Claim Missing Document
Check
Articles

Found 8 Documents
Search

Hypertension Detection via Tree-Based Stack Ensemble with SMOTE-Tomek Data Balance and XGBoost Meta-Learner Odiakaose, Christopher Chukwufunaya; Aghware, Fidelis Obukohwo; Okpor, Margaret Dumebi; Eboka, Andrew Okonji; Binitie, Amaka Patience; Ojugo, Arnold Adimabua; Setiadi, De Rosal Ignatius Moses; Ibor, Ayei Egu; Ako, Rita Erhovwo; Geteloma, Victor Ochuko; Ugbotu, Eferhire Valentine; Aghaunor, Tabitha Chukwudi
Journal of Future Artificial Intelligence and Technologies Vol. 1 No. 3 (2024): December 2024
Publisher : Future Techno Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/faith.3048-3719-43

Abstract

High blood pressure (or hypertension) is a causative disorder to a plethora of other ailments – as it succinctly masks other ailments, making them difficult to diagnose and manage with a targeted treatment plan effectively. While some patients living with elevated high blood pressure can effectively manage their condition via adjusted lifestyle and monitoring with follow-up treatments, Others in self-denial leads to unreported instances, mishandled cases, and in now rampant cases – result in death. Even with the usage of machine learning schemes in medicine, two (2) significant issues abound, namely: (a) utilization of dataset in the construction of the model, which often yields non-perfect scores, and (b) the exploration of complex deep learning models have yielded improved accuracy, which often requires large dataset. To curb these issues, our study explores the tree-based stacking ensemble with Decision tree, Adaptive Boosting, and Random Forest (base learners) while we explore the XGBoost as a meta-learner. With the Kaggle dataset as retrieved, our stacking ensemble yields a prediction accuracy of 1.00 and an F1-score of 1.00 that effectively correctly classified all instances of the test dataset.
Stacked Learning Anomaly Detection Scheme with Data Augmentation for Spatiotemporal Traffic Flow Binitie, Amaka Patience; Odiakaose , Christopher Chukwufunaya; Okpor, Margaret Dumebi; Ejeh, Patrick Ogholuwarami; Eboka, Andrew Okonji; Ojugo, Arnold Adimabua; Setiadi, De Rosal Ignatius Moses; Ako, Rita Erhovwo; Aghaunor, Tabitha Chukwudi; Geteloma, Victor Ochuko; Afotanwo, Anderson
Journal of Fuzzy Systems and Control Vol. 2 No. 3 (2024): Vol. 2, No. 3, 2024
Publisher : Peneliti Teknologi Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59247/jfsc.v2i3.267

Abstract

The digital revolution births transformation in many facets of today’s society. Its adoption in transportation to curb traffic congestion in major cities globally advances smart-city initiatives. Challenges of population growth, lack of datasets, and aging infrastructure have necessitated the need for traffic analytics. Studies have estimated an associated global annual loss of $583 billion to traffic congestion for 2023. This, caused fuel wastage, loss of time, and increased costs across congested areas. With the cost of building more road networks, cities must advance new ways to improve traffic flow via anomaly detection as an early warning in the flow pattern. Our study posits stacked learning with extreme gradient boost as a meta-learner to help address imbalanced datasets, yield faster model construction, and ensure improved performance via enhanced anomalous data detection.
Investigating an Anomaly-based Intrusion Detection via Tree-based Adaptive Boosting Ensemble Onoma, Paul Avweresuo; Agboi, Joy; Geteloma, Victor Ochuko; Max-Egba, Asuobite ThankGod; Eboka, Andrew Okonji; Ojugo, Arnold Adimabua; Odiakaoase, Christopher Chukwufunaya; Ugbotu, Eferhire Valentine; Aghaunor, Tabitha Chukwudi; Binitie, Amaka Patience
Journal of Fuzzy Systems and Control Vol. 3 No. 1 (2025): Vol. 3, No. 1, 2025
Publisher : Peneliti Teknologi Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59247/jfsc.v3i1.279

Abstract

The eased accessibility, mobility, and portability of smartphones have caused the consequent rise in the proliferation of users' vulnerability to a variety of phishing attacks. Some users are more vulnerable due to factors like personality behavioral traits, media presence, and other factors. Our study seeks to reveal cues utilized by successful attacks by identifying web content as genuine and malicious data. We explore a sentiment-based extreme gradient boost learner with data collected over social platforms, scraped using the Python Google Scrapper. Our results show AdaBoost yields a prediction accuracy of 0.9989 to correctly classify 2148 cases with incorrectly classified 25 cases. The result shows the tree-based AdaBoost ensemble can effectively identify phishing cues and efficiently classify phishing lures against unsuspecting users from access to malicious content.
Voice-based Dynamic Time Warping Recognition Scheme for Enhanced Database Access Security Onoma, Paul Avweresuo; Ugbotu, Eferhire Valentine; Aghaunor, Tabitha Chukwudi; Agboi, Joy; Ojugo, Arnold Adimabua; Odiakaose, Christopher Chukwufunaya; Max-Egba, Asuobite ThankGod; Niemogha, Star Umiyemeromesu; Binitie, Amaka Patience; Abdullahi, Mustapha Barau
Journal of Fuzzy Systems and Control Vol. 3 No. 1 (2025): Vol. 3, No. 1, 2025
Publisher : Peneliti Teknologi Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59247/jfsc.v3i1.293

