Procedia Computer Science最新文献

Strategies for Improved Out-of-Distribution Detection in Drone vs. Bird Classification 无人机与鸟类分类中改进的非分布检测策略

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.03.341

Ami Pandat , Punna Rajasekhar , Gopika Vinod , Rohit Shukla

The use of drones has expanded significantly across various applications over the past decade, leading to increased surveillance-related challenges. These challenges raised the necessity of developing Anti-Drone systems. One of the critical requirements for an effective Anti-Drone system is the ability to accurately distinguish drones from birds in the sky. While deep learning-based classification techniques have been employed for this task, they often suffer from high false positive rates. To address this challenge, Out-of-Distribution (OOD) detection is essential for enhancing the reliability and robustness of drone surveillance systems, particularly in differentiating drones from birds. This paper explores several techniques to improve OOD detection performance, focusing on Energy-Based Models (EBM) and Variational Autoencoders (VAE). We evaluate four loss functions within the EBM framework: Mean Squared Error (MSE) Loss, Mean Squared Error with OOD Penalty, Contrastive Loss, and Binary Cross-Entropy with Energy Regularization. Our results demonstrate that the Mean Squared Error with OOD Penalty function achieves the best performance, with an AUC of 0.9, providing clearer separation between in-distribution (drones) and out-of-distribution (birds) samples. However, the VAE approach did not yield significant results for the binary classification task. Future work could explore hybrid approaches to further enhance OOD detection in such applications.

在过去十年中，无人机的使用在各种应用中得到了显着扩展，导致与监视相关的挑战增加。这些挑战提高了开发反无人机系统的必要性。有效的反无人机系统的关键要求之一是能够准确区分无人机和天空中的鸟类。虽然基于深度学习的分类技术已被用于这项任务，但它们往往存在高误报率。为了应对这一挑战，out - distribution （OOD）检测对于提高无人机监视系统的可靠性和鲁棒性至关重要，特别是在区分无人机和鸟类方面。本文探讨了几种提高OOD检测性能的技术，重点是基于能量的模型（EBM）和变分自编码器（VAE）。我们在EBM框架中评估了四种损失函数：均方误差（MSE）损失，均方误差与OOD惩罚，对比损失和能量正则化的二元交叉熵。我们的研究结果表明，带OOD惩罚函数的均方误差达到了最好的性能，AUC为0.9，提供了更清晰的分布内（无人机）和分布外（鸟类）样本之间的分离。然而，对于二元分类任务，VAE方法并没有产生显著的结果。未来的工作可以探索混合方法，以进一步增强此类应用中的OOD检测。

{"title":"Strategies for Improved Out-of-Distribution Detection in Drone vs. Bird Classification","authors":"Ami Pandat , Punna Rajasekhar , Gopika Vinod , Rohit Shukla","doi":"10.1016/j.procs.2025.03.341","DOIUrl":"10.1016/j.procs.2025.03.341","url":null,"abstract":"<div><div>The use of drones has expanded significantly across various applications over the past decade, leading to increased surveillance-related challenges. These challenges raised the necessity of developing Anti-Drone systems. One of the critical requirements for an effective Anti-Drone system is the ability to accurately distinguish drones from birds in the sky. While deep learning-based classification techniques have been employed for this task, they often suffer from high false positive rates. To address this challenge, Out-of-Distribution (OOD) detection is essential for enhancing the reliability and robustness of drone surveillance systems, particularly in differentiating drones from birds. This paper explores several techniques to improve OOD detection performance, focusing on Energy-Based Models (EBM) and Variational Autoencoders (VAE). We evaluate four loss functions within the EBM framework: Mean Squared Error (MSE) Loss, Mean Squared Error with OOD Penalty, Contrastive Loss, and Binary Cross-Entropy with Energy Regularization. Our results demonstrate that the Mean Squared Error with OOD Penalty function achieves the best performance, with an AUC of 0.9, providing clearer separation between in-distribution (drones) and out-of-distribution (birds) samples. However, the VAE approach did not yield significant results for the binary classification task. Future work could explore hybrid approaches to further enhance OOD detection in such applications.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"259 ","pages":"Pages 398-407"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143936370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Transformative Tech and Social Dynamics: Redefining Customer Engagement in Industry 5.0 变革技术和社会动态：重新定义工业5.0中的客户参与

