检索结果-内蒙古大学图书馆

An efficient text-independent speaker verification for short utterance data from Mobile devices

MULTIMEDIA TOOLS AND APPLICATIONS 2020年第3-4期79卷 3049-3074页

作者： Arora, Sanghamitra V. Vig, Rekha DCE Dept Elect & Commun Engn Gurgaon India NCU Dept Elect Elect & Commun Engn EECE Gurgaon India

Speaker verification is the process used to recognize a speaker from his/her voice characteristics by extracting the features. Speaker verification with text-independent data is a process of verifying the speaker identity without limitation in the speech content. In the speaker verification process, long utterances are normally used but it contains lot of silences leading to complexity and more disruptions. So, we are performing speaker verification method based on short utterance data. The main objective of the research work is to extract, characterize, and recognize the information about speaker identity. Our proposed work contains four stages: 1) utterance partitioning, 2) feature extraction, 3) feature selection, and 4) classification. In our proposed model, an utterance partitioning approach is used to shorten the full-length speech into numerous short-length utterances before the pre-processing stage. In the feature extraction phase, noise removal is carried out with pre-emphasis filter in the pre-processing step. The Mel Advanced Hilbert-Huang Cepstral Coefficients (MAHCC) technique is used for extracting the features from the given input speech signal. Furthermore, the feature selection process is done with the help of a crow search algorithm (CSA) by ranking the given feature set to obtain optimal features for classification. In the classification stage, the Deep Hidden Markov Model (DHMM) method is introduced to classify the features for speaker verification with discriminative pre-training process. Thus, the proposed approach provides an accurate classification and the implementation results show that the performance of the proposed method is better than the existing methods.

关键词： crow search algorithm Deep hidden markov model Feature extraction Speaker verification Utterance partitioning

来源：评论

学校读者我要写书评

暂无评论

Optimal Feature Selection-Based Medical Image Classification Using Deep Learning Model in Internet of Medical Things

引用

IEEE ACCESS 2020年 8卷 58006-58017页

作者： Raj, R. Joshua Samuel Shobana, S. Jeya Pustokhina, Irina Valeryevna Pustokhin, Denis Alexandrovich Gupta, Deepak Shankar, K. CMR Inst Technol Dept Informat Sci & Engn Bengaluru 560037 India Knowledge Univ Dept Comp Sci Coll Sci Erbil 446015 Kurdistan Regio Iraq Plekhanov Russian Univ Econ Dept Entrepreneurship & Logist Moscow 117997 Russia State Univ Management Dept Logist Moscow 109542 Russia Maharaja Agrasen Inst Technol Dept Comp Sci & Engn New Delhi 110086 India Alagappa Univ Dept Comp Applicat Karaikkudi 630002 Tamil Nadu India

Internet of Medical Things (IoMT) is the collection of medical devices and related applications which link the healthcare IT systems through online computer networks. In the field of diagnosis, medical image classification plays an important role in prediction and early diagnosis of critical diseases. Medical images form an indispensable part of a patient & x2019;s health record which can be applied to control, handle and treat the diseases. But, classification of images is a challenging task in computer-based diagnostics. In this research article, we have introduced a improved classifier i.e., Optimal Deep Learning (DL) for classification of lung cancer, brain image, and Alzheimer & x2019;s disease. The researchers proposed the Optimal Feature Selection based Medical Image Classification using DL model by incorporating preprocessing, feature selection and classification. The main goal of the paper is to derive an optimal feature selection model for effective medical image classification. To enhance the performance of the DL classifier, Opposition-based crow search (OCS) algorithm is proposed. The OCS algorithm picks the optimal features from pre-processed images, here Multi-texture, grey level features were selected for the analysis. Finally, the optimal features improved the classification result and increased the accuracy, specificity and sensitivity in the diagnosis of medical images. The proposed results were implemented in MATLAB and compared with existing feature selection models and other classification approaches. The proposed model achieved the maximum performance in terms of accuracy, sensitivity and specificity being 95.22 & x0025;, 86.45 & x0025;and 100 & x0025;for the applied set of images.

