检索结果-内蒙古大学图书馆

25th Interspeech Conference

作者： Choi, Jeong-Hwan Jeoung, Ye-Rin Kim, Ilseok Chang, Joon-Hyuk Hanyang Univ Dept Elect Engn Seoul South Korea

This paper proposes an efficient speaker embedding (SE) extraction method that employs a twofold sliding window algorithm (SWA) for speaker diarization (SD) systems. Nonoverlapping short segments are obtained through the first SWA and fed into the frame-level neural networks of a pre-trained SE model to extract frame-level representations. The neighboring frame-level representations are concatenated along the time axis through the second SWA, which enables an overlap between representations. The concatenated representations are used to extract multiple SEs. Additionally, we propose a fine-tuning strategy that employs a residual adapter and knowledge distillation techniques on a pre-trained SE model to refine the frame-level representation. Experimental results using two SD benchmarks show the effectiveness of the proposed extraction method with a fine-tuned SE model in terms of floating-point operations while maintaining the diarization error rate.

关键词： speaker diarization speaker embedding segmentation sliding window algorithm

来源：评论

学校读者我要写书评

暂无评论

Prediction of Temperature for next three days using Decision Tree algorithm by comparing sliding window algorithm for better Accuracy 1

Prediction of Temperature for next three days using Decision...

引用

1st International Conference on Technologies for Smart Green Connected Society 2021, ICTSGS 2021

作者： Yaswanth Reddy, G. Saravanan, M.S. Saveetha University Tamil Nadu Chennai602105 India

ISBN: (纸本)9781607685395

Aim: The aim of this study is to predict the temperature for the next three days using a machine learning algorithm and the objective of this research is to improve the accuracy of temperature prediction. Materials and methods: The study used 36 samples with two groups of algorithms with the g-power value of 80%. To predict the temperature the sliding window algorithm (SLA) machine learning algorithm has found 89% of accuracy and this study will find better accuracy for temperature prediction with the Decision Tree (DT) algorithm. Results: This research study found 92% of accuracy for temperature prediction using a DT algorithm. Conclusion: This study concludes that the DT algorithm on temperature prediction is significantly better than the SLA. © The Electrochemical Society

关键词： Decision Tree algorithm Innovative weather prediction Machine learning sliding window algorithm

来源：评论

学校读者我要写书评

暂无评论

Anti-Motion Interference Wearable Device for Monitoring Blood Oxygen Saturation Based on sliding window algorithm

引用

IEEE ACCESS 2020年 8卷 124675-124687页

作者： Jiao, Bin Zhengzhou Univ Aeronaut Sch Art & Design Zhengzhou 450000 Peoples R China

Wearable multi-physiological parameter monitoring technology can realize non-intrusive, non-invasive daily health monitoring of the human body. It has the characteristics of convenient operation, long-term continuous work, display of results, abnormal physiological condition alarm and wireless data transmission. First, an improved sliding window confidence propagation algorithm is proposed, which reduces the inefficient iterative process in the decoding process by correlation coefficients and reduces the running time of the algorithm by more than 1.5 times;designing a channel noise estimation method based on sliding window confidence propagation algorithm, and The source coding is similar. In order to estimate the local statistical parameters of the channel noise, the confidence propagation, the decoder uses the local partial probability of the adjacent accompanying child nodes to estimate the noise parameters of the channel noise. Secondly, the design and research of the blood oxygen saturation monitor, including filtering out high-frequency glitch noise, low-frequency baseline drift noise, and mixing and sudden motion interference noise, also achieved rapid and accurate extraction of pulse wave feature points, and calculated the blood oxygen value and pulse rate value. Compared with the traditional oximeter, the filtering algorithm and pulse wave feature point extraction algorithm on the oximeter software have less calculation amount and higher real-time performance. Finally, the relevant test experiment of the blood oxygen saturation monitor proves that the blood oximeter has good stability, accuracy and sensitivity. The practical use of anti-motion interference wearable devices not only shows that the wearable PPG sensor in this paper can stably obtain high-quality PPG signals, but also reflects its many applications in the field of real-time blood.

