检索结果-内蒙古大学图书馆

International Conference on Artificial Intelligence and Computer Information technology (AICIT)

作者： Qihao Li Data Science and Big Data Technology Lanzhou City University LanZhou China

ISBN: (数字)9798350365443

ISBN: (纸本)9798350365450

This study aims to improve the accuracy of click-through rate prediction for push ads through machine learning methods. Using the dataset released by Tianchi, we synthesized basic user information, ad features and user behavior logs to enrich the training data. In feature selection, we sieve out the fields that have less impact on prediction, while weighting the user behavior data. For the sample imbalance problem, a Random OverSampling strategy is used to ensure the fairness and effectiveness of the model. Through these methods, we significantly improve the model's prediction accuracy of push ad clicking behavior, and provide a practical solution for the optimization of the ad push system.

关键词： Accuracy Computational modeling Training data Machine learning Predictive models Feature extraction data models Information technology Optimization Testing

来源：评论

学校读者我要写书评

暂无评论

An accurate cell segmentation method for high-resolution complex brain image 9

An accurate cell segmentation method for high-resolution com...

引用

9th International Conference on Image, Vision and Computing, ICIVC 2024

作者： Xu, Xiaofeng Li, Shuai Zheng, Wei Henan Engineering Research Center of Intelligent Processing for Big Data of Digital Image School of Computer Science and Technology Nanyang Normal University Nanyang China

ISBN: (纸本)9798350385991

Cell segmentation is a key step in medical image analysis and cell biology research, and is important in areas such as understanding cell behaviour, cancer detection and cell cycle analysis. This task is challenging due to the presence of overlap between cells and image noise. With the development of imaging techniques, the amount and quality of acquired cellular image data have increased significantly and even contain diverse information such as neuronal structures and blood vessels, which puts higher demands on segmentation techniques. Real rat brain cell images obtained using micro-optical slice tomography (MOST) contain complex and diverse information such as neuronal structure, cell morphology, and the distribution of blood vessels, demonstrating the characteristics of real biological samples, for which segmentation of the target region is more difficult. In order to solve these challenges, this paper proposes a new cell image segmentation method that combines watershed order differential feature extraction and fuzzy entropy spatial theory. Due to the complexity of image information, we first use traditional image enhancement techniques. Then feature extraction is performed and fuzzy entropy processing is added to achieve accurate identification of cell boundaries. Then distance transform and curvature analysis are added to further segment the cell clusters into individual cells while avoiding over-segmentation and missed detection. Experiments demonstrate that the method not only shows high performance on our dataset, but also validates our method on several publicly available cellular datasets respectively, proving that the segmentation algorithm shows great potential for segmentation on different types of cells. Future research will further optimise the model to improve segmentation performance on complex and irregularly shaped cells and explore the application of these methods to other types of medical image segmentation tasks. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Research on the Detection Methods for Changes in the Mangroves of Dongzhaigang, Hainan from 2019 to 2023 24

Research on the Detection Methods for Changes in the Mangrov...

引用

8th International Conference on Electronic Information technology and Computer Engineering, EITCE 2024

作者： Zhang, Xianyao Lu, Huajun Qiu, Shuailong Liu, Lijin Wu, Zhongqiang School of Information Science and Technology Hainan Normal University Hainan Haikou China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Guangdong Shenzhen China

ISBN: (纸本)9798400710094

This study introduces the research methods for detecting changes in the mangroves of Dongzhaigang, Hainan from 2019 to 2023. Mangrove ecosystems play a crucial role in providing habitats and have ecological and economic value, but in recent years, they have faced degradation and management issues. In order to effectively monitor mangrove changes, researchers have utilized remote sensing techniques and machine learning methods for mangrove cover information extraction and classification. The paper describes the methods of estimating mangrove cover using remote sensing satellite imagery and spectral indices and explores the applications of clustering, decision trees, support vector machines, and deep learning in mangrove change detection. The research results indicate that the support vector machine algorithm performs well in classifying mangrove and non-mangrove areas. This study provides a basis for the management and decision-making of Dongzhaigang Mangrove Nature Reserve in Hainan Province and highlights the future research potential in the application of multi-source and multi-resolution data and deep learning techniques. © 2024 Copyright held by the owner/author(s).

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A Method of Operating Express/Local Mode Under Unidirectional Tidal Passenger Flow 3rd

A Method of Operating Express/Local Mode Under Unidirectiona...

