检索结果-内蒙古大学图书馆

Development of a diacritic-aware large vocabulary automatic speech recognition for Hausa language

International Journal of Speech Technology 2024年第3期27卷 687-700页

作者： Abubakar, Abdulqahar Mukhtar Gupta, Deepa Vekkot, Susmitha Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bengaluru India Department of Electronics & amp Communication Engineering Amrita School of Engineering Amrita Vishwa Vidyapeetham Bengaluru India

Research on voice recognition for African languages is limited due to the scarcity of digital resources for training and adaptation, despite its broad usefulness. The Hausa language, spoken by almost fifty million inhabitants in West and Central Africa, is an example of a linguistic domain that has not been thoroughly studied. The Hausa language employs diacritics, which are symbols located above alphabetical characters to convey further information. By removing diacritics, the number of homographs increases, making it difficult to distinguish between similar words. This paper presents a study on speech recognition in the Hausa Language, specifically focusing on diacritized words. The study utilises the state-of-the-art wave2vec2.0 and Whisper deep learning architecture models, for transcribing audio signals into corresponding Hausa text. According to the results obtained in the study, the Whisper-large deep model emerged as the best, achieving a word error rate of 4.23% representing a considerable improvement of 43.9% when compared to the existing state-of-the-art model for Hausa language speech recognition. Additionally, the Whsiper-large model demonstrated a diacritic coverage of 92%, precision of 98.87%, with a diacritic error rate of 2.1%. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

An Automatic Diagnostic Tool for Autism Spectrum Disorder using Structural Magnetic Resonance Imaging and a Tailored Binary Coded Genetic Algorithm

引用

IEIE Transactions on Smart Processing and computing 2024年第3期13卷 236-242页

作者： Sachnev, Vasily Suresh, Mahanand Belathur Department of Information Communication and Electronics Engineering Catholic University of Korea Korea Republic of Department of Information Science and Engineering Computing Sri Jayachamarajendra College of Engineering JSS Science and Technology University Mysuru India

In this paper, we propose an automatic diagnostic tool for autism based on machine learning and structural MRI. A set of 989 relevant features extracted from structural magnetic resonance imaging (MRI) present in the ABIDE database was used. The tool has two steps: searching for the best set of features using a tailored binary coded genetic algorithm and then using the selected set of features to train a classifier using an extreme learning machine. The trained classifier can efficiently identify autism spectrum disorder (ASD) versus healthy controls. The proposed tailored binary coded genetic algorithm uses a statistical selective crossover designed specifically for the classification problem, which significantly speeds up the genetic algorithm and improves the classification of ASD versus healthy controls. Extensive experiments with the proposed automatic autism diagnostic tool clearly indicate the advantages of the proposed method compared to previous approaches. Copyrights © 2024 The Institute of electronics and Information Engineers.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Crypto Trend Prediction Based on Wavelet Transform and Deep Learning Algorithm 2

Crypto Trend Prediction Based on Wavelet Transform and Deep ...

引用

2nd International Conference on Machine Learning and Data engineering, ICMLDE 2023

作者： Parameswaran, Sumesh Eratt Ramachandran, Vidhyalavanya Shukla, Swati School of Electronics Engineering VIT-AP University Vijayawada India Department of Computing and Electronics N Engineering Middle East College Muscat Oman

Crypto currency price prediction is one of the biggest challenges for any prediction system owing to its volatility and dependence on many factors. Artificial intelligence powered by the development of sophisticated computing technology and developments in research has opened up a wide variety of applications and new possibilities to solve complicated problems. Extraction of appropriate features and input with features affecting the output is equally important in time series prediction problems. Wavelet transform is considered an effective method to split a non-stationary signal into uncorrelated components. This work attempts to extract the advantage of wavelets to extract features from an available data set and to predict the trend of the crypto price for next 16 days using the Bi-LSTM network. Historic data of crypto currencies available in Yahoo finance through the yFinance API with the python code is used as the primary data for this work. Due to the gaining importance of SHIB crypto due to its multiplication of value from its start, we have taken SHIB trend analysis as a case study with the use of a data set provided by Yahoo Finance API. The novelty of the proposed system lies largely on the combination of wavelet and deep learning system to improve the performance measures. The proposed systems evaluates the wavelet transform coefficients of the historic data available from the Yahoo Finance API and feed the wavelet transform values to the input of Bi-LSTM network to predict the wavelet transform coefficients for the next 16 days. Finally an inverse wavelet transform is used to get the price of the crypto currency for next 16 days. The results obtained shows that the model was able to estimate the trend in crypto price. The trend analysis shows whether there is an increase or decrease of prices in upcoming days and this could be useful to invest or to withdraw the funds to improve the returns. The proposed scheme resulted in lower Mean Square Error (MSE), Me

