检索结果-内蒙古大学图书馆

XGCN:a library for large-scale graph neural network recommendations

Frontiers of computer science 2024年第3期18卷 247-249页

作者： Xiran SONG Hong HUANG Jianxun LIAN Hai JIN National Engineering Research Center for Big Data Technology and System Services Computing Technology and System LabCluster and Grid Computing LabSchool of Computer Science and TechnologyHuazhong University of Science and TechnologyWuhan 430074China Microsoft Research Asia Beijing 100080China

1 Introduction Graph Neural Networks(GNNs)have gained widespread adoption in recommendation systems,and nowadays there is a pressing need to effectively manage large-scale graph data[1].When it comes to large graphs,GNNs may encounter the scalability issue stemming from their multi-layer messagepassing ***,scaling GNNs has emerged as a crucial research area in recent years,with numerous scaling strategies being proposed.

关键词： scaling gained operations

来源：评论

学校读者我要写书评

暂无评论

An Innovative Approach for Detection of Motorcyclists Helmet Non-Compliance using Edge computing 5

An Innovative Approach for Detection of Motorcyclists Helmet...

引用

5th International Conference on Data Intelligence and Cognitive Informatics, ICDICI 2024

作者： Jithendra, Mulpuri Krishna, M. Madhava Charan Sai, M. Harsha, P. Sri Supriya, M. Amrita School of Computing Dept. of Computer Science and Engineering Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (纸本)9798350389609

Using innovative technologies and creative strategies can help reduce fatalities and lessen the financial impact of accidents. This suggests a significant improvement in global traffic law enforcement and road safety advocacy. The study proposed in this paper uses an innovative approach to address the challenge of motorcyclists neglecting traffic regulations through a deep learning-driven framework. Using advanced pre-trained object detection models, the system accurately identifies helmetless motorcyclists and number plates. In addition, the proposed model does the Optical Character Recognition (OCR) to extract the license plate information from the identified motorcyclists who are not wearing helmets. The architecture was tested using an NVIDIA Jetson as the edge device, delivering quick performance, efficiently managing large data volumes, and optimizing resource use. Extensive testing and in-depth analysis show that the proposed approach is robust and highly beneficial. The proposed system integrates OCR algorithms, edge computing capabilities, and advanced deep learning techniques to offer a scalable solution for assisting law enforcement entities and enhancing traffic safety initiatives. © 2024 IEEE.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Time-Series-Based Stock Market Analysis using Machine Learning 15

Time-Series-Based Stock Market Analysis using Machine Learni...

引用

15th International Conference on computing Communication and Networking Technologies, ICCCNT 2024

作者： Amruth, S. Jaya Nigelesh, T.M. Shruthik, V. Sai Reddy, Valluru Sateesh Venugopalan, Manju Amrita Vishwa Vidyapeetham Amrita School of Computing Dept. of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798350370249

The stock market is a very complex and volatile entity. Predicting it is a very challenging task but the emergence of Machine Learning and the availability of large-scale stock market data provide an opportunity to leverage data-driven approaches for improving prediction accuracy. In this paper, a framework for stock market analysis using up-to-date financial data extracted from the Yahoo Finance library and machine learning models is proposed. This paper's approach is to capture the complex relationships and patterns in the data by using machine learning techniques. Specifically, machine learning algorithms like XGBoost, Gradient Boost, and Random Forests are employed. By analyzing diverse data sources from various companies, the aim is to extract valuable insights for stock prediction. The experiments involve processing contemporary stock data, implementing machine learning models on it, and comparing the outcomes. The predominant motives are to enhance stock prediction accuracy, manage inherent risks, enable real-time decision-making, and maximize profitability by providing valuable insights for investors that help optimize their portfolios. However, the results are real-time and the optimal algorithm for a particular stock is decided on a dynamic basis due to the stock market being volatile and unpredictable in nature. © 2024 IEEE.

关键词： Decision Tree Gradient Boost Machine Learning Random Forest Stock Market Analysis XGBoost Yahoo Finance

来源：评论

学校读者我要写书评

暂无评论

Prompt Palate: A Prompt-Based Restaurant Recommendation System with Scheduling and Chatbot Integration 8

Prompt Palate: A Prompt-Based Restaurant Recommendation Syst...

引用

8th IEEE International Conference on Computational System and Information Technology for Sustainable Solutions, CSITSS 2024

作者： Vidula, N.A. Babu, T.Hemanth Kiran, V. Gopi Balakrishnan, Roshni M. Bhaskaran, Sreebha Amrita School of Computing Amrita Vishwa Vidyapeetham Dept.of Computer Science & Engineering Bengaluru India

