Browsing by Author "Isiaka, M. Rafiu"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    A Comparative Analysis of Feature Selection and Feature Extraction Models for Classifying Microarray Dataset
    (Gale Academic OneFile, 2018-10-10) Arowolo M. Olaolu; Abdulsalam, O. Sulaiman; Isiaka, M. Rafiu; Gbolagade A. Kazeem
    The purpose of this research is to apply dimensionality reduction methods to fetch out the smallest set of genes that contributes to the efficient performance of classification algorithms in microarray data. Design/Methodology/Approach: Using colon cancer microarray dataset, One-Way- Analysis of Variance is used as a feature selection dimensionality reduction technique, due to its robustness and efficiency to select relevant information in a high-dimension of colon cancer microarray dataset. Principal Component Analysis (PCA) and Partial Least Square (PLS) are used as feature extraction techniques, by projecting the reduced high-dimensional data into efficient low-dimensional space. The classification capability of colon cancer datasets is carried out using a good classifier such as Support Vector Machine (SVM). The study is analyzed using MATLAB 2015. Findings: The study obtained high accuracies and the performances of the dimension reduction techniques used are compared. The PLS-Based attained 95% accuracy having edge over the other dimension reduction methods (One-Way- ANOVA and PCA). Practical Implications: The major implication of this research is getting the local dataset in the environments which lead to the usage of an open resource dataset. Originality: This study gives an insight and implications of high dimensional data in microarray gene analysis. The application of dimensionality reduction helps in fetching out irrelevant information that halts the performance of a microarray data technology. Keywords: Dimension Reduction, One-Way-ANOVA, PCA, PLS, Classification. Paper Type: Research work
  • Thumbnail Image
    Item
    An Enhanced Differential Homomorphic Model using N-Prime Scheme for Privacy Preservation
    (FUOYE Journal of Engineering and Technology, 2023-03-20) Sanni, O. Hafiz; Isiaka, M. Rafiu; Babatunde, N. Akinbowale; Jimoh, K. Muhammed
    In preserving individual privacy in data publishing, several efforts have been made by scholars globally to develop an individual privacy-preserving model and hybridized models which harness the strength of the individual model to increase privacy preservation in data Publishing (PPDP). The Differential homomorphic model (DHM) was among the hybridized models developed that combine differential and homomorphic models. Though is one of the state-of-the-art hybridization methods for privacy preservation because of the Differential model and Homomorphic model strengths of the two hybridized models which are the ability to prevent composition problems and database attacks respectively. However, applying this model is challenging because of the high computational complexities due to the modular exponentiation problem available in the pailler encryption scheme used in DHM. In this research, an N-PRIME homomorphic encryption scheme was proposed to replace the Pailler encryption scheme in the differential homomorphic model (DHM). The designed model was 51% faster than the existing model (Differential Homomorphic Model) in terms of computation time and 48.5% faster when generating the graphical data set, though the designed model consumed 4% more storage space than the existing model.
  • Thumbnail Image
    Item
    Diacritic Restoration for Yoruba Text with under dot and Diacritic Marks Based on LSTM
    (FUOYE Journal of Engineering and Technology, 2023-09-18) Ogheneruemu, L. Kingsley; Ajao, F. Jumoke; Isiaka, M. Rafiu; Asahiah, O. Franklin; Orimogunje, K. Olumide
    Yoruba is a tonal language spoken primarily in Nigeria, some West African countries, and other parts of the world by over 40 million people. Many Yoruba texts written online lack tone marks, which can be confusing, ambiguous, and difficult for Natural Language Processing. This paper presents a method, which combines syllable-based approach and long short-term memory (LSTM) for diacritics restoration of standard Yoruba text. By enhancing the built-in varnishing gradient of RNN, the aim is intended to recover lost diacritics in Yoruba text for both characters that carry diacritic signs and underdot and return it with the proper diacritics. Data were acquired from Yoglobavoice, BBC Yoruba new and Yoruba words collected from literate indigenous writers. 27050 Yoglobalvoice datasets, 2000 Yoruba words extracted from BBC Yoruba news, and 1470 Yoruba words collected from a Yoruba language teacher. In addition, syllabic module was developed to group the tokenized word into different syllables. The output of the syllabication algorithm was fed into the Long Short-Term Memory (LSTM) module for training, the LSTM model was trained using 70% of the dataset and validated using 30% of the dataset. The result obtained showed 96% accuracy. From the result, it was observed that the use of LSTM for restoring diacritic gave an improved restoration of both character with under dot and character that contains tone-marks.
  • Thumbnail Image
    Item
    Phishing Websites Detection based on Meta heuristic Optimization and Ensemble Algorithms
    (International Journal of Security and Networks, 2025-07-14) Adewole, S. Kayode; Hambali, A. Moshood; Saheed, Yakub; Adeyemo, E. Victor; Jimoh, M. Kamaldeen; Isiaka, M. Rafiu
    Phishing attacks pose a significant threat to web- and mobile-based applications, aiming to compromise network security and hijack legitimate resources. Illegitimate websites, fake emails, and deceptive web/mobile solutions are used to collect user credentials and compromise accounts. Traditional blacklist-based methods for detecting phishing websites suffer from a lack of zero-day attack detection. To address this, machine learning (ML) models have been widely explored. However, meta-heuristic approach that covers studies on feature selection to create compact ML models for phishing detection has not been extensively studied. This paper fills that gap by analyzing ten meta-heuristic algorithms to identify discriminating features for detecting phishing websites. The study evaluates three ensemble-based classifiers - Random Forest, AdaBoost, and Rotation Forest. The results show that Firefly Optimization Algorithm (FOA) produced the best results. Using the features selected by FOA, Random Forest outperforms other ML algorithms with 95.36% accuracy and AUC-ROC of 98.9%. This result is followed by Rotation Forest (accuracy 94.89%) and AdaBoost performs the least with accuracy of 91.42%. The findings highlight the efficacy of the FOA meta-heuristic algorithm in phishing website detection.