关键词： Wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Enhanced Detection and Characterization of Pulmonary Nodules 4

Deep Learning for Enhanced Detection and Characterization of...

引用

4th International Conference on Intelligent Technologies, CONIT 2024

作者： Lavanya, G. Muthulakshmi, M. Latha, Manohar Keerthinathan, A. Vishal Krishh, P. Sulthana, Sheba Meghana, Meka Kavya Uma Amrita School of Computing Department of Computer Science and Engineering Amrita Vishwa Vidyapeetham Chennai India Amrita School of Computing Department of Electronics and Communication Engineering Amrita Vishwa Vidyapeetham Chennai India Rajalakshmi Engineering College Department of Electronics and Communication Engineering India

ISBN: (纸本)9798350349900

This study aims to develop, an effective customized Convolutional Neural Network model that can accurately identify areas on CT scans that have lung nodules and those that do not. The study utilizes two datasets, the LUNA-16 dataset for training and the IQ-OTH/NCCD dataset for testing. Additionally, we employ a U-Net model for precise lung nodule segmentation. This segmentation approach aimed to achieve a comprehensive characterization of nodules based on their spatial characteristics. The testing phase involves evaluating the performance of our trained models on the IQ-OTH/NCCD dataset, particularly focusing on localizing nodules in malignant cases. Overall, the research highlights the effectiveness of deep learning in early detection and diagnosis of lung cancer. This work contributes to improving the accuracy and reliability of automated pulmonary nodule detection systems, with a focus on identifying nodules in patients with suspected malignancies. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Impact of Different Voltage Source Models on Radiated Field and Incident Power Density within the Boundary Element Method Formalism

Impact of Different Voltage Source Models on Radiated Field ...

引用

2024 IEEE International Symposium on Electromagnetic Compatibility, Signal and Power Integrity, EMC+SIPI 2024

作者： Nejašmić, Anna Šušnjara Poljak, Dragan University of Split Faculty of Electrical Engineering Mechanical Engineering and Naval Architecture Department of Electronics and Computing Split Croatia

ISBN: (纸本)9798350360394

This paper deals with the impact of different voltage source models on computed radiated electromagnetic (EM) field and incident power density (IPD) in the framework of boundary element method formalism. Quarter, half and full wavelength dipole antennas are placed in free space. Antenna current is obtained by solving the Pocklington's integro-differential equation via Galerkin-Bubnov Indirect Boundary Element Method (GB-IBEM). Delta gap (DG) and magnetic frill (MF) are used as excitation source models. Frequency, distance, antenna radius and segment number are varied. Relative errors in current, EM field and IPD values are computed for MF source with respect to DG source. It is found that definition of antenna radius plays an important role in the difference between the results obtained with DG and MF sources. The relative error does not depend on frequency nor on distance from the antenna for EM field and IPD. © 2024 IEEE.

关键词： Dipole antennas

来源：评论

学校读者我要写书评

暂无评论

Cost-Efficient Feature Selection for Horizontal Federated Learning

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第12期5卷 6551-6565页

作者： Banerjee, Sourasekhar Bhuyan, Devvjiit Elmroth, Erik Bhuyan, Monowar Umeå University Department of Computing Science UmeåSE-90187 Sweden Tezpur University Department of Electronics and Communication Engineering Assam Tezpur784028 India