引用

3rd International Conference on Smart Manufacturing, Industrial and Logistics Engineering, SMILE 2023 and the 7th Asian Conference of Management science and Applications, ACMSA 2023

作者： Wang, Hao Wang, Zongshui Beijing Information Science and Technology University Beijing China Beijing Key Laboratory of Green Development and Decision Making Based on Big Data Beijing China

ISBN: (纸本)9789819701933

The existing methods of operating express/local mode require additional railway lines. However, renovation projects of existing lines are often limited by engineering technology and funding. Meanwhile, tidal passenger flow in rush hour is obvious. To sum up, this paper proposes a method of operating express/local mode under existing railway lines and unidirectional tidal passenger flow. This method sets that all trains in the same direction of unidirectional tidal passenger flow are express trains, and all trains in the opposite direction of unidirectional tidal passenger flow are local trains. After simulation, this method has been proven to effectively reduce total passenger travel time in specific scenarios, providing references for subsequent research and engineering applications. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Travel time

来源：评论

学校读者我要写书评

暂无评论

Representation learning via an integrated autoencoder for unsupervised domain adaptation

引用

Frontiers of Computer science 2023年第5期17卷 75-87页

作者： Yi ZHU Xindong WU Jipeng QIANG Yunhao YUAN Yun LI School of Information Engineering Yangzhou UniversityYangzhou 225127China Key Laboratory of Knowledge Engineering with Big Data(Ministry of Education of China) Hefei University of TechnologyHefei 230009China School of Computer Science and Information Engineering Hefei University of TechnologyHefei 230601China

The purpose of unsupervised domain adaptation is to use the knowledge of the source domain whose data distribution is different from that of the target domain for promoting the learning task in the target *** key bottleneck in unsupervised domain adaptation is how to obtain higher-level and more abstract feature representations between source and target domains which can bridge the chasm of domain ***,deep learning methods based on autoencoder have achieved sound performance in representation learning,and many dual or serial autoencoderbased methods take different characteristics of data into consideration for improving the effectiveness of unsupervised domain ***,most existing methods of autoencoders just serially connect the features generated by different autoencoders,which pose challenges for the discriminative representation learning and fail to find the real cross-domain *** address this problem,we propose a novel representation learning method based on an integrated autoencoders for unsupervised domain adaptation,called *** capture the inter-and inner-domain features of the raw data,two different autoencoders,which are the marginalized autoencoder with maximum mean discrepancy(mAE)and convolutional autoencoder(CAE)respectively,are proposed to learn different feature *** higher-level features are obtained by these two different autoencoders,a sparse autoencoder is introduced to compact these inter-and inner-domain *** addition,a whitening layer is embedded for features processed before the mAE to reduce redundant features inside a local *** results demonstrate the effectiveness of our proposed method compared with several state-of-the-art baseline methods.

关键词： unsupervised domain adaptation representation learning marginalized autoencoder convolutional autoen-coder sparse autoencoder

来源：评论

学校读者我要写书评

暂无评论

Correcting Pronoun Homophones with Subtle Semantics in Chinese Speech Recognition 30

Correcting Pronoun Homophones with Subtle Semantics in Chine...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Zhang, Zhaobo Gan, Rui Yuan, Pingpeng Jin, Hai National Engineering Research Center for Big Data Technology and System Service Computing Technology and System Laboratory Cluster and Grid Computing Laboratory Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9782493814104

Speech recognition is becoming prevalent in daily life. However, due to the similar semantic context of the entities and the overlap of Chinese pronunciation, the pronoun homophone, especially "他/她/它 (he/she/it)", (their pronunciation is "Tā") is usually recognized incorrectly. It poses a challenge to automatically correct them during the post-processing of Chinese speech recognition. In this paper, we propose three models to address the common confusion issues in this domain, tailored to various application scenarios. We implement the language model, the LSTM model with semantic features, and the rule-based assisted Ngram model, enabling our models to adapt to a wide range of requirements, from high-precision to low-resource offline devices. The extensive experiments show that our models achieve the highest recognition rate for "Tā" correction with improvements from 70% in the popular voice input methods up to 90%. Further ablation analysis underscores the effectiveness of our models in enhancing recognition accuracy. Therefore, our models improve the overall experience of Chinese speech recognition of "Tā" and reduce the burden of manual transcription corrections. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Microwave Signal Replicator Design for Testing of the Multi-Channel Transcranial Magnetic Stimulator

引用

International Journal of Crowd science 2024年第2期8卷 83-87页

作者： Chen, Qiuju Zhang, Kai Pang, Jian Zhang, Shancai Feng, Guangyao Wu, Congfeng Li, Wei National Synchrotron Radiation Laboratory University of Science and Technology of China Hefei230029 China Laboratory for Big Data and Decision Hefei230027 China Advanced Manufacturing Engineering College Hefei University Hefei230601 China

The multi-channel transcranial magnetic stimulator is commonly used for rehabilitation treatment of ischemic cerebrovascular disease, neurosis, and brain injury diseases in the elderly. And multiple high coherency signals are required as inputs for its test, which are with consistent frequency, adjustable amplitude, equal magnitude, and long-time phase consistency ⩽±1°. But ordinary signal sources have only one output channel, which is far from meeting the test requirements. If the traditional power division is used for the duplication, the phase difference between the channels will be larger than ±1°. If the multi-throw Radio Frequency (RF) switch is used, it will seriously affect the phase consistency of the input signal and increase the distortion of the signal. In order to solve the above problems, the method to extend the output of the microwave signal source has been proposed in the paper by using the integrated transceiver AD9361. The proposed method can realize the duplication of the number of signal channels according to user requirements. Then a signal replicator has been designed by combining AD9361 chip and Field-Programmable Gate Array (FPGA) control module. The experimental platform has been built to test the performance of the designed device. Experimental results show that the duplicated signals are highly consistent with the source signal in the power amplitude and frequency, meanwhile the phase of the duplicated signals are all highly consistent with each other. © The author(s) 2024.

