检索结果-内蒙古大学图书馆

3rd International Conference on Machine Learning and Data engineering, ICMLDE 2024

作者： Kumari, Annu Edlaa, Damodar Reddy Department of Computer Science and Engineering National Institute of Technology Goa Goa Cuncolim403703 India

The rapid evolution of deep learning in the current era has prompted extensive exploration in the realm of Brain-computer Interface (BCI). Despite the substantial progress made in enhancing the accuracy of BCI systems employing Motor Imagery (MI) through deep learning techniques, especially when compared to certain traditional algorithms, there remains a substantial challenge in clearly interpreting these sophisticated models. This holds particular importance due to the labor-intensive and costly nature of the data collection process. This challenge is addressed in our manuscript through the introduction of a unique methodology to improve the effectiveness and precision of BCI systems using EEG data. The outlined methodology comprises five fundamental steps. The signal pre-processing begins with band-pass filtering, succeeded by Empirical Mode Decomposition (EMD) to capture underlying oscillatory components. Subsequently, correlation calculation is conducted for all the Intrinsic Mode Functions (IMFs) with the original signal. The IMF exhibiting the highest correlation factor is selected for feature extraction, and classification is executed using the adaptive Graph Convolutional Network (GCN). It plays a pivotal role in feature extraction, leveraging its capabilities to capture intricate patterns within EEG signals. Implemented in Python, this model, achieved an accuracy level of 91.3%, underscoring its significance in augmenting the overall functionality of BCI systems based on EEG data. © 2025 The Author(s).

关键词： Empirical mode decomposition

来源：评论

学校读者我要写书评

暂无评论

Successful Integration of IoT and Blockchain Technologies Using Several Machine Learning Algorithms 5

Successful Integration of IoT and Blockchain Technologies Us...

引用

5th IEEE International Conference on Computing, Power, and Communication Technologies, IC2PCT 2024

作者： Calarany, C. Indumathy, M. Senthilraja, P. Suganya, D. Gopi Vardhan, V. Rahul, Yelisetty Shanmukh Tagore Government Arts and Science College Department of Computer Science Puducherry India Rajiv Gandhi College of Engineering & Technology Department of Information Technology Kirumampakkam India K.S.Rangasamy College of Technology Department of Computer Science Tamil Nadu India Rajiv Gandhi College of Engineering & Technology Department of Computer Science Kirumampakkam India Data Science and Business Systems Srm Institute of Science and Technology Kattankulathur India

ISBN: (纸本)9798350383522

Internet of Things (IoT) produces massive amounts of data that need to be processed and saved securely. The strong features of Blockchain makes it as a best candidate for storing the data received from IoT sensors. However, there is a need of concern to take care of the challenges associated with both IoT and blockchain paradigms. Firstly, the enormous amount of data should be effortlessly handled by Blockchain network, without adding much complexity. Secondly, the heterogeneous nature of data that are received from various IoT Sensors should be stored within the blockchain. In this paper, a conceptual framework based on Multimodal Machine Learning (MML) algorithm is proposed as an interface between the IoT sensors and Blockchain networks. The MML algorithm is used to understand the different modalities of the data and provides a way to manage the heterogeneity in that data. The MML algorithms provides an intelligent data processing that allows for real-time data storage in the blockchain networks. The conceptual frame assumes that audio, video, and text are the three types of data as received from IoT. These inputs are processed in simplistic Recurrent neural network RNN that acts as MML and then passed to Blockchain networks for secure storage. Simulation of the above said conceptual framework is performed assuming all other relevant requirements of IoT, MML and Blockchain are taken into account with default parameters and values. The results obtained provide a promising note for future scope in this direction. © 2024 IEEE.

关键词： IoT MML RNN

来源：评论

学校读者我要写书评

暂无评论

Poetry in Pixels: Prompt Tuning for Poem Image Generation via Diffusion Models 31

Poetry in Pixels: Prompt Tuning for Poem Image Generation vi...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Jamil, Sofia Reddy, Bollampalli Areen Kumar, Raghvendra Saha, Sriparna Joseph, K.J. Goswami, Koustava Department of Computer Science & Engineering Indian Institute of Technology Patna India Adobe Research India

