Abou Bakary Ballo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Abou Bakary Ballo has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Kacoutchy Jean Ayikpa
Unité de Recherche et d’Expertise Numérique (UREN), Université Virtuelle de Côte d’Ivoire, Abidjan 28 BP 536, Côte d’Ivoire
3 shared publications - Brou Médard Kouassi
Laboratoire de Mécanique et Informatique, Université Félix Houphouët-Boigny, Abidjan 22 BP 801, Côte d’Ivoire
2 shared publications - Diarra Mamadou
Laboratoire de Mécanique et Informatique, Université Félix Houphouët-Boigny, Abidjan 22 BP 801, Côte d’Ivoire
2 shared publications - Diarra Mamadou
Laboratoire Imagerie et Vision Artificielle (ImViA), Université Bourgogne Europe, 21000 Dijon, France
1 shared publication - Minfonga Zié Jérôme Coulibaly
Laboratoire de Mécanique et Informatique, Université Félix Houphouët-Boigny, Abidjan 22 BP 801, Côte d’Ivoire
1 shared publication - Pierre Gouton
Laboratoire Imagerie et Vision Artificielle (ImViA), Université Bourgogne Europe, 21000 Dijon, France
1 shared publication
Latest Publications
- Open Access
- 19 November 2025
- Published by MDPI AG in Future Internet
- Vol. 17 (11) , 529
Top-K Feature Selection for IoT Intrusion Detection: Contributions of XGBoost, LightGBM, and Random Forest
The rapid growth of the Internet of Things (IoT) has created vast networks of interconnected devices that are increasingly exposed to cyberattacks. Ensuring the security of such distributed systems requires efficient and adaptive intrusion detection mechanisms. However, conventional methods face limitations in processing large and complex feature spaces. To address this issue, this study proposes an optimized intrusion detection approach based on Top-K feature selection combined with ensemble learning models, evaluated on the CICIoMT2024 dataset. Three algorithms, XGBoost, LightGBM, and Random Forest, were trained and tested on IoT datasets using three feature configurations: Top-10, Top-15, and the complete feature set. The results show that the Random Forest model provides the best balance between accuracy and computational efficiency, achieving 91.7% accuracy and an F1-score of 93% with the Top-10 subset while reducing processing time by 35%. These findings demonstrate that the Top-K selection strategy enhances the interpretability and performance of IDSs in IoT environments. Future work will extend this framework to real-time adaptive detection and edge computing integration for large-scale IoT deployments.
- 19 November 2025
- Published by MDPI AG in Future Internet
- Vol. 17 (11) , 529
Open AccessTop-K Feature Selection for IoT Intrusion Detection: Contributions of XGBoost, LightGBM, and Random Forest
The rapid growth of the Internet of Things (IoT) has created vast networks of interconnected devices that are increasingly exposed to cyberattacks. Ensuring the security of such distributed systems requires efficient and adaptive intrusion detection mechanisms. However, conventional methods face limitations in processing large and complex feature spaces. To address this issue, this study proposes an optimized intrusion detection approach based on Top-K feature selection combined with ensemble learning models, evaluated on the CICIoMT2024 dataset. Three algorithms, XGBoost, LightGBM, and Random Forest, were trained and tested on IoT datasets using three feature configurations: Top-10, Top-15, and the complete feature set. The results show that the Random Forest model provides the best balance between accuracy and computational efficiency, achieving 91.7% accuracy and an F1-score of 93% with the Top-10 subset while reducing processing time by 35%. These findings demonstrate that the Top-K selection strategy enhances the interpretability and performance of IDSs in IoT environments. Future work will extend this framework to real-time adaptive detection and edge computing integration for large-scale IoT deployments.
- Open Access
- 30 September 2025
- Published by MDPI AG in Technologies
- Vol. 13 (10) , 441
Optimized Intrusion Detection in the IoT Through Statistical Selection and Classification with CatBoost and SNN
With the rapid expansion of the Internet of Things (IoT), interconnected systems are becoming increasingly vulnerable to cyberattacks, making intrusion detection essential but difficult. The marked imbalance between regular traffic and attacks, as well as the redundancy of variables from multiple sensors and protocols, greatly complicates this task. The study aims to improve the robustness of IoT intrusion detection systems by reducing the risks of overfitting and false negatives through appropriate rebalancing and variable selection strategies. We combine two data rebalancing techniques, Synthetic Minority Over-sampling Technique (SMOTE) and Random Undersampling (RUS), with two feature selection methods, LASSO and Mutual Information, and then evaluate their performance on two classification models: CatBoost and a Simple Neural Network (SNN). The experiments show the superiority of CatBoost, which achieves an accuracy of 82% compared to 80% for SNN, and confirm the effectiveness of SMOTE over RUS, particularly for SNN. The CatBoost + SMOTE + LASSO configuration stands out with a recall of 82.43% and an F1-score of 85.08%, offering the best compromise between detection and reliability. These results demonstrate that combining rebalancing and variable selection techniques significantly enhances the performance and reliability of intrusion detection systems in the IoT, thereby strengthening cybersecurity in connected environments.
