检索结果-内蒙古大学图书馆

kProtoClust:Towards Adaptive k-Prototype Clustering without Known k

computers, Materials & Continua 2025年第3期82卷 4949-4976页

作者： Yuan Ping Huina Li Chun Guo Bin Hao School of Information Engineering Xuchang UniversityXuchang461000China Henan Province Engineering Technology Research Center of Big Data Security andApplications Xuchang461000China College of Computer Science and Technology Guizhou UniversityGuiyang550025China Here Data Technology Shenzhen518000China

Towards optimal k-prototype discovery,k-means-like algorithms give us inspirations of central samples collection,yet the unstable seed samples selection,the hypothesis of a circle-like pattern,and the unknown K are still challenges,particularly for non-predetermined data *** propose an adaptive k-prototype clustering method(kProtoClust)which launches cluster exploration with a sketchy division of K clusters and finds evidence for splitting and *** behalf of a group of data samples,support vectors and outliers from the perspective of support vector data description are not the appropriate candidates for prototypes,while inner samples become the first candidates for instability reduction of *** from the representation of samples in traditional,we extend sample selection by encouraging fictitious samples to emphasize the representativeness of *** get out of the circle-like pattern limitation,we introduce a convex decomposition-based strategy of one-cluster-multiple-prototypes in which convex hulls of varying sizes are prototypes,and accurate connection analysis makes the support of arbitrary cluster shapes *** by geometry,the three presented strategies make kProtoClust bypassing the K dependence well with the global and local position relationship analysis for data *** results on twelve datasets of irregular cluster shape or high dimension suggest that kProtoClust handles arbitrary cluster shapes with prominent accuracy even without the prior knowledge K.

关键词： Prototype finding convex hull support vector data description geometrical information

来源：评论

学校读者我要写书评

暂无评论

Low Birth Weight Prediction Using Machine Learning 5th

Low Birth Weight Prediction Using Machine Learning

引用

5th International Conference on Innovative Product Design and Intelligent Manufacturing Systems, ICIPDIMS 2023

作者： Harika, G. Venkata Lakshmi, T. Santhosh Prudhvi Raj, P. Prasan, U.D. Jayanthi Rao, M. Department of Computer Science & Engineering Aditya Institute of Technology and Management Tekkali532201 India

ISBN: (纸本)9789819767311

In recent decades, machine learning has its increased problem solving methodologies and applications in various fields of business, marketing, education and medical diagnostics. Among all the ML techniques, some have been employed in the fields of medical sciences to predict various health conditions which require complete analysis of the patient’s health by considering some parameters which impact the health condition of an individual. Low birth weight of infants has been a constant recurring problem which has its visible impacts only after the birth of the baby. It acts as an indicator of sickness in newborn baby’s weight. This can be avoided using suitable ML techniques which predict required values from the information passed to the trained model. The prediction algorithm is implemented using random forest regression algorithms. We are aiming to develop a website which considers the health indicators as inputs from the user and provides them with estimated value by the trained ML model. The measures of metrics used here are the number of gestational days, age of mother, height, weight, plurality, and smoking status of the mother. All of these parameters can be used to estimate the accurate value of the weight of the baby within the web application. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： CSS HTML LBW prediction Machine learning Python

来源：评论

学校读者我要写书评

暂无评论

Correction to: Transfer learning and self-distillation for automated detection of schizophrenia using single-channel EEG and scalogram images (Physical and engineering sciences in Medicine, (2024), 10.1007/s13246-024-01420-1)

引用

Physical and engineering sciences in Medicine 2025年第1期48卷 19-20页

作者： Mostafavi, Mohammadreza Ko, Seok-Bum Shokouhi, Shahriar Baradaran Ayatollahi, Ahmad Department of Electrical Engineering Electronic Engineering Iran University of Science and Technology Tehran Iran Division of Biomedical Engineering Department of Electrical and Computer Engineering University of Saskatchewan Saskatoon Canada

