Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169738
P. Shimna, A. Shirly Edward, T. Roshini
Lung cancer is a serious health issue that requires early detection. Machine Learning has figured prominently in the health sector in general, and in analyzing histopathological images and detecting illnesses in particular, because it may eliminate many mistakes that may arise when radiologists analyse image data. Traditional healthcare imaging techniques such as x-rays, CT scans, MRIs, and so on have little promise for detecting lung tumours. Convolutional Neural Networks have piqued the interest of doctors and academics due to their ability to analyse images accurately. The current study examines the role of CNN in lung cancer detection. Findings presented in the literature provide prospective researchers with a deeper understanding of the issue. We examined most of the features and includes extensive recommendations for future study. The primary purpose of this study is to detect malignant lung nodules in a lung image and to categorize pulmonary cancer. This work concentrates on novel Deep Learning techniques used in literature to locate cancerous lung nodules.
{"title":"A Review on Diagnosis of Lung Cancer and Lung Nodules in Histopathological Images using Deep Convolutional Neural Network","authors":"P. Shimna, A. Shirly Edward, T. Roshini","doi":"10.1109/ICAIA57370.2023.10169738","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169738","url":null,"abstract":"Lung cancer is a serious health issue that requires early detection. Machine Learning has figured prominently in the health sector in general, and in analyzing histopathological images and detecting illnesses in particular, because it may eliminate many mistakes that may arise when radiologists analyse image data. Traditional healthcare imaging techniques such as x-rays, CT scans, MRIs, and so on have little promise for detecting lung tumours. Convolutional Neural Networks have piqued the interest of doctors and academics due to their ability to analyse images accurately. The current study examines the role of CNN in lung cancer detection. Findings presented in the literature provide prospective researchers with a deeper understanding of the issue. We examined most of the features and includes extensive recommendations for future study. The primary purpose of this study is to detect malignant lung nodules in a lung image and to categorize pulmonary cancer. This work concentrates on novel Deep Learning techniques used in literature to locate cancerous lung nodules.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130305988","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}
Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169306
R. Vallikannu, V. Kanpur Rani, B. Kavitha, P. Sankar
Situational awareness is the sense and knowledge of one’s immediate surroundings. In safety-critical sectors, maintaining situational awareness is essential for performance and error prevention. Situational awareness (SAW) is crucial for the success of activities in many different domains, such as surveillance, humanitarian aid, and search and rescue efforts. SAW is however susceptible to enemy attacks. By giving users enhanced coverage, it increases survivability and mission capability. Recently, Smart gadgets used data to address crisis scenarios and provide real-time tracking to protect law enforcement personnel out in the field. Despite these developments, it might be challenging for first responders to get a precise feel of their surroundings due to an abundance of field data. Security teams need to be able to quickly transform this data into actionable intelligence using a few instruments at their disposal, including body cameras, fingerprint scanners, and facial recognition software. Officers can cut through the noise to acquire actual real-time situational awareness by integrating heterogeneous information into a cohesive platform. Therefore, the proposed work examines potential mitigation measures while considering hostile threats and assaults against SAW systems. Additionally, information and alarms can be instantly sent between operators and field officers using vital interface features. The optimization of the AutoML system is proposed for fusion of sensor data. AutoML classification with Bayesian and ASHA (Asynchronous successive halving algorithm) is used for situational forecasting and decision-making awareness, IoT is used to monitor data gathered from Kaggle and sensor readings, while thingspeak cloud is used to monitor sensor output.
