International Journal of Software Engineering and Computer Systems

PERFORMANCE ANALYSIS OF SELECTED CLASSIFICATION ALGORITHMS ON ANDROID MALWARE DETECTION

Ganiyat Kemi Afolabi-Yusuf — 2024-03-17

Android mobile devices are widely used across all platforms and the development of malicious apps can compromise a user’s mobile system. Considering the large amount of new malicious apps, there is a need for a detection system that can operate efficiently to identify these apps. The study analyzes and compares the performance of DREBIN and MALGENOME data sets with the dataset’s SMOTE version on selected machine learning algorithms using WEKA tools. The performance of bayesian, function, rule, and tree-based classification algorithms on the two datasets was explored in this work. WEKA tool was used in pre-processing and SMOTE class balancing of the datasets before the model training using different classification algorithms on the two datasets and the performance evaluation. In the performance evaluation, parameters such as accuracy, precision, f-measure, the area under cover, true positive, recall, and false positive rate were employed. According to the study, tree-based classifiers (Recursive Tree, Decision Tree and Classification and Regression Tree) algorithms have 97.24%, 98.21% and 98.21% accuracy on the Malgenome dataset and 97.30%, 97.33% & 97.28% of accuracy on Drebin dataset and functionbased classifiers (Support Vector Machine (SVM) and Logistic Regression) algorithms has 97.81% & 96.87% of accuracy on Malgenome dataset and 97.00% & 97.81% of accuracy on Drebin dataset which concludes that classifier algorithms in these groups proofed to be promising for the detection of android malware. The function-based classifier is the most outstanding method for the two datasets as it outperforms all other classifiers for both classes with 97.81% and 97.33%. SVM and Logistic Regression, are highly effective in detecting malicious Android apps, outperforming other classifier types with accuracy rates up to 97.81%. Tree-based classifiers also showed strong performance across DREBIN and MALGENOME datasets. This research underscores the potential of function-based algorithms as robust tools for enhancing mobile security against malware threats.

OPTIMAL SELECTION OF THE CLUSTER HEAD IN WIRELESS SENSOR NETWORKS BY COMBINING PARTICLE SWARM OPTIMIZATION AND EFFICIENT GENETIC ALGORITHM

Aliyu Zakariyya — 2024-01-11

Wireless Sensor Networks (WSNs) have become a crucial component of numerous applications, including the military, healthcare, and environmental monitoring. A promising approach to increasing the lifespan of the sensor network is cluster-based WSNs. In WSNs, choosing the best cluster head is a crucial task that has an impact on the network's performance and energy efficiency. There are various issues with current methods for choosing the cluster head, including nodes dying too soon, uneven energy usage, and shorter network lifetimes. Moreover, traditional methods such as Randomized Clustering and Fixed Cluster Head are not effective in prolonging the network lifetime as they do not consider the energy consumption and residual energy of nodes. In this paper, an optimal selection of cluster head is presented where we combine the Particle Swarm Optimization (PSO) and Efficient Genetic Algorithm (EGA). Firstly, PSO is used to randomly select the cluster head and update the position of each cluster. Thereafter, EGA invokes its fitness values to select the best cluster head that transmits information to the base station. The simulation result shows that the performance improvement of the proposed method PSO_EGA in terms of network lifetime is 0.10% against Improve Cuckoo Search Algorithm (ICSA) and 0.20% against Hybrid Crow Search Algorithm (HCSA), packet to cluester head is 7% against ICSA and 16% against HCSA, packet to sink is 11% against ICSA and 22% against HCSA and number of alive node is 28% against ICSA and 48% against HCSA. Therefore, our proposed method outperforms ICSA and HCSA in terms of the aforementioned parameters.

The MOBILE AUGMENTED REALITY APPLICATION FOR IMPROVING LEARNING OF ELECTRONIC COMPONENT MODULE IN TVET

Nalienaa Muthu — 2023-07-20

Teens and young adults may get training in anything from the basics to advanced skills in various workplace and academic settings at Technical and Vocational Education Training and Education (TVET) institutions. Some aspects of teaching and learning in TVET cannot be articulated clearly, and trainees cannot perceive how things fit together. The study was conducted to determine the optimal platform to develop mobile Augmented Reality applications for TVET trainees and, to assess the TVET trainee’s readiness for AR-based mobile application training deployment. An online questionnaire was sent to trainees at Industrial Training Institute in Malaysia via the online system. A marker-based Augmented Reality application was created for the Basic Electronic Components module utilizing Unity software, the Vuforia engine, and C# script. Finally, the trainees were allowed to test the generated application. The trainees were interviewed to obtain data on their responses. The results indicate that 83% of the TVET trainees own and use android as the application platform. The results of the pre-test and post-tests used to gauge the success of the Augmented Reality application show that its usage in the sub-learning module significantly improved memory recalls for the TVET trainees. The outcomes showed that the Augmented Reality application suited the participants' learning needs and improved the effectiveness of their learning. The result from this project will serve as a pre-test for determining the most suitable platform to deploy the Augmented Reality application to be developed in the future.

