A Machine Learning Approach to Detecting Deepfake Videos: An Investigation of Feature Extraction Techniques

Preeti; Khyati; Gopal; Manju; Bharat

doi:https://doi.org/10.54216/JCIM.090204

Fusion: Practice and Applications | 2023

Physical Activity Monitoring for Older Adults through IoT and Wearable Devices: Leveraging Data Fusion Techniques

The emergence of low-cost individual sensing devices has facilitated the application of data fusion methods to yield insights useful for score-level, rank-level, or hybrid-level fusion. Intelligent tools for fusion processing, such as fuzzy methods and optimization algorithms, may be used to the deluge of raw data generated by these devices. The use of numerous sensors allows for multi-level/hybrid-level fusion, and the combination of several models for intelligent systems allows for fusion system design optimized for score improvement. Multimedia data fusion applications and machine learning methods can be used to accomplish data fusion in cloud settings. For older people in independent living conditions, a physical activity assessment framework (PAAF) that uses deep learning models for fusion to identify activity and evaluate progress based on the spectral domain of each window is needed. This study highlights the significance of data fusion in outlining the needs for IoT devices in networked computers for distant patient monitoring. In order to provide for the health of the elderly without compromising their comfort or freedom of choice, we need a seniors network based on the Internet of Things and wearable health technology. The sensors' functionality was investigated by analyzing data gathered from the environment and the organisms within it. The proposed PAAF-IoT architecture has many layers, each one connected to a different device, with the most important part being the integration of data from all of them to classify types of physical activity. Cloud services geographically close to the customer are used to process the resulting mountain of data, reducing end-to-end delay and facilitating prompt responses from healthcare professionals. Data fusion in healthcare and remote patient monitoring are demonstrated through the deployment of an app that allows doctors to remotely administer prescriptions and maintain track of patients' medical histories.

groups

Hayder Mahmood Salman mail -

Hasan Faleh Hamdan mail -

Raed Khalid mail -

Sanaa Al-Kikani mail

link https://doi.org/10.54216/FPA.110204

Volume & Issue

Vol. Volume 11 / Iss. Issue 2

Details open_in_new

Fusion: Practice and Applications | 2023

Multi-Level Fusion Optimization in Cyber-Physical Systems Using Computer Vision-Based Fault Detection

The healthcare sector's use of cyber-physical systems to provide high-quality patient treatment highlights the need for sophisticated security solutions due to the wide range of attack surfaces from medical and mobile devices, as well as body sensor nodes. Cyber-physical systems have various processing technologies to choose from, but these technical methods are as varied. Existing technologies are not well-suited for managing complex information about problem identification and diagnosis, which is distinct from technology. To address this issue, intelligent techniques for fusion processing, such as multi-sensor fusion system architectures and fusion optimization, can be used to improve fusion score and decision-making. Additionally, the use of deep learning models and multimedia data fusion applications can help to combine multiple models for intelligent systems and enhance machine learning for data fusion in E-Systems and cloud environments. Fuzzy approaches and optimization algorithms for data fusion can also be applied to robotics and other applications.. In this paper, a computer vision technology-based fault detection (CVT-FD) framework has been suggested for securely sharing healthcare data. When utilizing a trusted device like a mobile phone, end-users can rest assured that their data is secure. Cyber-attack behavior can be predicted using an artificial neural network (ANN), and the analysis of this data can assist healthcare professionals in making decisions. The experimental findings show that the model outperforms with current detection accuracy (98.3%), energy consumption (97.2%), attack prediction (96.6%), efficiency (97.9%), and delay ratios (35.6%) over existing approaches.

groups

Mustafa Altaee mail -

Anwar Ja’afar M. Jawad mail -

Mohammed A. Jalil mail -

Noor Sami mail -

Zaid Saad Madhi mail

link https://doi.org/10.54216/FPA.110205

Volume & Issue

Vol. Volume 11 / Iss. Issue 2

Details open_in_new

Fusion: Practice and Applications | 2023

Fusion System for Blockchain Asset Securitization Risk Control Using Adaptive Deep Learning-Based Framework

Feature engineering methods, which entail identifying and extracting useful features from big datasets, can be used to enhance the precision of asset securitization. It might be difficult to securitize assets that produce multiple receivables, such as consumer or company debt. In order to overcome these difficulties, companies might think about adopting a fusion system that integrates feature engineering with distributed ledger technologies such as blockchain. Businesses can benefit from implementing a fusion system like the Deep learning-based Adaptive Online Intelligent Framework (DLAOIF) since it allows for better decision-making, less wasted time and money, and less chance of fraud. Financial asset tracking on a blockchain can help investors keep a closer eye on asset performance and related risks, while also decreasing their reliance on credit rating agencies. Blockchain's high data security standards and elimination of regulatory bottlenecks in the securitization process also make it a useful tool for easing the burden of due diligence.  