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.03.325

Bhawana Kothari , Ambica Prakash Mani , V M Tripathi

In the era of Industry 5.0, this paper offers a thorough assessment of how transforming technologies have changed consumer involvement. We start by looking at the change from conventional, linear methods to client participation towards current, technologically driven techniques. The emphasis is on how digitization is drastically changing consumer expectations and the policies companies have to follow to quickly fit these developments. Examining the ability of modern technologies such as artificial intelligence (AI), machine learning, the Internet of Things (IoT), and blockchain to influence modern customer engagement methods takes up a good amount of this research. We look at how machine learning and artificial intelligence improve predictive analytics, therefore allowing companies to proactively satisfy consumer wants and customize experiences. We also explore the key part IoT plays in creating a consistent and flawless consumer experience at many points of contact. This talk also addresses the creative use of virtual assistants and chatbots, assessing their efficiency in providing real-time consumer help. These AI-driven solutions are evaluated for their capacity to provide a degree of customizing like that of human interactions, hence improving client connections. This paper explores several aspects of customer commitment in the computerized era using approaches, innovations, and best practices that enable companies to create further associations, enhance customer interactions, and propel supportable development in an era marked by mechanical disturbance.

在工业5.0时代，本文对转型技术如何改变消费者参与进行了全面评估。我们首先着眼于从传统的线性方法到客户参与到当前的技术驱动技术的变化。重点是数字化如何彻底改变消费者的期望，以及公司必须遵循的政策以迅速适应这些发展。研究人工智能（AI）、机器学习、物联网（IoT）和区块链等现代技术对现代客户参与方法的影响，占据了这项研究的很大一部分。我们着眼于机器学习和人工智能如何改进预测分析，从而使公司能够主动满足消费者的需求并定制体验。我们还探讨了物联网在许多接触点创造一致和完美的消费者体验方面所起的关键作用。这次演讲还讨论了虚拟助手和聊天机器人的创造性使用，评估了它们在提供实时消费者帮助方面的效率。评估这些人工智能驱动的解决方案的能力，以提供一定程度的定制，如人类互动，从而改善客户关系。本文探讨了计算机时代客户承诺的几个方面，使用方法，创新和最佳实践，使公司能够建立进一步的联系，加强客户互动，并在以机械干扰为标志的时代推动可支持的发展。

{"title":"Transformative Tech and Social Dynamics: Redefining Customer Engagement in Industry 5.0","authors":"Bhawana Kothari , Ambica Prakash Mani , V M Tripathi","doi":"10.1016/j.procs.2025.03.325","DOIUrl":"10.1016/j.procs.2025.03.325","url":null,"abstract":"<div><div>In the era of Industry 5.0, this paper offers a thorough assessment of how transforming technologies have changed consumer involvement. We start by looking at the change from conventional, linear methods to client participation towards current, technologically driven techniques. The emphasis is on how digitization is drastically changing consumer expectations and the policies companies have to follow to quickly fit these developments. Examining the ability of modern technologies such as artificial intelligence (AI), machine learning, the Internet of Things (IoT), and blockchain to influence modern customer engagement methods takes up a good amount of this research. We look at how machine learning and artificial intelligence improve predictive analytics, therefore allowing companies to proactively satisfy consumer wants and customize experiences. We also explore the key part IoT plays in creating a consistent and flawless consumer experience at many points of contact. This talk also addresses the creative use of virtual assistants and chatbots, assessing their efficiency in providing real-time consumer help. These AI-driven solutions are evaluated for their capacity to provide a degree of customizing like that of human interactions, hence improving client connections. This paper explores several aspects of customer commitment in the computerized era using approaches, innovations, and best practices that enable companies to create further associations, enhance customer interactions, and propel supportable development in an era marked by mechanical disturbance.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"259 ","pages":"Pages 240-249"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143936372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Intelligent Analysis Method of E-commerce Data Based on Various Machine Learning Algorithms 基于多种机器学习算法的电子商务数据智能分析方法

