Scholarly Publication
Permanent URI for this collection
Browse
Recent Submissions
- ItemDevelopment of a Multi-Level Data Encryption Standard with Residue Number System for Data Security(Faculty of Communication and Information Sciences, University of Ilorin, Ilorin, Nigeria - Proceedings of the 3rd International Conference on ICT for National Development and its Sustainability, May 21-24, 2024 (ICT4NDS2024), 2024) Adebayo, A.; Adeniyi, A.E.; Ajao, J.; Isiaka, R.; Gbolagade, K.; Abdulsalam, S.Data security is critical in ensuring the privacy of data including sensitive material that ought to only be known by a few people. Every society needs secured data to maintain the integrity and authentication of the data. Data Encryption Standard (DES) is a block cipher algorithm that has been used to secure data or information over the years. Despite the tremendous efforts made by researchers on DES algorithm and efforts to reduce its computational complexity, DES is still susceptible to brute force attack. The need to increase the degree of security of DES algorithm led to the introduction of Residue Number System (RNS) to the DES algorithm as proposed in this study. The method for DES-RNS multilevel encryption uses a 64bits plaintext message which was encrypted using the DES technique. The 64bits plaintext message was divided into two equal halves; 32bit left plaintext (LPT) and 32bit right plaintext (RPT). The RPT was encrypted using 48bit sub-keys and the result was XORed with LPT. The transformation of RPT and LPT was performed for sixteen (16) rounds to produce encrypted text of 64bits. The encrypted text of DES was converted to American Standard Code for Information Interchange (ASCII) and passed through RNS forward conversion. The RNS made use of the moduli set 〖m_1=2 〗^n+1,m_2= 2^n and m_3=2^n-1. The decryption was performed using Chinese Remainder Theorem (CRT). The result was evaluated when it comes to cryptographic time, encryption/decryption memory, encryption/decryption throughput and security on three varying text sizes (256, 800 and 1472 bit) for DES only and DES-RNS multilevel cyrptosystem. Using DES, time and throughput shows a better performance for 256bit, 800bit and 1472bit message size but lesser performance in memory and security for 256bit, 800bit and 1472bit message size. On the other hand, using DES-RNS, time and throughput gives a lesser performance for 256bit, 800bit and 1472bit message size. Therefore, DES-RNS multilevel encryption model outperformed the conventional DES model in regard to storage utilization and safety, thereby achieving the aim of this research. Consequently, this DES-RNS model can be employed where security and memory conservation is of utmost concern. Keywords: Cryptography, Block Cipher, Data Encryption Standard (DES), Residue Number System (RNS)
- ItemFrequent Pattern and Association Rule Mining from Inventory Database Using Apriori Algorithm(Institute of Electrical & Electronics Engineers (IEEE) - African Journal of Computing and ICT, A Journal of the Institute of Electrical & Electronics Engineers (IEEE) Computer Chapter Nigeria Section. 7(3), 35 – 42, 2014) Adewole, K.S.; Akintola, A.G.; Ajiboye, A.R.; Abdulsalam, S.O.Recently, data mining has attracted a great deal of attention in the information industry and in a Society where data continue to grow on a daily basis. The availability of huge amounts of data and the imminent need for turning such data into useful information and knowledge is the major focus of data mining. The information and knowledge obtained from large data can be used for applications ranging from market analysis, fraud detection, production control, customer retention, and science exploration. A record in such data typically consists of the transaction date and the items bought in the transaction. Successful organizations view such databases as important pieces of the marketing infrastructure. This paper considers the problem of mining association rules between items in a large database of sales transactions in order to understand customer-buying habits for the purpose of improving sales. Apriori algorithm was used for generating strong rules from inventory database. It was found that for a transactional database where many transaction items are repeated many times as a superset in that type of database, Apriori is suited for mining frequent itemsets. The algorithm was implemented using PHP, and MySQL database management system was used for storing the inventory data. The algorithm produces frequent itemsets completely and generates the accurate strong rules. Keywords: Apriori Algorithm, data mining, database, strong rules & inventory
- ItemPerformance Evaluation of ANOVA and RFE Algorithms for Classifying Microarray Dataset Using SVM(Springer Nature Switzerland, Lecture Notes in Business Information Processing 402: 480 – 492 - Proceedings of 17th European, Mediterranean, and Middle Eastern Conference, EMCIS 2020, Dubai, United Arab Emirates, November 25–26, 2020, 2020) Abdulsalam, S.O.; Abubakar, A.M.; Ajao, J.F.; Babatunde, R.S.; Ogundokun, R.O.; Nnodim, C.T.; Arowolo, M.O.A significant application of microarray gene expression data is the classification and prediction of biological models. An essential component of data analysis is dimension reduction. This study presents a comparison study on a reduced data using Analysis of Variance (ANOVA) and Recursive Feature Elimination (RFE) feature selection dimension reduction techniques, and evaluates the relative performance evaluation of classification procedures of Support Vector Machine (SVM) classification technique. In this study, an accuracy and computational performance metrics of the processes were carried out on a microarray colon cancer dataset for classification, SVM-RFE achieved 93% compared to ANOVA with 87% accuracy in the classification output result. Keywords: SVM-RFE; ANOVA; Microarray; SVM; Cancer