关键词： IoMT classification deep learning medical image features crow search algorithm optimization

来源：评论

学校读者我要写书评

暂无评论

Reinforcement-Learning based energy efficient optimized routing protocol for WSN

引用

PEER-TO-PEER NETWORKING AND APPLICATIONS 2022年第3期15卷 1685-1704页

作者： Keerthika, A. Hency, V. Berlin Vellore Inst Technol Chennai Tamil Nadu India

Wireless Sensor Networks (WSN) is an efficient network for monitoring and recording the physical environment and transfers the monitored data into the central location using widely distributed sensor nodes. One of the main problems in WSN is the issue of developing an energy-efficient routing protocol that achieves less energy consumption and enhances the lifetime of the network. During the past decades, a researcher uses the mobile sink to reduce the energy problem and hotspot problems. In this paper, a dynamic routing protocol named Reinforcement-Learning based energy Efficient Optimized Routing (RLER) is proposed to reduce the energy consumption of the nodes and to protract the lifetime of the network. In the proposed work (RLER), Grid- Tree-based clustering is employed, and the root node(RN) selection is done by using the RLFIS algorithm, which uses the three parameters Bipartivity Index (BI), Neighbourhood Overlapping (NOVER) and Algebraic Connectivity (AC) to select the root node. Then the tree structure is formed by the root node inside the grid based on the energy and transmission range of the nodes, which supports the inter-cluster communication within the grids. Hybrid BAT-crow (BCSA) search algorithm is used to relocate the sink node based on the fitness value of the nodes. The experimental results show that the proposed methodology (RLER) provides better results in energy consumption and improves the lifetime of the network. The proposed RLER simulation is done on the OMNeT + + platform.

关键词： Wireless sensor network crow search algorithm Reinforcement learning-fuzzy inference system Clustering methods Bipartivity index QoS constraints Neighborhood overlapping

来源：评论

学校读者我要写书评

暂无评论

An intelligent hybrid deep belief network model for predicting students employability

引用

SOFT COMPUTING 2021年第14期25卷 9241-9254页

作者： Bai, Anita Hira, Swati CVR Coll Engn Dept Comp Sci & Engn Hyderabad India Shri Ramdeobaba Coll Engn & Management Dept Comp Sci & Engn Nagpur Maharashtra India

In recent times, the question of employability has become a critical concern not only for degree holders but for educational organizations. Hence, employability prediction models play an important role in analyzing the student's capability to get employment. In this paper, a hybrid model of deep belief network and soft max regression (DBN-SR) is proposed for student employability prediction. Initially, pre-processing is performed on student's data for removing irrelevant attributes to achieve data consistency. Further, to enhance the accuracy of the prediction model, the crow search algorithm-based feature selection model is employed to select the optimal subset of features from original features. Then, the selected subset of features is taken as the input of the deep belief network (DBN) for intrinsic feature learning to obtain high-level feature representation. Finally, the soft max regression (SR) is used to predict the class of students as employed or unemployed. The proposed employability prediction model achieves above 98% of accuracy which is comparatively 2.5%, 5% higher than the deep autoencoder and deep neural network-based models. The performance outcomes proved that the proposed DBN-SR model has been well suitable for predicting student's employability.

关键词： Employability prediction crow search algorithm Deep belief network-softmax regression Data mining

来源：评论

学校读者我要写书评

暂无评论

Tsallis entropy of uncertain random variables and its application

引用

SOFT COMPUTING 2021年第17期25卷 11735-11743页

作者： He, Zhenhua Ahmadzade, Hamed Rezaei, Kamran Rezaei, Hassan Naderi, Habib Guangxi Univ Finance & Econ Sch Informat & Stat Nanning 530003 Peoples R China Univ Sistan & Baluchestan Dept Stat Zahedan Iran Univ Sistan & Baluchestan Dept Comp Sci Zahedan Iran

Tsallis entropy is a flexible extension of Shanon (logarithm) entropy. Since entropy measures indeterminacy of an uncertain random variable, this paper proposes the concept of partial Tsallis entropy for uncertain random variables as a flexible devise in chance theory. An approach for calculating partial Tsallis entropy for uncertain random variables, based on Monte Carlo simulation, is provided. As an application in finance, partial Tsallis entropy is invoked to optimize portfolio selection of uncertain random returns via crow search algorithm.