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.04.008

Bo Yang

Under the current rapid development of the e-commerce industry, most e-commerce companies are pursuing to enhance the clicks of products and its conversion rate to buy. And there are many machine learning algorithms for the intelligent analysis of e-commerce data, among which, the most widely used is the recurrent neural network (RNN) and collaborative filtering algorithm. Based on the use of multiple machine learning algorithms, this paper compares the differences in the clicks of products and the purchase conversion rates between the RNN algorithm and the collaborative filtering algorithm. The RNN algorithm can make full use of the behavior sequence time dependence and context information and the collaborative filtering algorithm is based on the similarities between user and product. The evaluation results are as follows: the products clicked by the RNN algorithm are between 18,000 and 25,000, which is significantly higher than the products clicked by the collaborative filtering algorithm. In order to improve user purchase decisions and overall sales efficiency, e-commerce operators can select the RNN algorithm to fully understand the user’s interests and needs, and provide accurate personalized product recommendations.

在当前电子商务行业快速发展的情况下，大多数电子商务企业都在追求提高产品的点击量及其购买转化率。而针对电子商务数据智能分析的机器学习算法有很多，其中应用最广泛的是递归神经网络（RNN）和协同过滤算法。本文在使用多种机器学习算法的基础上，比较了RNN算法与协同过滤算法在产品点击量和购买转化率上的差异。RNN算法充分利用了行为序列的时间依赖性和上下文信息，协同过滤算法基于用户与产品之间的相似性。评价结果如下：RNN算法的产品点击量在18000 ~ 25000之间，显著高于协同过滤算法的产品点击量。为了提高用户的购买决策和整体销售效率，电商运营商可以选择RNN算法，充分了解用户的兴趣和需求，并提供准确的个性化产品推荐。

{"title":"Intelligent Analysis Method of E-commerce Data Based on Various Machine Learning Algorithms","authors":"Bo Yang","doi":"10.1016/j.procs.2025.04.008","DOIUrl":"10.1016/j.procs.2025.04.008","url":null,"abstract":"<div><div>Under the current rapid development of the e-commerce industry, most e-commerce companies are pursuing to enhance the clicks of products and its conversion rate to buy. And there are many machine learning algorithms for the intelligent analysis of e-commerce data, among which, the most widely used is the recurrent neural network (RNN) and collaborative filtering algorithm. Based on the use of multiple machine learning algorithms, this paper compares the differences in the clicks of products and the purchase conversion rates between the RNN algorithm and the collaborative filtering algorithm. The RNN algorithm can make full use of the behavior sequence time dependence and context information and the collaborative filtering algorithm is based on the similarities between user and product. The evaluation results are as follows: the products clicked by the RNN algorithm are between 18,000 and 25,000, which is significantly higher than the products clicked by the collaborative filtering algorithm. In order to improve user purchase decisions and overall sales efficiency, e-commerce operators can select the RNN algorithm to fully understand the user’s interests and needs, and provide accurate personalized product recommendations.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"259 ","pages":"Pages 591-597"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143936924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Advanced Video-Based Deep Learning Framework for Comprehensive Detection, Diagnosis, and Classification of Dermatological Conditions in Real-Time Datasets 先进的基于视频的深度学习框架，用于在实时数据集中全面检测、诊断和分类皮肤病

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.03.344

Syed Thouheed Ahmed , Amogh S Guthur , Pratyush Kumar Rai , Pranava Swaroop N

The advanced acne detection model showcased in this project utilizes deep learning methods to accurately classify skin conditions, including blackheads, dark areas, blemishes, and creases. It employs a YOLOv5 format annotation scheme to analyze spatial and temporal information from video sequences, resulting in exceptional performance in detecting seven distinct classes. The model’s resilient performance indicates high accuracy, with a mean Average Precision (mAP) of about 0.85-0.9 at an IoU threshold of 0.5. It also demonstrates generalization and robustness with an mAP of 0.5-0.55 across IoU thresholds from 0.5 to 0.95, making it suitable for real-world dermatological assessments. The proposed method enables early detection and more effective treatment by monitoring skin conditions over time, significantly impacting dermatological image analysis. The goal is to improve patient outcomes and provide personalized skincare recommendations using deep learning techniques, benefiting clinicians and researchers in analyzing and categorizing skin conditions. Additionally, incorporating additional data sources like clinical images or medical histories can enhance the model’s diagnostic capabilities and accuracy. Expanding the dataset will enhance the model’s generalizability and robustness for new skin conditions