关键词： sliding window algorithm blood oxygen saturation anti-motion interference wearable

来源：评论

学校读者我要写书评

暂无评论

A global grid model for the correction of the vertical zenith total delay based on a sliding window algorithm

引用

GPS SOLUTIONS 2021年第3期25卷 98-98页

作者： Huang, Liangke Zhu, Ge Liu, Lilong Chen, Hua Jiang, Weiping Guilin Univ Technol Coll Geomat & Geoinformat Guilin 541004 Peoples R China Guangxi Key Lab Spatial Informat & Geomat Guilin 541004 Peoples R China Wuhan Univ Sch Geodesy & Geomat Wuhan 430079 Peoples R China Wuhan Univ GNSS Res Ctr Wuhan 430079 Peoples R China

Reanalysis products have been applied to calculate the tropospheric delay for Global Navigation Satellite System (GNSS) positioning purposes widely. It is necessary to obtain high-precision tropospheric delay information from GNSS users with a high-precision tropospheric vertical stratification model because the height of the grids of the atmospheric reanalysis data is inconsistent with that of GNSS users, especially in regions with high terrains. In addition, the variation of the tropospheric delay in the vertical direction is much higher than that in the horizontal direction. The zenith total delay (ZTD) vertical stratification model is also key to the development of real-time and high-precision ZTD models. A new approach, the sliding window algorithm, is proposed to develop a ZTD vertical stratification model. In this work, a ZTD vertical stratification model considering spatiotemporal factors is developed based on the second Modern-Era Retrospective analysis for Research and Applications (MERRA-2) data, which is named the GZTD-H model. Radiosonde and International GNSS Service (IGS) data are treated as reference values to evaluate the performance of the GZTD-H model, which is compared to the model GPT2w. The results show that the GZTD-H model realizes the highest performance in ZTD layered vertical interpolation against ZTD layered profiles obtained at radiosonde sites, which achieves an improvement of 10% over the model GPT2w. Compared to model GPT2w, the GZTD-H model attains a spatial interpolation improvement of 8% for the Global Geodetic Observing System (GGOS) Atmosphere gridded ZTD over the surface ZTD calculated from radiosonde profiles. Furthermore, compared to model GPT2w, the model GZTD-H also attains improvements of 11% over the precise ZTD products acquired at IGS sites. In terms of model parameters, the GZTD-H model is greatly reduced and optimized over model GPT2w. Hence, the applicability of this model is enhanced in terms of GNSS atmospheric soun

关键词： sliding window algorithm Tropospheric delay model Zenith total delay Vertical stratification

来源：评论

学校读者我要写书评

暂无评论

Ship Spatiotemporal Key Feature Point Online Extraction Based on AIS Multi-Sensor Data Using an Improved sliding window algorithm

引用

SENSORS 2019年第12期19卷 2706.-2706.页

作者： Gao, Miao Shi, Guo-You Dalian Maritime Univ Nav Coll Dalian 116026 Peoples R China Dalian Maritime Univ Collaborat Innovat Res Inst Autonomous Ship Dalian 116026 Peoples R China Key Lab Nav Safety Guarantee Liaoning Prov Dalian 116026 Peoples R China

Large volumes of automatic identification system (AIS) data provide new ideas and methods for ship data mining and navigation behavior pattern analysis. However, large volumes of big data have low unit values, resulting in the need for large-scale computing, storage, and display. Learning efficiency is low and learning direction is blind and untargeted. Therefore, key feature point (KFP) extraction from the ship trajectory plays an important role in fields such as ship navigation behavior analysis and big data mining. In this paper, we propose a ship spatiotemporal KFP online extraction algorithm that is applied to AIS trajectory data. The sliding window algorithm is modified for application to ship navigation angle deviation, position deviation, and the spatiotemporal characteristics of AIS data. Next, in order to facilitate the subsequent use of the algorithm, a recommended threshold range for the corresponding two parameters is discussed. Finally, the performance of the proposed method is compared with that of the Douglas-Peucker (DP) algorithm to assess its feature extraction accuracy and operational efficiency. The results show that the proposed improved sliding window algorithm can be applied to rapidly and easily extract the KFPs from AIS trajectory data. This ability provides significant benefits for ship traffic flow and navigational behavior learning.