- 30 September 2025
- Published by MDPI AG in Technologies
- Vol. 13 (10) , 441
Open AccessOptimized Intrusion Detection in the IoT Through Statistical Selection and Classification with CatBoost and SNN
With the rapid expansion of the Internet of Things (IoT), interconnected systems are becoming increasingly vulnerable to cyberattacks, making intrusion detection essential but difficult. The marked imbalance between regular traffic and attacks, as well as the redundancy of variables from multiple sensors and protocols, greatly complicates this task. The study aims to improve the robustness of IoT intrusion detection systems by reducing the risks of overfitting and false negatives through appropriate rebalancing and variable selection strategies. We combine two data rebalancing techniques, Synthetic Minority Over-sampling Technique (SMOTE) and Random Undersampling (RUS), with two feature selection methods, LASSO and Mutual Information, and then evaluate their performance on two classification models: CatBoost and a Simple Neural Network (SNN). The experiments show the superiority of CatBoost, which achieves an accuracy of 82% compared to 80% for SNN, and confirm the effectiveness of SMOTE over RUS, particularly for SNN. The CatBoost + SMOTE + LASSO configuration stands out with a recall of 82.43% and an F1-score of 85.08%, offering the best compromise between detection and reliability. These results demonstrate that combining rebalancing and variable selection techniques significantly enhances the performance and reliability of intrusion detection systems in the IoT, thereby strengthening cybersecurity in connected environments.
- Open Access
- 12 May 2025
- Published by MDPI AG in Electronics
- Vol. 14 (10) , 1976
Intelligent Selection of Spectral Bands from High-Precision Spectroradiometer Measurements for Optimizing Cocoa Bean Classification
Evaluating the spectral properties of cocoa beans based on their fermentation state (fermented, in a poor state, unfermented) is essential for ensuring their quality in the cocoa industry. This study examined the spectral response of beans in the range of 380 nm to 780 nm using the Konica-Minolta CS-2000 spectrophotometer comes from Dijon, France, a device designed to measure the spectrum of objects and sources in the visible range. Different spectral band selection methods have been applied to identify the most discriminating wavelengths for their classification. Several techniques were used: ANOVA, F-score, Lasso, Linear Discriminant Analysis (LDA), Mutual Information, and Partial Least Squares (PLS). A band selector voting process was implemented to determine standard wavelengths identified using the different methods. The selected spectral bands were then leveraged to train classification models, including Random Forest, SVM, and XGBoost. The results show that a restricted subset of wavelengths allows for effective class separation, thereby improving model performance. Among the approaches tested, ANOVA and F-score combined with Random Forest achieved an accuracy of 92.59%, while F-score and Mutual Information coupled with SVM and voting associated with SVM obtained an accuracy of 96.30%. These feature selection methods have effectively reduced dimensionality while maintaining high classification accuracy. These results open up promising prospects for the automation of quality control of cocoa beans, thus contributing to the optimization of industrial processes.
- 12 May 2025
- Published by MDPI AG in Electronics
- Vol. 14 (10) , 1976
Open AccessIntelligent Selection of Spectral Bands from High-Precision Spectroradiometer Measurements for Optimizing Cocoa Bean Classification
Evaluating the spectral properties of cocoa beans based on their fermentation state (fermented, in a poor state, unfermented) is essential for ensuring their quality in the cocoa industry. This study examined the spectral response of beans in the range of 380 nm to 780 nm using the Konica-Minolta CS-2000 spectrophotometer comes from Dijon, France, a device designed to measure the spectrum of objects and sources in the visible range. Different spectral band selection methods have been applied to identify the most discriminating wavelengths for their classification. Several techniques were used: ANOVA, F-score, Lasso, Linear Discriminant Analysis (LDA), Mutual Information, and Partial Least Squares (PLS). A band selector voting process was implemented to determine standard wavelengths identified using the different methods. The selected spectral bands were then leveraged to train classification models, including Random Forest, SVM, and XGBoost. The results show that a restricted subset of wavelengths allows for effective class separation, thereby improving model performance. Among the approaches tested, ANOVA and F-score combined with Random Forest achieved an accuracy of 92.59%, while F-score and Mutual Information coupled with SVM and voting associated with SVM obtained an accuracy of 96.30%. These feature selection methods have effectively reduced dimensionality while maintaining high classification accuracy. These results open up promising prospects for the automation of quality control of cocoa beans, thus contributing to the optimization of industrial processes.