本项目展示的先进的痤疮检测模型利用深度学习方法对皮肤状况进行准确分类，包括黑头、暗区、瑕疵、折痕等。它采用了YOLOv5格式的注释方案来分析视频序列的时空信息，从而在检测七种不同的类别时获得了出色的性能。该模型的弹性性能表明了较高的精度，在IoU阈值为0.5时，平均平均精度（mAP）约为0.85-0.9。它还显示了泛化和稳健性，在IoU阈值从0.5到0.95之间的mAP为0.5-0.55，使其适用于现实世界的皮肤病学评估。所提出的方法可以通过长期监测皮肤状况来实现早期检测和更有效的治疗，显著影响皮肤学图像分析。目标是使用深度学习技术改善患者的治疗效果并提供个性化的护肤建议，使临床医生和研究人员能够分析和分类皮肤状况。此外，结合临床图像或病史等其他数据源可以增强模型的诊断能力和准确性。扩展数据集将增强模型的泛化性和对新皮肤状况的鲁棒性

{"title":"Advanced Video-Based Deep Learning Framework for Comprehensive Detection, Diagnosis, and Classification of Dermatological Conditions in Real-Time Datasets","authors":"Syed Thouheed Ahmed , Amogh S Guthur , Pratyush Kumar Rai , Pranava Swaroop N","doi":"10.1016/j.procs.2025.03.344","DOIUrl":"10.1016/j.procs.2025.03.344","url":null,"abstract":"<div><div>The advanced acne detection model showcased in this project utilizes deep learning methods to accurately classify skin conditions, including blackheads, dark areas, blemishes, and creases. It employs a YOLOv5 format annotation scheme to analyze spatial and temporal information from video sequences, resulting in exceptional performance in detecting seven distinct classes. The model’s resilient performance indicates high accuracy, with a mean Average Precision (mAP) of about 0.85-0.9 at an IoU threshold of 0.5. It also demonstrates generalization and robustness with an mAP of 0.5-0.55 across IoU thresholds from 0.5 to 0.95, making it suitable for real-world dermatological assessments. The proposed method enables early detection and more effective treatment by monitoring skin conditions over time, significantly impacting dermatological image analysis. The goal is to improve patient outcomes and provide personalized skincare recommendations using deep learning techniques, benefiting clinicians and researchers in analyzing and categorizing skin conditions. Additionally, incorporating additional data sources like clinical images or medical histories can enhance the model’s diagnostic capabilities and accuracy. Expanding the dataset will enhance the model’s generalizability and robustness for new skin conditions</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"259 ","pages":"Pages 424-432"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143936476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The Application of Neural Network Algorithm in Computer Mathematical Modeling 神经网络算法在计算机数学建模中的应用

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.04.012

Wenying Zhao , Xiaohong Li , Mingjie Shi

The application research of neural network algorithm in computer mathematical modeling has made extensive development. With its strong learning and approximation ability, it has shown great potential and application prospect in many fields. Through literature review and comparative analysis, this study compared the performance of CNN and RNN in the mathematical model of financial risk. The two algorithms have different performances under the same mathematical model. The results of the study showed that the accuracy of risk assessment of CNNS was between 93% and 98%, while the accuracy of RNN was between 89% and 96%, and the performance of CNNS was between 3-6s and RNN was between 4-8s on the assessment time. For the same neural network algorithm model, the two algorithms show different performance in financial risk assessment, because the weight parameter sharing in CNN can significantly reduce the number of parameters in the model, thus reducing the risk of overfitting.