关键词： big data mining spatiotemporal ship behavior key feature point feature online extraction AIS sliding window algorithm

来源：评论

学校读者我要写书评

暂无评论

New Regular sliding window algorithms for Elliptic Curve Scalar Point Multiplication

引用

AUTOMATIC CONTROL AND COMPUTER SCIENCES 2021年第8期55卷 1029-1038页

作者： Shenets, N. N. Petushkov, A. S. Peter Great St Petersburg Polytech Univ St Petersburg 195251 Russia

Simple side channel attacks on the implementation of elliptic curves scalar point multiplication algorithms are considered. New regular sliding window algorithms for calculating multiple points are proposed. Their opt... 详细信息

关键词： side channel attack point multiplication on elliptic curves regular algorithm sliding window algorithm

来源：评论

学校读者我要写书评

暂无评论

A framework for ship abnormal behaviour detection and classification using AIS data

引用

RELIABILITY ENGINEERING & SYSTEM SAFETY 2024年 247卷

作者： Rong, H. Teixeira, A. P. Soares, C. Guedes Univ Nova Lisboa Inst Super Tecn Ctr Marine Technol & Ocean Engn CENTEC Lisbon Portugal

This paper proposes a method for detecting and classifying ship abnormal behaviour in ship trajectories. The method involves generating parameter profiles for the ship's trajectory and applying a sliding window algorithm to detect the ship's abnormal behaviour. Then, several features are adopted to effectively describe the characteristics of each ship's abnormal behaviour, such as the standard deviation of speed, detour factor, maximum drift angle, accumulative change of Course Over Ground and maximum lateral distance to the ship route. A density -based clustering algorithm is applied to group similar abnormal behaviour patterns according to the feature similarity, and the Random Forest Classification method is used to train a classification model based on the features extracted from the clusters. The proposed method is then tested on historical ship trajectory data provided by the Automatic Identification System. The results suggest that the method effectively identifies and classifies different abnormal behaviours in the ship trajectories.

关键词： Maritime Traffic AIS data Ship abnormal behaviour sliding window algorithm Feature extraction Random forest classification

来源：评论

学校读者我要写书评

暂无评论

Identification of growth-related genes based on BSA in Pacific white shrimp Litopenaeus vannamei

引用

AQUACULTURE 2025年 596卷

作者： Bao, Zhenning Yu, Yang Lin, Pengfei Li, Fuhua Chinese Acad Sci Inst Oceanol Key Lab Breeding Biotechnol & Sustainable Aquacult Qingdao 266071 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China Qingdao Marine Sci & Technol Ctr Lab Marine Biol & Biotechnol Qingdao 266071 Peoples R China Chinese Acad Sci Inst Oceanol Shandong Prov Key Lab Expt Marine Biol Qingdao 266071 Peoples R China

The development of SNP genotyping techniques has facilitated the application of genome-wide association studies (GWAS) and genomic selection (GS) in plants and animals. However, the cost of high-throughput SNP genotyping is particularly high in the case of aquaculture animals which can produce a large number of offspring. Consequently, a cost effective method for trait-associated SNP identification and application was urgently needed in aquaculture breeding. In this study, a Bulk Segregation Analysis sequencing (BSA-seq) method was employed to identify growth-related single nucleotide polymorphisms (SNPs) and genes in the Pacific white shrimp Litopenaeus vannamei. Two independent families were used for this analysis to reduce the false positive. The DNA of individuals with the highest and lowest body weights was pooled separately. Two methods including the ED (Euclidean distance) method and the chi-square test were applied to calculate the allele differences between high and low growth rate individuals. A total of 24,325,437 SNPs were identified and consequently 1,299 SNPs and 163 genes were identified to be associated with growth traits. KEGG enrichment analysis showed that these genes were significantly enriched in phosphatidylinositol signaling system and phospholipase D signaling pathway. The diacylglycerol kinase genes, phosphatidylinositol 3-kinase genes, inositol polyphosphate 5-phosphatase genes, mitogen-activated protein kinase genes, myotubularin-related protein genes and epidermal growth factor receptor genes in these pathways were considered as candidate growth genes. This study demonstrated that the BSA-seq based on whole genome resequencing is cost-effective and useful in aquaculture. The results not only provide useful information for understanding the genetic basis of growth traits, but also offer a source of SNPs for marker-assisted selection (MAS) and genomic selection (GS) in shrimp.

