Donatien Koulla Moulla

Dr. Donatien Koulla Moulla

Computer Science, University of South Africa, Pretoria, Gauteng, South Africa
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
22 May 2024 in Sustainability
Verified
Journal Article
Open access
PDF file
DMDonatien Koulla Moulla
Donatien Koulla Moulla
ASAttipoe David Sena
Attipoe David Sena
Ernest MnkandlaErnest Mnkandla
Ernest Mnkandla
Alain AbranAlain Abran
Alain Abran
The recurrent load shedding crisis in South Africa has highlighted the need to accurately predict electricity consumption for residential buildings. This has significant ramifications for daily life and economic productivity. To address this challenge, this study leverages machine learning models to predict the hourly energy consumption of residential buildings in South Africa. This study evaluates the performance of various regression techniques, including Random Forest (RF), Decision Tree (DT), Extreme Gradient Boosting (XGBoost), and Adaptive Boosting (AdaBoost) machine learning models, using a national residential dataset that contains measurements collected every hour. The objective is to determine the most effective models for predicting next-hour residential building consumption. These models use historical patterns of energy usage to capture temporal details such as seasonal variations and rolling averages. Feature engineering methods are further employed to enhance their predictive capabilities. The performance of each individual model was evaluated using criteria such as the mean squared error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and coefficient of determination (R2). The results show that both RF and DT achieve the best accuracy for the prediction of residential electricity consumption (because the MSE, MAE, and MAPE for RF and DT are very close to 0). These findings offer actionable insights for households, businesses, and policymakers. By enabling more accurate and granular energy consumption forecasts, this can mitigate the effects of load shedding. This study contributes to the discourse on sustainable energy management by combining advanced machine learning models with real-world energy challenges.

31 October 2024 in IoT
Verified
Journal Article
Open access
PDF file
AOAmira Housam ElDin Omar
Amira Housam ElDin Omar
Hassan SoubraHassan Soubra
Hassan Soubra
DMDonatien Koulla Moulla
Donatien Koulla Moulla
Alain AbranAlain Abran
Alain Abran
The exponential growth and widespread adoption of Internet of Things (IoT) devices have introduced many vulnerabilities. Attackers frequently exploit these flaws, necessitating advanced technological approaches to protect against emerging cyber threats. This paper introduces a novel approach utilizing hardware honeypots as an additional defensive layer against hardware vulnerabilities, particularly hardware Trojans (HTs). HTs pose significant risks to the security of modern integrated circuits (ICs), potentially causing operational failures, denial of service, or data leakage through intentional modifications. The proposed system was implemented on a Raspberry Pi and tested on an emulated HT circuit using a Field-Programmable Gate Array (FPGA). This approach leverages hardware honeypots to detect and mitigate HTs in the IoT devices. The results demonstrate that the system effectively detects and mitigates HTs without imposing additional complexity on the IoT devices. The Trojan-agnostic solution offers full customization to meet specific security needs, providing a flexible and robust layer of security. These findings provide valuable insights into enhancing the security of IoT devices against hardware-based cyber threats, thereby contributing to the overall resilience of IoT networks. This innovative approach offers a promising solution to address the growing security challenges in IoT environments.

1 December 2024 in International Journal of Electrical and Computer Engineering (IJECE)
Verified
Journal Article
Open access
PDF file
DMDonatien Koulla Moulla
Donatien Koulla Moulla
Ernest MnkandlaErnest Mnkandla
Ernest Mnkandla
MAMoussa Aboubakar
Moussa Aboubakar
AAAdo Adamou Abba Ari
Ado Adamou Abba Ari
Alain AbranAlain Abran
Alain Abran
Predictive maintenance contributes to Industry 4.0, as it enables a decrease in maintenance costs and downtime while aiming to increase production and return on investment. Despite the increasing utilization of machine learning techniques in predictive maintenance in industrial systems over the past few years, several challenges remain to be addressed in the implementation of ML, including the quality of the data collected, resource constraints, and equipment heterogeneity. This study proposes an adaptive framework for predictive maintenance in the context of Industry 4.0, specifically in internet of things (IoT) systems, using machine learning (ML) models. In particular, this study introduces PdM-FSA, a new framework based on an ensemble classifier that takes advantage of four widely adopted ML models in the predictive maintenance literature: random forest (RF), support vector machine (SVM), extreme gradient boosting (XGBoost), and k-nearest neighbors (KNN). The performance evaluation results showed that the PdM-FSA framework can perform well for predictive maintenance according to the severity of equipment malfunctions in a smart factory. The results of this study provide significant knowledge to researchers and practitioners on predictive maintenance in the context of Industry 4.0. and enables the optimization of processes and improves productivity.