ISBN: (纸本)9798891761964

The task of text-to-image generation has encountered significant challenges when applied to literary works, especially poetry. Poems are a distinct form of literature, with meanings that frequently transcend beyond the literal words. To address this shortcoming, we propose a PoemToPixel framework designed to generate images that visually represent the inherent meanings of poems. Our approach incorporates the concept of prompt tuning in our image generation framework to ensure that the resulting images closely align with the poetic content. In addition, we propose the PoeKey algorithm, which extracts three key elements in the form of emotions, visual elements, and themes from poems to form instructions which are subsequently provided to a diffusion model for generating corresponding images. Furthermore, to expand the diversity of the poetry dataset across different genres and ages, we introduce MiniPo, a novel multimodal dataset comprising 1001 children's poems and images. Leveraging this dataset alongside PoemSum, we conducted both quantitative and qualitative evaluations of image generation using our PoemToPixel framework. This paper demonstrates the effectiveness of our approach and offers a fresh perspective on generating images from literary sources. The code and dataset used in this work are publicly available. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

A Real-Time Road Crash Prediction Model by Hybridizing Multiple Learning Classifiers 4th

A Real-Time Road Crash Prediction Model by Hybridizing Multi...

引用

4th International Conference on Computing, Communications, and Cyber-Security, IC4S 2022

作者： Arun, G. Anuguraju, K. Sangeetha, A. Babu, K. Trichy India Department of Computer Science and Engineering Gnanamani College of Technology Pachal Namakkal India Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai India

ISBN: (纸本)9789819914784

The majority of critical injuries caused by road accidents occur in developing nations, where they are a major global concern. Due to these traffic incidents, many individuals have lost loved ones. Consequently, a system that has the potential to save lives is needed. The approach identifies key contributing factors for accidents or establishes a connection between accidents and different causes of accidents. This study suggests a system for predicting accidents that can assist in analyzing potential safety concerns and foretelling whether an event will happen or not. In order to identify key qualities and forecast accident risk in single- and multiple-crashes, this research provides a hybrid feature selection-based classification-based machine learning (ML) method. It took into account classifiers like the random forest, k-nearest neighbor, and XGBoost. A few evaluation criteria, including precision, root mean square error, accuracy, recalls, and receivers operating output characteristics, were used to assess these classifiers. When the accuracy of the ML algorithms was compared, the accuracy of the proposed model attained as 99.78%. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

GraphXAI: a survey of graph neural networks (GNNs) for explainable AI (XAI)

引用

Neural Computing and Applications 2025年第17期37卷 10949-11000页

作者： Nandan, Mauparna Mitra, Soma De, Debashis Department of Computer Applications Techno Main Salt Lake West Bengal Kolkata India Department of Computational Science Brainware University West Bengal Barasat India Department of Computer Science and Engineering Maulana Abul Kalam Azad University of Technology West Bengal West Bengal Kolkata India

Graphs find wide applications in numerous domains, ranging from simulating physical systems to learning molecular fingerprints, predicting protein interfaces, diagnosing diseases, etc. These applications encompass simulations in non-Euclidean space, in which a graph serves as an ideal representation, and are also an indispensable means of illustrating the connections and interdependencies among its various constituents. Graph neural networks (GNNs) are specific types of neural networks that are specifically built to handle data possessing a graph structure. They are highly effective at capturing intricate relationships among different entities. Nonetheless, their "black-box" characteristics pose difficulties in transparency, trust, and interpretability, especially in critical sectors like heath care, banking, and autonomous systems. Explainable artificial intelligence (XAI) has emerged to clarify these obscure decision-making processes, thus enhancing trust and accountability in AI systems. This survey paper delves into the intricate interplay between GNNs and XAI, including an exhaustive taxonomy of the various explainability methods designed for graph-structured data. It classifies the existing explainability methods into post hoc and self-interpretable models. The paper analyzes their practical applications in diversified fields, highlighting the significance of transparent GNNs in essential sectors such as fraud detection, drug development, and network security. The survey also delineates evaluation parameters for assessing explainability along with addressing persistent issues in scalability and fairness. The paper concludes by addressing prospective advancements in the subject, including the creation of innovative XAI methodologies tailored for GNN architectures, integration with federated learning, and utilization of these models in interdisciplinary fields. This study bridges the gap between GNNs and XAI, providing an essential resource for researchers and p

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Cyber Attack Detection System in University Private Cloud Using Machine Learning

Cyber Attack Detection System in University Private Cloud Us...

