检索结果-内蒙古大学图书馆

Forecasting stock market volatility using social media sentiment analysis

Neural Computing and Applications 2025年第17期37卷 10771-10794页

作者： Saravanos, Christina Kanavos, Andreas Department of Computer Engineering and Informatics University of Patras Patras26504 Greece Department of Informatics Ionian University Corfu49100 Greece

In the era where social media significantly influences public sentiment, platforms such as Twitter have become vital in predicting stock market trends. This paper presents a cutting-edge predictive model that integrates historical stock market data, Twitter sentiment analysis, and an extensive array of tweet-related features. Utilizing advanced regression techniques and deep neural networks, our model forecasts the daily closing prices of the U.S. stock market indices with notable accuracy and demonstrates a strong link between market values, sentiment scores, and social media activities. Our analysis particularly emphasizes the importance of tweet diffusion and the influence of prominent Twitter users in refining prediction accuracy. Contrary to conventional wisdom, we discovered that incorporating a wide range of tweet-derived features significantly improves the model’s performance without leading to sparsity challenges. This study not only questions established paradigms but also underscores the potential of social media analytics in financial market forecasting, with substantial implications for investors, market analysts, and policy makers. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Throughput and Delay Performance of Slotted Aloha in SmartBANs Under Saturation Conditions

IEEE Networking Letters

引用

IEEE Networking Letters 2024年第3期6卷 168-172页

作者： Politis, Anastasios C. Hilas, Constantinos S. Department of Computer Informatics and Telecommunications Engineering Serres621 24 Greece

This letter evaluates the performance of the slotted Aloha protocol defined by the European Telecommunication Standard Institute (ETSI) SmartBAN specification, under saturation conditions. For this purpose, we develop a two-dimensional discrete time Markov chain (DTMC) to model the operational details of the protocol and assess its performance in terms of saturation throughput and average end-to-end delay. The accuracy of the proposed model is validated by means of simulation which reveals a very good match among theoretical and simulation results. The model can be used for protocol performance prediction and optimization purposes. © 2019 IEEE.

关键词： Markov chains

来源：评论

学校读者我要写书评

暂无评论

Machine Learning in e-Commerce: Trends, Applications, and Future Challenges

引用

IEEE Access 2025年 13卷 99048-99067页

作者： Dritsas, Elias Trigka, Maria University of West Attica Department of Informatics and Computer Engineering Egaleo Park Campus Athens12243 Greece

The rapid evolution of e-commerce has been significantly influenced by the integration of machine learning (ML) and data science techniques. The present survey provides a comprehensive overview of how ML methods are applied across various functional domains in e-commerce, including personalized recommendations, dynamic pricing, fraud detection, customer segmentation, and behavioral analysis. We categorize and evaluate a wide range of ML paradigms, namely supervised, unsupervised, reinforcement, and hybrid learning, as well as emerging approaches such as neurosymbolic artificial intelligence (AI), federated learning (FL), and quantumML(QML).Key challenges related to scalability, interpretability, cold-start problems, data sparsity, and privacy are critically analyzed. Additionally, we highlight underexplored areas, such as continual learning (CL) and multi-agent architectures in commerce. The survey incorporates comparative tables, real-world use cases, and a taxonomy of methods to support both academic and industrial perspectives. Ultimately, by analyzing trends and gaps in the literature, we provide a forward-looking research roadmap that bridges ML innovations with the evolving demands of e-commerce ecosystems. © 2013 IEEE.

关键词： Marketplaces

来源：评论

学校读者我要写书评

暂无评论

The Evolution of Generative AI: Trends and Applications

引用

IEEE Access 2025年 13卷 98504-98529页

作者： Trigka, Maria Dritsas, Elias University of West Attica Egaleo Park Campus Department of Informatics and Computer Engineering Athens12243 Greece

Generative artificial intelligence (AI) has revolutionized AI by enabling high-fidelity content creation across text, images, audio, and structured data. This survey explores the core methodologies, advancements, applications, and challenges of generative AI, covering key models such as Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), Diffusion Models, and Transformer-based architectures. These innovations have driven breakthroughs in healthcare, scientific computing, Natural Language Processing (NLP), computer vision, and autonomous systems. Despite its progress, generative AI faces challenges in bias mitigation, interpretability, computational efficiency, and ethical governance, necessitating research into scalable architectures, explainability, and AI safety mechanisms. Integrating Reinforcement Learning (RL), multi-modal learning, and self-supervised techniques enhances controllability and adaptability in generative models. Additionally, as AI reshapes industrial automation, digital media, and scientific discovery, its societal and economic implications demand robust policy frameworks. This survey provides a comprehensive analysis of generative AI's current state and future directions, highlighting innovations in efficient generative modelling, AI-driven scientific reasoning, adversarial robustness, and ethical deployment. By consolidating theoretical insights and real-world applications, it offers a structured foundation for researchers, industry professionals, and policymakers to navigate the evolving landscape of generative AI. © 2013 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Development of a machine-learning-based method for early fault detection in photovoltaic systems