Abstract

Rapid transformation with database security has remained imperative as unauthorized access exposes sensitive data to adversaries. To curb this, we suggest using a secured dynamic time-warp scheme to improve access to the database schemas. The study integrates voice biometrics with two-factor authentication to yield a robust, user-friendly platform, which utilizes time-warping to authenticate voice patterns against the variability in utterance speed. Results showcase high accuracy and resiliency in its usage against spoofing attacks as compared to state-of-the-art voice recognition systems. The model ensures the minimal possibility of credential theft by binding the access of databases to the voice features of authorized users. The study shows the system's architecture, implementation, and performance evaluation, highlighting its potential to revolutionize database security in various applications. The findings underscore the importance of leveraging advanced biometric techniques to safeguard critical information systems.
Integrating Hybrid Statistical and Unsupervised LSTM-Guided Feature Extraction for Breast Cancer Detection Setiadi, De Rosal Ignatius Moses; Ojugo, Arnold Adimabua; Pribadi, Octara; Kartikadarma , Etika; Setyoko, Bimo Haryo; Widiono, Suyud; Robet, Robet; Aghaunor, Tabitha Chukwudi; Ugbotu, Eferhire Valentine
Journal of Computing Theories and Applications Vol. 2 No. 4 (2025): JCTA 2(4) 2025
Publisher : Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/jcta.12698

Abstract

Breast cancer is the most prevalent cancer among women worldwide, requiring early and accurate diagnosis to reduce mortality. This study proposes a hybrid classification pipeline that integrates Hybrid Statistical Feature Selection (HSFS) with unsupervised LSTM-guided feature extraction for breast cancer detection using the Wisconsin Diagnostic Breast Cancer (WDBC) dataset. Initially, 20 features were selected using HSFS based on Mutual Information, Chi-square, and Pearson Correlation. To address class imbalance, the training set was balanced using the Synthetic Minority Over-sampling Technique (SMOTE). Subsequently, an LSTM encoder extracted non-linear latent features from the selected features. A fusion strategy was applied by concatenating the statistical and latent features, followed by re-selection of the top 30 features. The final classification was performed using a Support Vector Machine (SVM) with RBF kernel and evaluated using 5-fold cross-validation and a held-out test set. Experimental results showed that the proposed method achieved an average training accuracy of 98.13%, F1-score of 98.13%, and AUC-ROC of 99.55%. On the held-out test set, the model reached an accuracy of 99.30%, precision of 100%, and F1-score of 99.05%, with an AUC-ROC of 0.9973. The proposed pipeline demonstrates improved generalization and interpretability compared to existing methods such as LightGBM-PSO, DHH-GRU, and ensemble deep networks. These results highlight the effectiveness of combining statistical selection and LSTM-based latent feature encoding in a balanced classification framework.
Investigating a SMOTE-Tomek Boosted Stacked Learning Scheme for Phishing Website Detection: A Pilot Study Ugbotu, Eferhire Valentine; Emordi, Frances Uchechukwu; Ugboh, Emeke; Anazia, Kizito Eluemunor; Odiakaose, Christopher Chukwufunaya; Onoma, Paul Avwerosuoghene; Idama, Rebecca Okeoghene; Ojugo, Arnold Adimabua; Geteloma, Victor Ochuko; Oweimieotu, Amanda Enaodona; Aghaunor, Tabitha Chukwudi; Binitie, Amaka Patience; Odoh, Anne; Onochie, Chris Chukwudi; Ezzeh, Peace Oguguo; Eboka, Andrew Okonji; Agboi, Joy; Ejeh, Patrick Ogholuwarami
Journal of Computing Theories and Applications Vol. 3 No. 2 (2025): JCTA 3(2) 2025
Publisher : Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/jcta.14472