31 March 2023 in Applied Computer Science
Verified
Journal Article
Open access
PDF file
DMDonatien Koulla Moulla
Donatien Koulla Moulla
Ernest MnkandlaErnest Mnkandla
Ernest Mnkandla
Alain AbranAlain Abran
Alain Abran
The Internet of Things (IoT) touches almost every aspect of modern society and has changed the way people live, work, travel and, do business. Because of its importance, it is essential to ensure that an IoT system is performing well, as desired and expected, and that this can be assessed and managed with an adequate set of IoT performance metrics. The aim of this study was to systematically inventory and classifies recent studies that have investigated IoT metrics. We conducted a literature review based on studies published between January 2010 and December 2021 using a set of five research questions (RQs) on the current knowledge bases for IoT metrics. A total of 158 IoT metrics were identified and classified into 12 categories according to the different parts and aspects of an IoT system. To cover the overall performance of an IoT system, the 12 categories were organized into an ontology. The findings results show that the category of network metrics was the most discussed in 43% of the studies and, with the highest number of metrics at 37%. This study can provide guidelines for researchers and practitioners in selecting metrics for IoT systems and valuable insights into areas for improvement and optimization.

1 January 2025 in Applied Computational Intelligence and Soft Computing
Verified
Journal Article
Open access
ASAttipoe David Sena
Attipoe David Sena
DMDonatien Koulla Moulla
Donatien Koulla Moulla
Ernest MnkandlaErnest Mnkandla
Ernest Mnkandla
Alain AbranAlain Abran
Alain Abran
This study presents ensemble machine learning (ML) models for predicting residential energy consumption in South Africa. By combining the best features of individual ML models, ensemble models reduce the drawbacks of each model and improve prediction accuracy. We present four ensemble models: ensemble by averaging (EA), ensemble by stacking each estimator (ESE), ensemble by boosting (EB), and ensemble by voting estimator (EVE). These models are built on top of Random Forest (RF) and Decision Tree (DT). These base predictor models leverage historical energy consumption patterns to capture temporal intricacies, including seasonal variations and rolling averages. In addition, we employed feature engineering methodologies to further enhance their predictive abilities. The accuracy of each ensemble model was evaluated by assessing various performance indicators, including the mean squared error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and coefficient of determination R 2 . Overall, the findings illustrate the efficiency of ensemble learning models in providing accurate predictions for residential energy consumption. This study provides valuable insights for researchers and practitioners in predicting energy consumption in residential buildings and the benefits of using ensemble learning models in the building and energy research domains.

28 January 2026 in Bioengineering
Verified
Journal Article
Open access
PDF file
VMVincent Majanga
Vincent Majanga
Ernest MnkandlaErnest Mnkandla
Ernest Mnkandla
DMDonatien Koulla Moulla
Donatien Koulla Moulla
STSree Thotempudi
Sree Thotempudi
ASAttipoe David Sena
Attipoe David Sena
Recently, deep learning methods have seen major advancements and are preferred for medical image analysis. Clinically, deep learning techniques for cancer image analysis are among the main applications for early diagnosis, detection, and treatment. Consequently, segmentation of breast histology images is a key step towards diagnosing breast cancer. However, the use of deep learning methods for image analysis is constrained by challenging features in the histology images. These challenges include poor image quality, complex microscopic tissue structures, topological intricacies, and boundary/edge inhomogeneity. Furthermore, this leads to a limited number of images required for analysis. The U-Net model was introduced and gained significant traction for its ability to produce high-accuracy results with very few input images. Many modifications of the U-Net architecture exist. Therefore, this study proposes the watershed encoder–decoder neural network (WEDN) to segment cancerous lesions in supervised breast histology images. Pre-processing of supervised breast histology images via augmentation is introduced to increase the dataset size. The augmented dataset is further enhanced and segmented into the region of interest. Data enhancement methods such as thresholding, opening, dilation, and distance transform are used to highlight foreground and background pixels while removing unwanted parts from the image. Consequently, further segmentation via the connected component analysis method is used to combine image pixel components with similar intensity values and assign them their respective labeled binary masks. The watershed filling method is then applied to these labeled binary mask components to separate and identify the edges/boundaries of the regions of interest (cancerous lesions). This resultant image information is sent to the WEDN model network for feature extraction and learning via training and testing. Residual convolutional block layers of the WEDN model are the learnable layers that extract the region of interest (ROI), which is the cancerous lesion. The method was evaluated on 3000 images–watershed masks, an augmented dataset. The model was trained on 2400 training set images and tested on 600 testing set images. This proposed method produced significant results of 98.53% validation accuracy, 96.98% validation dice coefficient, and 97.84% validation intersection over unit (IoU) metric scores.