引用

Journal of engineering and Applied Science 2023年第1期70卷 27页

作者： Voutsinas, Stylianos Karolidis, Dimitrios Voyiatzis, Ioannis Samarakou, Maria Department of Informatics and Computer Engineering University of West Attica Athens Greece

In the process of the decarbonization of energy production, the use of photovoltaic systems (PVS) is an increasing trend. In order to optimize the power generation, the fault detection and identification in PVS is significant. The purpose of this work is the study and implementation of such an algorithm, for the detection as many as faults arising on the DC side of a photovoltaic system. A machine learning technique was chosen. The dataset used to train the algorithm was based on a year’s worth of irradiance and temperature data, as well as data from the PV cell used. The method uses logistic regression with cross validation as a new approach to detect and identify faults in PVS. It is applied to smart PV arrays, that can transmit voltage and current measurements from each PV cell of the array individually. The results are satisfactory since the algorithm can detect the majority of faults that occur on the DC side of a photovoltaic (open-circuit fault, short-circuit fault, mismatch faults). The accuracy of the algorithm (97.11%) is comparable to other methods presented by the literature. Moreover, the computational cost of the proposed method is significantly lower than the methods presented in the literature. In summary, the performance of the implemented algorithm is considered particularly satisfactory and can be easily applied to PVS. © 2023, The Author(s).

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

An effective intrusion detection system based on feature selection and regularized long short-term memory classifier

引用

Multimedia Tools and Applications 2025年 1-24页

作者： Chikkalwar, Sudha Rani Garapati, Yugandhar Department of Computer science and Engineering GITAM Deemed to be University Rudraram Hyderabad India Department of Computer science and Informatics Mahatma Gandhi University Telangana Nalgonda India

In recent days, the expansion of Internet of Things (IoT) and the quick advancement of computer system applications contribute to the current phenomenon of data growth. The field of intrusion detection has expanded considerably as a result of the need to protect networks from various attacks. Previously, security has been provided by combining conventional intrusion monitoring systems with a high number of security measures. However, the ability of combined systems to analyze enormous amounts of data has been restricted. Therefore, in this research, open-source and freely available datasets known as NSL-KDD and CSE-CIC-IDS2018 datasets are analyzed. Those datasets contain a class column that splits the attack detection into two classes (normal and anomalous) and four categories (DoS, probe, R2L, and U2R). Then, feature encoding and scaling are used for pre-processing. After that, a SelectPercentile-based feature selection algorithm is suggested to identify the numerous various intrusion attacks and choose the best features. Later, the Regularized Long Short-Term Memory (LSTM) is employed to classify the data. The result analysis verified that Regularized LSTM surpasses the existing models in the NSL-KDD based on accuracy (99.95%), false alarm rate (0.05%), attack detection rate (99.82%), and f1-measure (99.94%). While in the CSE-CIC-IDS2018 dataset, the Regularized LSTM method achieves better accuracy (99.33%), false alarm rate (0.07%), attack detection rate (99.96), and f1-measure (99.95%). © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Wireless Sensor Networks: From Fundamentals and Applications to Innovations and Future Trends

引用

IEEE Access 2025年 13卷 96365-96399页

作者： Trigka, Maria Dritsas, Elias University of West Attica Department of Informatics and Computer Engineering Egaleo Park Campus Athens122 43 Greece

Wireless Sensor Networks (WSNs) have transformed data acquisition and processing across diverse domains such as environmental monitoring, healthcare, and industrial automation. Recent advancements in areas such as federated intelligence, blockchain-based trust mechanisms, and quantum sensing are reshaping the landscape of WSNs. Yet, their full implications for next-generation applications demand deeper exploration. This survey is driven by the need to synthesize these emerging directions through a forward-looking and interdisciplinary lens, highlighting their transformative potential in evolving WSNs into intelligent, autonomous systems. It examines key solutions in energy harvesting, adaptive communication, and edge intelligence, which address persistent constraints like limited energy availability, dynamic topologies, and scalability. Moreover, it highlights the integration of WSNs with disruptive technologies such as the Internet of Things (IoT), artificial intelligence (AI), and digital twins, enabling intelligent, autonomous systems capable of real-time analytics. Critical energy optimization, security, and data management challenges are explored alongside cutting-edge solutions like federated learning (FL), quantum sensing, and blockchain-based trust mechanisms. Future research directions emphasize bio-inspired architectures, hybrid communication in extreme environments, and efficient multi-modal data fusion. This survey serves as a foundational reference for researchers and practitioners seeking to design resilient, intelligent WSN-based systems for the next generation of applications. © 2013 IEEE.

