← Back to VOLUME 13, ISSUE 10, OCTOBER 2026
This work is licensed under a Creative Commons Attribution 4.0 International License.
A Multilayer Ensemble Learning Framework for Cloud Job Failure Prediction
Vibha Vishwakarma, Ms. Dolly Verma
👁 7 views📥 4 downloads
Abstract: To achieve high performance, efficient resource utilization, and better fault tolerance, it is important for cloud data centers to predict job failures accurately; it is essential for the cloud data center to correctly predict job failures. This paper proposes a novel multilayer ensemble-based approach for cloud job failure prediction using the Google Cluster 2019 dataset. The framework uses several classification algorithms, including Decision Tree (DT), K- Nearest Neighbors (KNN), Artificial Neural Network (ANN), Extreme Gradient Boosting (XGBoost), and AdaBoost. These models are combined using a hard Voting Classifier to increase the prediction stability and robustness of the prediction. Furthermore, a Stacking Classifier is constructed with the base learners: Random Forest, KNN, and MLP, and the meta estimator: LR to improve predictive accuracy further. The experimental results show strong predictive performance. The Voting Classifier achieved 97.83% accuracy, while the Stacking Classifier achieved 97.98% accuracy in predicting cloud job outcomes. The use of explainable AI methods, such as LIME and SHAP, ensures transparency and reliability, and helps to explain the prediction and promote the value of the features. The trained models are integrated into a Flask-based web application that supports user sign-up and sign-in through SQLite, along with real-time processing and presentation of user inputs. The system provides an understandable judgement of “job will successfully complete” or “job failure predicted” for reliable, interpretable and userfriendly cloud job failure prediction.
Keywords: Cloud computing, job failure prediction, ensemble learning, stacking classifier, voting classifier, XGBoost, explainable AI, fault detection.
Keywords: Cloud computing, job failure prediction, ensemble learning, stacking classifier, voting classifier, XGBoost, explainable AI, fault detection.
How to Cite:
[1] Vibha Vishwakarma, Ms. Dolly Verma, “A Multilayer Ensemble Learning Framework for Cloud Job Failure Prediction,” International Advanced Research Journal in Science, Engineering and Technology (IARJSET), DOI: 10.17148/IARJSET.2026.131044
