LAUTECH JOURNAL OF COMPUTING AND INFORMATICS

A COMPREHENSIVE REVIEW ON MACHINE LEARNING TECHNIQUES FOR THE IDENTIFICATION OF RANSOMWARE ATTACKS IN COMPUTER NETWORKS

O.M. AYINLA, A.M. OYELAKIN, J.O. OLOMU — Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

Ransomware attacks have been identified as one of the serious threats in the cyber space. The malware poses serious security challenges to corporate networks and internet users worldwide. In response, several machine learning techniques have gained popularity for the classification of ransomware in the internet space when compared with signature-based approaches. This paper presented a comprehensive review of various studies that focus on the use of machine learning techniques for the identification of ransomware attacks in computer networks. The study collected relevant literature from various research databases by using some specific keywords and search strings that are deeply related to the topic. A good number of literatures that were obtained, were sorted and studied. The literatures were organised in different sections, arranged chronologically from the most recent to relatively older works. The publication years for the reviewed papers ranges from 2017 to 2023. The review began by exploring some relevant concepts and then shifted ground to machine learning algorithms that have been proposed for ransomware attacks identification. Thereafter, the performances of the different learning techniques used for the identification of ransomware attacks in computer networks were reported. The study argued that the review can serve as insights for future researches in this cyber security area.

SMOTE-TomekLink Super-Learner Ensemble Model (STL-SLEM) for the Prediction of Parkinson’s Disease

O.A. AROWOLO, O.J. EMUOYIBOFARHE, E.A. AMUSAN — Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

Parkinson's Disease (PD) is a progressive neurodegenerative disorder that affects millions of people worldwide. Early detection and prediction of Parkinson's Disease can significantly improve patient outcomes by enabling timely intervention and personalized treatment. Over the years, many Parkinson Disease (PD) prediction models have been developed using machine learning algorithms. Some of these existing models suffer over-fitting of data due to unavailability of sufficient dataset in PD as well as data imbalance. Hence, this work developed a Super Learner Ensemble Model (SLEM) that aggregated several machine learning models configurations to overcome the challenge of over-fitting thereby enhancing the performance of PD prediction. The dataset used for this research is Parkinson disease datasets obtained from Kaggle website and also local datasets from Federal Medical Center, Abeokuta, Nigeria for the validation of the developed model. The dataset from Kaggle website consists of 195 biomedical voice measurements from 31 people taken severally, 23 out of the 31 have Parkinson's disease and 8 without Parkinson's disease, while the local datasets consists of 13 people, 9 with PD and 4 without PD. The acquired dataset has class imbalance, and to handle this issue, Synthetic Minority Over Sampling Technique with TomekLink (SMOTE-TomekLink) was adopted to resample the dataset for class-balancing. For computational efficiency, six base learners were used to develop the Super Learner model, which includes Logistic Regression (LR), Decision Tree (DT), Naïve Bayes (NB), Adaptive Boosting (AB), Bagging Ensemble (BE), and Random Forest (RF) algorithms. The performances of each base model were measured, and the performance of the Super Learner ensemble model was also obtained using the following performance metrics: Accuracy, Precision, Recall, F1-Score, Matthews Correlation Coefficient (MCC), and Balanced Accuracy Score (BAS). However, Accuracy for LR, DT, NB, AB, BE, and RF with SMOTE-TomekLink-resampled datasets were 95.0%, 94.0%, 91.0%, 93.0%, 95.0%, and 96.5%, respectively, while the corresponding Accuracy for Super Learner Ensemble model was 99.0%.. The developed model showed an improvement in the performance metrics.

DEVELOPMENT OF A NEAR FIELD COMMUNICATION-BASED ATTENDANCE AND COURSE VERIFICATION SYSTEM

Adebimpe R. AJAYI, Esther T. Olawole , Mark O. AJAYI, H.O. Mahmud — Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

The task of taking class attendance using the conventional paper-based method can be a challenge with increased number of students’ enrolment for a course. The method is time consuming, prone to errors and unfairness in the process. An NFC-based attendance and course verification system is presented in this study. The system reads student’s data from the school ID card when brought in close proximity with an NFC module and sends the data to a web application where the attendance is taken and course(s) is verified as applicable. The technique is simple, cost effective and saves time. The developed system takes less than 3seconds to take a student’s attendance while compared with the manual paper-based method which took an average of 17seconds

Assessment of Users Perception of a English to Yoruba Computer Aided Translation Teaching System (CATTS) for Climate Change Literacy

Ifeoluwani A. Jenyo, Justice O. Emuoyibofarhe, Adebisi A. Baale — Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

Increasing environmental awareness and education on climate change has been identified as potential tools to combat climate change crisis. According to the United Nations Framework Convention on Climate Change (UNFCCC), all parties to the convention are responsible for launching educational and public awareness initiatives on climate change, as well as making sure that the general public accesses programs and information on the subject.