神经网络算法在计算机数学建模中的应用研究得到了广泛的发展。它具有较强的学习能力和近似能力，在许多领域显示出巨大的潜力和应用前景。通过文献综述和对比分析，本研究比较了CNN和RNN在金融风险数学模型中的表现。在相同的数学模型下，两种算法具有不同的性能。研究结果表明，CNNS的风险评估准确率在93% ~ 98%之间，而RNN的准确率在89% ~ 96%之间，在评估时间上，CNNS的表现在3 ~ 6s之间，RNN的表现在4 ~ 8s之间。对于相同的神经网络算法模型，两种算法在金融风险评估中表现出不同的性能，因为CNN中的权重参数共享可以显著减少模型中的参数数量，从而降低过拟合的风险。

引用次数: 0

Review of Segmentation Techniques for Weed Detection in Agricultural Crops 农作物杂草检测的分割技术综述

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.03.307

Akanksha Bodhale , Seema Verma

This study explores the role of deep learning in identifying and managing weeds in agriculture, a critical challenge for enhancing crop productivity. As the global population grows, increasing food production is essential. Weeds significantly hinder crop growth, making accurate identification vital. Deep learning techniques, which analyze elements like color, form, texture, and spectrum, offer promising solutions for distinguishing between crops and weeds. This review examines various segmentation techniques used in weed identification, comparing their effectiveness and potential for practical application. The findings aim to advance weed management strategies, contributing to improved agricultural productivity and the development of automated systems for precise weed detection.

本研究探讨了深度学习在农业杂草识别和管理中的作用，这是提高作物生产力的关键挑战。随着全球人口的增长，增加粮食产量至关重要。杂草严重阻碍作物生长，使准确识别至关重要。深度学习技术可以分析颜色、形状、纹理和光谱等元素，为区分作物和杂草提供了有希望的解决方案。本文综述了用于杂草识别的各种分割技术，比较了它们的有效性和实际应用潜力。研究结果旨在推进杂草管理策略，有助于提高农业生产力和开发精确检测杂草的自动化系统。

引用次数: 0

Advanced AI-Powered Intrusion Detection Systems in Cybersecurity Protocols for Network Protection 网络安全协议中基于ai的入侵检测系统

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.03.315

Hari Mohan Rai , Aditya Pal , Rashidov Akbar Ergash o’g’li , Bobokhonov Akhmadkhon Kholmirzokhon Ugli , Yarmatov Sherzojon Shokirovich

Conventional rule-based network intrusion detection systems (NIDS) find it difficult to remain with the increasing complexity of cyber-attacks. To solve these issues, this study examines the development of NIDS as well as the transformative potential of artificial intelligence (AI). AI-enhanced NIDS can efficiently identify and respond to known and unknown threats in real-time by utilizing machine learning (ML) techniques. The system can differentiate between typical network behavior and abnormalities using both supervised and unsupervised learning techniques, as opposed to depending exclusively on pre-established rules. The accuracy and adaptability of the system are further improved by deep learning (DL) architectures like recurrent neural networks (RNNs) and convolutional neural networks (CNNs). The paper explores the past developments of intrusion detection, comparing rule-based approaches to modern AI-driven systems. It discusses cutting-edge techniques like anomaly detection, ensemble methods, and hybrid models. While Recognizing issues such as adversarial attacks and interpretability, the article underlines the importance of AI-enhanced NIDS in protecting digital infrastructure. This study provides a complete overview, unique insights, and practical advice for cybersecurity experts looking to install and optimize AI-powered intrusion detection solutions.

传统的基于规则的网络入侵检测系统难以适应日益复杂的网络攻击。为了解决这些问题，本研究考察了NIDS的发展以及人工智能（AI）的变革潜力。通过利用机器学习（ML）技术，ai增强的NIDS可以有效地识别和实时响应已知和未知的威胁。该系统可以使用监督和非监督学习技术区分典型的网络行为和异常，而不是完全依赖于预先建立的规则。通过循环神经网络（rnn）和卷积神经网络（cnn）等深度学习（DL）架构，进一步提高了系统的准确性和适应性。本文探讨了入侵检测的过去发展，比较了基于规则的方法和现代人工智能驱动的系统。它讨论了前沿技术，如异常检测、集成方法和混合模型。在认识到对抗性攻击和可解释性等问题的同时，文章强调了人工智能增强的NIDS在保护数字基础设施方面的重要性。本研究为希望安装和优化人工智能入侵检测解决方案的网络安全专家提供了完整的概述、独特的见解和实用的建议。