引用

2023 International Conference on Self Sustainable Artificial Intelligence Systems, ICSSAS 2023

作者： Chethan, M.S. Channakrishnaraju Rajeswari, R. Selvam, M. Shridevi Institute of Engineering and Technology Department of Computer Science and Engineering Karnataka Tumkur India Sri Siddhartha Institute of Technology Department of Computer Science and Engineering Karnataka Tumkur India Alliance University Department of Computer Science and Engineering Karnataka Bangalore India

ISBN: (纸本)9798350300857

As cloud computing adoption in colleges continues to rise, the security of private cloud systems has become a paramount concern. Data breaches resulting from cyber attacks can inflict severe damage to a university's revenue and reputation. This research proposes a novel machine learning-based cyber threat detection system tailored to the university's private cloud environment. The system's main objective is to continuously monitor the cloud infrastructure and employ advanced machine learning algorithms to analyze network traffic, identify and prevent unusual activities that may indicate potential cyber-attacks. Here, the challenges posed on two sides of known possible threats and attack worldwide come across, and administrative defaults leads to security hole. By leveraging the power of machine learning, this innovative system aims to enhance the university's cyber defence capabilities. It considers the dynamic and evolving nature of cyber threats, enabling real-time detection and proactive measures against malicious activities. The integration of cutting-edge machine learning models and feature extraction techniques empowers the system to identify patterns of anomalous behaviour, even in the face of sophisticated attacks. Key components of the proposed system include network traffic analysis, anomaly detection and threat intelligence integration. Through the analysis of network packets and access logs, the system can effectively detect signs of unauthorized access, data exhilaration, and other cyber threats. Additionally, threat intelligence feeds provide the system with up-to-date information on emerging threats, enabling quick responses to potential risks. Moreover, the system's implementation adheres to privacy and data protection regulations, ensuring secure handling of sensitive information within the private cloud environment. Regular updates and adaptive learning capabilities enable the system to evolve with changing cyber threats, ensuring continued robustn

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Q-Learning-Based Approach to Detect Tumor in Human -Brain

Q-Learning-Based Approach to Detect Tumor in Human -Brain

引用

2024 International Conference on Communication, computer sciences and engineering, IC3SE 2024

作者： Mani, Chandra Aeron, Anurag Rajput, Kapil Kumar, Sandeep Jain, Arpit Manwal, Manika Raj Kumar Goel Institute Of Technology Department Of Computer Science & Engineering U.P. Ghaziabad India Meerut Institute Of Engineering And Technology Department Of Computer Science And Engineering U.P. Meerut India Graphic Era Hill University Computer Science And Engineering Uttarakhand Dehradun248002 India Tula's Institute Department Of Computer Science And Engineering Uttarakhand Dehradun India Koneru Lakshmaiah Education Foundation Department Of Computer Science And Engineering AP Vaddeswaram India Graphic Era Deemed To Be University Computer Science And Engineering Uttarakhand Dehradun India

ISBN: (纸本)9798350366846

The brain's own aberrant and unregulated cell division is what causes brain tumors. In the tumor growth exceeds 50%, the patient will not be able to recover. As a result, rapid and precise brain tumor identification is required. Brain image processing research started to focus heavily on brain tumors. Hence, a novel methodology of image processing is proposed for fast and accurate brain tumor detection. Reinforcement-based Q-learning technique is proposed for classifying brain tumor MRI images. Every MRI image is divided into 8 × 8 blocks and the features like mean, variance, skew ness, kurtosis, entropy, contrast, correlation, energy, and homogeneity are extracted. Ground truth matrix is generated by using discrete wavelet transform with which MRI image divided blocks are classified as 0 and 1 where 1 is the tumored block and 0 is the normal block. Reward mechanism in Q-learning is used to further classify the tumored and normal brain images. The efficiency of the classification is compared with other existing algorithms like Dense Net, ensemble methods, and SVM and CNN in accordance with performance parameters like accuracy, sensitivity, and specificity. The accuracy of the proposed algorithm is 97%, sensitivity is 96% and specificity is 97% proving that the proposed algorithm is highly efficient in classifying brain tumor MRI images. © 2024 IEEE.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

computer Vision and Convolutional Neural Networks (CNNs) for Advanced Face Detection and Recognition 2

Computer Vision and Convolutional Neural Networks (CNNs) for...