groups

Raed Khalid mail -

Omar Saad Ahmed mail -

Talib A. Al-Sharify mail -

Wasfi Hameed mail -

Riyam K. Marjan mail

link https://doi.org/10.54216/FPA.110206

Volume & Issue

Vol. Volume 11 / Iss. Issue 2

Details open_in_new

International Journal of Neutrosophic Science | 2023

Neutrosophic Environment and E-Learning: An Investigation into Student Satisfaction and Attitudes in the College of Engineering

Ever since the transition to online learning, students from across the globe have familiarized themselves with the use of e-learning. This research paper aims to investigate students’ attitudes towards e-learning and measure students’ level of experience in using the e – system among students in the College of Engineering in the United Arab Emirates (UAE). Furthermore, this paper explores the predictors of students’ perceived satisfaction with the e-learning system quality that is integrated to facilitate e-learning. Descriptive analysis and stepwise multiple regression were chosen to achieve the paper’s objectives. Data analysis demonstrated that Engineering students showed a positive attitude towards the e-learning system and were identified with high self-efficacy and a high level of experience in using the e-learning system. In addition, multimedia instructions, self–efficacy, and e-learning system quality were found as significant predictors of students’ satisfaction with the e-learning system; however, the interactive learning activities as a predictor of perceived satisfaction did not reveal any statistical significance. Also, this paper used the DEMATEL method to analysis the student attributes and compute the weights of the attributes. The DEMATEL method is a MCDM method due to the attribute of student are neuromas so the concept of the MCDM is used. The DEMATEL method integrated with the neutrosophic sets. The neutrosophic sets is used to overcome the uncertainty and the vague data.

groups

Nahla Moussa mail -

Bashria Yousef mail -

Shahin Abdel Naby mail -

Sondus Al Qudah mail

link https://doi.org/10.54216/IJNS.200413

Volume & Issue

Vol. Volume 20 / Iss. Issue 4

Details open_in_new

American Journal of Business and Operations Research | 2021

Toward sustainable development: Construction and appraising enhanced amalgamation framework of Business Intelligence based sustainability report

The purpose of this study is to strength information of sustainability Reports (SusRs) toward achieving sustainable development (SusD) according to SustainableDevelopment Goals (UN_SDGs). So, we proposed a robust framework for enhancing and appraising corporate sustainability performance by integrating BI in SusRs to provides a structured approach for nterprises.Due to ability of BI to facilitate  collect, analyze, and report processes on sustainability data. Additionality it leads to informed decision-making and improved sustainability outcomes. Our framework highlights the importance of SusR and its impact on stakeholder engagement and transparency. Moreover, it emphasized the potential of BI in enhancing the accuracy, reliability, and completeness of sustainability reporting. The framework's practical application is demonstrated through a case study, which showcases how it can be used to identify sustainability risks and opportunities and develop effective sustainability strategies. Generally, our work contributes to the growing body of research on corporate sustainability by providing a comprehensive framework that can help companies to achieve their sustainability goals while driving business performance.

groups

Mona Mohamed mail -

Ahmed M. AbdelMouty mail

link https://doi.org/10.54216/AJBOR.050204

Volume & Issue

Vol. Volume 5 / Iss. Issue 2

Details open_in_new

International Journal of Neutrosophic Science | 2023

Exploring the Role of ChatGPT and social media in Enhancing Student Evaluation of Teaching Styles in Higher Education Using Neutrosophic Sets

This paper provides an in-depth analysis of how Chat GPT and social media can be used as tools for capturing real-time student feedback on teaching styles in higher education. The study employs neutrosophic sets to deal with the uncertainties and ambiguities that arise in student evaluation data. Traditional methods of evaluating teaching styles in higher education, such as paper-based surveys, may not fully capture the nuanced experiences of students in the classroom. Recent advancements in chatbots, such as ChatGPT, and the growing use of social media platforms offer new opportunities for more efficient and effective methods of evaluating teaching styles. However, there are significant challenges in using these technologies, including the handling of indeterminate and uncertain data. Neutrosophic sets provide a mathematical framework for handling ambiguity and uncertainty in data and can be used to better capture the complex and multifaceted aspects of student experiences in the classroom. Additionally, the use of chatbots and social media platforms raises practical and ethical concerns that must be addressed in the evaluation process. This study aims to explore the role of ChatGPT and social media in enhancing student evaluation of teaching styles in higher education using neutrosophic sets, while also addressing the practical and ethical challenges that arise from their use.