{"title":"Advanced AI-Powered Intrusion Detection Systems in Cybersecurity Protocols for Network Protection","authors":"Hari Mohan Rai , Aditya Pal , Rashidov Akbar Ergash o’g’li , Bobokhonov Akhmadkhon Kholmirzokhon Ugli , Yarmatov Sherzojon Shokirovich","doi":"10.1016/j.procs.2025.03.315","DOIUrl":"10.1016/j.procs.2025.03.315","url":null,"abstract":"<div><div>Conventional rule-based network intrusion detection systems (NIDS) find it difficult to remain with the increasing complexity of cyber-attacks. To solve these issues, this study examines the development of NIDS as well as the transformative potential of artificial intelligence (AI). AI-enhanced NIDS can efficiently identify and respond to known and unknown threats in real-time by utilizing machine learning (ML) techniques. The system can differentiate between typical network behavior and abnormalities using both supervised and unsupervised learning techniques, as opposed to depending exclusively on pre-established rules. The accuracy and adaptability of the system are further improved by deep learning (DL) architectures like recurrent neural networks (RNNs) and convolutional neural networks (CNNs). The paper explores the past developments of intrusion detection, comparing rule-based approaches to modern AI-driven systems. It discusses cutting-edge techniques like anomaly detection, ensemble methods, and hybrid models. While Recognizing issues such as adversarial attacks and interpretability, the article underlines the importance of AI-enhanced NIDS in protecting digital infrastructure. This study provides a complete overview, unique insights, and practical advice for cybersecurity experts looking to install and optimize AI-powered intrusion detection solutions.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"259 ","pages":"Pages 140-149"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143937852","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fake News Detection in Indian Languages: A Case Study with Hindi Using CNN-LSTM 印度语言中的假新闻检测：以CNN-LSTM为例

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.03.316

Rajeev Kumar Gupta , Vaibhav Sharma , R.K. Pateriya , Vasudev Dehalwar , Punit Gupta

The increasing spread of false information also known as fake news has had a negative impact on the society and the political system hence the need for detection tools. This research work presents a hybrid CNN-LSTM deep learning model for detecting fake news in Hindi, a language that lacks adequate dataset and resources. A new dataset of 6,724 Hindi news articles (2,704 fake and 4,020 real) was collected from the trusted sources which are members of International Fact Checking Network (IFCN). The model uses FastText pretrained embeddings, a Conv1D layer for local feature extraction and LSTM units for sequential feature extraction, and is able to achieve 97% accuracy on the proposed dataset and an F1 score of 89% on CONSTRAINT2021 dataset.

This paper also presents a new dataset for future research and the first work done towards developing a system for detecting fake news in Hindi language. In the future, the work will be continued by trying to apply this approach to other sparse Indian languages and by using transformer-based models to improve results.

虚假信息也被称为假新闻的日益传播对社会和政治制度产生了负面影响，因此需要检测工具。这项研究工作提出了一种混合CNN-LSTM深度学习模型，用于检测缺乏足够数据集和资源的印地语中的假新闻。从国际事实核查网络（IFCN）成员的可信来源收集了6724篇印度语新闻文章（2704篇是假的，4020篇是真的）的新数据集。该模型使用FastText预训练嵌入，Conv1D层用于局部特征提取，LSTM单元用于顺序特征提取，并且能够在提出的数据集上达到97%的准确率，在CONSTRAINT2021数据集上达到89%的F1分数。本文还为未来的研究提供了一个新的数据集，并为开发一个检测印地语假新闻的系统做了第一项工作。在未来，这项工作将继续进行，尝试将这种方法应用于其他稀疏的印度语言，并使用基于转换器的模型来改进结果。