关键词： Field programmable gate arrays (FPGA)

来源：评论

学校读者我要写书评

暂无评论

Real-Time Hand Gesture Detection Based on YOLOv5s 41

Real-Time Hand Gesture Detection Based on YOLOv5s

引用

第41届中国控制会议

作者： Guangxiang Li Dequan Li Anni Yang School of Mathematics and Big Data Anhui University of Science and Technology School of Artificial Intelligence Anhui University of Science and Technology

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

Since the breakout of Corona Virus Disease 2019(COVID-19),the global fight against influenza has *** technologies have been developed to support the fast-growing contactless service market,and hence contactless services are rapidly becoming a new growth *** particular,the retail service industry most urgently needs contactless service technology.A representative technical case is the self-checkout machine,which can reduce labor costs and provide customer *** present a solution in this *** propose a hand gesture recognition contactless self-checkout system,which is a hand gesture recognition model based on YOLOv5 *** hand gesture recognition mAP(0.5) value reaches 0.995,the mAP(0.5:0.95) value reaches 0.865,and the F1 score is 0.96,together with the accuracy and recall rate is close to *** with the excellent algorithm YOLOx-s,the FPS value of YOLOv5 s can reach 123(YOLOx-s is 108).In addition,the model can be used to detect recorded static and dynamic hand gestures in *** results show that the YOLOv5 s can effectively recognize hand gestures and realize the contactless checkout process.

关键词： Hand Gestures recognization YOLOv5 Computer Vision Contactless services

来源：评论

学校读者我要写书评

暂无评论

GHVC-Net: Hypervolume Contribution Approximation Based on Graph Neural Network

GHVC-Net: Hypervolume Contribution Approximation Based on Gr...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Wu, Guotong Nan, Yang Shang, Ke Ishibuchi, Hisao Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen518060 China Southern University of Science and Technology Department of Computer Science and Engineering Shenzhen518055 China

ISBN: (纸本)9781665410205

This paper proposes a framework called GHVC-Net that uses the graph neural network (GNN) model to approximate each solution's hypervolume contribution (HVC). GHVC-Net is permutation invariant and can handle solution sets of arbitrary size, similar to the properties of GNN. Compared to HVC-Net (i.e., a machine learning model for HVC approximation), GHVC-Net achieves better accuracy with less training time. GHVC-Net is also compared with traditional approximation methods, such as line-based and point-based methods, to demonstrate its ability to identify the solution with the smallest (largest) HVC. © 2024 IEEE.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Research on Complementarity of LiDAR and YOLOv5 Object Detection and Localization 6

Research on Complementarity of LiDAR and YOLOv5 Object Detec...

引用

6th IEEE International Conference on Power, Intelligent Computing and Systems, ICPICS 2024

作者： Xu, Jiale He, Qiuyan Zhang, Qinghui Li, Junqiu College of Big Data and Intelligent Engineering Southwest Forestry University Kunming650224 China University of Electronic Science and Technology of China Chengdu611731 China

ISBN: (数字)9798350374315

ISBN: (纸本)9798350374315

In view of the existing environment sensing LiDAR and machine vision fusion technology, there are still some problems, such as high calculation requirements, system accuracy affected by the background environment, real-time limitation and so on. In order to achieve a more accurate and efficient object detection and spatial location system than a single sensor, and to integrate the information of LiDAR and vision sensor more accurately and stably, an object detection and spatial location algorithm based on the fusion of LiDAR and machine vision information is proposed in this paper. Firstly, after the joint calibration of LiDAR and camera, the laser point cloud is clustered through DBSCAN and fused with the boundary frame based on YOLOv5 object detection according to the signal intensity information of point cloud data. Then, the laser point cloud based on a specific included Angle and the spatial distance positioning parameters based on YOLOv5 object detection are calculated respectively. Finally, the data is compared to realize the data complementation of LiDAR and machine vision fusion. Self-built data sets, this paper uses the method of laser radar and machine vision fusion and validation test, and to explore the complementary relationship and effect of fusion, experimental results show that the proposed object detection algorithm of object detection range and spatial orientation recognition average recognition accuracy of 92.3%, the longitudinal distance within 1 meter distance average error less than 3.2%. An object detection and spatial location ranging system with high accuracy and small error is constructed, which verifies the effectiveness and accuracy of the algorithm in multi-sensor fusion object detection and location. © 2024 IEEE.

关键词： Machine vision

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：