引用

2nd International Conference on Machine Learning and Autonomous Systems, ICMLAS 2025

作者： Ajay, A. Javith Ahamed, M. St. Joseph's Institute Of Technology Department of Computer Science Engineering Chennai India

ISBN: (数字)9798331505745

ISBN: (纸本)9798331505745

Modern security and law enforcement involve face detection and recognition technologies, which are crucial instruments in criminal identification. It is a comprehensive system integrating advanced face detection algorithms and robust recognition techniques with a view of providing accurate and efficient identification of an individual relating to a criminal activity. It uses the latest CNNs technology for automatic extraction of facial features, matching, and classification, with an advanced algorithm for the matching that compares all identified faces to a database of criminal individuals. Its large processing databases improve the identification speed rate with fewer false positives or negatives. Other issues the paper deals with are lighting conditions, facial expression, etc. © 2025 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

computer Vision and Control Algorithm for Sustainable and Precision Agriculture

引用

SN computer science 2025年第4期6卷 1-13页

作者： Adebayo, Segun Aworinde, Halleluyah O. Ojerinde, Stephen A. Oyedoja, Kayode O. Adeniyi, Abidemi Emmanuel Olanloye, Odunayo Dauda Osueke, Christian O. Ogundokun, Roseline Oluwaseun Das, Abhishek Mechatronics Engineering Bowen University Iwo Nigeria Computer Science Bowen University Iwo Nigeria Computer Science Western Illinois University IL United States Technology Education Emmanuel Alayande University of Education Oyo Nigeria Department of Software Engineering Kaunas University of Technology Kaunas Lithuania Department of Computer Science Landmark University Omu-Aran Kwara State Nigeria Department of Computer Science and Engineering Aliah University Kolkata India Tshwane University of Technology Pretoria South Africa Durban University of Technology Durban South Africa Chitkara University Institute of Engineering and Technology Chitkara University Punjab India

The world population continues to increase with a large proportion of this increase happening in Africa just as predicated by Food and Agriculture Organization of the United Nation. To meet the demand of feeding this increasing population, innovative farming practices are being developed to increase the yield of the major stable food supply. This research developed a computer vision and control algorithm on a quadcopter for farm surveillance, crop protection from the invasion of pest, early disease detection as well as crop yield estimation. A process of developing a model capable of detecting and classifying these pests in the environment was presented in this study. The process involves selecting, preprocess and transforming the sensed data from a vision sensor to train classification algorithm and to detect and track object of interest in the environment. This study addresses a notable research gap in computer vision, control systems, big data analytics, and robotics, specifically within the context of smart farming. It primarily utilizes conventional learning algorithms, including Artificial Neural Networks and Support Vector Machines, typically found in widely-used commercial software. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Algorithm computer vision Machine learning Quadcopter UAV

来源：评论

学校读者我要写书评

暂无评论

Detection of pre-cluster nano-tendency through multi-viewpoints cosine-based similarity approach

引用

Nanotechnology for Environmental engineering 2022年第1期7卷 259-268页

作者： Basha, M. Suleman Mouleeswaran, S.K. Prasad, K. Rajendra Department of Computer Science and Engineering Dayananda Sagar University Bangalore India Department of Computer Science and Engineering Rajeev Gandhi Memorial College of Engineering and Technology Nandyal India

Pre-clusters assessment is a significant problem in data clustering. It found that visual cluster tendency assessment (VAT) is majorly focused on addressing the problem of pre-clusters assessment. This visual technique initially derives the similarity features of data objects using either cosine or Euclidean distance metrics. Cosine is considering both magnitudes and direction of the vectors;thus, it greatly succeeded in data clustering applications. Only a single viewpoint (i.e., origin) is used in the cosine metric. Finding the similarity features using multiple viewpoints is more accurate than a single viewpoint cosine metric. This paper presents the multi-viewpoints cosine-based similarity VAT (MVS-VAT) which considers the multi-viewpoints for an effective assessment of nano-pre-clusters (or nano-cluster tendency). Clustering accuracy (CA) and normalized mutual information (NMI) are taken for measuring the performance of the existing and proposed methods. It is proved that the efficiency of the proposed MVS-VAT is improved from 20 to 40% compared to VAT and cVAT concerning the parameters of CA and NMI. Therefore, the quality of data clusters is obtained through the proposed technique MVS-VAT. Experimental is conducted on several benchmarked datasets for illustration of an empirical study of the existing and proposed techniques. © 2022, The Author(s), under exclusive licence to Springer Nature Switzerland AG.

关键词： Cluster analysis

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：