ISBN: (纸本)9798331505462

In today's fast-paced world, individuals often find themselves craving a culinary adventure but struggle to decide on the perfect restaurant to satisfy their desires. Traditional restaurant search platforms typically offer rigid filters such as location, cuisine, and price range, limiting users' ability to explore options tailored to their specific cravings or preferences. We introduce Prompt Palate, a novel prompt-based restaurant recommendation system that uses natural language processing (NLP) techniques to address this challenge. Prompt Palate allows users to input their culinary desires via human-like prompts whose details and preferences are extracted through NLP algorithms. Moreover, the integration of a chatbot interface enhances user engagement and provides additional information about recommended restaurants. Users can engage in interactive conversations with the chatbot to learn more about menu offerings, ambiance, and special features of preferred restaurants. Additionally, the chatbot facilitates food ordering directly from the recommended restaurants, streamlining the dining or home delivery experience. To ensure a seamless and uncrowded dining experience, we incorporate intelligent scheduling algorithms for efficient table management, staff allocation, and delivery coordination. These algorithms optimize resource allocation, minimize wait times, and prevent overcrowding, thereby enhancing customer satisfaction. © 2024 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

A Boosting-Based Machine Learning Approach for Intrusion Detection in Vehicle Address Verification Systems 5

A Boosting-Based Machine Learning Approach for Intrusion Det...

引用

5th International Conference on Electronics and Sustainable Communication Systems, ICESC 2024

作者： Naga, Siva Gayatri Venkata Ammanamanchi, Datta Sai Chadha, Angela Raj Reddy, Kethamreddy Karthikeya Shanmugam, Udhaya Kumar Amrita School of Computing Amrita Vishwa Vidyapeetham Dept. of Computer Science & Engineering Chennai India

ISBN: (纸本)9798350379945

Autonomous vehicles (AVs) are susceptible to cyberattacks due to their reliance on vehicle-to-everything (V2X) communication. This research proposes a Gradient Boosting-based intrusion detection system to safeguard AV networks. By employing advanced machine learning techniques, including Adaboost, Decision Trees, and Random Forests, the system effectively identifies various cyberattacks. Evaluated on standard datasets, the proposed IDS achieves a remarkable accuracy of 99.49%, surpassing existing models. The combination of feature selection and ensemble learning enhances the system's detection rate while maintaining computational efficiency. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing Data Migration Efficiency for Cloud-Based Databases 6th

Enhancing Data Migration Efficiency for Cloud-Based Database...

引用

6th International Conference on Recent Innovations in computing, ICRIC 2023

作者： Daripelli, Manish Rithani, M. Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Tamil Nadu Chennai India

ISBN: (纸本)9789819778614

Effective data migration is essential for businesses to handle massive datasets between cloud databases in the digital age. In order to address the challenges of data migration, this paper presents a thorough and novel method that focuses on effective data migration with security and integrity in four main steps: data processing, uploading to MongoDB, moving data from MongoDB to AWS S3, and moving data from AWS S3 to Snowflake. The data processing stage ensures data integrity, maximizes resource utilization, and supports a range of data formats. It has methods for handling image data that involve compressing and converting it to a CSV format. Afterwards, the transition from MongoDB to AWS S3 guarantees safe data transport by employing IAM policies and the Amazon Web Services (AWS) SDK for Python (Boto3) to preserve data security and scalability. AWS IAM policies control the last migration step to cloud-based data warehousing platform Snowflake in order to improve data security and integrity. Impressive results are obtained with 7 Mbps throughput and a 57.1 ms latency reduction which makes this approach more efficient and secure form some of the existing methods. This method sets a new benchmark for data migration, enabling businesses to quickly adjust to changing data environments and enabling more dependable and quick cross-database transfers. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Data integrity

来源：评论

学校读者我要写书评

暂无评论

Exploring and Predicting Top streamed spotify songs using Lazy learners

Exploring and Predicting Top streamed spotify songs using La...

引用

2024 Asia Pacific Conference on Innovation in Technology, APCIT 2024

作者： Asmitha, M. Kavitha, C.R. Amrita Vishwa Vidyapeetham Amrita School of Computing Dept. of Computer Science and Engineering Bengaluru India

ISBN: (数字)9798350361537

ISBN: (纸本)9798350361537

Stream prediction holds great significance for analysing and recommending the song based on the viewer or most played category. Prediction involves instances that can be recommended based on playlist presence high viewer count, adding complexity to the prediction process. The paper mainly helps to unfold the convergence of machine learning, song prediction, and multimedia analysis within the context of the Spotify dataset. This paper focuses on the application of data exploration, ML and Data science techniques to unravel patterns and insights within the Spotify dataset. Leveraging the rich and diverse data provided by Spotify, encompassing user interactions, audio features, and textual information, the study aims to develop robust classification models for music-related tasks. The project explores various machine learning algorithms like lazy classifier and lazy regressor a combination of seventy algorithms trained into multiple classes to discern intricate patterns in the data and enhance the accuracy of genre classification. In addition to EDA analysis of features, the project also endeavors to predict Artist Highlights, Playlist Dynamics, Regional Variations. The anticipated outcomes include the development of accurate models. Evaluation metrics such as precision, R-Score, R Sq Score, recall, mean-R Sq Score are employed to assess the performance of the models in handling prediction data associated with diverse music genres. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Real-time Urban Parking Space Management using Fog computing 1