关键词： Chance theory Uncertain random variable Partial Tsallis entropy Portfolio selection Monte Carlo simulation crow search algorithm

来源：评论

学校读者我要写书评

暂无评论

Swarm-based optimally selected histogram computation system for image enhancement

引用

NEURAL COMPUTING & APPLICATIONS 2022年第9期34卷 7053-7067页

作者： Bhandari, Ashish Kumar Singh, Neha Singh, Anurag Natl Inst Technol Patna Dept Elect & Commun Engn Patna 800005 Bihar India Int Inst Informat Technol Dept Elect & Commun Engn Naya Raipur India

In this paper, two newly proposed evolutionary computational algorithms (ECAs) are joined with Otsu thresholding method to achieve brightness preserving image contrast enhancement. The selected ECAs are the whale optimization algorithm (WOA) and the crow search algorithm (CSA) which embrace more contrast and minimum entropy alteration corresponding to the original image. The proposed algorithm employs histogram equalization method, based on an improved cumulative distribution to calculate a mapping function. Therefore, a 3-stage procedure has been assumed to change the original histogram. The primary step of the proposed method is to sub-divide the image histogram by using the Otsu thresholding technique. Further, both of histograms are weighted and thresholded in order to control the level of enhancement. The constraint parameters are attained by WOA and CSA algorithms for modification. After constraining the histograms, mean shift modification is executed to slight altering the position of mean shifting from input to output image. The results reflect that proposed technique accomplishes balanced contrast enhancement and better color preservation in comparison with surviving techniques. Through the proposed technique, the enhanced images achieve low contrast boosting, a good trade-off between detail improvement, and brightness conservation with naturalness of the input image.

关键词： Image enhancement Otsu thresholding Whale optimization crow search algorithm Histogram equalization

来源：评论

学校读者我要写书评

暂无评论

Design optimization of a hybrid system subject to reliability level and renewable energy penetration

引用

ENERGY 2020年第0期193卷 416-427页

作者： Ghaffari, Abolfazl Askarzadeh, Alireza Grad Univ Adv Technol Fac Elect & Comp Engn Kerman Iran Grad Univ Adv Technol Inst Sci & High Technol & Environm Sci Dept Energy Management & Optimizat Kerman Iran

In recent years, design optimization of hybrid power generation systems which utilize renewable energy resources has received a significant attention. One of the most challenging problems in design of such systems is optimal sizing. To tackle this drawback, this paper presents an efficient and robust optimization approach for size optimization of a hybrid system composed of photovoltaic (PV) panel, diesel generator and fuel cell (FC). The proposed method is a modified version of crow search algorithm (CSA) in which awareness probability is adjusted by an adaptive manner. In the sizing framework, total net present cost (TNPC) is minimized subject to two main constraints: loss of power supply probability (LPSP) and renewable energy portion (REP). Simulation results show that (1) REP has a considerable impact on TNPC of the designed system, (2) optimal combination of PV, diesel generator, FC, electrolyzer and hydrogen tank leads to establishing a reliable and cost-effective hybrid power generation system and (3) the proposed method (CSA(adaptive-AP)) finds more promising and robust results than original CSA, genetic algorithm (GA) and particle swarm optimization (PSO). (C) 2019 Elsevier Ltd. All rights reserved.

关键词： Design optimization Renewable energy portion crow search algorithm Adaptive awareness probability

来源：评论

学校读者我要写书评

暂无评论

Geometric optimization of the space around the buildings based on the improvement of thermal comfort efficiency of the building

引用

INTERNATIONAL JOURNAL OF LOW-CARBON TECHNOLOGIES 2024年第19期19卷 1288-1300页

作者： Cui, Guolian Shangqiu Inst Technol Sch Educationy & Modern Art Shangqiu 476000 Peoples R China

This stud introduces a pioneering investigation into the geometric optimization of building surroundings to elevate thermal comfort efficiency and foster sustainable development within the construction industry. Employing the ENVI-met simulation tool and a novel crow search optimization algorithm, the research rigorously quantifies occupants' dissatisfaction utilizing the predicted percentage dissatisfied model, meticulously considering various environmental parameters. The simulations are based on the weather data of Nanjing, China, and the numerical results were validated against the observed data. The results reveal a remarkable 7% reduction in energy consumption and a corresponding 7.2% decrease in CO2 emissions compared to baseline configurations, underscoring the substantial impact of the proposed crow search algorithm. Notably, the study identifies an optimal configuration, characterized by a 30% vegetation cover, west-east orientation, and a three-story building, highlighting the algorithm's effectiveness in identifying configurations that concurrently enhance thermal comfort and mitigate energy consumption. These findings highlight the critical influence of geometric factors on thermal performance and underscore the significance of integrating innovative methodologies to address contemporary challenges in sustainable building design. By offering novel insights and practical solutions, this research contributes to advancing sustainable practices in architecture and urban planning, ultimately promoting occupant well-being and fostering energy-efficient construction practices.