In this article the legend of Fig. 6 was presented without a reference. The legend of Fig. 6 has been changed from "The general framework for knowledge distillation involving a teacher-student relationship" to "The generic teacher-student framework for knowledge distillation [33]." In Sharma et al.’s research, two results obtained from the "KNN with ten-fold cross-validation" and "KNN with leave-one-subject-out cross-validation (LOSOCV)" methods are reported as the final results. Unfortunately, we mistakenly attributed the accuracy of one method. While we correctly listed the name of the method used, which is KNN with ten-fold cross-validation, we inadvertently entered the precision of the KNN with leave-one-subject-out cross-validation (LOSOCV) method. Therefore, we updated our results table (Table 7). The correct Table 7 is shown below. (Table presented.) Comparison of the performance of the proposed method with existing methods Study Method Dataset Accuracy Sensitivity Specificity Shim et al. [38] SVM 34 SZ-34 healthy 88.24% – – Siuly et al. [7] EBT – EMD (Empirical Mode Decomposition) 49 SZ-32 healthy 89.59% 0.8976 0.8932 Phang et al. [39] Multi-domain connectome CNN 45 SZ-39 healthy 91.69% 0.9111 0.925 Jahmunah et al. [40] Non-linear features, decision-tree, neural network, SVM 14 SZ and 14 healthy 92.90% 0.9345 0.9224 Akbari et al. [41] Phase space dynamic & graphical features 14 SZ and 14 healthy 94.8% 0.943 0.952 Shalbaf et al. [20] ResNet-18-SVM 14 SZ and 14 healthy 94.84% 0.980 0.9239 Bagherzadeh et al. [42] DenseNet121 model 14 SZ and 14 healthy 96.26% 0.9548 0.9702 Sharma et al. [43] KNN with ten-fold 14 SZ and 14 healthy 99.21% 0.9942 0.9905 LOSOCV 97.20% 0.9649 0.9806 Proposed method CWT—Self distillation- VGG16 14 SZ and 14 healthy 97.81 0.970 0.986 The original article has been corrected. © Australasian College of Physical Scientists and Engineers in Medicine 2024.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Robustness of Smart Contracts Through Declarations 7th

Enhancing Robustness of Smart Contracts Through Declaration...

引用

7th International Conference on Blockchain, ICBC 2024, Held as Part of the Services Conference Federation, SCF 2024

作者： Shyamasundar, R.K. Borse, Snehal Ummair, Mohammad Department of Computer Science and Engineering Indian Institute of Technology Bombay Mumbai400076 India

ISBN: (纸本)9783031770944

Run-time monitoring has been one of the widely used techniques to realize robust smart contracts. In this paper, we show how we can abstract aspects of run-time monitoring through declarations of programming languages so that a majority of run-time monitoring code required for preserving the intended semantics of Solidity and thereby overcome vulnerabilities can be instrumented automatically at compile time itself. For illustrative purposes, we use Solidity and call the augmented Solidity language SolidityD. SolidityDsyntactically is identical to Solidity except for declarations. The declarations preserve the intended semantics of various constructs and thus overcome a large number of vulnerabilities. We illustrate how a vast variety of vulnerabilities encountered in programming smart contracts in Solidity are overcome through declarations. Declarations envisaged in SolidityDare similar to that used in classic concurrent programming languages and are easily visualizable by the programmer. Declarations lead to automatic introduction of run-time code at compilation time needed to satisfy various executional constraints needed to satisfy the semantic requirements of declarations. Further, we demonstrate that SolidityD, can be automatically transformed to pure Solidity: thus, enabling an effective debugging at source level rather than at the byte or EVM level. Another important outcome of using SolidityD, is that it can be effectively used for asserting coarse-grained properties of declarations. Such a feature leads to gaining trust in the usage of smart contracts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Program debugging

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Classification of Hyper Spectral Images using Feature Extraction and Fuzzy Logic

Unsupervised Classification of Hyper Spectral Images using F...

引用

2024 International Conference on Optimization Computing and Wireless Communication, ICOCWC 2024

作者： Kannagi, A. Chaudhary, Chetan Seth, Jyoti Department of Computer Science and Information Technology Bangalore India Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh India Vivekananda Global University Department of Computer Science & Engineering Jaipur India

