Clive Asuai

Dr. Clive Asuai

Computer Science, Delta State University, Abraka, Nigeria
orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 December 2023 in Journal of Computing Theories and Applications
Verified
Journal Article
Open access
PDF file
MAMaureen Ifeanyi Akazue
Maureen Ifeanyi Akazue
IDIrene Alamarefa Debekeme
Irene Alamarefa Debekeme
AEAbel Efe Edje
Abel Efe Edje
Clive AsuaiClive Asuai
Clive Asuai
UOUfuoma John Osame
Ufuoma John Osame
Fraud detection is used in various industries, including banking institutes, finance, insurance, government agencies, etc. Recent increases in the number of fraud attempts make fraud detection crucial for safeguarding financial information that is confidential or personal. Many types of fraud problems exist, including card-not-present fraud, fake Marchant, counterfeit checks, stolen credit cards, and others. An ensemble feature selection technique based on Recursive feature elimination (RFE), Information gain (IG), and Chi-Squared (X2) in concurrence with the Random Forest algorithm, was proposed to give research findings and results on fraud detection and prevention. The objective was to choose the essential features for training the model. The Receiver Operating Characteristic (ROC) Score, Accuracy, F1 Score, and Precision are used to evaluate the model's performance. The findings show that the model can differentiate between fraudulent transactions and those that are not, with an ROC Score of 95.83% and an Accuracy of 99.6%. The F1 Score of 99.6%% and precision of 100% further sustain the model's ability to detect fraudulent transactions with the least false positives correctly. The ensemble feature selection technique reduced training time and did not compromise the model's performance, making it a valuable tool for businesses in preventing fraudulent transactions.

1 January 2024 in International Journal of Advanced Computer Science and Applications
Verified
Journal Article
Open access
PDF file
SOSebastina Nkechi Okofu
Sebastina Nkechi Okofu
KAKizito Eluemunor Anazia
Kizito Eluemunor Anazia
MAMaureen Ifeanyi Akazue
Maureen Ifeanyi Akazue
MOMargaret Dumebi Okpor
Margaret Dumebi Okpor
AOAmanda Enadona Oweimieto
Amanda Enadona Oweimieto
EOEmmanuel Obiajulu Ojei
Emmanuel Obiajulu Ojei
User behaviour about an item is a choice predicated on their perception of the item in order to satisfy the intent of such a purchase pattern/choice as made. With virtual stores to improve consumer coverage, monetization and ease of product delivery, users' trust is lowered with the non-delivery of advertised products as items purchased are often replaced with new/similar products. To resolve the issues of lowered consumer trust and preference for products purchased via online shops – each transaction reflects a user buying behaviour. This, if harnessed – will aid businesses to reshape their inventory to handle various challenges arising from feature evolution, feature drift, product replacement, and concept evolution. Our study seeks to resolve these issues via a Bayesian network with trust, preference and intent as features of the virtual store to investigate their effectiveness in the design and usefulness to promote e-commerce in Nigeria. Data consists of 8,693 records collected via Google Play Scraper Library for Jumia as retrieved from over 586 respondents. Expert evaluation ranked the design choice in the use of the parameters as high.

23 July 2025 in American Journal of Mathematical and Computer Modelling
Verified
Journal Article
Open access
PDF file
Clive AsuaiClive Asuai
Clive Asuai
AAAyigbe Arinomor
Ayigbe Arinomor
CACollins Atumah
Collins Atumah
IKImarah Kowhoro
Imarah Kowhoro
DODaniel Ogheneochuko
Daniel Ogheneochuko
The rapid advancement of AI-generated synthetic speech poses significant threats, including identity fraud and misinformation, as deepfake audio becomes increasingly indistinguishable from genuine recordings. While existing detection methods have achieved high accuracy on specific datasets, they often struggle with generalization across diverse audio samples and real-world conditions. To address this limitation, this paper proposes a hybrid Deep CNN-LSTM model that leverages both Mel Frequency Cepstral Coefficients (MFCCs) and spectrogram analysis to capture complementary spatial and temporal artifacts indicative of synthetic speech. The model was evaluated on the Fake-or-Real (FoR) dataset, achieving a classification accuracy of 94.7%, surpassing standalone CNN (87.3%) and LSTM (82.7%) models. Crucially, the model demonstrated strong generalization capabilities with an AUC-ROC score of 97.3%. Further cross-dataset evaluation on ASVspoof 2019 confirmed its robustness, achieving an accuracy of 93.2%. The results indicate that the fusion of spectral and temporal features through a hybrid architecture provides a more robust solution for detecting AI-generated audio, contributing to the development of reliable deepfake detection systems for cybersecurity and digital forensics applications.