关键词： Data reduction

来源：评论

学校读者我要写书评

暂无评论

Efficient Energy Disaggregation Using DBSCAN: A Novel Approach for Enhanced Energy Management 13th

Efficient Energy Disaggregation Using DBSCAN: A Novel Approa...

引用

13th Mining Humanistic Data Workshop, MHDW 2024, 9th Workshop on 5G-Putting Intelligence to the Network Edge, 5G-PINE 2024 and 1st Workshop on AI in Applications for Achieving the Green Deal Targets, AI4GD 2024 held as parallel events of the IFIP WG 12.5 International Workshops on Artificial Intelligence Applications and Innovations, AIAI 2024

作者： Economopoulou, Emmanouela-Electra Vonitsanos, Gerasimos Mylonas, Phivos Kanavos, Andreas Computer Engineering and Informatics Department University of Patras Patras Greece Department of Informatics and Computer Engineering University of West Attica Athens Greece Department of Informatics Ionian University Corfu Greece

ISBN: (纸本)9783031632266

In the rapidly evolving technology landscape, smart homes are becoming increasingly common, driven by the demand for integrated management of information and services. However, despite advancements, managing electricity consumption efficiently remains a significant challenge, primarily due to the lack of detailed usage data and the complexity of predicting device behavior. This study addresses these challenges by utilizing the DinRail Cerberus meter for granular data collection on household electricity use and applying the DBSCAN clustering algorithm for unsupervised learning. Our research aims to develop a forecasting system that accurately discerns the operational status of household devices-active or inactive-based on energy consumption patterns. This innovative approach promises to revolutionize energy management in smart homes, offering detailed insights into device usage that facilitate more informed decisions for efficient electricity consumption. © IFIP International Federation for Information Processing 2024.

关键词： Energy management

来源：评论

学校读者我要写书评

暂无评论

Enhancing Aviation Efficiency Through Big Data and Machine Learning for Flight Delay Prediction 4th

Enhancing Aviation Efficiency Through Big Data and Machine ...

引用

4th International Conference on Novel and Intelligent Digital Systems, NiDS 2024

作者： Vonitsanos, Gerasimos Gounaridis, Ioannis Kanavos, Andreas Mylonas, Phivos Computer Engineering and Informatics Department University of Patras Patras Greece Department of Informatics Ionian University Corfu Greece Department of Informatics and Computer Engineering University of West Attica Athens Greece

ISBN: (纸本)9783031733437

Flight delays pose significant challenges to the aviation industry, leading to increased operational costs and passenger dissatisfaction. This paper explores the use of machine learning (ML) and big data analytics to enhance the accuracy and efficiency of flight delay predictions. Utilizing data from the Federal Aviation Administration (FAA) covering the period from 2018 to 2022, we analyze critical factors influencing delays and develop predictive models employing techniques such as Random Forest, Gradient Boosting Machines, Decision Trees, and k-Nearest Neighbors. Our analysis demonstrates that these ML techniques significantly outperform traditional models, improving the accuracy of delay predictions and thereby supporting airlines in optimizing operational efficiency and enhancing passenger satisfaction. The paper also discusses the practical implementation of these findings in real-time airline operations and outlines future research directions to further improve predictive accuracy. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Evaluating Machine Learning Techniques for Enhanced Prediction of Building Energy Consumption 9

Evaluating Machine Learning Techniques for Enhanced Predicti...

引用

9th South-East Europe Design Automation, computer engineering, computer Networks and Social Media Conference, SEEDA-CECNSM 2024

作者： Vonitsanos, Gerasimos Kanavos, Andreas Mylonas, Phivos University of Patras Computer Engineering and Informatics Department Patras Greece Ionian University Department of Informatics Corfu Greece University of West Attica Department of Informatics and Computer Engineering Athens Greece

ISBN: (纸本)9798350342482

Accurate prediction of energy usage is crucial for optimizing resource allocation, enhancing energy efficiency, and reducing environmental impact, pivotal for sustainable development. This study examines electricity consumption in three Cornell University buildings, utilizing advanced machine learning techniques to tackle the challenges of sustainable energy management effectively. We specifically evaluated the performance of Support Vector Machine (SVM), Random Forest, Decision Tree, and K-Nearest Neighbors (KNN) in forecasting electricity usage. Our findings reveal that SVM consistently outperforms the other models across various performance metrics, including accuracy and efficiency. These results provide vital insights into the efficacy of these algorithms in predicting energy consumption, thereby supporting strategic energy management decisions in educational institutions and potentially other similar settings. © 2024 IEEE.

关键词： Sustainable building

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：