{"title":"An Analysis of Situational Intelligence for First Responders in Military","authors":"R. Vallikannu, V. Kanpur Rani, B. Kavitha, P. Sankar","doi":"10.1109/ICAIA57370.2023.10169306","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169306","url":null,"abstract":"Situational awareness is the sense and knowledge of one’s immediate surroundings. In safety-critical sectors, maintaining situational awareness is essential for performance and error prevention. Situational awareness (SAW) is crucial for the success of activities in many different domains, such as surveillance, humanitarian aid, and search and rescue efforts. SAW is however susceptible to enemy attacks. By giving users enhanced coverage, it increases survivability and mission capability. Recently, Smart gadgets used data to address crisis scenarios and provide real-time tracking to protect law enforcement personnel out in the field. Despite these developments, it might be challenging for first responders to get a precise feel of their surroundings due to an abundance of field data. Security teams need to be able to quickly transform this data into actionable intelligence using a few instruments at their disposal, including body cameras, fingerprint scanners, and facial recognition software. Officers can cut through the noise to acquire actual real-time situational awareness by integrating heterogeneous information into a cohesive platform. Therefore, the proposed work examines potential mitigation measures while considering hostile threats and assaults against SAW systems. Additionally, information and alarms can be instantly sent between operators and field officers using vital interface features. The optimization of the AutoML system is proposed for fusion of sensor data. AutoML classification with Bayesian and ASHA (Asynchronous successive halving algorithm) is used for situational forecasting and decision-making awareness, IoT is used to monitor data gathered from Kaggle and sensor readings, while thingspeak cloud is used to monitor sensor output.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121315109","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}
Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169521
T. Saravanan, D. Vinotha
Pervasive computation plays an integral part in WBANs. Along with pervasive methodologies, bio-sensors are available in a range of shapes and sizes, and depending on the state of the patient, multiple sensors can be inserted in, on, or around the human body to monitor, store, and relay vital signs for further investigation, judgments, and treatment. The tracking of patients’ vital signs, as well as the time it takes to generate results, are essential components of the WBAN’s integration into ubiquitous computing technologies. To ensure low power consumption, high precision of collected data, low latency, high efficiency, higher throughput with efficient bandwidth utilization, and synchronization with other systems and at the same time data must be stored and exchanged with care. To function successfully, a WBAN must first measure the quantity of electricity the device utilizes and then impose energy-efficient operating strategies. Current routing processes, such as the Stable Increased-Throughput Multi-hop Protocol for Link Efficiency (SIMPLE) and Mobility-supporting Adaptive Threshold-based Thermal-aware Energy-efficient Multi-hop Protocol (M-ATTEMPT), can be employed in WBANs by incorporating confidence measures into both the sensor data being monitored and the power levels needed for effective data broadcast to reach the sink. In contrast to Expected Transfers (ETX), this protocol avoids continuous communications and only forwards data of interest to the sink, resulting in minimal power usage and thereby increasing network reliability time, overall network lifetime, throughput, and end to end latency to 0.915 mw, 290 bits/s, and 250 ms, respectively.
{"title":"Trust Value-Based Energy-Efficient Routing Protocol to Improve Lifetime in Heterogeneous WBAN","authors":"T. Saravanan, D. Vinotha","doi":"10.1109/ICAIA57370.2023.10169521","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169521","url":null,"abstract":"Pervasive computation plays an integral part in WBANs. Along with pervasive methodologies, bio-sensors are available in a range of shapes and sizes, and depending on the state of the patient, multiple sensors can be inserted in, on, or around the human body to monitor, store, and relay vital signs for further investigation, judgments, and treatment. The tracking of patients’ vital signs, as well as the time it takes to generate results, are essential components of the WBAN’s integration into ubiquitous computing technologies. To ensure low power consumption, high precision of collected data, low latency, high efficiency, higher throughput with efficient bandwidth utilization, and synchronization with other systems and at the same time data must be stored and exchanged with care. To function successfully, a WBAN must first measure the quantity of electricity the device utilizes and then impose energy-efficient operating strategies. Current routing processes, such as the Stable Increased-Throughput Multi-hop Protocol for Link Efficiency (SIMPLE) and Mobility-supporting Adaptive Threshold-based Thermal-aware Energy-efficient Multi-hop Protocol (M-ATTEMPT), can be employed in WBANs by incorporating confidence measures into both the sensor data being monitored and the power levels needed for effective data broadcast to reach the sink. In contrast to Expected Transfers (ETX), this protocol avoids continuous communications and only forwards data of interest to the sink, resulting in minimal power usage and thereby increasing network reliability time, overall network lifetime, throughput, and end to end latency to 0.915 mw, 290 bits/s, and 250 ms, respectively.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125900738","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}
Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169318
Souvik Chowdhury, B. Soni
Thanks to the emergence and continued devel-opment of machine learning, particularly deep learning, the research on visual question and answer, also known as VQA, has advanced dramatically, with great theoretical research significance and practical application value. This field of study makes use of multimodal learning, computer vision, and natural language processing techniques. Except for a few academics who presented different types of optimized bi-linear fusion approaches that integrate text and image characteristics in an efficient way, there haven’t been many efforts to optimize the VQA framework. In order to optimize the VQA problem, we offer a unique Visual Question Answering framework in this research. Because both 16-bit and 32-bit floating points provide automatic mixed precision, deep learning architectures can now be optimized with less computation and execution time. Using the VQA 2.0 and CLEVR datasets, the proposed framework has been tested against two models. In terms of overall accuracy and execution time, the experimental findings demonstrated a significant improvement.