Real-time Urban Parking Space Management using Fog Computing

引用

1st International Conference on Cognitive, Green and Ubiquitous computing, IC-CGU 2024

作者： Thangam, S. Snehitha, Munaga Sai Kumar, Kotyada Mohan Kiran Reddy, Koti Leela Sai Praneeth Deepthi, C. Amrita School of Computing Dept. of Computer Science and Engineering Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (纸本)9798350361780

The rise in the number of private automobiles, especially in populated zones leads to many issues like traffic congestion, waste of time and fuel as they spend more time finding a parking spot. Vehicles release toxic gases, as they are ideal in traffic thus the number of pollutants gets increased. Smart parking systems help with these problems. They reduce traffic congestion which leads to less environmental pollution and saves fuel. A cloud-based parking system has been proposed, but it requires high latency and high bandwidth. Fog computing gives us a better solution than cloud computing. With the help of fog computing smart car parking systems, the proposed model can clear all these issues. The driver gets information about the parking spots with minimal latency and energy consumption. By guiding the drivers where the parking spot is free it uses the space optimally. It also reduces the need to expand the parking area since land acquisition in urban areas is tough. A latency-reducing approach was developed using iFogSim and metrics like latency, total energy consumed, and total time taken were considered for a comparative analysis between the fog environment and the cloud environment. The developed fog-based system demonstrated better results than the cloud-based approach. © 2024 IEEE.

关键词： Fog computing

来源：评论

学校读者我要写书评

暂无评论

Pattern Recognition for Identifying a String Within a Text File Using Finite Automata 15

Pattern Recognition for Identifying a String Within a Text F...

引用

15th International Conference on computing Communication and Networking Technologies, ICCCNT 2024

作者： Prajana, Mamidi Rajkumar, Harsha Sannidhi, Akepati Sai Vidyasri, Kammari Panda, Niharika Amrita Vishwa Vidyapeetham Amrita School of Computing Dept. of Computer Science Engineering Bengaluru India

ISBN: (纸本)9798350370249

Pattern recognition is crucial across diverse domains, including retrieval of information, data mining, and bioinformatics. Numerous algorithms exist for string matching, and finite state machines (FSM) provide a robust framework for implementing these pattern recognition algorithms. This study introduces an efficient pattern recognition application utilizing a finite state machine, using the SSFSM library in Python. The SSFSM library facilitates the creation and management of FSMs, offering methods to define states and transitions efficiently. An algorithm is proposed, that identifies strings within a text file with the help of the SSFSM library in Python. It employs deterministic finite automata (DFA) for pattern recognition, scanning text files word by word and then character by character, and recording execution times to evaluate efficiency. The algorithm's effectiveness is demonstrated through a comparative analysis of its performance for the number of characters and size of the input file. Results demonstrate the algorithm's effectiveness in accurately identifying and quantifying specific string occurrences within large text files. It also shows that the search time of a string increases with an increase in the frequency of the character, size of the text file, and length of the search key, underscoring its potential for broader applications. Further, our results also show that there is approximately a 20% increase in search time when the search key length increases by 1. Future enhancements may include extended functionalities, support for additional document types, and improved complexity management. © 2024 IEEE.

关键词： Finite automata

来源：评论

学校读者我要写书评

暂无评论

A Systematic Survey on Diabetic Retinopathy and Deep Learning Approaches 2

A Systematic Survey on Diabetic Retinopathy and Deep Learnin...

引用

2nd International Conference on Self Sustainable Artificial Intelligence Systems, ICSSAS 2024

作者： Sadhana, U. Beena, B.M. Amrita Vishwa Vidyapeetham Amrita School of computing Dept. of Computer Science & Engineering Bengaluru India

ISBN: (纸本)9798350368413

Diabetic retinopathy (DR) is a significant threat to the vision of human beings;it is affected by patients and needs to be detected in the early stage to reduce the severity of vision. In recent research, the DR become interdisciplinary medical and computer science research. In this paper, based on the importance of this research, our research systematically reviewed the deep learning algorithms for automated DR detection, highlighting datasets, model implementations, and challenges. The study explored the Deep Learning approaches in DR management by assessing performance and clinical importance. In this review, our research also included the comparative study of various model outputs based on the different datasets, needful preprocessing techniques, and model refinement for accurate DR detection. We also discussed the integration of advanced models into clinical practice to improve DR management and reduce vision loss associated with diabetes. DR remains a significant global health problem, and advanced detection methods are used to mitigate vision loss. This literature review systematically examines the role of DL algorithms such as EfficientNet and MobileNet in enhancing DR detection. Through rigorous analysis of IDRiD (Indian Diabetic Retinopathy Image Dataset) datasets, model implementations, and challenges, the study synthesizes current research to explain critical directions and understandings. Our research concluded that integrating Deep learning into clinical research holds a guarantee for improving the accuracy and efficiency of DR detection, thereby improving patient results and quality of life. This review delivers valuable wisdom for researchers and clinicians aiming to advance DR detection methodologies and implement innovative solutions for this condition's early diagnosis and management. © 2024 IEEE.

关键词： Clinical research

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：