- ItemPredicting Drug Addiction Using Ensemble Learning: A Majority Voting Classifier Approach(Faculty of Information and Communication Technology, Kwara State University, Malete, Nigeria - KWASU Journal of Information, Communication and Technology, 3(1), 17 – 25, 2024) Akuji, I.I.; Ogunrinde, M.A.; Ahmed, T.O.; Abdulsalam, S.O.; Babatunde R.S.Globally, drug intake has claimed the lives of half a million people due to its complexity and devastating effects. Prompt intervention and prevention are crucial to curbing the fatalities and mortality caused by drug addiction. Leveraging machine learning for drug intake prediction and building on the existing research becomes imperative to reinforcing the effectiveness of machine learning utilisation, ultimately helping to curb the prevalence of drug intake. Therefore, this study proposes an ensemble learning approach, using a majority voting classifier to predict drug addiction from the analysis of facial images. The choice of the majority voting classifier over other ensemble methods, such as boosting, bagging, and stacking, is due to its simplicity, interpretability, and reduced potential for overfitting, as it averages out individual model predictions to produce more robust outcomes. A dataset containing 400 facial images was acquired from the Mendeley repository, comprising 240 addicted and 160 non-addicted images. Combining the predictions of multiple base models, namely ResNet, VGG, and Inception models, this study’s approach ensures improved accuracy and robustness of the drug addiction prediction model. The Majority Voting classifier model was evaluated on the acquired dataset, achieving a 0.6573 loss and 69% accuracy rate on the training set and 2.8133 loss and 60% accuracy on the validation set. The discrepancy between training and validation performances indicates that the model is overfitting to the training data. As a result, further study can address the overfitting issue by considering data augmentation to ensure the model performs well in real-world scenarios. Keywords: Drug addiction, Facial image analysis, Ensemble learning, Majority Voting classifier, Machine learning
- ItemTowards a New Hybrid Synthetic Minority Oversampling Technique for Imbalanced Problem in Software Defect Prediction(IEEE Xplore - Proceedings of 5th International Conference on Data Analytics for Business and Industry (ICDABI), University of Bahrain, October 23-24, 2024, 2024) Saheed, Y.K.; Abdulsalam, S.O.; Ibrahim, M.B.; Baba, U.A.The software industry strives to improve software quality through continuous bug prediction, bug elimination, and module fault prediction. This issue has piqued researchers’ interest because of its significant relevance in the software industry. Frequently, Software Defect Prediction (SDP) models contain significantly skewed data, making it difficult for classifiers to recognize defective occurrences. The machine learning (ML) community has put a lot of effort into solving the problem of learning from imbalanced SDP data, though less so in empirical software engineering. The over-sampling strategy is one of many recent solutions to this problem. This strategy balances the number of defective and non-defective cases by creating new defective instances. Unfortunately, these methods would result in non-diverse synthetic instances as well as a large number of unneeded noise instances, creating an imbalanced class problem. As a result, we propose a hybrid synthetic minority oversampling (HSMOTE) to address the problem of imbalance in SDP. In this paper, we introduce the hybrid Synthetic Minority Oversampling Technique (HSMOTE), a method that utilizes Extra Tree, Random Forest (RF), and Extreme Gradient Boosting (Xgboost) for classification. We develop and deploy the proposed method on the National Aeronautics and Space Administration (NASA) dataset, evaluating its performance on three datasets: JM1, KC1, and PC3. We compared the parameters for accuracy, precision, AUC, recall, F-measure, and Mathew Correlation Coefficient to those of the existing SDP. The findings from the simulations on the JM1 data showed that the proposed techniques work better than the current best models. The proposed SMOTE+RF technique surpasses the existing techniques with an accuracy of 93.69%, an AUC of 82.70%, and an F-measure of 32.98%. Similarly, the proposed SMOTE+Xgboost method outperforms the existing techniques with an accuracy of 93.432%, an AUC of 82.64%, and an F-measure of 34.13%, while SMOTE+ET achieved an accuracy of 93.43%, an AUC of 77.68%, and an F-measure of 31.90%. Keywords—Synthetic Minority Oversampling Technique, Class Imbalance, Oversampling Method, Software Defect Prediction, Imbalance Data