{"title":"Fake News Detection in Indian Languages: A Case Study with Hindi Using CNN-LSTM","authors":"Rajeev Kumar Gupta , Vaibhav Sharma , R.K. Pateriya , Vasudev Dehalwar , Punit Gupta","doi":"10.1016/j.procs.2025.03.316","DOIUrl":"10.1016/j.procs.2025.03.316","url":null,"abstract":"<div><div>The increasing spread of false information also known as fake news has had a negative impact on the society and the political system hence the need for detection tools. This research work presents a hybrid CNN-LSTM deep learning model for detecting fake news in Hindi, a language that lacks adequate dataset and resources. A new dataset of 6,724 Hindi news articles (2,704 fake and 4,020 real) was collected from the trusted sources which are members of International Fact Checking Network (IFCN). The model uses FastText pretrained embeddings, a Conv1D layer for local feature extraction and LSTM units for sequential feature extraction, and is able to achieve 97% accuracy on the proposed dataset and an F1 score of 89% on CONSTRAINT2021 dataset.</div><div>This paper also presents a new dataset for future research and the first work done towards developing a system for detecting fake news in Hindi language. In the future, the work will be continued by trying to apply this approach to other sparse Indian languages and by using transformer-based models to improve results.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"259 ","pages":"Pages 150-160"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143937853","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Strategic Insights into Customer Diversity: Unraveling Purchase Patterns, Income Disparities, and Relationship Dynamics through K-means Clustering for Enhanced Engagement and Loyalty 顾客多样性的战略洞察：通过k均值聚类分析购买模式、收入差异和关系动态，以增强参与和忠诚度

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.03.301

Shraddha Sharma , Rupali Satsangi , Preeti Manani , Priti Sharma , Jyoti Gupta

This research paper explores the integral role of customer segmentation in modern marketing strategies, emphasizing its significance in understand-ing and catering to diverse customer needs. Acknowledging Exploratory Data Analysis (EDA) as a crucial preliminary step, the study investigates how EDA acts as a catalyst for effective segmentation algorithms by un-veiling hidden patterns in raw data. Through empirical evidence and case studies, the paper demonstrates the transformative impact of incorporat-ing EDA before deploying segmentation models. The results underscore the necessity of a robust EDA framework to extract actionable insights, enhancing the precision of targeted marketing efforts and aligning seg-mentation strategies with real-world customer dynamics. This research contributes valuable insights for practitioners and researchers seeking to optimize marketing strategies through a holistic approach that combines customer segmentation with exploratory data analysis.

本文探讨了顾客细分在现代营销策略中不可或缺的作用，强调了它在理解和满足不同顾客需求方面的重要性。该研究承认探索性数据分析（EDA）是关键的初步步骤，研究了EDA如何通过揭开原始数据中的隐藏模式，作为有效分割算法的催化剂。通过实证和案例研究，本文展示了在部署分割模型之前结合EDA的变革性影响。结果强调了一个强大的EDA框架的必要性，以提取可操作的见解，提高目标营销工作的准确性，并使细分战略与现实世界的客户动态保持一致。这项研究为从业者和研究人员提供了宝贵的见解，通过将客户细分与探索性数据分析相结合的整体方法来优化营销策略。

{"title":"Strategic Insights into Customer Diversity: Unraveling Purchase Patterns, Income Disparities, and Relationship Dynamics through K-means Clustering for Enhanced Engagement and Loyalty","authors":"Shraddha Sharma , Rupali Satsangi , Preeti Manani , Priti Sharma , Jyoti Gupta","doi":"10.1016/j.procs.2025.03.301","DOIUrl":"10.1016/j.procs.2025.03.301","url":null,"abstract":"<div><div>This research paper explores the integral role of customer segmentation in modern marketing strategies, emphasizing its significance in understand-ing and catering to diverse customer needs. Acknowledging Exploratory Data Analysis (EDA) as a crucial preliminary step, the study investigates how EDA acts as a catalyst for effective segmentation algorithms by un-veiling hidden patterns in raw data. Through empirical evidence and case studies, the paper demonstrates the transformative impact of incorporat-ing EDA before deploying segmentation models. The results underscore the necessity of a robust EDA framework to extract actionable insights, enhancing the precision of targeted marketing efforts and aligning seg-mentation strategies with real-world customer dynamics. This research contributes valuable insights for practitioners and researchers seeking to optimize marketing strategies through a holistic approach that combines customer segmentation with exploratory data analysis.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"259 ","pages":"Pages 1-10"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143937858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Performance Evaluation of Advanced Classification Models Combined with Feature Selection for Credit Risk Performance 结合特征选择的高级分类模型信用风险绩效评价

