Research Article
Multilayer Perceptron (MLP) and Convolutional Neural Network (CNN) for the Prediction of Quartz Grain Color from the Ivorian Onshore Basin
Akoua Clarisse Kra*
,
Assie Francois Kouao,
Fori Yao Paul Assale
Issue:
Volume 12, Issue 1, February 2026
Pages:
1-9
Received:
26 March 2026
Accepted:
7 May 2026
Published:
18 May 2026
Abstract: This study contributes to the digital transformation of geosciences by integrating artificial intelligence into sediment characterization, a field traditionally dominated by manual and visual techniques. Quartz grains collected from onshore drilling in the Ivorian basin were first subjected to granulometric analysis and then to morphoscopic study. The resulting photographs formed a novel database used to train two neural network models: the Multilayer Perceptron (MLP) and the Convolutional Neural Network (CNN). The main objective was to automatically predict quartz grain color, thereby reducing subjectivity and improving reproducibility in sedimentological analyses. Three categories were identified: translucent, oxidized, and transparent. These chromatic distinctions provide key insights into geological history, mineral composition, and depositional environments, allowing for more refined reconstructions of physico-chemical conditions during sediment transport and deposition. Performance evaluation confirmed the feasibility of applying AI to sediment analysis. While both models produced satisfactory results, the CNN consistently outperformed the MLP, demonstrating greater robustness and accuracy. This highlights the suitability of convolutional architectures for image-based geological tasks. By combining traditional petrographic methods with advanced computational techniques, this research demonstrates the potential of automated sediment characterization and underscores the relevance of digital approaches in modern sedimentology. It opens new perspectives for reproducibility and contributes to the ongoing digital transformation of geosciences.
Abstract: This study contributes to the digital transformation of geosciences by integrating artificial intelligence into sediment characterization, a field traditionally dominated by manual and visual techniques. Quartz grains collected from onshore drilling in the Ivorian basin were first subjected to granulometric analysis and then to morphoscopic study. ...
Show More
Research Article
Improved Stacked Ensemble Technique in Enhancing the Classification of Diabetes Mellitus Patients
Issue:
Volume 12, Issue 1, February 2026
Pages:
10-16
Received:
14 April 2026
Accepted:
30 April 2026
Published:
10 June 2026
DOI:
10.11648/j.ijdsa.20261201.12
Downloads:
Views:
Abstract: Diabetes mellitus is a global health challenge which is associated with various complications such as cardiovascular disease, vision impairment, and kidney failure. Therefore, early detection and accurate prediction of diabetes risk play a significant role in improving the management of the disease and minimising the long-term health complications. Individual machine learning methods that have been applied exhibit various limitations, such as overfitting, which negatively influence the performance due to reduced generalisation capability and high variance, making the model more sensitive to specific data features. The study aimed to solve this issue by applying a stacked ensemble learning technique in enhancing the classification performance of diabetes using the Pima Indian Diabetes Data. The study incorporated various base learners: Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), K-Nearest Neighbours (KNN), Gradient–Boosting Machine (GBM) and Logistic regression as a meta-learner. The base models were trained using a 10-fold cross-validation approach to ensure a robust model and minimise overfitting. The study showed that the stacked ensemble technique achieved an average AUC of 0.84 and a standard deviation of 0.05 across all folds, showing a stable predictive performance. To improve on interpretability SHapley Additive exPlanations (SHAP) analysed the contribution of individual features, such as Glucose and Body Mass Index (BMI), which were influential in predicting diabetes risk. Further, the SHAP analysed the contribution of base learners to meta-learner prediction and found Gradient Boosting and Random Forest exerted stronger influence on the stacked ensemble compared to others. Overall, the stacking ensemble provided a robust and reliable approach for an improved diabetes classification performance. Furthermore, the integration of explainable artificial intelligence, such as SHAP, improves model transparency and interpretability among healthcare professionals.
Abstract: Diabetes mellitus is a global health challenge which is associated with various complications such as cardiovascular disease, vision impairment, and kidney failure. Therefore, early detection and accurate prediction of diabetes risk play a significant role in improving the management of the disease and minimising the long-term health complications....
Show More
