MMU Press (Multimedia University)
Not a member yet
714 research outputs found
Sort by
Fiber Break Prevention Using Machine Learning Approaches
No full textModern fiber-optic communication systems are built around optical fiber, which allows data to be sent by emitting infrared light pulses. It is widely used by telecommunications firms and is essential to the smooth transmission of information in internet communication as well as the transmission of telephone signals. Nonetheless, optical fibers intrinsic fragility raises a problem, especially in areas where building projects are taking place. Especially nowadays construction-related impact and crushing pressures can cause physical damage that jeopardizes the fiber optic's integrity. Therefore, this research emphasizes the necessity of taking preventative and mitigating actions to reduce the possibilities of fiber optic breakages in response to these difficulties by using machine learning approaches. The data collected by an optical fiber sensor and a distributed acoustic sensing interrogator unit (DAS). Five tools are used to simulate fiber break threats on the road surface and the fiber optic signal is denoised by using the bandpass Butterworth filter. The filtered data is then transformed into spectrogram representation and trained by using the machine learning approaches. The results of the experiments in the research achieves the accuracy 99.78% which is a high accuracy which can be potentially applied in classifying the signals of the tools and preventing the breakage of the fiber optic cables
Early Identification of Parkinson's Disease Using Time Frequency Analysis on EEG Signals
No full textParkinson's Disease (PD) is a progressive neurological disorder. It affects movement and can significantly impact quality of life. Early and accurate diagnosis is crucial for effective management and intervention. Traditional diagnostic methods can be time-consuming and less effective in the early stages of the disease. This study aims to develop an automated approach for identifying PD using time-frequency image analysis of electroencephalogram (EEG) signals. The goal is to enhance diagnostic accuracy and efficiency, facilitating early detection. EEG signals, often contaminated with artifacts such as eye blinks and muscle movements etc., were first cleaned. Time-frequency images were then plotted from the cleaned signals, and Event-Related Spectral Perturbation (ERSP) plots were extracted. A customized deep learning model was employed to classify the ERSP plots, distinguishing PD patients from healthy controls. The deep learning model achieved an accuracy of 94.64% in separating PD patients from healthy controls. The approach demonstrated robustness against common EEG artifacts, ensuring reliable PD detection. The model's architecture was specifically designed to handle the complexities of EEG data, making it a powerful tool for PD classifications. This study highlights the potential of integrating deep learning with EEG analysis to explore PD diagnosis. The proposed method is faster and more accurate than traditional approaches, enabling early detection and timely intervention. By reducing the time required for analysis and enhancing diagnostic accuracy, this approach can significantly improve patient outcomes and support better management of Parkinson's Disease
The Impact of Deep Learning in Brain Tumour Analysis
No full textThe need for early and precise identification of abnormalities has made the detection and classification of brain tumours essential components of medical diagnosis. Because brain tumours are naturally complex and can have a wide range of sizes, shapes, and types, conventional diagnostic techniques like MRI interpretation and manual evaluations are difficult and time-consuming. Traditional methods frequently depend on human expertise, which is prone to errors, delays, and variability. Deep learning (DL) developments, on the other hand, have completely changed this field by providing increased automation, efficiency, and precision in tumour detection and classification because they can automatically extract pertinent features from MRI scans, Convolutional Neural Networks (CNNs) have shown impressive success in medical image analysis in recent years. CNNs improve the classification of tumour types like gliomas, meningiomas, and pituitary tumours by using multiple layers to find patterns in imaging data. Despite their efficiency, CNNs sometimes struggle with complex tumour patterns, requiring further enhancement in feature extraction. Vision Transformers (ViTs) have become a viable substitute to overcome this constraint. ViTs are especially good at identifying complex tumour structures because, in contrast to CNNs, they use self-attention mechanisms to capture global image dependencies. ViTs can perform better diagnostics by more thoroughly analysing entire MRI images. Additionally, hybrid methods that combine CNNs and ViTs have demonstrated better outcomes, taking advantage of both long-range spatial understanding (ViTs) and local feature extraction (CNNs). These developments allow for real-time medical applications, drastically improve diagnostic accuracy, and lower false positives. Neuro-oncology could undergo a revolution with the incorporation of DL models into clinical workflows, which would improve tumour detection's accuracy, speed, and accessibility. These techniques will be further developed in future studies, guaranteeing even higher accuracy and versatility in medical imaging
Improving Culinary Informatics through Meaningful Social Web Engineering