由于机器学习特别是深度学习的出现和不断发展，视觉问答(visual question and answer，简称VQA)的研究有了长足的进步，具有很大的理论研究意义和实际应用价值。这个研究领域使用了多模态学习、计算机视觉和自然语言处理技术。除了少数学者提出了不同类型的优化的双线性融合方法，有效地整合了文本和图像的特征，对VQA框架进行优化的努力并不多。为了优化VQA问题，我们在本研究中提供了一个独特的可视化问答框架。因为16位和32位浮点都提供自动混合精度，深度学习架构现在可以用更少的计算和执行时间进行优化。使用VQA 2.0和CLEVR数据集，对所提出的框架进行了两个模型的测试。在总体精度和执行时间方面，实验结果显示了显着的改进。
{"title":"Visual Question Answering Optimized Framework using Mixed Precision Training","authors":"Souvik Chowdhury, B. Soni","doi":"10.1109/ICAIA57370.2023.10169318","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169318","url":null,"abstract":"Thanks to the emergence and continued devel-opment of machine learning, particularly deep learning, the research on visual question and answer, also known as VQA, has advanced dramatically, with great theoretical research significance and practical application value. This field of study makes use of multimodal learning, computer vision, and natural language processing techniques. Except for a few academics who presented different types of optimized bi-linear fusion approaches that integrate text and image characteristics in an efficient way, there haven’t been many efforts to optimize the VQA framework. In order to optimize the VQA problem, we offer a unique Visual Question Answering framework in this research. Because both 16-bit and 32-bit floating points provide automatic mixed precision, deep learning architectures can now be optimized with less computation and execution time. Using the VQA 2.0 and CLEVR datasets, the proposed framework has been tested against two models. In terms of overall accuracy and execution time, the experimental findings demonstrated a significant improvement.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125465229","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}
Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169126
Hardik Saini, K. S. Saini
With the advancement in world wide web, the way to communicate among individuals, via internet, is changed and thus, various platforms become popular such as email. Numerous organizations and people make the deployment of email as major sources of communication. This platform is extensively utilized in spite of alternative means, such as electronic messages, and social networks. However, this technology is more prone to malicious activities. The malicious users target this free mail structure and send a huge number of useless messages, for attaining revenues, or stealing personal data or IDs, to harm its users. Thus, there is necessity to discover the methods for detecting the email spam. The spam is detected in email in different phases in which the data is pre-processed, features are extracted, and the mails are classified. This work introduced a new model to predict the email spam. This approach implements the random forest in order to extract the features. Eventually, the spam is predicted using logistic regression model. The proposed model is implemented in python using anaconda.