ISBN: (纸本)9798350383348

Hyperspectral pictures are complicated information items with high spectral resolution, making their categorization and analysis timeingesting and challenging. Traditional strategies for classifying hyperspectral pix may be unreliable and gradually attributable to the presence of diverse noise resources and a high range of pixels. This paper proposes a new unsupervised classification approach for hyperspectral pictures using function extraction and fuzzy common sense. The method starts by first using feature extraction techniques on the hyperspectral pictures to lessen the dimensionality of the facts. Numerous characteristic extraction algorithms, including primary thing analysis (PCA) and impartial component evaluation (ICA), are tested to determine which function extraction algorithms yield satisfactory effects. The reduced function area is then used as an entry for the fuzzy category system. The bushy common sense device is used to classify the hyperspectral pix into distinctive classes according to the extracted capabilities. Experimental results display that the proposed method achieves proper effects for the category venture with classification accuracy accomplishing as high as 79%. The proposed technique demonstrates advanced performance over conventional category strategies in terms of each accuracy and speed. Hyperspectral pics (HSI) offer valuable statistics approximately the environment and the functions gift inside it. But, the sheer quantity of facts present in HSI makes guide evaluation of those photos a time-eating and exhausting project. As such, there is a growing demand for robust and reliable automated techniques to analyze HSI. In this context, unsupervised tactics for classifying HSI have gained interest due to their ability to examine facts without requiring manually categorized education facts. Fuzzy logic is one method being explored for unsupervised HSI type due to its capability to assign more than one label to pixels of the image and its r

关键词： analysis classification hyperspectral primarily satisfactory

来源：评论

学校读者我要写书评

暂无评论

Optimizing Monitoring Utility of Unmanned Aerial Vehicles Considering Adverse Effects

引用

IEEE Transactions on Mobile Computing 2025年第7期24卷 5996-6013页

作者： Zhang, Haihan Dai, Haipeng Qiu, Yu Yu, Enze Zhou, Ruiben Wang, Weijun Wang, Jingwu Chen, Guihai Nanjing University State Key Laboratory for Novel Software Technology Nanjing210023 China South China University of Technology School of Computer Science and Engineering Guangzhou China Beijing China

For Unmanned Aerial Vehicles (UAVs) monitoring tasks, capturing high quality images of target objects is important for subsequent recognition. Concerning the problem, many prior works study placement/trajectory planning for UAVs to maximize the quality of captured images. However, all of them overlook a fact that UAV monitoring may cause a huge risk/annoyance on living objects. In this paper, we investigate the novel problem of oPtimizing unmanned aErial vehicles plAcement by Considering both monitoring utility and adverse Effects (PEACE). We propose an approach to solve PEACE, which is proved to be NP-hard. Overall, our approach achieves a 1-1/e- approximation ratio. First, we approximate the original problem of PEACE as a classical problem of Monotone Submodular function Maximization under a Uniform Matroid constraint (MSMUM) with a controlled gap. Then, for MSMUM, we propose a combination of algorithms achieving a 1-1/e approximation and O(n\log n) time complexity considering the correlation among the UAV monitoring strategies. The proposed algorithms outperform existing algorithms for MSMUM through theoretical analysis and experimental results. Extensive simulations and field experiments demonstrate the effectiveness of our approach, achieving performance gains of 9.0% to 1434.5% compared to existing methods. © 2002-2012 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Chronic Disease Prediction Using Different ML Techniques 11

Chronic Disease Prediction Using Different ML Techniques

引用

11th International Conference on Reliability, Infocom Technologies and Optimization, ICRITO 2024

作者： Singh, Shikha Badotra, Sumit Shelke, Nitin Arvind School of Computer Science Engineering and Technology Bennett University Greater Noida India

ISBN: (纸本)9798350350357

These days, emerging technologies like artificial intelligence (AI)is crucial to the developments in the medical field. Machine learning will become very helpful for experts in making their decisions fast, accurate and early disease predictions. Early disease prediction is very much needed for an easy cure and reducing the mortality rate for that specific disease. Machine Learning has various techniques for the prediction of different diseases, like some chronic diseases such as cardiovascular disease, Diabetes, Dementia, Cancer and Kidney failure. In this research study five different machine learning techniques, i.e. Decision Tree (DT), Logistic Regression (LR), Random Forest (RF), Gradient Boosting Model(GB) and eXtreme Gradient Boost (XGB) are analyzed for the mentioned diseases. For this analysis some parameters like Precision, Recall, f-Score, accuracy and Time complexity are considered. In the experimental setup the overall dataset is divided into two parts which is 70:30 as training and testing dataset. So the results obtained from the given experimental model show that the GB classifier will gives the most optimized results for heart disease prediction, LR gives a good result for diabetes prediction and XGBoost is best for dementia prediction while the time complexity is optimized for LR and DT. © 2024 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

An Experimental Analysis of Various Deep Learning Architectures for the Classification of Cognitive Stimuli based EEG Signals 15