No full textThis paper gives a good reflection about social web engineering aspirations to ardently see the truth to enlighten culinary concentration and wisdom. With the compassionate path to discover food and engage customers to share content with loving kindness spread to every corner. There are so many features realized, which include direct interfaces through intuitive interfaces to lead the daily food streamlining. May one engaged with constructive creation to grow compassions in every culinary living sharing. Contentment would develop in our heart involved in different phases put forward in this paper. The initial phase begins with honesty in planning and design. Next, the right livelihood of background study brings about requirements definition. This cast out diligently the interactive user conceptualization to arouse wholesome of good and wise design diagrams. With this, the users can reach the goal of registering and establishing profiles. This practice leads to steady mindful culinary owners as well as food enthusiasts. Watching the different choices in food, this social web engineering model delves to equanimity and attention to seek the right locality of certain cuisines. In addition, the model takes the opportunities for the effort to share fine speech with ratings and photos uploaded to assist in deciding dining choices. The peaceful location, especially with good healthy promotions, would help the culinary businesses to start the presentation of menu and events from the beginning. This research moves on to unlock the healthy choice filled with exploratory hats to make every day fantastic learning adventures to ready and learn social engineering guidance. In conclusion, the mobile mechanism can help to present important pieces of advice to learn and practice useful culinary informatics
Training the Brain: A Machine Learning Approach to Predicting Wellbeing Through Intentional Thought Pattern Modification
No full textThis study provides a quantitative framework for wellbeing outcome prediction through intentional cognitive pattern alteration. We demonstrated 81.67% accurate prediction of wellbeing states, in a three-level classification (Low, Medium, High), using a Random Forest classifier with 16 features from psychological, physiological, and behavioural metrics. Our model singles out the gratitude cultivation (21.3%) and peace duration (23.7%) as the strongest predictors of positive well-being outcomes, which provides empirical support to traditional approaches of cognitive training with empirical evidence. Analysis of 1,000 synthetic cases shows that consistent practice of positive thought patterns over 3-6 months can strongly shift wellbeing states, with key behavioural markers showing progressive improvement which include increased joy moments, reduced anxiety episodes, and enhanced sleep quality. Our results establish that cognitive training outcomes can be quantitatively tracked and predicted with meaningful accuracy, hence providing a data-driven approach to mental health intervention design. Additionally, the research shows machine learning for mental health analysis to present a scalable method for wellbeing prediction. Integrating multiple data modalities, our model presents an integrative view of cognitive transformation that covers the gap between qualitative opinion and quantitative prediction. The contribution of this research is in presenting the viability of applying artificial intelligence (AI) models to facilitate enhanced mental health interventions through adaptive and personalized cognitive training programs. More generally, our results add to the emerging science of neuroplasticity-based cognitive training by delivering an evidence-based method for evaluating and predicting wellbeing improvement. The findings have implications that reach outside the research clinic, to clinical interventions, self-help programs, and mobile phone health applications, to offer a new mechanism for improving mental resilience and world life satisfaction through rigorous cognitive training
Analysis of Forensic Disk Imaging Tools for Data Acquisition and Preservation
No full textThe identification, preservation, analysis, and presentation of electronic evidence to support legal or organizational inquiries constitute the discipline of digital forensics, which is crucial to contemporary investigations. A crucial component of forensic inquiry, disk imaging guarantees precision, dependability, and legal defensibility. To preserve the original evidence, disk imaging makes an identical, bit-by-bit duplicate of a digital storage device, capturing hidden data, deleted material, and active files. Given the critical role of disk imaging in forensic investigations, selecting the right tool is crucial for accuracy, efficiency, and compliance with forensic standards. This study assesses widely used tools, including AccessData FTK Imager, Guymager, X-Ways Forensics, OSForensics, and FTK Imager, to help researchers and industry professionals choose the most suitable option for their investigative needs. This research examines the usability, imaging speed, supported hashing techniques, supported output formats, and other aspects of each tool to assess their suitability for usage in various forensic scenarios. The shows that X-Ways Forensic is among the greatest imaging tools because of its wide range of supported operations, fast imaging speed, and format compatibility. The result of hash verification, perfectly matched with source data, again establishes the capability of AccessData FTK Imager, FTK Imager, Guymager, X-Ways Forensics, and OS Forensics to ensure forensic soundness. Its capability to generate a detailed report with comprehensive drive geometry and file segmentation establishes its applicability in forensic workflows. Besides, the time consumed for processing shows its applicability in time-critical investigations too
Enhancing Zero Trust Cybersecurity using Machine Learning and Deep Learning Approaches