Horizontal federated learning (HFL) exhibits substantial similarities in feature space across distinct clients. However, not all features contribute significantly to the training of the global model. Moreover, the curse of dimensionality delays the training. Therefore, reducing irrelevant and redundant features from the feature space makes training faster and inexpensive. This work aims to identify the common feature subset from the clients in federated settings. We introduce a hybrid approach called Fed-MOFS, utilizing mutual information (MI) and clustering for local FS at each client. Unlike the Fed-FiS, which uses a scoring function for global feature ranking, Fed-MOFS employs multiobjective optimization to prioritize features based on their higher relevance and lower redundancy. This article compares the performance of Fed-MOFS with conventional and federated FS methods. Moreover, we tested the scalability, stability, and efficacy of both Fed-FiS and Fed-MOFS across diverse datasets. We also assessed how FS influenced model convergence and explored its impact in scenarios with data heterogeneity. Our results show that Fed-MOFS enhances global model performance with a 50% reduction in feature space and is at least twice as fast as the FSHFL method. The computational complexity for both approaches is O(d2), which is lower than the state of the art. © 2024 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Multi-response Optimization of Process Parameters in Stretch Blow Molding of PET Plastic Bottles

引用

Journal of The Institution of Engineers (India): Series D 2024年 1-15页

作者： Merga, Henok Hailu Sinha, Devendra Kumar Getachew, Aklilu Workneh, Melese Rathee, Devender Singh Department of Mechanical Engineering School of Mechanical Chemical and Materials Engineering Center of Excellence Advanced Manufacturing Engineering Adama Science and Technology University Adama Ethiopia Department of Electronics Engineering School of Electrical Engineering and Computing Adama Science and Technology University Adama Ethiopia

Plastic waste in Polyethylene Terephthalate (PET) bottles is creating new challenges, which in today’s scenario is a major concern in every industry. The defect in PET bottles mainly occurs due to improper selection of levels of process parameters during the production process. Therefore, optimizations of process parameters are very important to minimize the defects produced due to improper stretch blow molding process parameter selections. The study is carried out on 2 L bottle since they have the highest demand and the highest defect rate in the PET bottling sector. The PET bottles have been developed from the preform at various blowing temperature, blowing speed and blowing angle on commercially available SBM setup. The design and plan of experiments have been done according to L9 orthogonal array of Taguchi method. The optimization of process parameters has been done through the hybrid Taguchi grey relational analysis optimization technique on four factors namely: blowing temperature, first blow angle, second blow angle and blowing speed. Each value of parameter was taken at three levels in combination. The testing of all samples such as tensile test, wall thickness measurement and hoop stretch ratio measurement have been carried out as per ASTM standards. The morphology of the samples has been carried out through scanning electron microscope. A maximum tensile strength of 65 MPa, minimum hoop stretch ratio of 5.45 and a maximum wall thickness 0.24 mm was found at varying process parameter combinations. It has been observed that among the process parameters: blowing temperature was the most significant factor for minimizing defects while second blow angle was the least significant parameter. It was found that optimized values of the process parameters yielded 15% reduction in defects on PET bottles. The present study is expected to contribute significantly for the industries working in this field for carrying out production at optimum levels of process parameter

关键词： Plastic bottles

来源：评论

学校读者我要写书评

暂无评论

Enhanced Fetal Development Assessment via Contour Detection and CRL Estimation 3

Enhanced Fetal Development Assessment via Contour Detection ...

引用

3rd IEEE International Conference on Control, Instrumentation, Energy and Communication, CIEC 2024

作者： Sriraam, Natarajan Chinta, Babu Suresh, Seshadhri Sudharshan, Suresh Ramaiah Institute of Technology Center for Medical Electronics and Computing Department of Medical Electronics Engineering Bangalore India Center for Medical Electronics and Computing Ramaiah Institute of Technology Bangalore India Mediscans Pvt Ltd Chennai India

ISBN: (纸本)9798350313703

Accurate fetal growth evaluation is critical for monitoring pregnancy health and delivering appropriate prenatal treatment. We provide a unique strategy for improving the precision of fetal growth assessment by merging sophisticated contour recognition algorithms with Crown-Rump Length (CRL) estimate in this work. We establish a more complete and reliable approach of measuring fetal growth and development by integrating cutting-edge computer vision and deep learning methods. Our technique uses powerful image processing methods to determine the shape of the fetal anatomy in ultrasound images, overcoming hurdles such as noise, artifacts, and fetal positional fluctuations. Following contour identification, exact CRL length estimate is performed, which is a critical predictor of gestational age and fetal well-being. The combination of these two components considerably improves accuracy of fetal growth assessment. © 2024 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Dynamic Data Placement Strategy with Network Security Issues in Distributed Cloud Environment for Medical Issues: An Overview