Abstract

The daily exchange of informatics over the Internet has both eased the widespread proliferation of resources to ease accessibility, availability and interoperability of accompanying devices. In addition, the recent widespread proliferation of smartphones alongside other computing devices has continued to advance features such as miniaturization, portability, data access ease, mobility, and other merits. It has also birthed adversarial attacks targeted at network infrastructures and aimed at exploiting interconnected cum shared resources. These exploits seek to compromise an unsuspecting user device cum unit. Increased susceptibility and success rate of these attacks have been traced to user's personality traits and behaviours, which renders them repeatedly vulnerable to such exploits especially those rippled across spoofed websites as malicious contents. Our study posits a stacked, transfer learning approach that seeks to classify malicious contents as explored by adversaries over a spoofed, phishing websites. Our stacked approach explores 3-base classifiers namely Cultural Genetic Algorithm, Random Forest, and Korhonen Modular Neural Network – whose output is utilized as input for XGBoost meta-learner. A major challenge with learning scheme(s) is the flexibility with the selection of appropriate features for estimation, and the imbalanced nature of the explored dataset for which the target class often lags behind. Our study resolved dataset imbalance challenge using the SMOTE-Tomek mode; while, the selected predictors was resolved using the relief rank feature selection. Results shows that our hybrid yields F1 0.995, Accuracy 0.997, Recall 0.998, Precision 1.000, AUC-ROC 0.997, and Specificity 1.000 – to accurately classify all 2,764 cases of its held-out test dataset. Results affirm that it outperformed bench-mark ensembles. Result shows the proposed model explored UCI Phishing Website dataset, and effectively classified phishing (cues and lures) contents on websites.
Phishing Website Detection via a Transfer Learning based XGBoost Meta-learner with SMOTE-Tomek Agboi, Joy; Emordi, Frances Uche; Odiakaose, Christopher Chukwufunaya; Idama, Rebecca Okeoghene; Jumbo, Evans Fubara; Oweimieotu, Amanda Enaodona; Ezzeh, Peace Oguguo; Eboka, Andrew Okonji; Odoh, Anne; Ugbotu, Eferhire Valentine; Onoma, Paul Avwerosuoghene; Ojugo, Arnold Adimabua; Aghaunor, Tabitha Chukwudi; Binitie, Amaka Patience; Onochie, Christopher Chukwudi; Ejeh, Patrick Ogholuwarami; Nwozor, Blessing Uche
Journal of Fuzzy Systems and Control Vol. 3 No. 3 (2025): Vol. 3 No. 3 (2025)
Publisher : Peneliti Teknologi Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59247/jfsc.v3i3.325

Abstract

The widespread proliferation of smartphones has advanced portability, data access ease, mobility, and other merits; it has also birthed adversarial targeting of network resources that seek to compromise unsuspecting user devices. Increased susceptibility was traced to user's personality, which renders them repeatedly vulnerable to exploits. Our study posits a stacked learning model to classify malicious lures used by adversaries on phishing websites. Our hybrid fuses 3-base learners (i.e. Genetic Algorithm, Random Forest, Modular Net) with its output sent as input to the XGBoost. The imbalanced dataset was resolved via SMOTE-Tomek with predictors selected using a relief rank feature selection. Our hybrid yields F1 0.995, Accuracy 1.000, Recall 0.998, Precision 1.000, MCC 1.000, and Specificity 1.000 – to accurately classify all 3,316 cases of its held-out test dataset. Results affirm that it outperformed benchmark ensembles. The study shows that our proposed model, as explored on the UCI Phishing Website dataset, effectively classified phishing (cues and lures) contents on websites.
EcoSMEAL: Energy Consumption with Optimization Strategy via a Secured Smart Monitor-Alert Ensemble Aghaunor, Tabitha Chukwudi; Agboi, Joy; Ugbotu, Eferhire Valentine; Onoma, Paul Avweresuoghene; Ojugo, Arnold Adimabua; Odiakaose, Christopher Chukwufunaya; Eboka, Andrew Okonji; Ezzeh, Peace Oguguo; Geteloma, Victor Ochuko; Binitie, Amaka Patience; Orobor, Anderson Ise; Nwozor, Blessing Uche; Ejeh, Patrick Ogholuwarami; Onochie, Christopher Chukwudi
Journal of Fuzzy Systems and Control Vol. 3 No. 3 (2025): Vol. 3 No. 3 (2025)
Publisher : Peneliti Teknologi Teknik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59247/jfsc.v3i3.319

Abstract

The global demand for automation that seeks the efficient consumption and usage of energy via the adoption of embedded-fit management solutions that yield improved performance with reduced consumption has become the new norm. These explore sensor-based units in their own right with eco-friendly platforms that raise germane environmental, health, and consumption regulation(s) concerns that have today become a global issue, even when they proffer improved life standards that replace traditional solutions. Our study posits an embedded sensor design to observe environmental conditions associated with energy consumption by residential or home appliances. It utilizes a machine learning scheme and algorithm to analyze the total energy consumed by each appliance and delivers optimal consumption that reduces energy waste. The system was tested across multiple parameters and found to yield desired effectiveness, reliability, and efficiency. Our utilization of the ESP8266 and ThingSpeak is able to handle extensive inputs without significant delays or data losses. Results affirms the system ability to maintain stable performance even with more devices connected to the unit.