Procedia Computer Science

Pub Date : 2025-01-01 DOI: 10.1016/j.procs.2025.04.265

Sarakam Tribuvan , Sandeep Deepak Kamath , Sparsh Mishra , Usha M , Shreyas J , Gururaj H L , Dayananda P , Karthik S A

In this study, we propose an advanced methodology utilizing machine learning models for predicting home equity credit risk on a real-world dataset. Traditional credit risk models often rely on outdated statistical methods that fail to capture complex, non-linear relationships in data, resulting in suboptimal accuracy and limited interpretability. Furthermore, existing models lack transparency, making it difficult for stakeholders to understand and act on the predictions. To address these issues, we employ state-of-the-art machine learning algorithms such as Decision Trees, AdaBoost, Support Vector Machine (SVM), Neural Networks, and Random Forest, along with feature selection techniques like Boruta and Principal Component Analysis (PCA) to enhance both accuracy and explainability. Our approach aims to provide improved credit risk assessment tools, offering better interpretability for loan companies, regulators, and applicants, while ensuring robust performance. The results demonstrate that our proposed models outperform traditional methods and offer actionable insights for stakeholders, enhancing decision-making processes.

在这项研究中，我们提出了一种先进的方法，利用机器学习模型来预测现实世界数据集上的房屋净值信用风险。传统的信用风险模型往往依赖于过时的统计方法，这些方法无法捕捉数据中复杂的非线性关系，导致准确性不理想，可解释性有限。此外，现有模型缺乏透明度，使得利益相关者难以理解预测并根据预测采取行动。为了解决这些问题，我们采用了最先进的机器学习算法，如决策树、AdaBoost、支持向量机（SVM）、神经网络和随机森林，以及Boruta和主成分分析（PCA）等特征选择技术，以提高准确性和可解释性。我们的方法旨在提供改进的信用风险评估工具，为贷款公司、监管机构和申请人提供更好的可解释性，同时确保稳健的绩效。结果表明，我们提出的模型优于传统方法，并为利益相关者提供可操作的见解，提高决策过程。

{"title":"Performance Evaluation of Advanced Classification Models Combined with Feature Selection for Credit Risk Performance","authors":"Sarakam Tribuvan , Sandeep Deepak Kamath , Sparsh Mishra , Usha M , Shreyas J , Gururaj H L , Dayananda P , Karthik S A","doi":"10.1016/j.procs.2025.04.265","DOIUrl":"10.1016/j.procs.2025.04.265","url":null,"abstract":"<div><div>In this study, we propose an advanced methodology utilizing machine learning models for predicting home equity credit risk on a real-world dataset. Traditional credit risk models often rely on outdated statistical methods that fail to capture complex, non-linear relationships in data, resulting in suboptimal accuracy and limited interpretability. Furthermore, existing models lack transparency, making it difficult for stakeholders to understand and act on the predictions. To address these issues, we employ state-of-the-art machine learning algorithms such as Decision Trees, AdaBoost, Support Vector Machine (SVM), Neural Networks, and Random Forest, along with feature selection techniques like Boruta and Principal Component Analysis (PCA) to enhance both accuracy and explainability. Our approach aims to provide improved credit risk assessment tools, offering better interpretability for loan companies, regulators, and applicants, while ensuring robust performance. The results demonstrate that our proposed models outperform traditional methods and offer actionable insights for stakeholders, enhancing decision-making processes.</div></div>","PeriodicalId":20465,"journal":{"name":"Procedia Computer Science","volume":"258 ","pages":"Pages 278-287"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143931403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0