groups

Najla M. Alnaqbi mail -

Walaa Fouda mail

link https://doi.org/10.54216/IJNS.200414

Volume & Issue

Vol. Volume 20 / Iss. Issue 4

Details open_in_new

International Journal of Neutrosophic Science | 2023

Pseudo Similarity of Neutrosophic Fuzzy matrices

In this paper, first we shall define Pseudo Similarity for Neutrosophic Fuzzy Matrices and prove that Pseudo Similarity relation on pair of Neutrosophic Fuzzy Matrices. Also, we derive some relation between Pseudo Similarity and Idempotent matrices. Finally, we give in varies inverse of Neutrosophic Fuzzy Matrices.

groups

M. Anandhkumar mail -

V. Kamalakannan mail -

S. M. Chithra mail -

broumi said mail

link https://doi.org/10.54216/IJNS.200415

Volume & Issue

Vol. Volume 20 / Iss. Issue 4

Details open_in_new

Journal of Cybersecurity and Information Management | 2022

Securing the Internet of Things (IoT) with Blockchain: A Proof-of-Concept Implementation and Analysis

The Internet of Things (IoT) has revolutionized the way we interact with everyday objects, enabling devices to collect and share data seamlessly. However, this increased connectivity has also increased the security risks associated with these devices, as they often lack the necessary security mechanisms to prevent malicious attacks. To address this issue, we propose using blockchain technology to secure IoT devices. In this paper, we present a proof-of-concept implementation of a blockchain-based IoT security system and analyze its effectiveness. Our system leverages blockchain's distributed ledger technology to ensure data integrity, decentralization, and transparency, making it more resilient to attacks. We evaluate our system's performance and compare it with other existing IoT security solutions. Our results show that our blockchain-based approach outperforms traditional security measures and is a viable solution for securing IoT devices. Finally, we discuss the limitations of our study and suggest future research directions for improving the security of IoT devices.

groups

Mahmoud A. Zaher mail -

Nabil M. Eldakhly mail

link https://doi.org/10.54216/JCIM.100203

Volume & Issue

Vol. Volume 10 / Iss. Issue 2

Details open_in_new

International Journal of Advances in Applied Computational Intelligence | 2022

Modelling Weather Conditions Using Encoder-Decoder and Attention Based on LSTM Deep Regression Model

In the rapidly evolving field of smart cities, the accurate prediction of weather patterns plays a crucial role in various industries such as agriculture, tourism, and socioeconomic development. This study utilizes Artificial Intelligence (AI) and Machine Learning (ML) through advanced machine learning techniques, including Encoder-Decoder LSTM and Attention LSTM models, to analyze daily climatic weather data in the Narmadapuram district. The research investigated the future patterns of key weather parameters, including maximum temperature, minimum temperature, morning relative humidity, evening relative humidity, and bright sunshine hours. The study analyzed daily data collected between November 1, 1977 and April 30, 2022, with 80% used for training and 20% for testing. Results showed that the Encoder-Decoder LSTM model outperformed the Attention LSTM model in forecasting maximum temperature, morning relative humidity, evening relative humidity, and bright sunshine hours, while the Attention LSTM model had better results in predicting minimum temperature. The findings provide valuable insights into climatic patterns and variability and have implications for the development of more precise weather forecasting models. This study demonstrates the potential of AI and ML in addressing the challenges of smart cities and highlights the significance of machine learning techniques in weather forecasting, a critical aspect of urban operations and decision-making.

groups

Khder Alakkari mail -

Mostafa Abotaleb mail -

Amr Badr mail -

Ammar Kadi mail -

A. M. Ghazi Al khatib mail -

Bayan Mohamad Alshaib mail -

El-Sayed M. El-Kenawy mail

link https://doi.org/10.54216/IJAACI.010201

Volume & Issue

Vol. Volume 1 / Iss. Issue 2

Details open_in_new

Journal of Cybersecurity and Information Management | 2022

A Machine Learning Approach to Detecting Deepfake Videos: An Investigation of Feature Extraction Techniques

Deepfake videos are a growing concern today as they can be used to spread misinformation and manipulate public opinion. In this paper, we investigate the use of different feature extraction techniques for detecting deepfake videos using machine learning algorithms. We explore three feature extraction techniques, including facial landmarks detection, optical flow, and frequency analysis, and evaluate their effectiveness in detecting deepfake videos. We compare the performance of different machine learning algorithms and analyze their ability to detect deepfakes using the extracted features. Our experimental results show that the combination of facial landmarks detection and frequency analysis provides the best performance in detecting deepfake videos, with an accuracy of over 95%. Our findings suggest that machine learning algorithms can be a powerful tool in detecting deepfake videos, and feature extraction techniques play a crucial role in achieving high accuracy.

groups

Preeti Singh mail -

Khyati Chaudhary mail -

Gopal Chaudhary mail -

Manju Khari mail -

Bharat Rawal mail

link https://doi.org/10.54216/JCIM.090204

Volume & Issue

Vol. Volume 9 / Iss. Issue 2

Details open_in_new