No full textThe recent Zero-Trust Architecture (ZTA) is progressively adopted to the develop network security by assuming no implicit trust within or outside an organization’s boundary. Though, ZTA faces substantial challenges in detecting sophisticated and developing cyber threats, particularly due to its trust on traditional security mechanisms that struggle to manage internal threats and sophisticated attack techniques. To report these shortcomings, the proposed study discovers the combination of advanced machine learning (ML) and deep learning (DL) performances to improve the anomaly detection proficiencies within ZTA environments. The study develops the CICIDS2017 dataset, which contains diverse and realistic network traffic patterns, to assess the efficiency of nine different models: Naïve Bayes, Logistic Regression, Random Forest, Decision Tree, Gated Recurrent Unit (GRU), Multi-layer Perceptron (MLP), Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory (Bi-LSTM), and Convolutional Neural Network (CNN). Concluded comprehensive investigation and performance evaluation, the study validates that ensemble methods such as Random Forest and Decision Tree, together with deep learning models like LSTM and GRU, significantly exceed conventional models in terms of accuracy and detection abilities. The best-performing models attained up to 99.99% accuracy in recognizing malicious network activity. This exceptional performance validates that the strong potential of participating intelligent learning-based methods into ZTA to create scalable and dynamic security solutions with high accuracy. These findings illustrate the value of ML/DL in enhancing the threat detection layer of ZTA, eventually providing a stronger resistance to advanced attacks cyber threats
Optimising Phishing Detection: A Comparative Analysis of Machine Learning Methods with Feature Selection
No full textPhishing is an act of cybersecurity attack that tricks people into sharing sensitive data. Due to the inefficiency of the current security technologies, researchers have been paying much attention to employing machine learning methods for phishing detection lately. In our proposed solution, the effectiveness of machine learning techniques with feature selection techniques for phishing detection is investigated. To be specific, Random Forest (RF) and Artificial Neural Network (ANN) are integrated with feature selection techniques, Principal Component Analysis (PCA) and Recursive Feature Elimination (RFE). The goal was to identify and classify the model with the highest accuracy. The experiments were evaluated using a dataset of 4,898 phishing sites and 6,157 legitimate sites, with the phishing data sourced from Kaggle.com. Our experiments demonstrate that the combination of RF model with PCA achieved 95.83% accuracy, while the ANN model with PCA reached 95.07% accuracy. The incorporation of PCA and RFE not only optimised the models' predictive performance but also improved computational efficiency. Overfitting can also be reduced. The experimental results also demonstrate that the proposed ANN with PCA method outperforms the state-of-the-art methods. Consequently, this research highlights the potential of combining advanced feature selection techniques with machine learning algorithms to develop robust solutions for phishing detection. Yet, this undoubtedly contributes to a safer internet environment
DebugProGrade: Improving Automated Assessment of Coding Assignments with a Focus on Debugging
No full textIn education, the evaluation of programming assignments is challenging, especially to do with the debugging aspect. Self-grading technologies are unable to capture the level of understanding of students and context-bound responses. In light of these, we created DebugProGrade to take what we normally know about grading and improve it with semantic analysis and keyword extraction. DebugProGrade identified 1000 first-year BCA students who in Google Forms provided their answers to evaluate error detection and solution proposals for a basic C programming assignment. For the explanations’ specificity and for the context-level evaluation, the system employs the SBERT embedding, namely, the sentence-transformer bidirectional encoding representations from transformers. We employ the methods with tuned parameters and apply academic criteria to the evaluations performed by them. Other key functionality in DebugProGrade that should be mentioned is the classification of debugging skills into competence levels providing more comprehensive view of student proficiency regarding bugs which remain unaddressed by traditional grading systems – that is the ability to identify or fix bugs. Upon optimizing the Gradient Boosting Regressor algorithm, it gives outstanding results in terms of evaluating and predicting redshift. The mean squared error is very low with a value of MSE = 0.025107 and the MAE is also quite low with the value 0.031335, overall the high R² score 0.99932 shows that the given dataset has been predicted with high accuracy with reference to the target variable. DebugProGrade precisely flips the paradigm of conventional grading and provides us with even greater understanding of where exactly students are strong
Social Engineering Threat Analysis Using Large-Scale Synthetic Data
No full textWe frequently hear news about compromised systems, virus attacks, spam emails, stolen bank account numbers, and loss of money. Safeguarding and protecting digital assets against these and other cyber-attacks are extremely important in our digital connected world today. Many organizations spend substantial amounts of money to protect their digital assets. One type of cyber threat that is rampant these days is social engineering attacks that work on human psychology. These attacks typically persuade, convince, trick and threaten naïve and innocent individuals to divulge sensitive information to the attackers. Consequently, traditional approaches have not been effective or successful in preventing these attack types. In this paper, we propose a machine learning model to detect these types of threats. The model is trained using a large synthetic dataset of 10,000 samples to simulate various types of real-world social engineering threats such as phishing, spear phishing, whaling, vishing, smishing, baiting, and pretexting. Our analysis on attack types, patterns, and characteristics revealed interesting insights. Our model achieved an accuracy of 0.8984 and an F1 score of 0.9253, demonstrating its effectiveness in detecting social engineering attacks. The use of synthetic data overcomes the problem of lack of availability of real-world data due to privacy issues, and is demonstrated in this work to be safe, scalable, ethics friendly and effective