关键词： geometric optimization thermal comfort efficiency sustainable development building surroundings crow search algorithm CO2 emission

来源：评论

学校读者我要写书评

暂无评论

Intrusion detection based on hybrid metaheuristic feature selection

引用

COMPUTER JOURNAL 2024年第1期68卷 13-22页

作者： Zhang, Fengjun Huang, Lisheng Shi, Kai Zhai, Shengjie Lan, Yunhai Li, Qinghua China Elect Technol Grp Corp Res Inst 30 Dept Integrat 8 Chuangye Rd Chengdu 610036 Sichuan Peoples R China Univ Elect Sci & Technol China Sch Comp Sci & Engn 2006 Xiyuan Ave Chengdu 611731 Sichuan Peoples R China Cent Southern Air Traff Control Bur Civil Aviat Network Ctr R&D Ctr 1633 Beitai Rd Guangzhou 510405 Guangdong Peoples R China

The multidimensional features of network flows are the main data source for intrusion detection, but excessively low-value features generate accuracy and efficiency challenges. Researchers have used redundant feature reduction to simplify intrusion detections, and feature selection algorithms are beginning to be widely used. This paper presents a novel hybrid feature selection algorithm, CSA-FPA, which combines both a crow search algorithm and a flower pollination algorithm. In this method, properties such as local pollination and the levy flight of FPA are used to balance the global search and local search efficiencies, and parameters such as group distance and probability thresholds are introduced to customize the model's appearance. The simulation results on the UNSW-NB15 and CIC-IDS2017 datasets show that the proposed CSA-FPA method achieves better detection accuracies than previous algorithms. Using the proposed feature selection method, the AdaBoost classifier achieved a detection accuracy of 99.14% on the CIC-IDS2017 dataset and 97.98% on the UNSW-NB15 dataset.

关键词： intrusion detection system crow search algorithm flower pollination algorithm feature selection

来源：评论

学校读者我要写书评

暂无评论

Wrapper-Based Feature Selection and Optimization-Enabled Hybrid Deep Learning Framework for Stock Market Prediction

引用

INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGY & DECISION MAKING 2024年第1期23卷 475-500页

作者： Patil, Pankaj Rambhau Parasar, Deepa Charhate, Shrikant Amity Univ Maharashtra Amity Sch Engn & Technol Dept Comp Sci & Engn Mumbai India Amity Univ Maharashtra Dept Comp Sci & Engn Mumbai India Amity Univ Maharashtra Dept Civil Engn Mumbai India

Stock market is a significant element of economic market. Accurate forecasting of stock market is very helpful to shareholders because future prediction of a stock value will elevate the profits of investors. The data acquired from the stock market is a time-series data, in which the values of the stock prices are inherently varied with respect to time. Due to its complexity nature and nonlinearity characteristics, the prediction of stock market becomes very difficult and still it remains a challenging task. In order to cope up with such limitation, this research proposes an effective strategy called Deep Recurrent Rider LSTM to provide an accurate detection of stock market values. The accurate forecasting of stock market is carried out with two classifiers, namely Rider Deep Long Short-Term Memory (Rider Deep LSTM) and Deep Recurrent Neural Network (Deep RNN). The Rider Deep LSTM is derived by the integration of Rider concept with Deep LSTM, whereas the Deep RNN is trained using the proposed Shuffled crow search Optimization (SCSO). Moreover, the SCSO is derived by the integration of Shuffled Shepherd Optimization (SSO) algorithm and crow search algorithm (CSA). Finally, the predicted output is determined based on the error condition. Furthermore, the proposed Deep Recurrent Rider LSTM achieved the MSE and RMSE of 0.018 and 0.132 that shows higher performance with better accuracy. The stock market prediction using the proposed classification model is accurate and improves the effectiveness of the classification.

关键词： Stock market prediction deep RNN rider deep LSTM crow search algorithm shuffled shepherd optimization 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：