Education, particularly digital education has been identified as a potential tool to combat climate change. However, a very huge percentage of instructional materials and digital contents on the internet on climate change are in English language which is not the native language understood by the majority of the population of Africans. Translation and interpretation are the bridges that make cross-cultural and cross-linguistic communication possible.

In this paper an assessment of user’s perception of a English to Yoruba Computer Aided Translation Teaching System (CATTS) for Climate Change Literacy was carried out. A questionnaire was developed and administered to different categories of respondents who understand both English language and Yoruba language with or without prior knowledge of climate change literacy and green innovation entrepreneurship. After interacting with the developed Computer Aided Translation Teaching System (CATTS) for Climate Change Literacy (CCL) in Yoruba language, the uses were evaluated on the level of translation accuracy of the system on CCL concepts, the system’s reliability, user friendliness, users understandability of CCL and the need to mitigate the activities that contributes to climate change. The system had an overall 91.3% translation accuracy and 87.0% translation error. Statistical analysis of respondent feedback from the survey conducted revealed that users were satisfied with the developed system with a significant majority, 89.0% agreeing that employing the Computer Aided Translation System for translation could enhance the teaching and understanding of Climate Change in Yoruba language, the translation efficiency, usability and reliability of the system were highly rated.

A Kohonen Self Organizing Map (KSOM) Technique for Classification of Electrocardiogram (ECG) Signals

Ronke S. Babatunde, Adigun Oyeranmi — Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

Electrocardiogram (ECG) signals are crucial in diagnosing cardiovascular diseases. Handling noisy ECG data, which is common in real-world situations makes accurate classification a critical task. Because ECG signals are faint and are quickly disrupted, classification accuracy can be poor, hence the need for improvement in the automatic ECG categorization system's recognition accuracy. The Kohonen Self Organising Map (KSOM) is known for its ability to cluster high-dimensional data in a low-dimensional space, hence its adoption in this research. The procedure employed include collection and pre-processing of diverse ECG data, including normal and abnormal cardiac rhythms. Inherent noise was removed from the data to ensure better-quality input data into the classification algorithm. A Kohonen SOM neural network (MiniSom model) was trained using the preprocessed ECG data. The KSOM organizes ECG signals into clusters on a topological map, preserving similarities and dissimilarities between different cardiac rhythms. Subsequently, the trained SOM serves as a reference model for classifying unseen ECG signals, indicating the corresponding cardiac rhythm. Benchmarked on two different dataset, evaluation of the classification performance of the technique was carried out. Cross-validation was done to assess the model's robustness and generalizability. Comparative analysis was conducted to measure the effectiveness and efficiency of the SOM-based approach against other common ECG signal classification techniques based on accuracy, precision, recall and fi-score. The result obtained shows that the average accuracy of 94.2%, precision of 83%, recall of 100% and f1-score of 91% achieved by the MiniSOM model outperformed the other models.

MESH PLOT- BASED ANALYTICAL SELECTION OF OPTIMAL MQ-SERIES GAS SENSOR PARAMETERS

Adebimpe R. AJAYI, Olalekan F. Adebayo, Mark O. Ajayi — Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

Metal Oxide Semiconductor or MOS-Type gas sensors are a type of sensors that can detect presence of some volatile, oxidizable or reducible substances in an environment. The sensitivity of these sensors depend on the selection of the appropriate parameters in terms of the sensor resistance in clean air () and the load resistance value (). A mesh plot analytical method for determing these parameters is presented in this paper. The model equation for MQ-135 sensor as a case study was derived and this was used to generate the mesh surface plot of the various relationships existing between the different parameters leading to the selection of the optimal value that can give a useful sensor output voltage. The mesh surface plot of (Ω), and concentration (ppm) and as well as the plot of, and were critically examined. The result of the simulation shows that, for optimal sensor output in MQ-135 sensor, the value must be selected to be at a value of 30kΩ and the load resistor at a value of 47kΩ for effective sensor output.

DEVELOPMENT OF A PNEUMONIA DETECTION SYSTEM USING CONVOLUTIONAL NEURAL NETWORKS

Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

Pneumonia is one of the world’s most lethal and life-threatening diseases today, affecting people of all ages. Early proactive treatment can significantly reduce the likelihood of death from this illness and also prevent circumstances from worsening. Chest X-rays imaging has been one of the most well-liked and well-known clinical approaches. However, diagnosing the condition using X-rays has become increasingly challenging due to pneumonia resembling other lung disorders. As a result, this study developed a system to detect pneumonia disease in chest X-ray images using convolutional neural network-based approach. The datasets used for this study consists of 5856 chest X-ray images was obtained from Kaggle to train three pre-trained CNN architectures: VGG16, ResNet50, and MobileNetV3. The dataset was cleaned, preprocessed, and divided using the 80:20 data split ratio. Early stopping and learning rate reduction were applied to each model to prevent overfitting. The performance of each model was evaluated using accuracy, precision, recall, and f1-score on the test data. The VGG16 model outperformed others with 94% accuracy, 91% precision, 95% recall, and 93% f1-score. The MobileNetV3 model, which had the second-best performance, had an accuracy of 93%, precision of 90%, recall of 94%, and f1 score of 92%, while ResNet50 had 92% accuracy, 89% precision, 93% recall, and 91% f1-score. The best performing model of the three which is VGG16 was chosen and implemented on a web application. This system will serve as a tool for the medical practioners in detecting pneumonia earlier and accurately for proper treatment.