4 June 2025 in Journal of Future Artificial Intelligence and Technologies
Verified
Journal Article
Open access
Clive AsuaiClive Asuai
Clive Asuai
AMAndrew Mayor
Andrew Mayor
PEPeace Oguoguo Ezzeh
Peace Oguoguo Ezzeh
HHHoussem HOSNI
Houssem HOSNI
AJAghoghovia Agajere Joseph-Brown
Aghoghovia Agajere Joseph-Brown
IDIrene Debekeme
Irene Debekeme
Distributed Denial of Service (DDoS) attacks remain a critical threat to network infrastructure, necessitating timely and precise detection techniques to mitigate their impact. This study presents a hybrid deep learning framework that integrates the Three Conditions for Feature Aggregation (3ConFA) framework with a one-dimensional Convolutional Neural Network (1D-CNN) for effective DDoS detection. Initially, salient features were selected using the 3ConFA approach, which combines multi-filter feature ranking—based on Chi-square, Information Gain, and Decision Tree Recursive Feature Elimination (DT-RFE) to extract robust and relevant features from raw network traffic data. The training samples were balanced using the Adaptive Synthetic Sampling Approach (ADASYN) to address the high-class imbalance typically present in DDoS datasets. The refined features were then passed through a 1D-CNN model designed to learn temporal and spatial attack behavior patterns. Feature fusion was applied by concatenating the aggregated statistical features and the deep features learned by the CNN, then re-selected the most informative features using Recursive Feature Elimination with Cross-Validation (RFECV). The final classification was performed using a Softmax output layer, and the model was evaluated using 5-fold cross-validation and a separate test set. Experimental results demonstrated an average training accuracy of 99.42%, an F1-score of 99.35%, and an AUC-ROC of 99.87%. On the test set, the model achieved a detection accuracy of 99.56%, a precision of 99.61%, and an F1-score of 99.50%, with an AUC-ROC of 0.9982. The proposed hybrid approach outperforms traditional models such as Random Forest, Decision Tree, XGBoost, and standalone CNN, validating the synergistic impact of integrating 3ConFA with deep temporal convolutional modeling for accurate and interpretable DDoS attack detection.

30 March 2024 in Advances in Multidisciplinary & Scientific Research Journal Publications
Verified
Journal Article
Closed access
AOAmanda Enaodona Oweimieotu
Amanda Enaodona Oweimieotu
AEAbel Efetobore Edje
Abel Efetobore Edje
MAMaureen Ifeanyi Akazue
Maureen Ifeanyi Akazue
Clive AsuaiClive Asuai
Clive Asuai
This research delves into the innovative application of feed-forward neural networks (FNNs) specifically the multi-layer perceptron (MLP). MLP is a flexible algorithm due to its ability to adapt to different real-world problems amongst other features, and this makes it a preferred machine learning algorithm in the early detection of mental health disorders. MLP’s number of layers and the number of neurons per layer changes to accommodate these abilities. MLP was chosen for this work because they can model non-linear relationships found in a dataset as well as the fact that the algorithm is efficient in accuracy detection which is lacking in other types of FNN. This study focused on developing and optimizing MLP architectures to achieve heightened accuracy in identifying mental health disorders. Original dataset that was used comprise 334 rows (datapoints) and 31 columns (features) and Only 27 features were quantifiable. We utilized the first 13 features in the dataset for this research work as too many features will affect the training time. The model performance was evaluated using Accuracy, Precision, F1 Score, Recall, and the results of the model evaluation showed that early detection of mental health disorders is predictable using this type of Feed-forward Neural network, with an Accuracy of 96%, Recall of 80%, F1 Score of 77%, Sensitivity 90%, and Specificity of 88% when compared to previous research with lower accuracy of 81.75% amongst the other result they got for the other parameters. Furthermore, by using widely collected datasets and employing advanced machine learning techniques such as feature importance technique for optimization of the initial result got, this approach contributes significantly to the field of early detection of mental health disorders. Keywords: Mental Health Disorder, Feedforward Neural Network, Machine Learning, Early Detection, Artificial Intelligence

15 May 2025 in International Journal of Latest Technology in Engineering Management & Applied Science
Verified
Journal Article
Open access
Clive AsuaiClive Asuai
Clive Asuai
CACollins Tobore Atumah
Collins Tobore Atumah
AJAghoghovia Agajere Joseph-Brown
Aghoghovia Agajere Joseph-Brown
Predictive Maintenance (PdM) plays a pivotal role in Industry 4.0 and 5.0 by minimizing equipment downtime and optimizing performance. However, limitations such as scarce fault data, data quality issues, and model interpretability hinder its effectiveness. This study presents a machine learning-based PdM framework tailored for Vortex Oil and Gas Nigeria Ltd., leveraging synthetic sensor data and eXtreme Boost (XGBoost) regression to predict Remaining Useful Life (RUL) of industrial equipment. Using simulated data from 50 machines over 300 operational cycles, the model achieved strong performance metrics, with an RMSE of 40.73 and MAE of 32.38. A four-layer system architecture—comprising data acquisition, edge processing, cloud analytics, and user interface—enabled real-time monitoring and decision-making. The results underscore the system’s capacity to detect early failure trends and support proactive maintenance, aligning with the goals of intelligent, sustainable, and human-centric industrial operations. This research contributes a scalable, data-driven PdM solution suitable for environments with limited real-world fault data.