{"title":"Hybrid Model for Email Spam Prediction Using Random Forest for Feature Extraction","authors":"Hardik Saini, K. S. Saini","doi":"10.1109/ICAIA57370.2023.10169126","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169126","url":null,"abstract":"With the advancement in world wide web, the way to communicate among individuals, via internet, is changed and thus, various platforms become popular such as email. Numerous organizations and people make the deployment of email as major sources of communication. This platform is extensively utilized in spite of alternative means, such as electronic messages, and social networks. However, this technology is more prone to malicious activities. The malicious users target this free mail structure and send a huge number of useless messages, for attaining revenues, or stealing personal data or IDs, to harm its users. Thus, there is necessity to discover the methods for detecting the email spam. The spam is detected in email in different phases in which the data is pre-processed, features are extracted, and the mails are classified. This work introduced a new model to predict the email spam. This approach implements the random forest in order to extract the features. Eventually, the spam is predicted using logistic regression model. The proposed model is implemented in python using anaconda.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130720839","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}
Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169347
P. Patwal, Amit Kumar Srivastava
Accurate prediction of stock market price is highly challenging. This paper presents a proposed model for prediction of stock market price of Netflix. We have considered a five–year data set (April, 2017 – April, 2022) of Netflix. An Exploratory Data Analysis (EDA) of Netflix’s stock price data for predicting its stock market prices using time series is done. The implementation of the model is done using Python language. We imported five-years data and applied several techniques: importing libraries, calculating stock return, line plot, plot all, plot return year wise, plot histogram, plot kernel density, plot box plot, differencing method, resample daily to monthly data etc. EDA proved that using time series technique achieved better results in prediction of stock price and visualizing.
{"title":"Proposed Model for Prediction of Stock Market Price of Netflix","authors":"P. Patwal, Amit Kumar Srivastava","doi":"10.1109/ICAIA57370.2023.10169347","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169347","url":null,"abstract":"Accurate prediction of stock market price is highly challenging. This paper presents a proposed model for prediction of stock market price of Netflix. We have considered a five–year data set (April, 2017 – April, 2022) of Netflix. An Exploratory Data Analysis (EDA) of Netflix’s stock price data for predicting its stock market prices using time series is done. The implementation of the model is done using Python language. We imported five-years data and applied several techniques: importing libraries, calculating stock return, line plot, plot all, plot return year wise, plot histogram, plot kernel density, plot box plot, differencing method, resample daily to monthly data etc. EDA proved that using time series technique achieved better results in prediction of stock price and visualizing.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133782017","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}
Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169373
H. S. Shashank, Aniruddh Acharya, E. Sivaraman
Image reconstruction and super resolution has various applications. Several deep learning techniques are being employed to constantly improve this space. The aim of this experiment is to showcase a unique deep learning approach to try and super resolve human faces from low resolution images. The experiment makes use of a machine learning framework designed to improve image quality called Super Resolution Generative Adversarial Neural (SRGANs) with a loss function based on the features accumulated from multiple layers of a trained Convolutional Neural Network named Visual Geometry Group-19 (VGG-19). The model super resolves lower quality image input and gives out image output of a superior quality
{"title":"Facial Image Super Resolution and Feature Reconstruction using SRGANs with VGG-19-based Adaptive Loss Function","authors":"H. S. Shashank, Aniruddh Acharya, E. Sivaraman","doi":"10.1109/ICAIA57370.2023.10169373","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169373","url":null,"abstract":"Image reconstruction and super resolution has various applications. Several deep learning techniques are being employed to constantly improve this space. The aim of this experiment is to showcase a unique deep learning approach to try and super resolve human faces from low resolution images. The experiment makes use of a machine learning framework designed to improve image quality called Super Resolution Generative Adversarial Neural (SRGANs) with a loss function based on the features accumulated from multiple layers of a trained Convolutional Neural Network named Visual Geometry Group-19 (VGG-19). The model super resolves lower quality image input and gives out image output of a superior quality","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132685010","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}
Cross-Site Scripting (XSS) represents one of the most prevalent application layer attacks perpetrated by an attacker, a client, and the web server. Cyber-attacks steal clients’ cookies / sensitive details and therefore associate the client with the web. Filtering user data in server-side scripts like ASP (Active Server Pages), PHP (Hypertext Preprocessor), or some other web-enabled programming language is a general solution to this which can be found floating around the internet. From the server perspective, we suggest a modular and extensible solution against XSS attack; the extensible solution can be used as an identity management solution for validating the users accessing the web application and testing for correct permissions for various web resources allocated to web users. Using deep learning, the research creates a secure ecosystem that may be used to provide efficient real-time detection and mitigation of cross-site scripting attacks in fog/cloud online applications. In this study, a deep learning model was used to detect XSS attacks, and its output was compared to that of three other deep learning models, namely Multilayer Perceptron, Long Short-Term Memory, and Deep Belief Network. This web-based system utilizes an MLP architecture for deep learning to detect inserted XSS attack scripts in web applications. The effectiveness of the algorithm for deep learning is assessed by utilizing evaluation metrics to evaluate the framework. Employing embedding as a feature, the MLP method performed the best in the evaluation for detecting XSS attacks, attaining an accuracy of 99.47%.