引用

Recent Advances in Computer Science and Communications 2024年第6期17卷 25-38页

作者： Pradhan, Devasis Behera, Majusha Gheisari, Mehdi Department of Electronics and Communication Engineering Acharya Institute of Technology Karnataka Bangalore560107 India School of Electronics Engineering Campus 12 KIIT deemed to be University Odisha Bhubaneswar India Department of Cognitive Computing Institute of Computer Science and Engineering Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai India

The rapid integration of distributed cloud systems in the healthcare industry has profoundly impacted the management of valuable medical data. While this advancement has significantly improved data handling, protecting sensitive healthcare information in such a complex environment remains daunting. This comprehensive study explores the crucial intersection between dynamic data placement strategies and network security concerns in distributed cloud environments, particularly healthcare. After establishing the significance and context of this research, the survey delves into the growing need to safeguard medical data within the everevolving landscape of cloud-based healthcare systems. It lays out fundamental concepts, such as dynamic data placement and network security, highlighting their unique implications in the medical domain. Ultimately, this survey sheds light on the most effective approaches for balancing dynamic data placement and network security in the healthcare sector. This research delves into examining many tactics, evaluating their effectiveness in handling delicate medical information, and presenting tangible use cases. A key focus of this investigation is the fusion of data organization and network safety within the healthcare industry. It investigates the adaptability of dynamic data positioning techniques in fortifying network security and safeguarding against potential threats unique to the healthcare sector. Case studies of the successful implementation of these strategies in healthcare establishments are also included. © 2024, Bentham Science Publishers. All rights reserved.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Big Data Security Framework Based on AEFS-KENN Algorithms for the Detection of Cyber-Attacks from Smart Grid Systems

引用

Big Data Mining and Analytics 2024年第2期7卷 399-418页

作者： Sankaramoorthy Muthubalaji Naresh Kumar Muniyaraj Sarvade Pedda Venkata Subba Rao Kavitha Thandapani Pasupuleti Rama Mohan Thangam Somasundaram Yousef Farhaoui Department of Electrical and Electronics Engineering CMR College of Engineering&TechnologyHyderabad 501401India Department of Electronics and Communication Engineering Vardhaman College of Engineering KacharamShamshabad 501218India Department of Electronics and Communication Engineering Sreenidhi Institute of Science and TechnologyHyderabad 501301India Department of Electronics and Communication Engineering Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and TechnologyChennai 600062India Department of Electrical and Electronics Engineering Bharat Institute of Engineering and TechnologyHyderabad 501510India Department of Computer Science and Engineering Amrita School of ComputingAmrita Vishwa VidhyapeethamBengaluru 560035India T-IDMS Department of Computer ScienceFaculty of Sciences and TechniquesMoulay Ismail UniversityErrachidia 52000Morocco

Big data has the ability to open up innovative and ground-breaking prospects for the electrical grid,which also supports to obtain a variety of technological,social,and financial *** is an unprecedented amount of heterogeneous big data as a consequence of the growth of power grid technologies,along with data processing and advanced *** main obstacles in turning the heterogeneous large dataset into useful results are computational burden and information *** original contribution of this paper is to develop a new big data framework for detecting various intrusions from the smart grid systems with the use of AI ***,an AdaBelief Exponential Feature Selection(AEFS)technique is used to efficiently handle the input huge datasets from the smart grid for boosting ***,a Kernel based Extreme Neural Network(KENN)technique is used to anticipate security vulnerabilities more *** Polar Bear Optimization(PBO)algorithm is used to efficiently determine the parameters for the estimate of radial basis ***,several types of smart grid network datasets are employed during analysis in order to examine the outcomes and efficiency of the proposed AdaBelief Exponential Feature Selection-Kernel based Extreme Neural Network(AEFS-KENN)big data security *** results reveal that the accuracy of proposed AEFS-KENN is increased up to 99.5%with precision and AUC of 99%for all smart grid big datasets used in this study.

关键词： smart grid Machine Learning(ML) big data analytics AdaBelief Exponential Feature Selection(AEFS) Polar Bear Optimization(PBO) Kernel Extreme Neural Network(KENN)

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：