefore, early identification of infected individuals and their isolation can help reduce community transmission of COVID-19. Despite being commonly used, the reverse transcription polymerase chain reaction (RT-PCR) test is costly, time-consuming, and requires suitable kits, which are not always readily available. An alternative solution for the traditional RT-PCR test is machine learning and deep learning-based COVID-19 detection that utilizes clinical features of chest X-rays and computed tomography (CT) images. In this study, we conducted a detailed review of more than 100 recently published works to detect COVID-19-infected patients. Thus, different data preprocessing, data augmentation, image enhancement, feature extraction, machine learning, deep learning, Explainable Artificial Intelligence (AI) methods, etc. were explored to understand how these techniques were used to process images for identifying COVID-19. Additionally, we identified the current challenges in this field and suggested further research directions. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Chest X-ray Computed tomography COVID-19 Deep learning Machine learning

来源：评论

学校读者我要写书评

暂无评论

Improved Bat Algorithm with Deep Learning-Based Biomedical ECG Signal Classification Model

引用

computers, Materials & Continua 2023年第2期74卷 3151-3166页

作者： Marwa Obayya Nadhem NEMRI Lubna A.Alharbi Mohamed K.Nour Mrim M.Alnfiai Mohammed Abdullah Al-Hagery Nermin M.Salem Mesfer Al Duhayyim Department of Biomedical Engineering College of EngineeringPrincess Nourah Bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Information Systems College of Science and Art at MahayilKing Khalid UniversitySaudi Arabia Department of Computer Science College of Computers and Information TechnologyTabuk UniversitySaudi Arabia Department of Computer Sciences College of Computing and Information SystemUmm Al-Qura UniversitySaudi Arabia Department of Information Technology College of Computers and Information TechnologyTaif UniversityP.O.Box 11099Taif21944Saudi Arabia Department of Computer Science College of ComputerQassim UniversitySaudi Arabia Department of Electrical Engineering Faculty of Engineering and TechnologyFuture University in EgyptNew Cairo11835Egypt Department of Computer Science College of Sciences and Humanities-AflajPrince Sattam Bin Abdulaziz UniversitySaudi Arabia

With new developments experienced in Internet of Things(IoT),wearable,and sensing technology,the value of healthcare services has *** evolution has brought significant changes from conventional medicine-based healthcare to real-time observation-based *** Electrocardiogram(ECG)signals are generally utilized in examination and diagnosis of Cardiovascular Diseases(CVDs)since it is quick and non-invasive in *** to increasing number of patients in recent years,the classifier efficiency gets reduced due to high variances observed in ECG signal patterns obtained from *** such scenario computer-assisted automated diagnostic tools are important for classification of ECG *** current study devises an Improved Bat Algorithm with Deep Learning Based Biomedical ECGSignal Classification(IBADL-BECGC)*** accomplish this,the proposed IBADL-BECGC model initially pre-processes the input ***,IBADL-BECGC model applies NasNet model to derive the features from test ECG *** addition,Improved Bat Algorithm(IBA)is employed to optimally fine-tune the hyperparameters related to NasNet ***,Extreme Learning Machine(ELM)classification algorithm is executed to perform ECG classification *** presented IBADL-BECGC model was experimentally validated utilizing benchmark *** comparison study outcomes established the improved performance of IBADL-BECGC model over other existing methodologies since the former achieved a maximum accuracy of 97.49%.

关键词： Data science ECG signals improved bat algorithm deep learning biomedical data data classification machine learning

来源：评论

学校读者我要写书评

暂无评论

Nutri-Smober: A Preliminary Study to Investigate the Domain Dependency of Behaviour Change Strategies in Persuasive Games 22

Nutri-Smober: A Preliminary Study to Investigate the Domain ...

引用

9th ACM SIGCHI Annual Symposium on computer-Human Interaction in Play, CHI PLAY 2022

作者： Ndulue, Chinenye Melis, Marsha Orji, Rita Faculty Of Computer Science Dalhousie University Canada Department Of Computer Science The University Of Hong Kong Hong Kong

ISBN: (纸本)9781450392112

Although persuasive strategies have been shown to be effective at promoting behaviour change across various domains of health and wellness, the domain dependency of the effectiveness of these strategies, when applied to persuasive games targeting different domains remains vague. In this preliminary study, we investigate the domain dependency of eight persuasive strategies implemented in two persuasive games. In a between-study of 31 participants, we investigated the perceived effectiveness of two similar persuasive games, implementing the same set of strategies but focusing on two different domains: healthy eating and smoking cessation, with only changes in their context to suit each domain. Our results revealed that although there was no overall significant difference in the overall effectiveness of the two interventions, seven of the persuasive strategies implemented were significantly effective for healthy eating while only two of them were significantly effective for smoking cessation. We also present discussions from participants' comments of how to improve these features for future game design. © 2022 Owner/Author.

关键词： Human computer interaction

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：