跨站点脚本(XSS)是由攻击者、客户端和web服务器共同实施的最常见的应用层攻击之一。网络攻击窃取客户端的cookie /敏感细节，从而将客户端与网络联系起来。在服务器端脚本中过滤用户数据，如ASP (Active Server Pages)、PHP (Hypertext Preprocessor)或其他一些支持网络的编程语言是解决这个问题的通用解决方案，在互联网上随处可见。从服务器的角度来看，我们建议采用模块化和可扩展的解决方案来抵御XSS攻击;可扩展解决方案可以用作身份管理解决方案，用于验证访问web应用程序的用户，并测试分配给web用户的各种web资源的正确权限。利用深度学习，该研究创建了一个安全的生态系统，可用于在雾/云在线应用程序中提供有效的实时检测和缓解跨站点脚本攻击。本研究使用深度学习模型检测XSS攻击，并将其输出与其他三种深度学习模型(多层感知器、长短期记忆和深度信念网络)的输出进行比较。这个基于web的系统利用深度学习的MLP架构来检测web应用程序中插入的XSS攻击脚本。利用评价指标对框架进行评价，评估算法在深度学习中的有效性。采用嵌入作为特征，MLP方法在检测跨站攻击的评估中表现最好，准确率达到99.47%。
{"title":"Web Server Security Solution for Detecting Cross-site Scripting Attacks in Real-time Using Deep Learning","authors":"Monika Sethi, J. Verma, Manish Snehi, Vidhu Baggan, Virender, Gunjan Chhabra","doi":"10.1109/ICAIA57370.2023.10169255","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169255","url":null,"abstract":"Cross-Site Scripting (XSS) represents one of the most prevalent application layer attacks perpetrated by an attacker, a client, and the web server. Cyber-attacks steal clients’ cookies / sensitive details and therefore associate the client with the web. Filtering user data in server-side scripts like ASP (Active Server Pages), PHP (Hypertext Preprocessor), or some other web-enabled programming language is a general solution to this which can be found floating around the internet. From the server perspective, we suggest a modular and extensible solution against XSS attack; the extensible solution can be used as an identity management solution for validating the users accessing the web application and testing for correct permissions for various web resources allocated to web users. Using deep learning, the research creates a secure ecosystem that may be used to provide efficient real-time detection and mitigation of cross-site scripting attacks in fog/cloud online applications. In this study, a deep learning model was used to detect XSS attacks, and its output was compared to that of three other deep learning models, namely Multilayer Perceptron, Long Short-Term Memory, and Deep Belief Network. This web-based system utilizes an MLP architecture for deep learning to detect inserted XSS attack scripts in web applications. The effectiveness of the algorithm for deep learning is assessed by utilizing evaluation metrics to evaluate the framework. Employing embedding as a feature, the MLP method performed the best in the evaluation for detecting XSS attacks, attaining an accuracy of 99.47%.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"64 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132703884","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}
Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169773
Ayush Dodia, Sumit Kumar
The use of vehicle object detection in intelligent video surveillance and vehicle-assisted driving has expanded as science and technology have advanced. Traditional car object detection algorithms have some limitations in their generalization capacity and recognition rate. The primary goal of this survey is to detect the vehicle, which forms managing crucial traffic data, including vehicle detection, vehicle count, and vehicle movement. This research compares modern object detectors that incorporate traffic situation estimations To determine which version of the YOLO algorithm is the best for detecting the vehicle explained here. Process of the YOLO algorithm the dataset is the first clustered using the clustering analysis approach, and the network structure is improved to increase the vehicle prediction capacity and the final numbers of output grids. In the second process, it maximizes both input image and dataset collection. This research suggests a better vehicle identification technique based on YOLO (You Only Look Once) to address this issue. Three versions of the YOLO (You Only Look Once) algorithm are evaluated to detect the vehicle.