关键词： Litopenaeus vannamei Bulk segregation analysis (BSA) Growth trait Single nucleotide polymorphisms (SNPs) sliding window algorithm

来源：评论

学校读者我要写书评

暂无评论

An improved global grid model for calibrating zenith tropospheric delay for GNSS applications

引用

GPS SOLUTIONS 2023年第1期27卷 17页

作者： Huang, Liangke Zhu, Ge Peng, Hua Liu, Lilong Ren, Chao Jiang, Weiping Guilin Univ Technol Coll Geomat & Geoinformat Guilin 541004 Peoples R China Guangxi Key Lab Spatial Informat & Geomat Guilin 541004 Peoples R China Wuhan Univ GNSS Res Ctr Wuhan 430079 Peoples R China

Accurate modeling of zenith tropospheric delay (ZTD) is beneficial for high-precision navigation and positioning. Many models with good performance have been developed for calibrating ZTD, such as the GPT3 model, which is recognized as an excellent global model and is widely used. However, certain limitations still remain in current models, such as the adoption of only single gridded data for modeling, and the model parameters need to be further optimized. In our previous research, a new approach based on the sliding window algorithm was proposed and applied to develop the GZTD-H model to address some of these limitations. However, this model is only suitable for the vertical adjustment of ZTD, not for estimating ZTD directly. In this study, an improved global grid ZTD model considering height scale factor (GGZTD-H) is derived from the initial GZTD-H model for estimating ZTD. The RMSs of the GGZTD-H model are 4.11 cm and 3.29 cm as validated by radiosonde data and IGS data, respectively. Compared with the UNB3m model and the canonical GPT3 model, the new model exhibits better performance. Moreover, three resolutions of the GGZTD-H model have been developed to reduce the quantity of gridded data delivered to users and optimize the ZTD computation process. Compared with the GPT3 model, the GGZTD-H model shows better performance with lower resolution and requires fewer model parameters for ZTD estimation, greatly optimizing ZTD computation. Users may select the best model that meets their needs in terms of the balance between resolution and accuracy. The high-precision GGZTD-H model could be used as a ZTD vertical stratification model for the vertical adjustment of atmospheric data and as an empirical model for ZTD estimation, which has potential applications in GNSS precise positioning, such as for the establishment and maintenance of the global terrestrial reference frame.

关键词： Zenith tropospheric delay Global model sliding window algorithm Global navigation satellite system (GNSS) Vertical adjustment

来源：评论

学校读者我要写书评

暂无评论

Event-triggered Relearning Modeling Method for System with Non-stationary Variable Operating Conditions 36

Event-triggered Relearning Modeling Method for System with N...

引用

36th Chinese Control and Decision Conference (CCDC)

作者： Liu, Jiyan Sun, Qinjiang Zhou, Yuyang Zhang, Yong Inner Mongolia Univ Sci & Technol Sch Informat Engn Baotou 014010 Peoples R China CNOOC China Ltd Tianjin Branch Tianjin 300459 Peoples R China Edinburgh Napier Univ Sch Comp Engn & Built Environm Edinburgh EH10 5DT Midlothian Scotland

ISBN: (纸本)9798350387780;9798350387797

This paper proposes an innovative event-triggered relearning strategy for neural network modeling, to enhance prediction accuracy under the non-stationary and variable conditions frequently encountered in complex industrial environments. Firstly, the proposed approach divides the input and output datasets across different operational conditions, by applying a sliding window algorithm along with entropy to identify the boundaries of the input data. Following this, we utilize the prediction error from the current model, which is relevant to variable operational states, to construct an array of event-triggered conditions. Once these conditions are triggered, we employ the corresponding dataset to re-establish the model and predict the data under this operating condition. Compared with the traditional neural network model, the proposed model can produce a better estimation accuracy. Finally, numerical simulation tests are provided to verify the effectiveness of the proposed method. The results show that the proposed method can effectively improve the prediction accuracy of neural network model under non-stationary variable working conditions.

关键词： Neural network Non-stationary and variable conditions Event-triggered conditions sliding window algorithm Information entropy

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：