An Experimental Analysis of Various Deep Learning Architectu...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Sarkar, Prashant Srinivasan Kanaga, E. Grace Mary Bhuvaneshwari, M. Mathew, Joel Stephen, Caleb Karunya Institute of Technology and Sciences Department of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798331300579

The human brain functions through electrical signals. By measuring these signals, one can monitor brain activity and gain insights into the brain function of the subject. An electroencephalogram (EEG) allows one to monitor brain activity by having the subject wear an array of sensors on their head. This process is frequently used to diagnose medical conditions such as epilepsy. In recent years, there have been efforts to use EEG signals in concert with deep learning to create a brain computer interface (BCI). Such a device would enable the wearer to communicate to a system via brain signals. While such a system would not be so advanced as to enable the translation of complex thoughts, it would enable a user to command a machine to perform a small number of functions. The objective of this paper was to develop and optimize recurrent neural network architectures for use with a brain computer interface. Using EEG data collected from subjects, a variety of neural network models were created to learn from the data. The models that were used were simple recurrent neural networks (RNN), long short-term memory (LSTM), and gated recurrent units (GRU). This paper proposes a novel approach to EEG signal classification, demonstrating the capabilities of recurrent networks which are seldom explored for this purpose. This study produced promising results for recurrent models, obtaining a 91% accuracy with the 4-layer LSTM architecture. This presents a solid foundation for the argument that LSTM and similar architectures are feasible for BCI applications. © Grenze Scientific Society, 2024.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

WENSA: WhatsApp Emotion and Sentiment Analyzer Using LSTM and BI-LSTM 3rd

WENSA: WhatsApp Emotion and Sentiment Analyzer Using LSTM an...

引用

3rd International Conference on Intelligent Systems and Sustainable Computing, ICISSC 2023

作者： Kumar, N. A. Abhinav Sanjay, R.V. Vani, V. Karthik, N. Department of Computer Science and Engineering National Institute of Technology Puducherry Karaikal609609 India

ISBN: (纸本)9789819783540

WhatsApp is a widely popular messaging app used for communication. This study focuses on analyzing WhatsApp group chats, aiming to assess member activity and engagement. It seeks to determine the most active day, the highest number of messages sent on that day, the most active user, active group members, the total user count, individual message contributions, and the most frequently used phrase. Additionally, it conducts daily, monthly, and annual analyses of the top 15 members’ messages. Furthermore, the study aims to predict emotions within the text, such as joy, sadness, and anger, using a deep learning algorithm known as LSTM. The model's performance is evaluated using a dataset from Kaggle comprising 16,000 English language records with various emotions, with validation accuracy as the key metric for emotion analysis. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Active learning

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Multi-view Pedestrian Detection 24

Unsupervised Multi-view Pedestrian Detection

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Liu, Mengyin Zhu, Chao Ren, Shiqi Yin, Xu-Cheng School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China

ISBN: (纸本)9798400706868

With the prosperity of the intelligent surveillance, multiple cameras have been applied to localize pedestrians more accurately. However, previous methods rely on laborious annotations of pedestrians in every frame and camera view. Therefore, we propose in this paper an Unsupervised Multi-view Pedestrian Detection approach (UMPD) to learn an annotation-free detector via vision-language models and 2D-3D cross-modal mapping: 1) Firstly, Semantic-aware Iterative Segmentation (SIS) is proposed to extract unsupervised representations of multi-view images, which are converted into 2D masks as pseudo labels, via our proposed iterative PCA and zero-shot semantic classes from vision-language models;2) Secondly, we propose Geometry-aware Volume-based Detector (GVD) to end-to-end encode multi-view 2D images into a 3D volume to predict voxel-wise density and color via 2D-to-3D geometric projection, trained by 3D-to-2D rendering losses with SIS pseudo labels;3) Thirdly, for better detection results, i.e., the 3D density projected on Birds-Eye-View, we propose Vertical-aware BEV Regularization (VBR) to constrain pedestrians to be vertical like the natural poses. Extensive experiments on popular multi-view pedestrian detection benchmarks Wildtrack, Terrace, and MultiviewX, show that our proposed UMPD, as the first fully-unsupervised method to our best knowledge, performs competitively to the previous state-of-the-art supervised methods. Code is available at https://***/lmy98129/UMPD. © 2024 ACM.

关键词： Visual languages

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：