{"title":"A Comparison of YOLO Based Vehicle Detection Algorithms","authors":"Ayush Dodia, Sumit Kumar","doi":"10.1109/ICAIA57370.2023.10169773","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169773","url":null,"abstract":"The use of vehicle object detection in intelligent video surveillance and vehicle-assisted driving has expanded as science and technology have advanced. Traditional car object detection algorithms have some limitations in their generalization capacity and recognition rate. The primary goal of this survey is to detect the vehicle, which forms managing crucial traffic data, including vehicle detection, vehicle count, and vehicle movement. This research compares modern object detectors that incorporate traffic situation estimations To determine which version of the YOLO algorithm is the best for detecting the vehicle explained here. Process of the YOLO algorithm the dataset is the first clustered using the clustering analysis approach, and the network structure is improved to increase the vehicle prediction capacity and the final numbers of output grids. In the second process, it maximizes both input image and dataset collection. This research suggests a better vehicle identification technique based on YOLO (You Only Look Once) to address this issue. Three versions of the YOLO (You Only Look Once) algorithm are evaluated to detect the vehicle.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130596731","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}
Pub Date : 2023-04-21DOI: 10.1109/ICAIA57370.2023.10169396
Devi Naveen, M. D Nirmala, T. J. T. Maladhkar, M. Serena, Rahmath Mohis
Physical limitations are and always will be a barrier to daily progress. Technology advancements are assisting everyone in leading simpler lives. Using the same technologies, we can provide a significant and beneficial answer to the issues associated with physical impairment. This essay discusses the use of technology to improve daily life for those who have physical or sensory disabilities. Not just for those who have hearing loss, but also as a tool for those who have speech disability, sign language is a vital means of communication. People without disabilities have a hard time understanding sign language, and specialists are frequently the only ones who can. Hence, a tool for sign language interpretation becomes necessary. Although Braille is a reading and writing system used by people who are blind. Braille is less popular among persons who are visually impaired, as it is time-consuming to manually translate every text into braille. Our study examines the issues raised by these two deficits and looks for technical remedies. Text to audio conversion is a piece of technology that can revolutionize the way visually impaired individuals communicate currently. It is simple and has been done effectively for the past ten years to convert written text to audio. In addition to sign language interpreters, a relatively new concept for assisting the education of the blind is to translate speech into sign language. The technologies stated above are anticipated to significantly improve the daily lives of people with physical disabilities, and this project can be further customized to match any suitable smart object.
{"title":"Tech-It-Easy: An Application for Physically Impaired People Using Deep Learning","authors":"Devi Naveen, M. D Nirmala, T. J. T. Maladhkar, M. Serena, Rahmath Mohis","doi":"10.1109/ICAIA57370.2023.10169396","DOIUrl":"https://doi.org/10.1109/ICAIA57370.2023.10169396","url":null,"abstract":"Physical limitations are and always will be a barrier to daily progress. Technology advancements are assisting everyone in leading simpler lives. Using the same technologies, we can provide a significant and beneficial answer to the issues associated with physical impairment. This essay discusses the use of technology to improve daily life for those who have physical or sensory disabilities. Not just for those who have hearing loss, but also as a tool for those who have speech disability, sign language is a vital means of communication. People without disabilities have a hard time understanding sign language, and specialists are frequently the only ones who can. Hence, a tool for sign language interpretation becomes necessary. Although Braille is a reading and writing system used by people who are blind. Braille is less popular among persons who are visually impaired, as it is time-consuming to manually translate every text into braille. Our study examines the issues raised by these two deficits and looks for technical remedies. Text to audio conversion is a piece of technology that can revolutionize the way visually impaired individuals communicate currently. It is simple and has been done effectively for the past ten years to convert written text to audio. In addition to sign language interpreters, a relatively new concept for assisting the education of the blind is to translate speech into sign language. The technologies stated above are anticipated to significantly improve the daily lives of people with physical disabilities, and this project can be further customized to match any suitable smart object.","PeriodicalId":196526,"journal":{"name":"2023 International Conference on Artificial Intelligence and Applications (ICAIA) Alliance Technology Conference (ATCON-1)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131381711","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：