SENTIMENT CLASSIFICATION OF TWEETS WITH EXPLICIT WORD NEGATIONS AND EMOJI USING DEEP LEARNING

Mdurvwa Usiju Ijairi — 2023-07-20

The widespread use of social media platforms such as Twitter, Instagram, Facebook, and LinkedIn have had a huge impact on daily human interactions and decision-making. Owing to Twitter's widespread acceptance, users can express their opinions/sentiments on nearly any issue, ranging from public opinion, a product/service, to even a specific group of people. Sharing these opinions/sentiments results in a massive production of user content known as tweets, which can be assessed to generate new knowledge. Corporate insights, government policy formation, decision-making, and brand identity monitoring all benefit from analyzing the opinions/sentiments expressed in these tweets. Even though several techniques have been created to analyze user sentiments from tweets, social media engagements include negation words and emoji elements that, if not properly pre-processed, would result in misclassification. The majority of available pre-processing techniques rely on clean data and machine learning algorithms to annotate sentiment in unlabeled texts. In this study, we propose a text pre-processing approach that takes into consideration negation words and emoji characteristics in text data by translating these features into single contextual words in tweets to minimize context loss. The proposed preprocessor was evaluated on benchmark Twitter datasets using four deep learning algorithms: Long Short-Term Memory (LSTM), Recurrent Neural Network (RNN), and Artificial Neural Network (ANN). The results showed that LSTM performed better than the approaches already discussed in the literature, with an accuracy of 96.36%, 88.41%, and 95.39%. The findings also suggest that pre-processing information like emoji and explicit word negations aids in the preservation of sentimental information. This appears to be the first study to classify sentiments in tweets while accounting for both explicit word negation conversion and emoji translation.

PROTOCOL EFFICIENCY USING MULTIPLE LEVEL ENCODING IN QUANTUM SECURE DIRECT COMMUNICATION PROTOCOL

Nur Syuhada Mohamad Rodzi — 2023-10-18

One of the objectives of information security is to maintain the confidentiality and integrity of the information by ensuring that information is transferred in a way that is secure from any listener or attacker. There was no comparison experiment conducted in earlier studies regarding different level encoding performance towards multiphoton technique. Multiphoton technique in the earlier study is particular to transmission time for data transfer encoding and extra time for polarizers to change polarisation angles, both of which contribute to longer transmission times. With four different size of qubits, the three simulation experiments are carried out using Python coding with 2,4 and 8 levels of encoding. Experiment results demonstrate that the most efficient average photon transmission derived from 18 qubit size ranges from 98.71% to 98.73% depending on encoding level. With 18 qubit size, the four-level encoding result has the highest average efficiency, followed by the eight-level and two-level encodings, respectively. 4-level encoding exhibits the highest average photon efficiency between 2 and 8 level encoding.

SECURING IOT HEALTHCARE APPLICATIONS AND BLOCKCHAIN: ADDRESSING SECURITY ATTACKS

MUHAMMAD UMAR DIGINSA — 2023-12-04

The Internet of Things (IoT) describes the connection of bodily devices as "things" that can communicate with other systems and devices through the Internet and exchange statistics (data or information), facilitating the exchange of data with other systems and devices. These devices have sensors, software, and various components designed to exchange data seamlessly within the IoT network. Securing and protecting the data transmitted over the Internet from unauthorized access is imperative to ensuring the integrity and confidentiality of the information. IoT Smart health monitoring systems, integral components of the IoT landscape, are susceptible to various attacks. These include denial of service (DoS), fingerprint, router, select, forwarding, sensor, and replay attacks, all of which pose significant threats to the security of these systems. As such, there is a pressing need to address and mitigate the vulnerabilities associated with IoT healthcare applications. This paper aims to explore the significant role of IoT devices in healthcare systems and provide an in-depth review of attacks that threaten the security of IoT healthcare applications. The study analyses the existing literature on the vulnerabilities present in smart health monitoring systems and the potential application of blockchain technology as a robust solution to enhance the security of IoT healthcare applications. This research reveals critical vulnerabilities in IoT healthcare applications and highlights blockchain's effectiveness in mitigating them, providing insights for robust security measures and strategic decision-making in secure healthcare systems. This paper provides valuable insight and recommendations for policymakers, researchers, and practitioners involved in the domain of the IoT healthcare system.

A SYSTEMATIC REVIEW ON ANDROID-BASED PLATFORM FOR SMART INVENTORY SYSTEM

Noorihan Abdul Rahman — 2023-07-20

Inventory tracking is one of the most crucial aspects in business strategy. Effective inventory system can help the prevention of stockouts, effective management of different locations, as well as the maintenance of accurate records in a business. Nowadays, digitalization is a critical component of business operations. Digitalization is the process of implementing new digital technology into all aspects of a company's operations, resulting in a significant change in how the business operates. A systematic mapping has been performed on Android-based for smart inventory system by using digitalized technology which is barcoding technology. The mapping are done by conducting systematic mapping process for analyzing related research areas on barcode and inventory system. Two research questions and related keywords are initiated for identifying possible operating system platforms in developing a smart inventory system with barcoding technology for tracking product items.