A Review on Attack Landscape and Machine Learning Techniques for the Classification of Attacks in Internet of Medical Things (IoMT)

J.O. Olomu , A.M. Oyelakin, O.M. Ayinla , H.A. Ibrahim — Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

Healthcare systems globally are struggling to handle the increasing number of patients, partly due to busy work schedules. To address this issue and enhance healthcare services, the Internet of Medical Things (IoMT) is gaining popularity. IoMT refers to internet-connected devices used in healthcare processes. However, the widespread adoption of IoMT devices has led to new security vulnerabilities and cyber threats. Protecting these devices from cyberattacks is vital for patient safety and data integrity. This study focuses on examining trends in cyber-attacks and the use of machine learning for attack classification in the Medical Internet of Things. The research involved a comprehensive analysis of relevant articles written in English between 2016 and 2023. The study established a search strategy and exclusion criteria to identify highly relevant works from reputable research databases. A significant number of papers were carefully chosen, organized, and reviewed. The reviewed articles delve into the threat landscape and assess the strengths and limitations of machine learning-based techniques for classifying security attacks in IoMT systems and networks. This study believes that this review can pave the way for the development of improved machine-learning methods for classifying attacks in the IoMT environment.

A Multi-level Effective Wireless Sensor Networks (WSNs) with Residue Number Systems and Intelligent Multi-Agent Technologies

Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

Wireless Sensor Networks (WSNs) consist of numerous sensor nodes spread across a targeted area, complemented by one or more base stations that monitor environmental and physical conditions. The primary challenges faced by WSNs include ensuring their energy efficiency, speed, and reliability. These networks operate on a limited energy reserve, complicating the process of energy replenishment. Additionally, the critical nature of the data collected from the environment demands that WSNs operate swiftly, reliably, and maintain functionality even in the event of component failures. Hence, designing WSNs to be reliable, energy-efficient, and fault-tolerant is crucial for enhancing their lifespan and overall performance. To address these challenges, the integration of the Residue Number System (RNS) and intelligent multi-agent technologies has been pursued to create WSNs that excel in fault tolerance, energy efficiency, speed, and reliability. Techniques such as the Chinese Remainder Theorem (CRT)-based packet division and the Low-Energy Adaptive Clustering Hierarchy (LEACH) algorithm have been developed to reduce energy consumption significantly. Additionally, the implementation of a High Spread REverse Converter (HISPREC) based on Mixed Radix Conversion for the moduli set {2ⁿ⁺¹-1, 2ⁿ, 2ⁿ-1} facilitates the swift conversion of collected residues back into the original message by the Cluster Head (CH). Moreover, the adoption of intelligent multi-agent technology enables the WSN to monitor and control its operation dynamically, ensuring continuous operation even in the presence of faults and efficiently isolating any nodes compromised by power depletion. This design approach results in a multi-layered, efficient WSN framework that leverages the strengths of RNS and intelligent multi-agent technologies. This strategy is among the pioneering efforts to incorporate these technologies into real-time WSN applications, representing a significant advancement in the design and functionality of WSNs.

A Comparative Study of Two Convolutional Neural Network Models for Detecting Rice Plant Diseases Using Online and Local Image Data

Thu, 11 Apr 2024 00:00:00 +0000

ABSTRACT

Rice is one of the most widely staple foods around the globe, however, rice fields are severely affected by diseases, which can disrupt global food security. Early and accurate detection of rice diseases is essential for the recovery of such rice plants. Manually identifying rice plant diseases can be tedious and error prone. Artificial intelligence (AI) driven models, such as Convolutional Neural Networks (CNN) have proven very successful in the identification or detection of various crop diseases. This study, therefore, presents a comparative study of the effectiveness of two popular CNN architectures; ResNet and AlexNet for the detection of rice plant disease. The data used to train the models include a combination of rice leaf images that were gathered locally from a rice field/farm in Ede, Osun State, Nigeria, and from an online repository. The dataset consists of 5200 images classified into four classes: Bacterial leaf blight, Brown spot, Blast, and Healthy, each containing 1300 images. The effectiveness of the two trained models was measured using classification performance metrics including Accuracy, Precision, Recall, and F1-Score. The finding from the study showed that The ResNet has a test accuracy of 95.25% as against 92.91% for the AlexNet. The ResNet had 0.93 precision, while AlexNet recorded a precision of 0.24. For recall, the ResNet model had 0.98 while the AlexNet model had 0.23 and for the f1-score, the ResNet model had 0.95 while the AlexNet model had 0.24. Generally, the ResNet model outperformed the AlexNet model in detecting rice plant diseases, most significantly, brown spot disease.